摘要:

1012 LXXXVII1.-CONTRIBUTIONS FROM THE LABORATORIES OF THE HERIOT WATT COLLEGE, EDINBURGH. Note o n the Formation of Anthraquinone ,from Orthobewoyl- bernzoic A c i d . By W. H. PERKIN, Jun., Ph.D., F.R.S. THE constitution of anthraquinone and its derivatives has been proved by a number of elegant synthetical experiments, and, among these, the formation of rnonobromant hraquinone by the dehydrating action of sulphuric acid on orthobromorthobenzoylbenzoic acid (Pech- mann, Ber., 12, 2127) is perhaps the most important- co CGH4Br*C0.C6HI*COOH = C6H,3Br<CO> C6Ha 4- H20. Liebermann (Ber., 7, 805) also showed that orthobenxoylbenzoic acid, when heated with fuming sulphuric acid, is converted into anthrayuinonesulphonic acid; but no mention has been made of tho formation of anthraqninone itself by the action of ordinary sulphuric acid on benzoylbenzoic acid, although it is well known that this change may be brought about by anhydrous phosphoric acid.I n the course of some experiments on benzoylbenzoic acid, 1 observed that this acid, in contact with ordinary sulphuric acid at loo", is converted readily, and apparently quantitatively, into anthra- quinone, co C6H5.CO*CG'E3[4*COOH = C6H4<CO>C6H4 i- H20. The benzoylbenzoic acid employed in these experiments was pre- pared by acting on a mixture of phthalic auliydride and benzene with aluminium chloride (Pechmann, Ber., 13, 1612) ; it was repeatedly recrystallised from water., and thus obtained in long, thick needles, which, after drying at 115-120°, melted at 127", and gave the following results on analysis :- 0.1420 gram substance gave 0.0568 gram H,O mid 0.3863 gram co,.Theory. C14H,,O,. Found. C ...... 74.34 per cent. 74-19 per cent. H.. .... 4.43 ,, 4-60 ,, 0 ...... 21.23 ,, 21.21 ,) I f this pure acid is dissolved in sulphuric acid, the yellow solu-THE KITRO-DEHIVATIVES OF ORTHOTOLWIDINE. 1013 tion heated at 100" for half an hour, and the product poured into water, a white precipitate separates which consists for the most part of anthraquinone. The crude substance was collected, washed with water, digested with sodium hydrate solution to remove m y un- changed acid, dried on a piece of porous plate, recrystallised from glacial acetic acid, and finally sublimed. I n this way deep-yellow needles were obtained which melted at 275", and consisted of pure anthraquinone. This conversion of benzoylbenzoic acid into anthraquinone by means of sulphuric acid may be carried out well on the small scale as a lecture experiment. Asmall quantity of the finely divided acid is mixed in a test tube with sulphuric acid, the mixtiire heated at about 150" for 8 few seconds, and until t,he whole of the benzoylbensoic acid has dissolved ; the resulting deep-hrown solution is then poured into water. If now an excess of sodium hydrate is added, the anthraquinone remains undissolved, and may be readily identified by warming with a little zinc-dust, when the well-known, characteristic, red-coloured liquid is ,obtained, which, if decanted and shaken with air, again deposits anthraquinone. Heriot Watt CollegP, Edittburgh.

ISSN:0368-1645    

DOI:10.1039/CT8915901012

出版商:RSC    

年代:1891

数据来源: RSC

摘要:

THE KITRO-DEHIVATIVES OF ORTHOTOLWIDINE. 1013 LXXXIX-The Odbo- and Para-nitro-derivatives of Orthntoluidine. By A i i t i ~ u ~ G. GREEN and THOS. A. LAWSON, Ph.D. NOLTING and Collin (Ber., 17, 265) have shown that by the nitration of orthotoluidine in presence of a large excess of sulphuric acid at a low temperature, the entering NO, group assumes the meta-position relatively to the NH, group, producing paranitro-orthotoluidine, CH3 . This nitrotolaidine appears to be regarded by Niilting (?a and Collin as the sole product of the reaction: this is not the case, however, as although it constitutes at least 75 per cent. of the whole, we have been able to prove the presence of from 3 to 4 per cent. of metanitro-orthotoluidine, C,H,( CH,) (NH,) *NO, [ 1 : 2 : 51, and to isolate a still larger quantity (about 20 per cent.) of the orthonitro- orthotoluidine [C6H,(CH3)(NHz)~NOz 1 : 2 : 61.The latter has, up to the present, only been obtained by the partial reduction of' 4 ! A 21014 GREEN AND LAWSON: THE ORTHO- AND orthodinitrotoluene, CsH3(C13',) (NO,), [l : 2 : 61 (Cunerth, ArznaZerz, 172, 223; Bernthsen, Ber,, 15, 3018), so that its formation by the direct nitration of orthotoluidine is a matter of some interest. The orthonitro-orthotoluidine presents considerable analogies t o the paranitro-orthotohidine, since the NO, group in each stands in the meta-position relatively to the NH, group, and from this fact also their simultaneous formation in the nitration of ortho t'oluidine might have been inferred. Having at our disposal considerable quantities of these two isomers, we considered it would he of interest t o make a more complete comparison of these bases and their derivatives than has yet been done.The orthonitro-orthotobuidine was separated from the paranitro orthotoluidine by means of its greater solubility in water, and was obtained pure by crystallising its hydrochloiaide. The orthonitro- orthotohidine thus obtained crystallises from alcohol in beautif 11 I, bright-yellow, slender needles, whereas the para-compound forms large, orange prisms. I n accord with Bernthsen, we have found the melting point to be 91.5". It is soluble in about 75 parts of boiling water, the para-com- pound requiring about 100 parts for solution. Its volatility with steam is about the same as that of the para-compound, namely, 1 part with 1250 parts of water.I t boils at about 305" under ordinary atmospheric pressure, but with great decomposition ; the para-com- pound boils a few degrees higher. It differs from the para-compound in being tasteless, tjhe latter having an intensely sweet taste which is not possessed by any of the other nitrotolmidines. Its hydrochloride forms flat needles or plates which are partly basified by water. The acetyl derivative was obtained in white needles melting a t 157-158" uncorr. The benzoyl derivative formed slender, white needles, sparingly soluble in cold alcohol, and melting a t 164-165" uncorr. These melting points agree with those found by Ullmann and by Bernthsen. The identity of our nitrotoluidine with that obtained by Cunerth, Bernthsen, and others from 1 : 2 : 6-dinitrotoluene was proved by the complete correspondence of the properties of the base and of its acetyl and benzoyl derivatives, by its giving the conseczitiue tolylenemetadi- amine on reduction, and the orthonitro-orthoeresol of melting point 143" on boiling the diazo-compound with water.For complete con- firmation, the NH, group was eliminated by warming the diazo- compound with alcohol : orthonitrotoluene mas thus obtained in good yield, and was identified by its boiling point (218" uncorr. at 754 mm.), and by its giving orthonitrobenzoic acid, of sweet taste, and melting point 146", on oxidation with potassium permanganate, An ahtempt t o eliminate the NH, group by the hydrazine methodPARA-NITRO-DERIVATIVES OF ORTHOTOLUIDINE.1015 (reduction of the diazo-compound with stanno us chloride and boiling with copper sulphate) gave, instead of orthonitrotol ue lie, an almost theoretical yield of ort hochloro-orthonitrotoluene, C6H3( cH,j C1.N02 [l : 2 : 61, identical with that obtained by the action of reduced copper upon the diazocldoride (see below). In correspondence with the fact that these two nitrotoluidines are amido-derivatives of ortho- and para-nitrotoluene respectively, we have found that they behave with alcoholic soda in a manner similar to the nitrotoluenes from which they are derived; for whilst the paranitro-orthotoluidine, when boiled with alcoholic soda, gives (like paranitrotoluene) a deep magenta coloration, the orthonitro-ortho- toluidine (like orthonitrotoluene) gives no colour . Although both nitrotoluidines can be readily brominated in hot acetic acid solution, they offer considerable resistance to other substi- tuting agents : thus, the orthonitro-orthotoluidine is only sulphonated by sulphuric anhydride at the high temperature of 140--150", whilst the paranitro-orthotoluidine, under the same conditions, is not sulph- onated, but undergoes complete decomposition. More over, by nitra- tion in sulphuric acid solution, we were unable to introduce a second nitro-group into either isomer.Alkaline Reduction. By reducing paranitro-orthotoludine with sodium amalgam, him- pricht and Graeff (Bey., 18, 1404; Annalen, 229, 340) obtained the corresponding azoxytoluidine, C,H,( CH,) (NH,)*N,O *c 6H3(NH~) *CH,, and azotoluidine, CsH3( CH,) (NH,).N,*C,HI,(NH2)*C~I,. In place oE sodium amalgam, most other alkaline reducing agents ean be used, but we have found sodium stannite to give the best resalts in the preparation of the azoxy- and azo-derivatives, both from the paranitro- and also from the orthonitro-orthotoluidine.For this purpose, the nitrotoluidine was dissolved in boiling water, and to the boiling solution there was slowly added a cold, aqueoub solution of sodium stannite, prepared by mixing solutions of rather more than tlhe calculated quantity of stannous chloride and an equal weight of caustic soda. The azoxytoluidine which separates is purified by crys tallisation from toluene. The azotoluidine was obtained by further reduction of the azoxytoluidine, by means of sodium stannite in alcoholic solution : the azoxytoluidine was dissolved in alcohol, and to the boiling solution an aqueous alcoholic solution of the calculated quantity of sodium stannite was added; after boiling for some time, the product was precipitated by the addition of water, and purified by crystallisation from toluene.By the same method of reduction, we have also prepared the azoxytoluidine and azotoluidine come-I016 GREEN AND LAWSON: THE ORTHO- AND aponding to orthonitro-orthotoluidine ; these have not before beerr obtained. The parahydrazo-orthotoluidine described by Limpricht and Graeff we have been unable t o obtain, neither were we able to isolate the orthohydrazo-orthotoluidine, on account of the great rapidity with which it is osidised to azotoluidine by the air.Puruzoxyortho to luidine, [l : 2 : 41 C,H,(C~,)(NH,).N,O.C,H,(NH,).CH, [4 : 2 : 11, forms yellow needles (from alcohol) o r orange prisms (from toluene). The melting point was found to be 168" uncorr., as given by Lim- pricht and Gmeff. The diacetyl derivative, obtained by adding acetic anhydride t o a boiling acetic acid solution of the base, forms small, white needles, moderately soluble i n hot acetic acid, bnt only very sparingly in alcohol. It melts at 290" uncorr. Parazo-~rtlzotoluidine, [l : 2 : 41 C6H,(CH,)(NH,)*N,oCsH,(NHZ)*CHS [4 : 2 : 11: forms dark-orange, prismatic needles (from toluene). The melting point was found to be 203" uncorr., whilst Linlpricht and Graeff give 197". The diacetyl derivative, prepared as above, foi*ms small, orange-yellow needles, moderately soluble in boiling acetic acid.It melts at 300" uncorr. Orthazoxy orthotol~ci~ine, [l : 2 : 61 C,H,(CH,)(NH,)'NzO*C,H,(NH,)*CH, [ 6 : 2 : 13, crystallises from toluene in straw-yellow plates, easily soluble in toluene, acetic acid, and alcohol, slightly soluble in water. It melts :It 149" uncorr. Its hFdrochloridc forms white plates, sparingly soluble in dilnte hydrochloric acid, easily iu water. The diacetyl tlerivative, obtained by adding acetic anhydride to the boiling alco- holic solution of the base, forms small, white, flattened needles, nearly insoluble in alcohol, acetic acid, toluene, &c. It melts at 307". Orthaao-orthotoluidine, [l : 2 : 61 c,H,(CH,)(NH,)*N,*CaHB(NH,)CH, [6 : 2 : 11, crystallises from toluene in orange plates, easily soluble in toluene, alcohol, &c.Its hydrochloride forms an orange, crystalline powder, sparingly soluble in water, especially in presence of hydrochloric acid. The diace tyl derivative crystallises from acetic acid in very small, orange-yellow needles. It is very sparingly soluble in boiling acetic acid, insoluble in alcohol, and melts ;tbove 340". Acid Bedzcction. It melts at 175" uncorr. Whilst the paranitro-orthotoluidine is converted by complete reduc- tion (with tin and hydrochloric acid, for example) into the ordinaryPARA-NlTRO-DERIVATIVES O F ORTHOTOLUIDINE. 101 7 tolylenediamine, C6H,( CH3)( NH,), [ 1 : 2 : 41, the orthonitro-ortho- tolnidine gives rise to the consecutive tolylenediamine, C6H3(CH,)(NH2), [l : 2 : 61 (Ullmann, Bey., 17, 1959).Wc have also prepared the latter diamine by reduction of the orthonitro-orthotoluidine formed by nitration of orthotoluidine ; the base crystallises in prismat'ic needles, apparently of the same form as the ordinary tolyleiiediamine, since the latter causes it to crystallise. We found the melting point to be 105", whilst Ullmann gives 103.5". The hydrochloride forms large, flat plates, moderately soluble i n water. Its diacetyl derivative melts at 202-203" uncorr., and has the very peculiar and distinctive property of subliming, when heated, in white, excessively light, woolly flocks, immediately over the melted substance, and these, becoming disengaged, float abont in the .air ; these flocks consist of masses of very fine needles. The diacetyl derivatives of other tolylenediamines which we have examined do not show this property in the slightest degree.Displacement of the NH, Group by Chlol-ine. The displacement of the NH, group in the nitrotoluidines by chlorine takes place very readily by employing Gatterman's modification of Sandmeyer's reaction. On adding precipitated copper to a cold solu- tion of the diazochloride containing excess of hydrochloric acid, 8 vigorous evolution of nitrogen takes place, and, when the reaction is complete, the chloronitrotoluene is obtained in nearly theoretical quantity by distilling the product with steam. OrthockZoroparanitrotoluene, C6H3(CH3) Cl-NO, [ 1 : 2 : 41, obtained a s above from paranitro-orthotoluidine, crystallises fyom alcohol in long, white, pointed needles, having an odour resembling that of paranitrotoluene.The same compound has been prepared by Wachendorff (AnnaZen, 185, 273) and by Lellmann ( B e y . , 17, 537) by chlorination of paranitrotoluene with antimony pentachloride, and its melting point is given as 65.5". Grthochloro.orthonitl-otoZuene, CsH,( CH,) Cl-NO, [ 1 : 2 : 61, obtained as above from orthonitro-orthotoluidine, crystallises from dilute alcohol in white needles of peculiar odour, resembling that of meta- chloronitrobenzene. It melts at 37" and is readily volatile with steam. As this melting point lies very close t o that of the chloronitrotoluene (m. p. 38") obtained by Beilstein and Kuhlberg (AnnaZen, 158, 338) from orthoparadinitrotoluene, and to which they assigned the eonsti- tution [CH, : C1 : NO, = 1 : 4 : 21, we prepared the latter for com- payison by treating the diazochloride of the orthonitroparatoluidine Its melting point was found to be 68".I E?2 NO,.CH, NHAc N,O. 290@ (G. and L.). Small, white needles. --- ----_- 307" (G. and L.). 3mal1, white, flat needles. ~ 107" 1 (Nolting l and Collin), ' 107 '5" 1 : 2 : 4 (G. and L.). Orange /prisms ; verj C& NH2 N,. 197" (Limpricht) 203" (G. and L.) Orange, prismatic needles. --- 1'75" (G. and L.) Orange plates. ~ 91 '5" (Bernthsen) '(G. and L.) . yellow needles ; I not sweet. 1 : 2 : 6 , Bright- --- 202-203° :G. and L.). Sublimes in very light flocks. :2 NO2. --- 10'7-108" (Noltinp and Collin). Fine, yellowish needles ; very sweet. _I_-- 142-143" 143-144" (Ullmann) , :G. and L.).Yellow needles ; slightly sweet. --- 149' (G. and L.). Pale-yellow plates. CH3 NH2 NIX2. 99". Prisni %tic needles. --- - ~ - _ - 103.5" (Ullmann) ! 105" (G. and L.) Prisms. I 221" (Tiemann), 224O (La- denburg). Does not sublime in Bocks. 168" (Limpricht ) (G. and L.). Yellow needles or orange prisms. CH3 CH:3 NHAc c 1 " L a NOz. -_-- 300" (G. and L.) Small, yellow needles. . ---- 65.5" dorff) , 68" (G. and L.). Long, white needles. . (Wachen- Above 340" 3T (G. and L.). (G. and L.). Small, Long, white yellow needles. needles. ITHORPE AND TUTTON : PHOSPHOROUS OXIDE. 1019 (m. p. 78") with precipitated copper in the same way as the pre- ceding. The chloronitro toluene thus prepared crystallises from dilute alcohol in white needles, is readily volatile with steam, and melts at 38", as given by Beilstein acd Kuhlberg. In appearance and proper- ties it closely resein bles the chloronitrotoluene of melting point 37', but differs slightly in odour. In order t o piaove beyond all question that they are not identical, the two substances were oxidised with potassium pernianganate under t h e same conditions. Beilstein and Kuhlberg's chloronitrotoluene of melting point 38" gave the chloronitrobenzoic acid, C,H,(COOH)Cl*NO, [l : 4 : 2J, prepared by Varnholt ( J . pi". Chem. [2], 36, 30). I t melted at 140-141" uncorr.: whilst Varnholt gives 1338-139". The new chloro- nitrotoluene of melting point 57" gave st new chloronitrobenzoic acid, C,H,(COOH)Cl*NO, [l : 2 : 61, which was easily soluble in water, and crystallised from it in whits, prismatic needles of melting point 161" uncorr. Our thanks are due to Messrs. Brooke, Simpson, and Spiller (Limited), in whose laboratory at the Atlas Works the above research was conducted. S umwzary I On the preceding page is a tabulated comparison of melting points and chief properties of the two series of compounds.

ISSN:0368-1645    

DOI:10.1039/CT8915901013

出版商:RSC    

年代:1891

数据来源: RSC

摘要:

THORPE AND TUTTON : PHOSPHOROUS OXIDE. 1019 XC.-Phospho~ous Oxide. Part 11. By T. E. THORPE, F.R.S., and A. E. TUTTON, Demonstrator of Chemistry at the Royal College of Science, South Kensington. Action of Light.-In our first communication on this subject (Trans., 1890, 553), we stated that phosphorous oxide was acted upon by light. Even diffused daylight turned it yellow, and in strong sun- shine it rapidly became dark-red. On melting the residual oxide and decanting it from the red product, and redistilling it, the clear, colour- less substance, on exposure t o sunshine, quickly became red as hef ore. Further observation, however, has rendered it doubtful whether PUTS phosphorous oxide is affected by light. We have been able, as a matter of fact, to prepare a sample of crystallised phosphorous10%) THORPE AND TUTTON : PROSPHOROUS OXlDE.oxide which, after a year’s continuous exposure to light, has not suffered the slightest change in colour. A few grams of the freshly- distilled and apparently pure oxide were sealed up in a tube pre- viously filled with dry carbon dioxide. It was then placed in the direct rays of the snn, when i t rapidly became red. After nearly three months’ exposure to light, the dark-red substance was warmed to the melting point of the oxide, passed through glass-wool to remove the suspended red powder, and the oxide again distilled into a fresh tube, in which i t solidified in large, transparent, colourless crystals. On again exposing the crystals to light, they ultimately became as deeply-red as before. The operations of melting, filtering, and dis- iilling were repeated a third time, but the result m-as the same.Finally, the oxide was simply melted at t’he lowest possible tempera- ture, filtered through glass-wool, and again exposed to sunshine. It was no longer acted upon by light, but has remained perfectly clear and colourless after nearly 12 months’ exposnre. The red substance is the so-called amorphous modification of phosphorus. Although the distilled oxide, when exposed t o light 9 eventually becomes dark-red, the actual extent of decomposition i s not very considerable. We have determined the weight of red phos- phorus formed in one case after four months, and in another after seven months’ exposure. I n the first instance it was 0.8 per cent. of the total weight of oxide ; in the other case it was 1.0 per cent.We have made many attempts to obtain preparations of the oxide which should be unaffected by light, by distillation, at the ordinary temperature, in a vacuum ; but however many times the same speci- men was redistilled in a Sprengel vacuuni, it invariably became red on exposure to light. Frequently, however, large, isolated crystals of the oxide of perfect form were obtained by slow spontaneous sublima- tion in a vacuum, which retained their clear, transparent appearance for days. If, however, the crystals were melted by the imrmth of the hand, the liquid drops, on solidification in the wax-like form, rapidly became red. Indeed, it seems not improbable that the permanency of the crystallised oxide is in some way connected with its crystalline character.I t is of course important to determine the bearing of 6hese observations on the quwtion of the phosphorescence of phosphorous oxide. We have repeated our original experiments with samples of oxide whicli have been several times exposed to light and afterwards distilled in a vacuum, but in no case did the phosphorescence show any diminution in intensity. Action. of Bmmine.-Liquid bromine acts violently upon phosphorous oxide, and the mass generally inflames. To study the nature of the change, a known quantity of the recently distilled oxide was placedTHORPE AKD TUTTOK : PHOSPHOROUS OXIDE. 1021. in a closed glass apparatus so arranged t h a t it could be continuously acted npon a t the ordinary temperature by bromine vapour.Small, lemon-yellow, and perfectly homogeneous crystals quickly made their appearance above the oxide. These continued to grow until they were about 2 mm. in diameter ; they consisted of short prisms, terminated at both ends by pyramids. It was impossible to determine their crystallographic characters with greater precision, as they deliquesced immediately when in contact with the air. Quantitative analysis showed that they consisted of pure phosphorus pentabromide. The numbers obtained were as follows :- 0.4350 gram of crystals gave 7.57 per cent. phosphorus and PBr5 contains 7.19 per cent. phosphorus and 92.81 per cent. The oxide in the flask quickly became covered with a white, amorphous powder, resembling phosphorus pentoxide. After a time both it and the crystals became dissolved in the excess of hromiiie which condensed npon them.When the reaction was apparently complete, the contents of the flask were distilled. A small quantity of bromine first appeared, after which the thermometer snd(1enly rose to 190", and a quantity of phosphoryl bromide passed over anti solidified in the receiver. It boiled constantly at 195" (tori=,), aid melted a t 45". The residue in the flask consisted of a dark, resinous mass, resembling the so-called " metaphosphoryl chloride " of Gus- tavson (compare G. N. Huntley, this vol., p. 202); it probably consisted of the bromine analogue of t h a t compound. The first action of bromine upon phosphorous oxide appears, therefore, to result i n the formation of phosphorus pentabromide and phosphoric oxide.93.32 per cent. bromine. bromine, The pentabromide crystals are afterwards w-ashecl down upon the pentoxide by the excess of bromine, and the two substances then react, forming the oxybromide and " metaphosphoryl broniide." Hence the ultimate action of bromine upon phosphorous oxide appears to be similar to that of the action of chlorine. The reaction may thus be represented :- P406 + 4Brz = 2POBr, + 2P0,Br. The yield of phosphoryl bromide actually obt'ained was very nearly that required by this equation. Action of Iodine.-Iodine reacts very slowly with phosphorous oxide, forming an orauge-red solid. Even when the substances are heated together in a sealed tube a t 150°, the reaction is far froni complete.1022 THORPE AND TUTTON : PHOSPHOROUS OXIDE.When the two substances are heated, under pressure, with a quantity of carbon bisulphide, phosphoric oxide is formed, and orange-red prisms of P,I, separate out from the concentrated solution. No for- mation of the tri-iodide could be detected. The main reaction is probably in accordance with the equation 5PaOs + 81, = 4P,I, + 6P20,. Action of Hydrogen 0hZoriJe.-This gas is rapidly absorbed by phosphorous oxide with the formation of a viscous mass and a clear, mobile liquid. The semi-solid substance is at first quite white, but as the reaction proceeds it changes t o yellow and orange. The clear liquid consists of phosphorus trichloride, boiling at 76" ; the semi- solid residue is, for the most part, soluble in water, and the solution contains phosphorous and phosphoric acids.The yellow solid was identified as free phosphorus. With a view of obtaining a quantita- tive determination of the nature of the change, a weighed quantity of phosphorous oxide, contained in a small distilling flask, was connected with a graduated gas eprouvette standing over mercury, and which could be replenished from time to time with dry hydrogen chloride as the absorption proceeded. By noting the volume of gas absorbed, and checking the amonut by occasionally weighing the flask, the amount of the reacting hydrogen chloride could be ascertained. The main action may be represented by the equation Pa06 + 6HC1 = 2PC13 + H&PO,. Geuther (J. pr. Chem. [2], 8, 359) has pointed out that phosphorus trichloride reacts with phosphorous acid t o form yellow phosphorus and orthophosphoric acid, and our own observation confirms the statement.The reaction is probably PCI3 + 4HSP03 = 3H3PO4 + P, + 3HC1. I n an actual experiment carried out by the above method, but in which the reaction was not quite complete even after a week's dura- tion, we obtained from 5 grams of phosphorous oxide a little over 4 grams of phosphorus trichloride, 1.5 grams of orthophosphoric acid, 0.4 gram of phosphorus and 2.7 grams of phosphorous acid. Action of XuZphur.-When phosphorous oxide and sulphur are heated together in an atmosphere of carbon dioxide or nitrogen, the two substances at first melt and form separate layers of liquid. At about 160°, however, a violent reaction occurs, and the mixture becomes solid.The solid substance is an addi- tion product, having, as we shall show, the formula P,O,S,. Direct experiments have proved that it is formed quantitatively in aucord- aime with the equation P,OS + 4s = PaO,S,. No gas is produced.THORPE AND TUTTON : PIIOSPHOROUS OXIDE. 1023 It mny be obtained in well-formed crystals by sublimation in i~ vacuum, or by crystallisation from carbon bisulp hide. We propose to call it phosphorus sdphoxide. Phosphorus sulphoxide may be readily prepared in the following manner :--From 3 to 5 grams of recently-distilled phosphorous oxide are transferred to the bottom of a stroiig glass tube, closed a t one end, and previously filled with dry carbon dioxide o r nitrogen. The requisite amount of sulphur, preferably in the form o€ small crystals, calculated for the above reaction is then added, and the tube sealed a t the blowpipe, and immersed for the lover half of its length in a bath of glycerin, the temperature of which is gradually raised.No. reactlion is apparent up to 155", the sulphur merely melting at 115", and forming a layer below the phosphorous oxide. At a temperature varying in different experiments from 154" to 168", the lower layer of sulphur is suddenly projected into the phosphorous oxide, and the whole mass is violently thrown up to the top of the tube, with a distinct rushing sound.* I n two or three seconds, during which time the contents of the tube are in rapid motion, the reaction is coniplet,e, and the internal walls of the tube are seen to be covered in the cooler portion with feathery, colourless crystals, together with compact masses of a yellowish-grey, crystalline solid, which fuses to a viscous.liquid in the more strongly heated poi.tion of the tube. The heating may, of course, be carried out in an air-bath, but in that case it is impossible to observe the curious phenomena accompanying the reaction. The product is next transferred to a similar tube, which is then exhausted by the Sprengel pump and sealed. On heating the lower half of the vacuous tube, the sulphoxide sublimes in perfectly colourless, strongly-refractive crystals. The sublimation begins at, about go", but the most favourable temperature is about 140-150". A portion of the sublimate a t first condeiises as a viscous liquid which subsequently solidifies to a colourless, vitreous or crystalline mass ; a large proportion, however, always condenses in the form of feathery aggregations, or long ueedles extending across the tube, or isolated rectangular crystals.Analysis has shown that these various forms of the sublimate possess the same composition. Phosphorus sulphoxide melts a t about 102", and boils constantlJ- at 295" (corr.). The melting point is not very sharp, as the sub- stance seems to become somewhat viscous before it actually liquefies, The distilled sulphoxide is pale-yellow, and almost wholly sublimes Jt: It is important that the quantitiee taken should not exceed the ainounts specified, for if more than 5 grams of phosphorous oxide and its equivalent of sulphur are employed, the reaction is so Tiolent that it usually results in a loud explosion, the tube being shattered into Iragments.The explosion is accompanied by the production of an intecsely bright flame.1024 THORPE AND TUTTON : PHOSPHOROUS OXIDE. in a vacuum in the form of the colourless crystals above described. Three distinct preparations, analysed by oxidation with bromine-water and determination of the phosphorus by magnesia mixture and the sulphuric acid by barium chloride, yielded the following results :- Calculated for I. 11. 111. p4°6s4. Phosphorus . , . 35.81 35.71 35.94 35.63 Sulphur . . . . . . 36.08 36.88 36.36 36.78 It was at first surmised that the substance might be a mixed anhydride, that is, phosphoric anhydride in which a portion of the oxygen had been replaced by an equivalent amount of sulphur, or P203S2; but that it actually has tbe composition P406S4, or, in other words, is an additive compound of phosphorous oxide and sulphur, is established by the determinations of its vapour density.These were made by Victor lleyer’s method, in an atmosphere of nitrogen, and at a temperature of 350-430’, using a bath of molten lead. No decom- position was evident; at this temperature ; on cooling, the unaltered substance was found ci+ystzlllised on the walls of the cylindrical tnbe. The following results were obtained 1- Calculated for I. 11. 111. p4°6s4. Wt. of substance.. 0.28.53 0.3558 0.2700 - Density, H = 1 , . . 180.4 171.9 170.8 174 Determinatioiis I1 and 111 were made with crystals obtained by Phosphorus sulphoxid e deliquesces rapidly in air, and hence smells I t is quickly dissolved by water, forming ,, air = 1 .., 12.5 11.9 11.8 12.1 subliming the distilled sulphoxide in a vacuum. of sulphuretted hydrogen. sulphuretted hydrogen and, at first, metaphosphoric acid : P,O6S, + GI-JZO = 4HPO3 + 4H2S. The Inetaphosphoric acid passes, eventually, rapidly if the solution is evaporated upon a water-bath, into orthophosphoric acid, Phos- phorus sulphoxide is readily soluble in twice its volume of carbon bisulphide, from which i t crystallises unchanged. It is also soluble in benzene, upon which, however, it reacts, as the liquid becomes dark and sulphuretted products axe formed. The isolated coloui-less crystals of phosphorus sulphoxide obtained by subliniatioii in a vacuum consist of rectangular prisms, frequently attaining a length of 2 to 3 mm., ahd R thickness of from 1 to 2 mm.A large number of crptals have been examined goniometrically and measured, in spite of their rapid deliquescence. This result has been obtained by employing tlie ingenious little arrangement supplied byTHORPE AND TUTTON : PHOSPHOROUS OXIDE. 1025 Fuess, of Berlin, with the large horizontal-circle goniometer. The essential part of the arrangemeut consists of a small closed glass chamber, so shaped as to carry in its lower portion a desiccating substance, such as calcium chloride or phosphoric anhydride, and 6tting on the adjusting table instead of the ordinary crystal holder. The crystal is enclosed in this dry chamber during the measurement, and the signal passes from the collimator to the crystal, and from the crystal face to the telescope through two adjacent plate-glass sides.I n the case of every crystal examined, no faces were observed besides those of the rectangular prism and basal plane, all the nurner- 011s angles measured being almost and in some cases exactly 90". On examining the crystals, under the microscope, in parallel polarised light, lookiiig through one of the three pairs of faces, the crystals are alwajs found to be isotropic, while the other faces extinguish parallel to the prism edges. In convergent light, the uniaxial circular rings and dark cross arc seen through the isotropic face, provided the crystal is at least 2 inm. thick i n this direction ; if thinner, only the cross is seen, owing to the feeble double refraction.On rotating the crystal, the cross appears stationary. and does not, unless badly adjusted, break up into hyperbolae. The crystals, therefore, appear to be uniaxial, or, if biaxial, the angle between the optic axes must be exceedingly small. Hence they are probably tetragonal prisms, terminated by the basal plane. Although several distinct preparations were made, no other faces were ever observed, the whole of the crystals being homogeneous and exhibiting nothing but pinacoidal faces and basal plane. In order to obtain, i f possible, crystals showing further forms, so as to enable the axial ratios to be determined, a considerable quantity of the sulphoxidc obtained by sublimation in a vacuum was dissolved in carbon bisulphide, arid the latter slowly evaporated in a vacuum from one limb of a v-tube t o the other by cooling the second limb with ice.Large, Fvcll-formed crystals were again obtained, but they consisted, like thosc formed by sublimation, of tetragonal prisms terminated by the basal plane and exhibited no other faces whatever. As previously mentioned, the sulphoxide is sometimes deposited, generally in that part of the tube nearest the surface of the glycerin of the bath, as a colourless, viscous liquid. On cooling, this solidifies t o a transparent glass, which sooner or later devitrifies into crystals of the same form as the isolated ones. Generally, the formation of crystals occurs an hour or so after cooling, but in one case devitri- fication suddeiily set in several days after preparation, with pro- duction of fantastic feathery aggregations of the rectangular crystals.'l'he fact that the isolated crystals, the vitreous modification, and t h e featheq and ncicular forms are composed of the same substance was1026 THORPE AND TUTTON : PHOSPHOROUS OXIDE. proved by direct analysis. The feathery forms are merely aggrega- tions of small, tetragonal prisms, and the needles, similar prisms largely developed in the direction of the vertical axis. Actiow, of XeEenium. -Selenium appears to form a crystalline com- pound with phosphorous oxide, similar. to the sulphoxide, bnt owing t o the fact that the phosphorons oxide is largely decomposed a t the tern- perature of the reaction, the sublimed substance is mixed with t h e products of the decomposition. Action of Xulpphzcr Trion icZe.-When t'he t-apours of sulphur trioxide and phosphorous oxide are allowed to act on one another in a closed apparatus at the ordinary temperature, white flocks of phosphoric anhydride are deposited upon the walls of the apparatus, and sulphur dioxide is liberated.When the t w o oxides are separately placed a t the two ends of a closed tube fitted with a side tube connected with a gas cylinder standing over mercury, and the phosphorons oxide is slowly melted down upon the sulphur trioxide, a, somewhat violeiit reaction occurs with evolution of heat, formation of phosphoric anhydride, and liberation of sulphur dioxide. Even when the re- action is moderated by cooling with ice, no compound of the two oxides appears to be formed, but merely an oxidation of the phos- phorous oxide a t the expense of the sulphur trioxicle.Action of Szdphuric Acid.-Concentrated sulphuric acid dropped upon phosphorous oxide occasions a great rise of temperature. Sulphur dioxide is liberated and the phosphorous oxide becomes oxidised to phosphoric acid. When quantities of a gram and upwards are em- ployed, the reaction is so violent that the mass generally ignites. Action of Xzdphzer Chloride.-Sulphur chloride, S,CI,, acts with great, violence on phosphorous oxide, forming phosphoryl and thio- phosphoryl chlorides, free sulphur, and sulphur dioxide : Action of Ammonia.-Ammonia slowly reacts in the cold with phosphorous oxide. A small quantity of phosphorous oxide placed in a tube over mercury in an atmosphere of ammonia absorbed a quan- tity of ammonia corresponding to between 7 and 8 mols.of ammonia per mol. of phosphorous oxide. When ammonia is led over phok- phorous oxide melted by the warmth of the hand, in an apparatus previously filled with nitrogen, a somewhat violent reaction occurs w i t h production of a white cloud; the mass ignites, and a considerable quantity of amorphous phosphorus or the red suboxide is formed. The violence of the reaction may, however, be controlled by sur- rounding the flask with iced water. On removing the cold bath and again melting by the hand, the same sudden combination with pro- duction of flame occurs.THORPE AND TUTTOS : PHOSPHOROUS OXIDE. LO2 7 The reaction between ammonia and phosphorous oxide is more easily regulated when the phosphorous oxide is dissolved in benzene or ether.When benzene was used as the solvent, the reaction was accompanied by a rise of temperature to about SO”, necessitating occasional cooling by a cold-water bath. The formation of a white solid begins with the passage of the first bubbles of ammonia, and i t is necessary to shake the flask continually so as to prevent, the formation of a solid layer upon the surface of the liquid. When no more heating is noticeable upon removing the colci bath, the re- action is at an end, and the benzene may then be evaporated away in a vacuum. If the benzene is distilled off over a water-bath, consider- able decomposition occurs. In two experiments performed in an apparatus previously filled with nitrogen, the amount of ammonia taken up corresponded to a little over 6YH3 and 7NH, respectively.The reaction is most readily carried out when ether is used as it solvent of the phosphorous oxide, there being but little heating effect even when large quantities are used. 15.63 grams of phosphorous oxide took up 8.46 grams of ammonia, corresponding to a little over 7NH,. The same white product is obtained as in the case of phos- phorous oxide dissolved in benzene, together with smaller quantities of a yellowish, viscous or deliquesced substance. The ether may readily be removed by gently warming in a current of nitrogen. An analysis of 0.843 gram of the solid white substance, separated as completely from the viscous substance as possible, showed the presence of 37.8 per cent. of phosphorus.This corresponds approximately with N H, the hitherto unknown diamide of phosphorous acid, O H O P < ~ ~ , , which contains 38.7 per cent. of phosphorus. A complete separation of the diamide from the viscous substance is impossible, owing to the highly inflammable nature of the product of the reaction, which takes fire at once in contact with the air. It appears probable that phosphorous oxide reacts with 8 mols. of ammonia with the formation of 4 mols. of phosphorous diamide and 2 mols. of water, which may further react with production of a molecule of the corresponding di-ammonium salt. The appearance of deliquescence or formation of small quantities of a viscous substance is probably due to the action of the water produced : PdO6 + 8NH3 = 40€3-P<::: + 2H20 ; 2H,O + OH-P<”H2 H2 0 1’ P d 0 6 + 8NH3 = 30H-P<;% + OH-P<ggZ.Phosphorous diamide is a white powder which dissolves instantly VOL. LIX. 4 B1028 THORPE AND TUTTON : PHOSPHOROUS OXIDE. in water with sufficient rise of temperature to indiice incandescence. When heated in a dry test-tube, ammonia is evolved, then fusion and ft partial sublimation occur. Treated with moderately dilute hydro- chloric acid, a TTiolent reaction occurs with liberation of non- spontaneously iiiflanimable phosphoretted hydrogen, separation of free phosphorus, and foi*mation of a solution of ammonium chloride and phosphorous and phosphoric acids. The fact that the gas evolved was phosphoretted hydrogen was established by a determination of its vapour density by the method described by Thorpe and Rodger in their paper on " Thiophosyhoryl Fluoride " (Trans., 1889, 55, 306).The density obtained was 17.4, that of pure phosphoret'ted hydrogen being 1.7*0. I n this reaction, the effect of the hydrochloric acid appears t o be first to form ammonium chloride and phosphorous acid, but the heat of the reaction is so great that the latter compound is partially converted into phosphoretted hydrogen, phosphoric acid, and free phosphorus : Substituted ammonias behave similarly with phosphorous oxide dissolved in ether, forming the corresponding substituted diamides of phosphorous acid, as white and somewhat viscous solids, which evolve phosphoretted hydrogen on the addition of hydrochloric acid, with separation of yellow phosphorus. Action of Nitrogen Peroxide.-A few grams of phosphorous oxide were placed in one V of a W-tube, and a corresponding quantity of liquid nitrogen peroxide in the other, and the open ends of the apparatus sealed. After some days, it was found that the phosphorous oxide had hecome converted into a voluminous mass of what was afterwards proved by analysis to be phosphoric anhydride.The red liquid nitrogen peroxide had simultaneously become green, On softening one end of t-he apparatus in a flame, comparatively little gas escaped. Hence it appears probable that the phosphorous oxide bad become oxidised to phosphoric anhydride and the nitrogen per- oxide reduced to nitrogen trioxide or nitric oxide, which changed the original colour of the nitrogen peroxide to green. No compound of phosphorous oxide and nitrogen peroxide appeared to be formed, for on dissolving the bulky white substance i n water riolent hissing occurred, b u t no red fumes escaped, and t'he solution gave no reactpion T\r.ith ferrous sulphate.Action of Phosphorus PentachZu,ride.-Phosphorus pen tachloride re- acts somewhat violently with phosphorous oxide, with considerable ex-olution OF heat. The product is a liquid which is raised t o its temperature of ebullition, unless the containing vessel is cooled by ice.O'SULLIVAN: THE GUMS OF THE ARABIN GROUP. 1029 The reaction is complete when the proportion of 6 mols. of phos- phorus pentachloride to 1 mol. of phosphorous oxide is employed. When the liquid has cooled t o the ordinary temperature, and no further rise of temperature occurs on shaking, it may safely be distilled, when it is found to be a mixture of phosphorus trichloride m d phosphoryl trichloride : P 4 @ 6 + 6PC1, = 6POC1, + 4PC13. Action of Phosphorus !l'&chlo&It.. --Phosphorus trichloride and liquid phosphorous oxide are miscible without action a t ordinary temperatures, and practically the whole of the phosphorus trichloride may be distilled off unchanged. When heated together in a sealed fube at 180" for some hours, however, provided the phosphorus tri- chloride is not present in more than the proportion of 4 mols. to 1 mol. of phosphorous oxide, the whole is converted into solid products, As the temperature is rising towards 180", a white solid commences to be deposited, which rapidly turns yellow and then red. The forma- tion of the solid continues until the whole mixture is solid. The red solid is found to consist of it mixture of phosphorus pentachloride, phosphorus pentoxide, and amorphous phosphorus. The phosphorus pentachloride may be readily extracted by carbon bisulphide. Phos- plzoiw trichIoride, therefore, appears to react with phosphorous oxide only at a temperature approaching the boiling point of the latter, and in a closed apparatus under these circumstances does not form phos- phoryl trichloride, but a mixture of pentachloride and pentoxide of phosphorus together with amorphous phosphorus. Hydrogen, phosphoretted hydrogen, carbon monoxide, carbon di- oxide, sulphur dioxide, nitrogen, nitric oxide, cyanogen, and ethylene have apparently no action upon either cold or warm phosphorous oxide.

ISSN:0368-1645    

DOI:10.1039/CT8915901019

出版商:RSC    

年代:1891

数据来源: RSC

摘要:

O’SULLIVAN: THE GUMS OF THE ARABIN GROUP. 1029 XCI.-Researches on t h e Gums of the.Arahi.it Group. Part 11. Geddic A c i d s , Gedda Gums ; t h e Dextrorotcctory Varieties. By C. O’SULLZVAK, F.R.S. h Part 1: of this paper (Trans., 1884, 45, dl), I described arabic acid and its decomposition products, and gave a general idea of the nature of the hvorotatory gums. Although many new facbs con- nected with some of the decomposition products, especially with arabinose, have since been brought to light, they do not materially alter the generail conclusions arrived at and described i n the paper in- 4 ~ 21030 O'SULLIVAN : RESEARCHES ox THE dicated. This will be obvioue from the facts with which I have to deal in this part, which is devoted to a description of some of the dextro- rotatory gums.Scheibbler (Bey., 6, 612) states that the gum obtained from the beet of one season frequently differs from that from the beet of another in optical activity, that yielded by one season being dextro- rotatory and of another laworotatory. He also shows that gum arabic from different sources does not always rotate the polarised rag in the same direction or with the same power. I have examined many samples of gum arabic-Lerantine, Senary, East Indian, and Turkey gum-and although 1: have found them vary in optical activity, I have not found one sample dextrorotatory. On the contraq, I have found but few samples of gedda gum laworotatory ; the dextrorotatory samples predominated. It is to the latter varieties I shall confine myself on this occasion. The gum known in commerce as gedcla gum is, in appearance, very similar to the inferior kinds of the arabin gums, but, as a rule, the pieces of the former are smaller, and more varied in colour and size than those of the latter.The pieces of the gedda gum are irregular in shape, imperfectly rounded, being evidently rounded masses broken. The general appearance of these gums gives one the idea of a rather heterogeneous mixture. Sample of Gedda Gum A. This, the first sample examined, was obtained from the dealers as gedda gum. A few of the whitest and most glassy pieces were selected. This portion was divided into two fractions by dissolving in a little water, partially precipitating with alcohol, in presence of hydrochloric acid, decanting the clear alcoholic solution, and obtaining a further precipitate by adding more strong alcohol.Each fraction was purified by redissolving in water and reprecipitating with alcohol three or four times, and the syrupy precipitates thus obtained were converted into a powder by treating them with alcohol sp. gr. 0.82 and rubbing them down with it. Each portion was then dried, first in a vacuum over sulphuric acid, and finally in a current of dry air at 160" under diminished pressure, and a determination of the optical activity made ; that of themore soluble I-raction, t'hat is, the one separated from the alcoholic solution by the further addition of strong alcohol, was found to be [ a ] D = +54", and that of the less soluble fraction faID = +45". From this it was obvious that even the apparently purest pieces of the gum were not homogeneous, but consisted of a mixture of two or more substances.It was consequently manifest that it was useless to hand-pick t'he gum, and that the most conve- nient method of working was t o fractionate the material a s a whole.GUMS OF THE ARABIN GROUP. Aa ...... A b ...... A c ...... 1031 2-56 1 0'18 2 . 5 8 i 0.38 2.30 ~ 0.27 I The Ash and Water in the Rample of Gedda Gum A. Before proceeding with the fractionation of the gum, I thought it desirable to acquire some knowledge of the amount of ash in the gum and the constitution thereof, and also of the amount of water. The gum was roughly hand-picked into three fractions, viz. :- A a, consisting of white pieces. A b, 7 9 amber ,, A c, 9 , reddish ,, 15.404 grams of A a gave 0.469 gram ash after treatment with am- 10.534 grams of' A b gave 0.341 gram ash after treatment with am- 24.146 grams of Ac gave 0.738 gram ash after treatment with am- An analysis was made of each of these ashes; the results are as monium carbonate.monium carbonate. monium carbonate. follows :- Percentage on Crude Gum. 1 Ash calculated given. K,C03. 1 Total. 1 from figures A determinahion of the water was made in each fraction with the following results :- 1.001 grams of A a, finely powdered, lost 0.117 gram in a vacuum, and a further 0.031 gram in a current of dry air, at 250 mm. pressure and 100". 1.515 grams of A b, finely powdered, lost 0.204 gram in a vacuum, acd a further 0.048 gram in a current of dry air, at 250 mm. pressure and 100".1.626 grams of A c, finely powdered, lost 0.212 gram in a vacuum, and a further 0.036 gram in a current of dry air at 250 mm. pressure and 100". The weight did not become constant in a vacuum over sulphuric acid, but when the 10% reached the stage given, the decrease was so slow that I did not consider i t of importance, under the circumstances,1032 O'SULLTVAN : RESEARCHES ON THE to continue the drying further; I have no doubt, however, the gum would lose all its water in a vacuum over sulphuric acid a t the or- dinary temperature. From these numbers the following percentages are calculated :- A a = 14.78 per cent. water. Ab = 16.63 ,, $ 7 A c = 15-25 ,, , 9 These figures, those obtained for the ash, and its constitution, show, again, that the gum is not homogeneous.I shall have to further consider the ash later on. The next step was to determine the number and nature of the or- ganic constituents into which the gum could be separated. The whole gum was dissolved in as little water as passible, the solution placed upon a dialyser and hydrochloric acid added a little at a time until the liquid on the dialyser was found to be free from calcium,. and then until the hydrochloric acid had also disappeared, or, a t least, until not more than a trace of i t remained. To the solution thus freed from ash, alcohol of sp. gr. 0.83 was added until about five- sixths of t'he solid matter therein was precipitated ; the precipitate thrown out of solution in this way separated as a syrup, thus differing completely from the arabin gum acids, which, under like conditions,. are thrown out as curdy precipitates.When the supernatant liyiiid became clear, it was decmted off the syrup, and stronger alcohol added t o it as long as a precipitate formed; this most soluble portion we shall call fraction a. Thc syrupy precipitate was redissolved in a little water, and alcohol added to the solution in sufficient quantity to throw down three-fourths of the solid matter i t contained ; fraction b was taken out of the clear supernatantt liquid by the addition of stronger alcohol. Proceeding in the same way, the second precipitate was divided into fractions c arid d. A portion of each fraction was converted into a, brittle, whitish powder by treatment with alcohol of sp. gr. 0.82. A few grams of each of the powders were dried in a vacuum over sulphuric acid for three or four days and then at 100" under a pressure reduced tu 200-250 mm.mercury, until the weight became constant, When these substances are heated at 100" before the greater part of the water and alcohol they contain is eliminated by drying in a vacuum over sulphuric acid, they, like the arabic acids, are so changed as to swell up when afterwards treated with water into a jelly-like mass, without dissolviug. They also undergo this change if they are kept in the undried state for a few months. This is the so-called meta- modification. On the contrary, when dried as indicated above,. they remain soluble and show no sign of being converted into the jelly-GUMS OF THE ARABIN GROUP. 1033 yielding modificaticn.of the dried fractions was made with the following results :- A determination of the optical activity of each a , 4.608 grams substance dissolved to 100 C.C. solution, sp. gr. = 1,01733 ; optical activity, in a 200-mm. tube, for sodium fla,me = +5*37". b. 8.569 grams substance, 100 C.C. solution, sp. gr. 1.03248 ; optical activity an = +9.29", in a 200-mm. tube. c. 5.090 grams substance, 100 C.C. solution, sp. gr. =: 1.01949; optical activity aD = +4*64", in a 2100-mm. tube. d. 8.270 grams substance, 100 C.C. solution, sp. gr. = 1.03237; optical activity cx, = +5.73", in a 200-mm. tube. These numbers lead to the following results :- A bnriuni sslt of each of these fractions was prepared by exactly neutralising a portion of a solution of each with clear bnryta-water, c;trefully neutralised litmus paper being used as an indicator, adding alcohol, and rubbing down the waxy precipitate with alcohol of sp.gr. 0.82 to a powder. Dried as usual, that is, first in a vacuum over sulphuric acid and finally in a current of dry air at 100" under a pressure of 100-120 mm. mercury, the salts yielded on analjses the following results :- a = 5.31 per cent. BaO. c = 6.29 3 , d = 7.30 9 9 b = 5.77 7 The optical activity of the organic portions of the salts was practi- cally the same as that, of the free acids before neutralisation. Froin these experiments, it is obvious that this sample 9f gedda gum contains a t least two acids of very differeiit optical activity axid neutralising power.. It now became necessary to determine if there were more than two acids in the gum.With this object in view, fraction a was first examined. To its solution in as little water as possible, sufficient * D = a number calculated from the sp. gr. by rejecting the unit, multipIy- ing the decimal by 1000, and dividing t,he product by the weight of substance in grams in 100 C.C. solution. It is a good comparative number. t High D, due to presence of ash.1034 O'SULLIVAN : RESEARCHES ON THE alcohol mas added to precipitate a very small quantity of it ; this was rejected, and a further addition of alcohol made to the clear super- natant liquid until a quantity estiniaked at one half the solid matter in solution was thrown out ; this fraction, a2, was allowed t'o settle until the supernatant liquid became clear, when the remainder of the gum acid was precipit'ated by adding an excess of strong alcohol; this is called fraction nl.These gum acids are soluble to a con- siderable extent in alcohol weaker than sp. gr. 0.87 ; hence, in dealing with them as indicated here, it is necessary to keep the solutions concentrated. When pure, they are easily " milked," that is, they form, when treated with alcohol, milky solutions in which the further addition of strong alcohol does not produce a precipitate ; a drop or two of hydrochloric acid separates the gum acid completely from these solutions. Thc ammonium, potassium, and sodium salts of these gum acids are not precipitated from their solutions by the careful addition of strong alcohol, but a milky liquid is produced from which most acids precipitate tlie gum acids.It may, therefore, be that the milking of the gum acid solution with alcohol, as described above, is due to the presence of ammonium salt ; I have little doubt, however, that the pure acids can be made to yield milky solutions with a1 c o h 01, yielded the following results :- A determination of the optical activity of fractions al and C L ~ . [RID = +58.5". a,. [.ID = +58*5". These numbers are the same as that obtained f o r the undivided fraction ; hence, that portion was a homogeneous substance. In order, however, t o have duplicate numbers, I prepared barium and calcium salts of the two fractions a, and for analyses. These salts were made in the same way as the barium salts of t'he fractions described above. They were dried as usual, and yielded, on analysis, the following results :- al.1.795 grams barium salt gave 0.1465 grams RaSOp. 1.585 9 7 7 , 0.1110 ,, BaC03. 7.092 7 ' 7 7 100 C.C. solution a t 15.5". 15.5" 1a.5 sp. gr. To* = 1.0%952. This solution, i n a 200-mm. tube, had an optical activitF tlD = 7.73". # The weight of tlie liquids a t 15.3 compared with the same bulk of water also at 15.5". !these are the conditions under which all the sp. gr. experiments were made,GUMS OF THE ARBBIN GROUP. 1035 a,. 1.996 grams of barinin salt gave 0.1625 gram BaSO,. 0.853 7 7 7 7 0.0585 ,, BaCO+ 8.158 7, 9, 100 C.C. solution at 15.5. Sp. gr. 1.03386. This solution, iii a 200-mm. tube, had an optical activity aD = 9.01". In determining the barium as sulphate in these salts, when the sulphuric acid is added t o the solution, the sulphate is precipitated in such it form as does not admit of its being filtered, nor does it settle out ; in fact, thrown on t o a filter of the very best Swedish paper, it passes through readily. On digesting, however, with an excess of sulphuric acid for some t h e , until, in fact, the gum acid is broken dcwn, the barium sulphate can be easily separated. The carbonate was prepared by igniting the salt, and treating the white ash with ammonium carbonate.The numbers lead to the following results, viz. : - al. 5.36 per cent. UaO from sulphate ; 5.42 per cent. from carbonate. a2. 5.34 37 9, 5.33 >> > > al. [aIn = 3-57-3" for organic acid.* a,. D = 4.16; a,. D = 4.15. a,. [ a ] D = +58*3" 7 7 ul.1.834 grams calcium salt gave 0-0714 gram CaCO,. 'The carbonate WRS prepared in t'he same way as the barium 7.490 grams calcium salt of a, p v e 100 C.C. solution at 15.5" of sp. gr. 1.03019. This solution had an optical activity, in a 200-mm. tube, ED = + 8-60'. 7.668 grams calcium salt of a? gave 100 C.C. solution at 15.5" of sp. gr. 1.03130. This solntion had m i optical activity, in a 200-mm. tube, = 8.84". az. 1.812 7 7 9 7 0.0641 9 , earbonate. These numbers lead t o the following results:- a,. 2.18 per cent. ClaO a,. 1.98 9 ) ~ 1 . [a]= = +58*6O. a2. [a]D = +58'8'. a l . D = 4.03 ; a2. D = 4-88. These results leave no doubt that we have a homogeneous substance, For reasons which I shall give in the i n a pure state, to deal with. sequel, 1. propose to call this body tetral.abiizavltrigaZacta?.Lgeddic acid.ash. * In all cases the specific rotatory power is calculated for the substance free from1036 O'SULLIVAN : RESEARCHES ON THE This acid, as I have already to some extent pointed out, differs very materially from arabic acid ; it is precipitated from its aqueous solu- tion by alcohol as a syrup, whilst arabic acid is thrown out as a curdy precipitate ; i t is soluble, to a considerable extent, in alco,hol of 0.87, arabic acid is almost insoluble in alcohol of that strength ; its optical activity is [a]D = + 58"to + 59", that of arabic acid [ a ] D = -24 to--25" ; and its barium salt contains 5.33 per cent. of barium, whilst barium arabate yields 6.00 per cent., showing that the geddic acid has a larger molecular weight.With these differences and the many agree- ments bet'ween the tm-o acids, it is a matter of some interesi to determine in what relationship they stand one t o the other; this shall be a question for future work. I must confine myself, on this occasion, to a consideration of the geddic acid. We have now to turn our attention to the b fraction described above. It was separated, by partial precipitation with alcohol, in the same manner as that described in dealing with fraction a, into three fractions, vie., bl, bS, and b,, of which b, was the largest and most soluble. A determination of the optical activity of each of these fractions was made after the manner already described when dealing with the u fractions, with the following results :- 73,. [a]= = +56.0".ba. [cc]D = +4Ym5". b,. [a]= = +46.8". On again fractionating b,, a large and most soluble fraction was obtained, the optical activity of which was fonnd to be [ a ] D = + 58" to +XI"; a small, least soluble fraction had [a]D = +48", and there were fractions with intermediate activities ; all of these were divisible into a fraction [.ID = +So, at the most soluble side, and one of [a]D = 48", at the most insoluble one, there being no indivisible fraction w-ith an activity between the two. From these figures and facts it is evident that the b fraction con- tained at least two compounds, tetrarabinantrigalactangeddic acid and a substance of lower optical activity. The whole of the fractions the optical activity of which was [ElI, = +48" to +49", from b, and bz, were mixed, and divided into two, by dissolving in water, adding sufficient alcohol to throw out about one half the solid ma,tter, allowing to stand until the super- natant liquid became clear, decantiug, and precipitating the remainder by adding strong alcohol; this, the more soluble portion, we shall call z; the first precipitate is 8.The optical act>ivity of these fractions was found to be- baa [a]D = +49 0". 6/3. [a]D = +48*6.GUMS OF THE ARBBIN GROUP, 1037 These, then, constitute a definite and distinct fraction. A barium salt was prepared of the a portion of it, in the way already described. 1.729 grams of this salt,, dried as usual, gave 0.160 gram RnSOa. 6.756 grams of this salc, dried as usual, gat-e 100 C.C. of sp. gr. This solution, in a 200-mm.tube, had an optical =1.02943. act8ivity CLD = + 5-14". These numbers lead to the following results :- BaO = 6.08" per cent. LccID = +49*5" for acid. T_) = 4.35. Further, the barium salts of several fractions, analyses of which will be given later on, gave percentages of barium oxide varying between 5.9 and 6.1. It is clear froni these factors that we have in this fraction a secoiicl acid, in the pure state. For reasons which shall be given, I propose to call it triarnbinaizti.igalactangeddic acid. This acid is equal in neutralising power to arabic acid, but it differs. from it in its optical activity ; in being, like the former geddic acid, precipitated from its aqueous solutions by alcohol as a syrup ; and in being more soluble in alcohol. I n fact, there is no difficulty in separating it from arabic acid by fractional precipitation, the arabic acid being thrown out first.As shown above, the c fraction of the gun1 acid had an optical activity [a]= = +45*5"; in the isolation of the geddic acid just described, many fractions were obtained, all with a lower optical activity than ti*iarabinantrigalactangeddic acid ; hence, there must be in the gum a lowcx- rotating acid then that just described. All those portionswith [a]o = +43" to +4Go,including the c fraction, were mixed, dissolved in a little water, and alcohol gradually added, with continual stirring, until a portion, estimated a t 90 per cent. of the total solid matter in solution, was precipitated. The precipitate was allowed to subside, the supernatant liquid, when clear, was decanted off, and strong alcohol added as long as a precipitate was produced.The optical activity of this fraction was found to be [aID = 48-49", hence it was triarabinantrigalactaiigeddic acid. The remaining 90 per cent. precipitated by the first addition of alcohol was again dissolved in a little water, and about 90 per cent. of it pre- cipitated by the gradual addition of alcohol ; fraction c1 was taken from the clear supernatant liquid by adding strong alcohol. The operation was again repeated with the precipitate un ti1 fractions c2, c3, c4, and c5, were obtained, c1 being, of course, the most soluble. The optical activity of these fractions was found to be1038 O'SULLIVAN : RESEARCHES OX THE cl. [a]D = +46". c2. [u]D = 4-43'. ~ 3 .[uJD = +42". c4. [all, = +43". ~ 5 . [ a ] D = +36". c1 admitted of being split up into two fractions, the one [a]D = +49" triarabinantrigalactangeddic acid on the most soluble side, and the other [a]= = +43" on the least soluble one, there being no interme- diate indivisible fraction. All the portions = +42" to +43", were again mixed and divided into three fractions, cu, c/3, and cy, ca: being the most soluble. An examination of the barinm salts of each of these fractions led to the following results :- 2.864 grams salt of cu, dyied as usiial, gave 100 C.C. solution, sp. gr. This solution, in a 200-mm. tube, rotated the polarised 2.908 grams salt, of c/3, dried as usual, gave 100 C.C. solution, sp. gr. This solution, in a 200-mm. tube, rotated the polarised 3.126 grams salt of CPJ, dried as usual, gave 100 C.C.solution, sp. gr. This solution, in a 200-mm. tube, rotated the polarised 1.01222. ray aD = +2*30". 1.01237. ray UD = +2*33". 1.01300. ray aD = 2.50". 0.716 gram salt of cu, dried, gave 0.072 gram BaS04. From these numbers we have the following :- ccr. [RID = +43.0" ; D = 4.27 ; BaO = 6.60 per cent. ~ y . [RID = $42.6"; D = 4.16. CP. [u]D = +42*8"; D = 4.25. The D of the salts of the cy fraction shows t,hat the preparation was not quite dry. Several other barium salts were found t o contain from 6.55 to 6.65 per cent. BaO. This portion which admitted of being thus divided is: therefore, a third acid in a pure state ; I shall call i t diarabinantrigalactangeddic acid. The precipitate of this acid thrown out of the aqueous solution by alcohol is less syrupy than that of the other gum acids described.Proceeding in the same way with the portions having an optical activity less than [KID = +43", I succeeded in eliminating a fraction [a]= = +36", which admitted of being divided into three fractions of the same optical activity. In doing this there was a less soluble small portion on one side, and the more soluble diarabinantrigalactan- geddic acid OE the other, This is a fourth gum acid ; we shall call itGUMS OF THE ARABIN GROUP. 1039 .1Pzonarabi.nantrigalactangeddic acid. lysis, the following results :- Its calcium salt yielded, on ana- 0.672 gram dried substance gave 0.0428 gram CaCO,. 2.688 grams ,, ,, 100 C.C. solntmion, sp. gr. 1.01124. This solution, in a 200-mm.tube, roLated the polarised ray ZD = +1*91". From these figures we have :- CaO = 3.5 per cent. [aID = +36.7" for acid. 3.50 CaO = 7.30 BaO. The portion of the dialysed gum less soluble than this acid, 1 neglected, as it was small in quantity, much coloured, and not easily worked. I, however, determined that it contained a nitro- genous compound, easily soluble i n acids and alkalis, which in alka- line solution gave a beautiful purple colour, when treated with a few drops of alkaline copper solution (Fehling's solution). On boiling this, a portion of the copper oxide was reduced to red oxide ; this is due, probably, rather to the presence of some gum acid than to the ni trogenous substance. I have so far shown that the gum consists of the calcium, mag- nesium, and potassium salts of four acids, which I have named and characterised as follows :- a Tetrarabinantrigalactangeddic acid, [ c c ] ~ = + 58-59" ; per cent.Triarabinantrigalactangeddic acid, [a] D = + 48--49' ; per cent. of Diarabinantrigalactangeddic acid, [a]D = +42-43" ; per cent. BaO Monarnbinantrigalnctangeddic acid, [a]= = + 36-37' ; per cent. BaO in barium salt = 5.33, BaO in barium salt = 6.00, in barium salt = 6.60, Ba0 in barium salt = 7.30, and a nitrogenous substance of the character of a prote'id. The names applied to the acids give some idea of their constitution, but, of course, a t this stage of the investigation, I did not possess the data which enabled me to apply these terms to them ; that knowledge was the result of an examination of the products of the action of sulphuric acid on the acids themselves.A solution containing 30 grams of tetrarabinantrigalactangeddic acid in 90 C.C. was heated to 92-95' in a water-bath, and 10 C.C. dilute. sulphuric acid containing 2 grams acid, heated t o the same tempera- ture, was added to it. The digestion was then continued for 13 minutes, when the solution was rapidly cooled. To the cold solution, alcohol of sp. gr. 0.82-0.83' was added as long as a precipitate1040 O'SULLIVAN : RESEARCHES ON THE formed. The sugar o r sugars rernained i n the alcoholic solution, whilst tlhe giim acid or acids were i n the precipitate. This was neutralised with strong baryta-water, the barium sulphate and a little barium salt of a gum acid allowed to settle, and the clear solution decanted.The precipitate was washed several times on a filter with alcohol of 0.53, and the washings added t o the original solution, in which they invariably produced a cloudiness, due t o the presence of a little barium salt. The alcoholic liquid was then evaporated, iu a vacuum, t o a syrup, which, on standing for a few hours, became a mass of crystals ; after standing for some days, these were washed with dry methyl alcohol and dried. An examination of the crystals showed that they had a higher optical activity than (3-arabinose* (Trans., 1884, 45, 41), namely, [RID = +115"; the reducing power being K = 105-106". They were redissolved in a little water and the solution allowed to stend for some time, when crystals appeared, which showed a habit somewhat different from the arabinose from arabic acid, although they undoubtedly belonged to the same system.The sngar agreed in all its other properties with arabinose, its optical activity, &c., being identical, as shown by the following experiments :- I shall deal with the alcoholic solution first. c . t D. Cal n. K. CL 108.3. lJ 109.3. 110.3. I. 4:960 3.828 + 104.f' 11. 4459 3.860 + 105.4'' 111. 4.543 3.860 + 1046' 110.3. IV. 4.846 3.861. + 104-7" 109.9. The mother liquors, those in the methyl alcohol washings as well i t s the aqueous ones, were mixed and evaporated in a vacuum to a strong syrup. This was treated with dry methyl alcohol, which separated a few grams of barium salt of gum acid ; this we shall deal with later. The solution was again concentrated to a syrup in a vacuum, and the syrup dissolved in it little methyl alcohol; this solution yielded a further crop of arabiiiose on standing. On freeing the mother liquor from alcohol, the optical activity of the solid matter in solution determined from the sp.gr. was found to he LaJD = + 160". Arabinon was isolated from this as described (Trans., 1890, 57, 59). There was 110 evidence of any other product ; all the factors of the solid niatter agreeing with mixtures of arabinoii and arabinose. * Since the publication of the paper quoted here, much tias been done to enable us to identify arabinose absolutely. Had I had the same means a t my disposal then, I should not hare used the tcrni B-arabinose : the sugar is undoubtedl? arabinose, a-nabinose bcing the sugar I have since described as ambinon.t c = gram. per 100 C.C.GUMS OF THE ARABIN GROUP. 1041 There was present, indeed, a small quantity of a non-crystallisable substance OF low rotating power, which yielded no mucic acid on treatment with nitric acid, showing the absence of galactose (y-arabinose) or a galactose-yielding substance. I n the paper just quoted, I showed that ttrabinose, when digested with sulphuric acid for a short time, yields a low-rotating, syrupy snbstance of the same kind ; hence, it can fairly be concluded that the small quantity of the substance mentioned is derived from arabinose by the action of the sulphuric acid. It is, therefore, clearly established that arabinon, arabinose, and the precipitate with alcohol are the only products oE the first qutcrter of an hour's action of sulphuric acid on the geddic acid. This was dissolved in a little water and again precipitated, the operation being repeated until the precipitate was free from sulphuric acid.The alcoholic solutions thus obtained were added to that the results of the examination of which are given above. A portion of the syrupy precipitate was dissolved in water, and the solution boiled until free from alcohol. The sp. gr. of this solution was found to be 1.03362, hence 8.8466 grams substance in 100 C.C. (taking mean value for sp. gr.), and its optical activity in 200-mm. tube aD = +4.45". This gives a specific rotatory power for the solid matt'er [a]n = + 25". The remainder of the syrup was dissolved in a little water and divided into two fractions, S,, more soluble, and S,, less soluble, 'by treatment with alcohol, as already described.Barium salts were made of a portion of each of these fractions in the usual way. On analysis .they yielded the following results :- 0.670 gram S,, dried a t loo", gave 0.080 gram BaS04. 2.682 grams S , ,, ,, 100 C.C. solution, sp. gr. 1.01201, 2.852 grams S,, dried a t lOO", gave 100 C.C. solution, sp. gr. 1.01254, We have now to turn to an examination of the precipitate. 0.713 ,, S:! 9 , 9 9 0'085 93 and rotated the ray i n 200-mm. tube = +1.21". and rotated the ray a;> in 200-mm. tube = +1.38". From these numbers we have- S1. 7*8$ per cent. BaO in barium salt. S,. 7.83 ,, 7, 7 3 S,. [.ID = +24.7" for organic acid; D = 4.48. S,. [aJn = 3-26.3" 7 D = 4-40.A consideratiou of these factors makes it clear that we had in the precipitate a pure, homogeneous compound ; I shall call it geddimsic1042 O'SULLIVAN : RESEARCHES ON THE acid ; i t differs completely,'as we have seen, from either of the natural gum acids in optical activity and neutralising power. The gum acid salt of barium precipitated on neutrnlising the alco- holic solution with baryta-water, and the small quantity of barium salt insoluble in dry methyl alcohol obtained when treating the syrup as described above were, upon examination, found to contain slightly more barium, and to have a somewhat less optical activity than the barium salts just described ; the meaning of this will be clear later on. It was difficult to sepwnte the organic barium salt precipitated with the barium sulphate from the latter ; the separation was accom- plished, however, by boiling the precipitate with water, filtering, washing the residue with boiling water (the filtrate was quite milky from barium sulphate), evaporating the filtrate till it became mode- rately concentrated, arid then adding alcohol drop by drop until a slight portion of the barium salt was rendered insoluble ; this took down with it the barium sulphate, and left the supernatant liquid clear.The organic barium salt was obtained by further addition of alcohol to the clear supernat,ant liquid, Geddinosic acid, prepared and purified as above described, was acted upon with sulphuric acid. A solution of it was prepared con- taining in 100 C.C. about 25 grams of the dry acid, heated in a water- bath, and 2.5 grams of sulphuric acid, previously diluted, added to it, The digestion was then continued for 15 minutes, after which the liquid was allowed to cool, when strong alcohol was added to it as long as a precipitate formed. The alcoholic solution and the precipitate were examined in the same way as is described above.The solution contained arabinon in very small quantity, arabinose, and a little gum acid. There was no galac- tose, no mucic acid being produced by the action of nitric acid on the sugars, and other sugars than arabinon and arabinose were proved to be absent by crystallising the arabinose, and finding that the solid matter in the mother liquors, when freed from arabinon, had an optical activity not far removed from arabinose.In this case, how- ever, it must be stated that altogether but very little sugar, in proportion to the geddinosic acid taken, was produced. The precipitate, purified from sngars and sulphuric acid by repeated precipitation with alcohol, and finally by dialysis, was divided into four fractions by partial precipitation with alcohol, after the manner already described, Gl being the most, and G, the least soluble fraction, The optical activity of these fractions was found to be- GI. [ a ] ~ = +26*8". GZ. [a]D = +d4*4". G3. [a]= +22*2". G,. [ a ] D = +22'2".GUMS OF THE ARABfN GROUP. 1043 Fractions GI and G2 are practically unaltered geddinosic acid. G, and Gq are products of the action; these were mixed and again divided into two fractions, G a and G p.Barium salts were prepared of portions of each of these fractions; they yielded on analysis the following results :- 1.828 grams of G a, dried as described, gave 0.201 gram BaGO,. 1.803 ,, G p 9 , 7 , 0.199 9 , 7.596 ,, G a 9, 7 9 100 C.C. solution, sp. gr. 1.03432, and had an optical activity an = +3.08" in a 200-mm. tube. 7.434 grams of G p, dried as described, gave 100 C.C. solution, sp. gr. 1.03351, and had an optical activity = +3.02" in a 200-mm. tube. Prom these numbers we have :- G a. 8.54 per cent. BaO in the barium salt. G p . 8.56 7, 9 , 9 , CJ a. [all, = +22*2" for acid ; D = 4.52 f o r barium salt. GIQ. [a]= = f21.9" ,, D = 4.51 9 9 These numbers show Ohat this is another acid in the pure state; I propose t o call i t trigalnctangeddic acid. Taking into consideration the very sinall proportion of sugar pro- duced by this further 15 minutes' action of sulphuric acid on geddinosic acid, this new acid may be looked upoii a9 a distinct resting stage in the action of sulphuric acid on tetrarabiiiaritrigalsctangeddic acid.That such is the case was proved by acting on that acid, in the mag described, for 40 minutes, when i t was found that trigalactan- geddic acid, with practically the same neutralising power and optical activity as given above, was the only substance formed in quantity besides the sugars. As can be seen from these experiments, neither of the acids found ready formed in the gum are amongst the products of even 15 minutes' action of sulphuric acid on the higher acid ; hence it would appear that neit'her of the loww gum acids is a resting stage in.the action of sulphuric acid on the higher one. It seemed to me, however, worth determining whether o r not it was possible t o so modify the action of the sulphuric acid on the higher acid as to obtain one or all three of the lower ones. 50 g~*ams of dl-y tetrara~inantriffalactan~e~ dic acid, 2.5 grams of sulphuric acid, and 60 grams of water were digested for five minutes at 98-99'. The gum acid was dissolved in a portion of the water, and heated to the temperature mentioned ; the sulphuric acid was added VOL. LIX. 4 c1044 O'SULLIVAN : RESE-4RCHES ON THE to the remainder of the water, and the dilutedacid boiled, the liquids were then mixed, and the digestion continued in a water-bath. At the end of five minutes, the solution was cooled as quickly as possible, and strong alcohol added as long as a precipitate was produced.The alcoholic solution and the precipitate were examined in the same way as has been described in treating of the other conversions. The sugars in the solutions were arabinon in large quantity, as might be expected, and arabinose ; there was no galactose. I n purifying the gum acids from the sugars, the low solubility of arabinon in strong alcohol had t o be taken into account; the pre- cipitate was dissolved in a little water and alcohol of sp. gr. 0.85 added in insufficient quantity, so as not t o precipitate the acids com- pletely. The clear supernatant liqnid was then neutralised with bnryta-water, upon which a precipitate of barium sulphate and the barium salt of a gum acid were precipitated ; these were collected and washed with dilute alcohol.The filtrate was evaporated in a vacuum to a syrup, which was dissolved in a little niethyl alcohol, and strong alcohol (0*820) added; some barium salt of a gum acid was precipitated, and the mixture put aside to allow the supernatant liquid to 'become clear. On examining it at the end of a few days, it was observed that the slight syrupy precipitate and the sides of the vessel were covered with sphero-crystals ; these, I have reason to believe, are arabinon. I mention this matter here simply to indicate the direction in which that sugar is to be lookedfor in quantity, and the possibility of obtaining it in the crystallised state. The precipitate, with insufficient alcohol, was purified on the lines indicated, and divided into three fractions, TI, T,, and TB, TI being the most soluble.The ED and D of these fractions were determined with the following results :- 3.198 grams TI, dried as usual, gave 100 C.C. solution, sp. gr. 1.01230; this, in a 200-mm. tube, had an optical activity 3.200 grams T,, dried as usual, gave 100 C.C. solution, sp. gr, 1.01210 ; this, in a 200-mrn. tube, had an optical activity 2-84,, grams T3, dried as usual, gave 100 C.C. solution, sp. gr. 1.01086 ; this, in a 200-mm. tube, had an optical activity aD = +2'17". aD = +2*00". a D = +1.84". Prom these numbers we have :-GUMS OF THE ARABIN GROUP. 1045 A portion of the purified undivided precipitate was convcrted into 3.470 grams dry salt gave 100 C.C.solution, sp. gr. 1.01526 ; this, in a 200-mm. tube, had an optical activity aD = + 1*97"* 0.867 gram dry salt gave 0.094 gram BaS04. Hence [a]= = +30.5"; D= 4.40; BaO = 7.12 per cent. These correspond sufficiently well with the factors obtained for monarabinantrigalactangeddic acid, and as the barium salt which sepa- rated on neutralizing the alcoholic solution, yiclded nnnibers agreeing pretty well with these, it i p clear that no higher acid than the mon- ambinan one was prodaced in any quantityby the five minutes' action of sulphuric acid on the high gum acid. I am, however, of opinion that by carefully regulating the strength of the solution, the quantity of sulphuric acid, and the temperature, the three lower natural acids of the series can be prepared from the higher one.From this it would, of course, naturally follow that the products of the final action of sulphuric acid on all four acids are the same. This I have proved t o be the case. A solution containing in 90 C.C. 25 grams of trinrabinantrigalactan- geddic acid was heated to 96-97' in a water-bath, 10 C.C. dilute sulphuric acid containing 2 grams H2S04 added, and the digestion continued for 20 minutes. The cooled liquid was treated with alcohol as described above so as to separate the sugars from the gum acids. On distilling off the alcohol from the final alcoholic solution freed from sulphuric acid, and concentrating to a syrup, a large crop of crystals was obtained; this was higher in optical activity than arabinose ; a second crop was obtained which approached very closely t o arabinose in optical activity, and a third crop was allowed to form ; this agreed with arabinose still more closely.The three crops were washed with methyl alcohol, and recrystallised from water ; the crystals were exactly the same in habit as those obtained from tetr- ~rabinantrigalactangeddic acid, and the optical activity of the sugar was found to be- a barium salt. [a], = +104*5" c = 3.169, which is the activity of pure arabinose. The methyl alcohol washings and the mother liquors were mixed, and the mixture freed from alcohol by distillation. A determination of the optical activity of the solid matter in solut'ion, which was little in proportion to the arabinose cry s t all ised , gave- [ a ] D = +140°, showing the presence of arabinon. no mucic acid being produced by the actiou of nitric acid.4 c 2 There was no galactose present, I may,1.046 O'SULLIVAN : RESEARCHES ON THE however, say that this test for galactose is unnecessai.y, for if it should be present, the first crop of crystals is always contaminated with it,, as is indicated by the crop possessing a lower. rotatory power than arabinose. The gum acids were purified from the sugars as much as possible and divided into two fractions, (1) more, and (2) less, soluble. Barinm salts of these were prepared ; they yielded on analysis the following results :- (1) = 7.55 per cent. BaO, and (2) = 8.03 ,, 9 9 (1) [aIn = +27.3" for organic acid, (2) [a]J) = +23.8" 9 ) A comparison of these numbers with those obtained for geddinosic acid and trigalactangeddic acid shows very clearly that we have in this product a mixture of both acids; it was not necessary to frac- tionate it further.Diarabinantrigalactangeddic acid was acted upon in the same way with the following results :- The sugars were found to be arabinon and arabinose ; there was no galactose. The purified gum acids were divided into three fractions, D1, Dz, and D3, of which D, was the most soluble, and of these, barium salts were prepared ; they yielded on analyses :- Per cent. BaO. [ a ] ~ for txicl free from BaO. D, .......... 7.56 + 26.3" D, .......... 7.85 + 25.8 D, .......... 7-90 + 18.8 These numbers also show that we have in the transformed product geddinosic and trigalactangeddic acid. Both the percentage of barium and the optical activity of the D, fraction are low, but when we consider that this fraction coiitained all the impnrity of the original acid, we may feel satisfied that, the indication is sufficiently distinct to obviate the necessity of further fractionation.~onarabinantrigalactangred~ic acid treated in the same way yielded similar results. The sugars were the same, arabinon, however, being present only in minute quantities. An analysis of the barium salts of the two fractions into which the gum acids were divided gave tha following results :- Per cent. BLLO. [ a ] ~ for acid free from BaO. MI.. ........ 7.92 + 25.1" M2.. . . . . . . . . 8-30 + 21.8GUMS OF THE ARABIN GROUP. 1047 There was a small, less soluble fraction than M,, but as its appear- ance left no doubt that it contained much impurity, it was neglected.Here, again, we have geddinosic and trigalactangeddic acids, the same products as in the three preceding cases ; it is clear, therefore, khat all four gum acids break down in the same way, and yield, as a distinct resting stage, the same compound, namely, triguhctangeddic mid. I n further support of the conclusion that the resting stage in the action of sulphuric acid on the four acids is the same, I determined the amount of mucic acid yielded by the action of nitric acid on hhe products from the four sources which agreed in optical activity and neutralising power with trigalactaiigeddic acid. To 1 part of each oE these products, 7.5 parts of nitric acid and 2.5 parts of water were added ; the mixture was heated in a beaker of 60 C.C.capacity, covered with a watch glass until red fumes. began t o be evolved, when the vessel was removed from the source of heat until the violence of the reaction had ceased. Heat was again applied until nitrous fumes were no longer evolved in quantity. On cooling, the contents of the beaker became solid, and after standing a few days the mucic acid was collected on a tared filter, washed with cold water, dried at loo", and weighed. To the weight thus obtained, minus the weight of tlie filter, a, quantity was added proportionate to tlie bulk of the filtrate ; this gave the mucic acid. It was found to vary from 50 to 54 per cent. of' the gum acid product employed, but 110 satisfactorily concordant results were obtained with the acids from either of the four sources, although they all agreed in their variations ; in fact, all four products behaved as if they were one and the same substance, and yielded a quantity of muck acid, not far removed, above or below, from 52 per cent.Although these experiments would not in them- selves be conclusive as to the identity of the four products, they materially add to the evidence already adduced, that trigalactangeddic is a distinct resting stage in the action of sulphuric acid on the four natural gum acids of the sample of gedda gum under examination. I must not omit to point out that acids other than trigalactan- geddic acid and geddinosic acids are at times found amongst the product's of the transformation ; they possess a higher rotatory power and neutralising power than either of these acids.When they appear amongst the products, galactose is found with the sugars. This will be accounted for later on. Before we proceed any further with this investigation, it would probably be as well to enquire into the meaning of the facts we have established. To begin with, we have tetrarabinantrigalactangeddic acid, with optical activity [ a ] U = +58", the barium salt of which contains 5.331048 O'SULLIVAN : RESEARCHES ON THE per cent. BaO, and we find that this is converted in a short time by the action of sulphuric acid into geddinosic acid, [ a ] , = +25", the barium salt of which yields 7.8 per cent. BaO, and finally into trigalactangeddic acid, [aln = +22", the barium salt of which gives 8.5 per cent.BaO, whilst, at the same time, a large quantity of arabinou and arabinose is formed. With the elimination of the sugars, we see that an acid is produced of a much lower molecular weight than the acid acted upon. If we calculate the molecular weight of each from the percentage of BaO in the barium salt, we have for tri- galactangeddic acid 153 loo - 153 = 1647, and for tetrarabinan- trigalactangeddic acid 153 loo - 153 = 2717; and if we subtract the former from the latter number, 2717 - 1647 = 1070, we get a figure which represents the molecular weight of the material from which the arabinon and arabinose were derived. Does this indicate the multiple of the unknown group arabinan, C10H160B, or of arabinon, CloH,,09 ? The molecular weight of the former is 264 and of the latter 282; then 8.5 5.33 1070 1070 ~ = 4.05 and 264 282 = 3.79; hence, it is more likely that it is four gi*oups of C,,H160B which are hydrolysed than 3-79 groups of CloHl,Og.If, therefore, we take Tg as a formula for trigalactangeddic acid, that of tetrarabinantrigalactan geddic acid will be 4CloH160~,Tg ; hence the term tetrarabirrzan. Had it been 4c,0&,0g that were removed, the formula of the higher acid would be 4CloH,,O,,Tg, and its barium salt 4CloH1,O9,Tg,Ba0 ; this requires 5.22 per cent. BaO, whilst 4CloH,60B,Tg,Ba0 requires 5-35 per cent. The mean result of many analyses of the carefully-prepared barium salt is 5-33 per cent. BaO, as given above, the variations being onlyin the second place of decimals ; hence, the tetrarabinan formula is well established. I have not been able t o determine whether BaO dis- places H,O in these salts or not; I believe, however, the salts are addition products ; they certainly are not of the nature of true salts, as I shall be able to show later on.Let us now see what relation the other acids hold to tetrarabinan- trigalactangeddic acid. We found that the barium salt of that acid containedabout 6 per cent. BaO, as the mean of the results of many con- cordant analyses of several different preparations, 3C10H,60,,Tg,Ba0, that is, tetrarabinantrigalactangeddic acid, rnilzus the arabinan group CloHl,O,, requires 5-90 per cent. BaO ; hence, the secoad natural gum acid is triarabiiiantrigalactangeddic acid. The barium salt of the third acid yielded 6% per cent. BaO; 2C,oH,,0B,Tg,Ba0, that is, the high acid, minus two CloHl,Os groups,GUMS OF THE ARABIN GROUP, 1049 requires 6.57 per cent.BaO ; hence, this acid is diarabinantrigalactan- geddic acid. The barium salt of the fourth acid yielded 7.3 per cent. of BaO; CloH1,O,Tg,BaO, that is, the high acid, minus three C10H16O8 groups, require 7.41 per cent, BaO ; hence, this is moraarabinantrigalactan- geddic acid. I showed above that an acid, which I called geddinosic acid, a marked resting stage in the action of sulphuric acid on the gum acids, yields a barium salt which was found to contain 7.84 per cent. BaO. This seems t o indicate an acid intermediate between monsrabinan- tsigalactanged.dic acid and trigalactangeddic acid ; CSHJ3ATg,BaO, requires 7.92 per cent. BaO. I t is not easy to understand how such a substance can exist and be a prodnct of the action of sulphuiic acid on the geddic acid, for it necessitates the representation of the breaking down not by the equations 4C,oH16O,Tg + 4HZO = Tg f 4C1oH,,O, Trigalactsn- Arabinon.geddic acid. and but by the equations 4CloH&6 + 4HZO = 8C,HlOO,, Arabinose. 4CioHi60srTg, + 4HzO = C,H,O+iTg + 3C10Hd-h -I- C5HioO5 Geddinosic acid. Aruhinon. Arabinose. and 3C,oH1,0, + 3HzO = 6C5HIoO5. There can, however, be no question as to the existence of the acid, for 1 have taken much trouble to purify it and determine its compo- sition. 1: record the fact as I found it, and leave to future investi- gation the task of explaining the matter. This is a convenient stage to summarise the results hitherto obtained.We have found that the four natural gum acids may be named and represented as follows :- Tetrarabinantrigalactangeddi2 acid. . . . . . . . Triarabinantrigalactangeddic acid . . . . . . . . . Diarabinantrigalactangedclic acid.. . . . . . . . . Monarabinantrigalactangeddic acid.. . . . . . . 4CloH1608,Tg. 3C&E&60e,Tg. 2C1J31,O,,Tg;. C10H16O8,Tg. Each of these, when acted upon by sulphuric acid, yields as a definite resting point Geddinosic acid. . . . . . . . . . . . . . . . . . . . . . . . . C,H,Oa,Tg, and finally Trigalactangeddic acid . . . . . . . . . . . , . . . . . . Tg.1050 O'SULLIVAN : RESEARCHES ON THE The next step in the investigation was to inquire into the constitu- We have seen that this acid is but slightly acted on when digested 25-30 minutes in a solution containing 2-3 per cent.sulphuric acid. On being oxidised with nitric acid, it yielded about 52 per cent. of mucic acid ; galactose, oxidised under similar conditions, yielded 85 per cent. of the same acid ; it is fair, therefore, Do conclude that trigalactangeddic acid cont>ains, at least, 85 : 100 :: 52 : 61 per cent. of a galactose-yielding constituent. Such being the case, it was t o be expected that the acid would yield galactose when broken down by digestion with sulphuric acid. A solution containing, in 100 c.c., 25 grams of trigalactangeddic acid and 3 grams of sulphuric acid, was heated at 98" in a water-bath, and the digestion continued for two hours. The liquid was then cooled, neutralised with baryta-water, and the barium sulphate separated by filtxation. When the gum acids are degraded to the ext,ent they are under under these circumstances, the bariuni sulphate is easily separated, and the solution filters clear.The filtrate was concen- trated in a vacuum t o n syrup, and alcohol (0.82) added as long as a precipitate was produced; this was freed from sugars by repeated dissolution in water and precipitation with alcohol, It contained the barium salt of the degraded gum acid; the sugars were in the alcoholic solutions, which, however, contained also some barium salt. This was eliminated by repeatedly evaporating to a syrup in a vacuum and taking up with strong alcohol. The final solution was freed from alcohol by distillation, and the optical activity of the sugars determined.Sp. gr. of solution 1.03214 ; optical activity in a 200-mm. tube ED = +12.50° ; hence on of the last-named acid. [ u J D = +74*9". This solution was not free from barium, and as the amount of sugar in 100 C.C. was inferred from the sp. gr., the specific rotatory power obtained for the sugars must be too low. It was again evapo- rated to a syrup, and treated with hot alcohol, which dissolved i t all except a small portion rich in barium. The clear alcoholic solution depasited, on fitanding, a small crop ( a ) of crgstds. This was sepa- rated, and the alcohol distilled off. The syrup thus obtained was dissolved in its own weight of dry methyl alcohol; this solution yielded a large crop ( b ) of crystals on standing. A further crop (c) was obtained from the mother liquids.Deterniinations of those three crops of crystals were made with the following results :- 2.5'70 grams n crop gave 100 c.c., sp. gr. 1.00988 ; this, in a 200-mm. tube, gave a rotation LXD = +4*0".GUMS OF THE ARABIN GROUP. 1051 3.413 grams b crop gave 100 c.c., sp. gr. 1.01314 ; this, in a 200-mm. 1.620 grams c crop gave 100 c.c., sp. gr. 1*00624 ; this, in a 200-mm. The crystals were anhydrous. Prom these figures we have, for the specific rotatory power of the tube, gave a rotation a D = +5*50". tube, gave a rotation u.D = +2*63". three crops :- a . [a]= = +77-8". b. [ a ] , = +80-6". c. [all) = +81*2". The optical activity of galactose at the temperature and degree of average concentration of these observations is [a], = +81.4", and, as the crops of crystals have all the same habitus as that sugar, and as they yield mucic acid on oxidation with nitric acid, they con- sist of galactose iii a moderately pure state.On recrystallisation from water, a product was obtained having the exact crystalline form and specific rotatory power OP galactose. Besides this sugar and a trace of barium salt, there was only a very small quantity of uncrystalli- sable syrup of low optical activit8y in the alcoholic solutions. Whether this syrup is tlie result of the action of the acid on the sugar, which is very probable, as I have shown that arabinose and, as we know, dextrose are acted upon in the same way, or on the gum acid, I must leave undecided f o r the present. On attempting to purify the gum acid or acids of the barium salt, as far as possible, by dialysis with hydrochloric acid, it was found that the greater portion of the gum acid diffused, only a small and somewhat coloured portion being left on the dialyser.This was too opalescent to admit of a n optical determination. It was divided into four fractions, and the barium salts prepared of the second (a) and fourth (8). These were aiialysed :- 0.746 gram ( a ) ga-re 0.114 gram BaS04. 0.522 ,, @) ,, 0.079 ,, Numbers corresponding to :- 0: = 10.03 per cent. BaO. 8 = 9-94 ,, 1 9 The fraction, therefore, left on the dialyser is a fairly pure homo- geneous snbstance. As, however, this was only a small portion of the acid produced by the degradation, a further quantity of tri- galactangeddic acid was acted upon, in the sams way as described1052 0'SULLIVA.N : RESEARCHES ON THE above.The sugar in the alcoholic: solution was found t o be galactose. The barium salt was divided into eight fractions: 1 being the most in- soluble. An optical activity, sp. gr., and barium oxide determination was made of each of these; 1, being a small and rather unsatis- factory fraction, was rejected :- 0.563 gram 2 gave 0.070 gram BaS04. 0-42G ,, 3 ,, 0.096 0.536 ,, 5 ,) 0.114 0.803 ,, 6 ,, 0.180 0.732 ), 7 ,, 0.110 0.698 ), 8 ,, 0.082 0.182 ,, 4 )) 0.039 2.214 grams 2 gave 100 C.C. solution, sp 200-mm. tube, deviated the ray a, = 1.706 grams 3 gave 100 C.C. solution, sp. 200-mm. tube, deviated the ray ED = 2.146 grams 5 gave 100 C.C. solution, sp. 200-mm. tube, deviated the ray ED = 7 ) 7 7 7 9 7 7 Y 7 9 9 gr.= 1.01048, which, in a + 1. *4", gr. = 1.00854, which, in a + 1.92". gr. = 1.01052, which, in a + 2.20". 3.212 grams 6 gave 100 C.C. solution, sp. gr. = 1.01658, which, in a 2.928 grams 7 gave 100 C.C. solution, sp. gr. = 1.01370, which, in a 2.794 grams 8 gave 100 C.C. solution, sp. gr. = 1.01270, which, in a 200-mm. tube, deviated the ray ED = +3.75". 200-mm. tube, deviated the ray CXD = +3*48". 200-mm. tube, deviated the ray aD = +3*15". From these numbers, we have the following factors :- Sp. rot. power for acid Per cent. BaO. free from BaO. D. 2.. .. .. .. 8.31 [.]I, = 4-34.5" 4.73 3. . . . . . . . 14.79 La]D = +65.5" 5-01 4.. .. .. .. 14.07 5.. . . . . . . 13.94 [aID = +60*0" 4-90 6.. .. .... 14.72 [a]D = +68.4" 5-16 7.. . . .. . . 9-87 [a]D = +65.9" +68 - - 8 ........ 7.71 [a]= = +61.1" 4.54 A further treatment of fraction 2 showed the high activity and low percentage of barium oxide to he due to a galactose-yielding body, probably gaZacton, the saccharon of galactose, contaminating a gun1 acid containing slightly over 10 per cent. BaO. Fractions 3,4,5, and 6 are clearly it mixture of two or more barium salts, contaminated prob- ably with sugar ; and fractions 7 and 8 are gum acid salts, evidently containing sugar. The latter were mixed, dissolved in water, and care-QUMS OF THE ARABIN GROUP. 1053 fully purified by repeated solution and reprecipitation with alcohol. The alcoholic solutions were all collected, concentrated t o a syrup, aud dissolved in a small quantity of methyl alcohol. On standing, the solution crystallised.The sugar thus obtained was found to have an optical activity and to yield muck acid on oxidation with nitric acid, so that it was pure galactose. The purified barium salt was divided into two fractions, one of which was analysed. It yielded 13.90 per cent. BaO, and had an optical activity for organic acid [XI= = +81.1", [&ID = 3-62.5". It is clear, therefore, that the barium salts in fractions 7 and 8 were of the same nature as those in the other fractions, and that it is difficult to separate these salts from sugar. The whole of the purified fractions were mixed and again submitted to careful treatment with alcohol. Some of the more soluble fractions were found to contain between 14 and 15 per cent. of BaO, and a more insoluble fraction was obtained, which, on analysis, yielded the following results :- 0.702 gram dry substance gave 0.163 gram BaS04.2.808 grams dry substance gave 100 C.C. solution, sp. gr. = 1.01410 ; and this, in a 200-mm. tube, deviated the ray = +2.95". These numbers lead to the factors 15.2.5 per cent. BaO ; [&ID = +69*0" for acid free from BaO. On referring to Part I of this paper (Trans., 1884, 45, 46), we find a barium salt described ; it was obtained under the same con- ditions from arabic acid as this salt was prepared from the gedda gum acid, containing 15.59 per cent. RaO, but haring little or GO optical activity. The acids neutralise practically the same quantity of barium oxide, but they are not identical, as is shown by the dif- ference in the optical activity.Do they contain the same proportion of carbon and hydrogen? The salt of the arabic series gave a per- centage agreeing with the formula C,,H4,02,,Ba0. The portion of the geddic salt of which an analysis is given above was mixed with all the fractions from the same source containing 14 to 15 per cent. BaO, and having an optical activity varying between [&ID = -t 62" and +66", and again purified by repeated treatment with alcohol. The salt thus obtained was burnt in the usual way in a current of oxygen.1054 O'SULLIVAN : RESEARCHES ON THE I. 0.3290 gram dry barium salt gave H,O = 0.147 gram, CO, = 11. 0.3732 gram dry barium salt gave HzO = 0.168 grain, COz = 0.417 gram: and BaCQ3 = 0.0665% gram. 0.474 gram, and RaCO, = 0,0752 gram. From these numbers we have :- Found.--J---7 Theory for I. 11. CyjI-I,,O,;BaO. C per cent.. . . , . 35.83 35.83 35.47 H ,, .. . . . 4.96 5.00 4.89 BaO per cent. . . 15.70 15.66 15.59 These numbers agree sufficiently closely to warrant the application of the arabiiiosic formula to the geddic derivative. The latter I pro- pose to call p-geddinosic acid. Further, in dealing with the arabic series, I showed (Zoc. cit.) that an acid having the composition C23H3s022 was obtained as the result of the digestion of arabic acid for three or four hours with sulpharic acid. An acid of this composition, no doubt, is also obtained by treating trigalactangeddic acid or the higher acids in the same way, I have not thought it necessary to isolate the acid, but the evidence I obta,ined of its existcrice is conclusive.One of the natural gum acids, in fact, triarabinanti-igalactangeddic acid, was digested in the usual way with sulphuric acid for three hours. The transformed solution was neutral ised with baryta-water, the barium sulphate filtered off , and the filtrate concentrated. The precipitate with alcohol was freed from sugar as much as possible by treatment with alcohol, and then divided into two fractions, a and b ; on analysis, these yielded the following results :- 0.835 gram of a, dried as usual, gave 0.21'7 gram BaS04. 1.312 ,, b ,, >? 0.338 > 9 3.340 ,, a ,, 9 7 100 C.C. solution, sp. gr. 1.01706, and this solution gave a, rotation OID = +3*80" in a, 200-mm. tube. a = 17.1" per cent. BaO ; [aID = +68*2" for free acid. b = 16.9" 29 9 , 7 9 22 Theory for C23H3,Q22,Ba0, 18-68 per cent.BaO. There can be no doubt that a salt of this composition could be ob- tained by carefully fractionating n or b, but when we take the facts already established into account, i t seemed to me unnecessar.y t o * There was no increase i n weight when the residue in the boat after burning the BaQ salts was treated with ammonium carbonate.GUMS OF' THE ARABISF OROUP. 1055 prepare the salt in the pure state. The a fraction is a mixture of barium P-geddinosate and the barium salt of the low acid in about equal parts ; taking the percentage of barium as the indicator, the optical activity of the low acid, which I propose to call geddic acid, would be = +71" about. The corresponding arabic acid has little o r no optical activity; the acids from the two sources are isomeric, they differ in optical activity.We may now turn to the acid obt'ained as the first marked resting stage of the action of sulphuric acid on the natural gum acids, namely, trigalactangeddic acid, represented by the symbol Tg, arid described above, and inquire what relation this acid holds to the final one, the existence of which I have just indicated, and which I have called geddic acid. As I showed that barium trigalactrtngeddate contains 8.4-8*5 per cent. of barium ; that by the action of sulphnric acid it yields geddic acid, C,,H,,O,,, and that no other product is formed cxcept galactose, it is clear that the higher acid must bc some compound of galactose, or a galactose-yielding substance, with ,geddic acid, Taking C,,H,,O,, to represent a hypothetical galactan, 3C,,H,,0,,,C,3H,,0,2,Ba0, that is, TgBnO, would represent the con- stitution of the salt.This formula requires 8.54 per cent. BaO, against the 8-4-4.5 per cent. found ; hence the term trigaZactangeddic acid. If this is the real relationship existing between the two acids, an elementary analysis of trigalactangeddic acid, or of its barium salts, should yield results confirmatory of the hypothesis. Combustions were made of several preparations of the barium salts with the following results :- 0.2820 gram 1 in dry state gave H,b = 0.141 gram, CO, = 0.415 0.3551 gram I1 in dry state gave H,O = 0.172 gram, CO, = 0.528 0.2126 gram 111 in dry state gave H,O = 0.104 gram, GO, = 0.312 0.2863 gram IV in dry state gave H,O = 0.144 gram, CO, = 0.494 0.3585 gram V in dry state gave H,O = 0.174 gmm, CO, = 0.528 gram, and BaCQ, = 0.0312 gram.gram, and BaCOs = 0.0391 gram. gram. and BaCO, = 0.0235 gram. gram, and BaC0, = 0.0317 gram. gram, and BaCO, = 0.0390. These results cnlculated for the free acid are :-- Theory for I. 11. 111. IV. V. ~ c ~ ~ ~ I ~ , o ~ ~ ~ . o , ~ T ~ C.. , . 44-64 45-02 44.50 44-92 44.62 43.2 H . . . 6.07 5-85 5-95 6.11 5.89 5.981056 O’SULLIVAK : RESEARCHES OH THE The BaO in the salt being- I. 11. 111. IV. v. BaO . 8.59 8.56 8.59 8.60 8.45 The carbon found is much higher tlhan that required by this theory; tbe hydrogen affords no marked evidence, but taking the general results of hydrogen determinations into consideration, the numbers obtained are too low, on the whole, for the theory.I t is clear, therefore, that the relationship between geddic acid and trigalac- tangeddic acid is not that mentioned above. I f , however, we suppose that each C,,H2,010 group hydrolysed occupied the place of a water molecule, H20, in C23H38022 we have a compound 3C,2H,,0,,,C,,H32018, w hie h re quire P C . . .......... 44-70 H.. .......... 5.81 numbers agreeing perfectly with those found. 3C12H2,01,,C23H32019,Ba0 requires 8-80 per cent. BaO, a number agreeing sufficiently closely with those Pound, especially when it is considered that the acid can only with difficulty be freed fi-om magnesium and potassium cxides, which take the place of a portion of the barium oxide. It is, therefore, clear that the formula of trigalactangeddic acid is 3C12H200107 C2sH3,01,, and that the equation represents the fin21 action of sulphnric acid upon it.I may here point out that in dealing with the breaking down of arabic acid, I showed that, after a certain stage of the degradation, the breaking down was accompanied with the fixation of more water t’han was required to convert a C,HinO, or a C,,H2,010 compound into a C,H,,O, one, in exact.ly the same way as is here indicated. Between trignlactangeddic acid and geddic acid, I have been able to obtain evidence of the existence of only two acids, namely, the acid left upon the dialyser, the barium salt of which cont’ained 10.03 and 9.94 per cent. of barium oxide, and the acid of the fractions con- taining between 14 and 15 per cent. BaO. Two preparations of the barium salt of the former were analysed with the following results :- 0.3512 gram I dry salt gave H20 = 0.183 gram, CO, = 0.500 gram, 0.3931 gram I1 dry salt gave Ef20 = 0.202 gram, C02 = 0.561 and RaC03 + 0.0469 gram.gram, and BaC03 = 0.520 p m .GUMS OF THE ARABIN GROUP. 1057 Found. 7--- -7 Theory for I. 11. 2C12H~0010,C,3H340,0,Ba0. C ...... 39.60 39-75 39-40 H. ...... 5.79 5-71 5-16 RaO .... 10.37 10.28 10.61 These numbers agree sufficiently well t o support the theory ; t,he agreement would, of course, be closer if the carbon and hydrogen were calculated on the acid free from ash. This acid I call di- qnlwtangeddic acid. The second acid, the barium salt of which contains between 13 and 14 per cent. BaO, I call 177onogalactangeddic acid, because i-equires 13-66 per cent.BaO. The optical activity of this acid would be somewhere about [a]= = +58-60". Having thus far secureIy established the relationship existing between the final acid, geddic acid, and trigalactangeddic acid, we may return to inquire whether or not the relationship of the natural gum acids to one another, and to trigalactangeddic acid, is exactly as was indicated by the composition of the barium sdts. This compo- sition indicated that the natural gum acids were compounds of tri- galactangeddic acid with hypothetical .ambinan, C,,H,,O,, so that, substituting ~3C,,HZ,O1,,Cz,H,,O,, for Tg, the formula of tetrarabinan- trigalactangeddic acid would be 4C~~H~~0~,3Cl,H~o0,0,C,,H,,0,,. Is this so ? Combustions were made of four preparations of this acid, namely, two calcium salts and two free acids, with the following results :- I.0.2481 gram dry CaO salt gave H,O = 0.134 gram, CO, = 11. 0-3315 gram dry CaO salt gave H20 = 0.174 gram, CO, = 111. 0.2641 gram dry acid gave H,O = 0.143 gram, CO, = 0.437 IV. 0-3'716 gram dry acid gave H,O = 0.202 gram, C02 = 0.616 Ci,H~oOIoCz3H'3602iBaO 0*39$* gram, CaCO, = 0.00947 gram. 0-536* gram, CaCO, = 0*0125.-if. gram. gram. Found. ---A-- 7 Theory for 1. 11. 111. IT. 4Cl0H,,0,3C,2H2o0,,,C23H3,0,,. C .... 44-71 45-05 45.13 45.21 45.00 H . . , . 6-13 5.96 6.52 6.04 5.91 CaO .. 2.12 2-11 - - 2-08 ~ C ~ O H ~ , ~ , ~ C ~ ~ ~ ~ O O ~ , , ~ * This number includes the C 0 2 weighed and the C 0 2 as Ca170, in boat, tlie t After treatment of ash with ammonium carbonate.boat being weighed before treatment of ash w i t h ammonium carbonate.1058 O'SULLIVAN : RESEARCHES ON THE These numbers prove conclusively that the relationship indicated by the barium salts is that existing betweeu trigalactangeddic acid and tet,rarabinantrigalactangeddic acid, that unlike the group yielding galactose, the group from which the arabinon and arabinose are derived is hydrated, without hydration of thc derived acid taking place at the same time. The facility with which the arabinan is removed, and the difficulty with which the galactan-containing acids are broken down, show clearly, too, that the loorid of union in both cases cannot be the same. The bai-ium salt of tetrarabinantrigalactangeddic acid yielded 5.33-5-42 per cent. BaO ; the formula 4Ci,H,,0,,3Ci,H,,Oi,,C,,H,,O,g,BnO requires 5.47 per cent.BaO. 5.9-6.07 per cent. BaO ; the formula The barium salt of r:riarabinantl.igalactangeddic acid yielded 3CioH,,0,,3Ci,~,,O~~,c~~H~zoi~)BaO requires 6-04! per cent. BaO. 6.6-6-7 per cent. BsO ; the formlnla The barium salt of diarabinantrigalactangeddic acid yielded 2c IOR1608,3 C1ZH20010 > C?3H22019>Ba0 requires 6.75 per cent. BaO. The barium salt of monarabinantrigalantangeddic acid yielded 7.3-7 6 per cent. BaO ; the formula [ ~,,H,~08,3~~~H,,01,,~~~requires 7.65 per cent. RnO. I n all these cases, the theory agrees well with the numbers found ; taking this into account, as well as the general facts demonstrated in this paper, it is clearly established that these formuh, attributed to the natural gum acids and to the products o f their degradation, are highly probable.Tlie gum is a mixture of the calcium, magnesium, and potassium salts of these acids in various proportions, together with a small quantity of prote'id as mentioned above. Tlie evidence afforded by the constitution of this gum, and of the acids it contains, seems t o me to point distinctly to these bodies as materials-matter being biiilt up, and to be employed in building LIP, rather than as dk'bris-matter produced by the breaking down of sub- stances of greater complexity. I f , too, they were JQbris, we can have no doubt that the gum woulcl contain other products of the degrada- tion. This we find is not the case, for, although there is undoubtedly a trace of sugar present, there is not anything like sufficient to acconnt, for example, f o r the degratiation of tctrarabinantrigalactan- geddic acid to monarabinantrigalnc tangeddic acid.It is not possible t o say whether the nitrogenous matter is de'bris o r material, for itGUMS OF THE ARABIN GROUP. 1059 has been but slightly examined, in consequence of its being'present in very small quantity, and of tlie difficulty of arriving at any simple factor of it which would serve as a criterioii of its homogeneity or purity. This work, so far, is complete in itself, for it gives us a clear iusight into the consbitlition of a dextrorotatory gum; but I felt that the paper would not be complete unless I was able to show that, all the dextrorotatory gums were like the one described, or held some well-defined relationship to it.The investigation was commenced in 1883, and in 1887, before I was able to quite satisfy myself upon many points described above, the supply of the A sample of gum becamme exhausted. I applied to the firm from whom the first sample was obtained, but a few experiments with the sample furnished showed me that I had an altogether different gum in my hands. It became necessary, there- fore, to submit the fresh sample to pretty well as full an investigation as the first. This I undertook the more readily, as I hoped t h e results would confirm and complete those already described. I had, therefore, to seek a fresh supply. Sample of Gedda Gurn B. Physically there was no apparent difference between this sample and the A sample. The colour was the same in both, and there was no material difference in the shape or size of the pieces.If anyt!hing, the darker pieces of the B sample were more contaminated with bark and sand than those of the A sample. Neglecting these pieces, the sample was easily divisible into B a, light glassy pieces, B b, dark- ruby pieces, and B c , deep-reddish ones. An optical activity deter- nlination of each of these divisions was made with the following results; the sp. gr. of the solution was employed as tlhe indicator of the amount of substance in solution. R a. [z]D = + 83.4" B b. [ a ] D = +80*7" B C . [ a ] D = +68*9". These numbers show that the second sample of gum is not the Ash determinations were also made :- 1.091 grams B a gave 0.022 gram ash after treatment with 5.678 grams B b gave 0.107 gram as11 after treatment with 5.466 grams Bc gave 0.093 gram ash after treatment with same as the first, and that, like it, it is not homogeneous.(NfI*) 2co3. (NH*)ZCO3. (NHa),C 0 3 . VOL. LlX. 4 D1OtiO O'SULLIVAN : RESEARCHES OX THE Hence B a = 2.0 per cent. ash. B b = 1-89 7 , B c = 1.70 3, On comparing these percentages with those obtained for the A sample, we find there is much less ash i n the B sample. The Teason of this will be clear later on. The constitution of the ash i n both samples was, however, pra~t~ically the same, namely, calcium (*arbonate in greater part, with a, small quantity of potassium and magnesium carbonates, arid a trace of silicic and phnsplioric acids. It was no use to attempt to fractionate the gum mechanically, because 1 found that even pieces of exactly the same colour con- tained more than one gum acid; hence, as in the case of the A sample, I had recourse to the precipitation method.I n dealing with it, however, a difficulty presented itself; a considerable quantity of the gum acid was soluble in strong alcohol. When a strong solution, one of water to one of gum, was acidified with hydrochloric acid, and the least soluble fraction thrown out by the addition of alcohol, of sp. gr. 0.85, further addition of this alcohol to the clear solution did not precipitate nearly the whole of the gum acid ; this was only effected on neutralisation with milk of lime ; the calcium salt of the most soluble acid was then precipitated. Pro- 'ceeding in this way, the gum was divided into four frachions : a, the least soluble, was thrown out first, b was precipitated by a further addition of alcohol to the clear solution, c was next thrown out,, and d, the most soluble fraction, was precipitated only as lime salt O n neutralising with milk of lime.Fractions a and b contained much nitrogenous matter ; they were again mixed and divided into fractions a,, a?, and a3, the first being the least soluble. After these substances had been dried, they were insoluble in water, and had to be dissolved in dilute soda to admit of an optical observation. They were all strongly acid to litmus, hydro- chloric acid being proved to be absent. I n alkaline solution the optical activity was found to be a,. [a]D = 4-38' ~ 2 . [ a ] D = +38" a3.[a]= = +46". Neither fraction was free from nitrogenous matter. On further treating a3, a soluble fi-action was obtained, free from nitrogen, the optical activity of which was [ a ] D = +80--83", the more insoluble portion having an optical activity not far removed from that of a, and a2. I must leave t o a future occasion a descrip-GUMS OF THE ARABIN GROUP. 106 1 t,ion of my attenipts to purify these substances, and an account of the nitrogenous matter they contain. I may, however, state here, thnt the nitrogenous matter is of the nature of a proteyd, and that it gives the following reactions :- I. It is diastatic, that is, it dissolves starch like malt diastase.* It does not invert sucrose. 11. When precipitated and dried, it does not dissolve in water to a clear solntion, yet the greater portion of the substance is evidently soluble.111. It dissolves easily in soda or potash to a perfectly clear solu- tion, with the production of a slight brown colour, if the alkali is in excess. On adding acid t o the strong solution, no cloudi- ness is produced, but the colour disappears. IV. No precipitate is produced by boiling the acid or alkaline solutions. V. An alkaline solution yields a beautiful purple colour on tho addition of a drop o r two of Fehling's solution ; the colour is much intensified on the application of heat. There is no reduction. except, on long boiling, and then it is only slight, the solution still retaining its purple or rose colour. VI. When boiled with strong nitric acid, it yields a dark-brown solution; potash discharges the colour to some extent, and produces a flocculent precipitate.VII. When boiled with nitric acid not sufEciently strong to give rise to nitrous fumes, a yellowish-red solution is obtained, which on cooling deposits a yellowish, flocculent precipitate. VIIL With Millon's reagent i t yields a rose-pink coloration, which becomes yellow on adding an excess of the reagent. IX. I n strong acetic acid solution, no precipitate is produced on the addition of potassium ferrocyanide. We have again to turn to the portions of the gum which were Fraction c was found to have an found to be free from nitrogen. a p t ical activity [.ID = +goo, and d [ a ] D = +102". Proceeding after the manner described in dealing with the A sample of gum, I found it possible to divide the portion of this gum free from nitrogen into four fractions, viz.:- A. [a]D = +107-110", B. [%ID = +99-101", C. [ a ] D = +89-91", D. [a]= = +80-83". * This reaction was observed for the whole gum. 4 ~ 21062 O'SULLIVAN : RESEARCHES ON THE A is the most, and D the least, soluble fraction. The B fraction constituted the greater portion of the gum free from nitrogen ; A and D were small fractions. As mentioned above, the high solubility of these gum acids in alcohol was a great obstacle to their separation. The aqueous solu- tion had to be made concentrated, and, when alcohol was added, it not nnfrequently happened that a portion of a higher rotating fraction was precipitated, and a portion of a lower rotating one held in soh- kion ; it was t'hns exceedingly difficult t o effect a separation.Feeling, however, that this was chiefly due to too rapid precipitation, and t o the concentration of the solution, I found that by guarding against these sources of complication, I was able to so arrange matters as to ensure t)hat the most soluble fraction was the most optically active, and did not contain any substance of less optical activity than the most soluble portion of the fraction next in solubility. There was very great difficulty in separating these acids from lime ; this necessi- tated taking the percentage of lime in the neutral lime salts as an! indication of the neutralising power. A fraction.-The most soluble fraction was dialysed until as free f ~ o m ash as possible ; it was then divided into three parts : Al, free acid ; A?, converted into lime salt; and, AS, into baryta salt.These were analysed, with the following results :- 2.773 grams free acid gave 100 C.C. solution, sp. gr. 1.01040, and this solution gave a rotation aD = +.5.97" in a 200-mm. tube. 0.765 gram dry calcium salt gave 0.015 gram CnC03. 3.060 grams dry calcium salt gave 100 c.c., sp. gr. 1.01160, an4 0.658 gram dry barium salt gave 0.028 gram BaS04. 2.634 grams dry barium salt gave 100 C.C. solution, sp. gr. 1.1040, this solution gave a rotation tcD = +6*6" in a 200-mm. tube. and this solution gave a rotation = +5*5". From these numbers we have the following factors :- Free acid .. . . . . . . . . . . " . . Free acid in calcium salt.. CnO in calcium salt = 1.1 per cent. BaO in barium salt = 2.79 pel.cent. [RID = +1@7.6", [a]D = +109*0", ,, in barium salt . . [a]* = +107.4". B fraction.-T his being the largest portion, a thorough fractiona- tion was made of it. All the preparations with an optical activity [alD = + 100 (about) were mixed, a small, less soluble portion taken out a t one end, a very sniail, more soluble one a t the other, and the great bulk of t'he fraction, after being dialysed and repeatedly pre-GUNS OF THE ARABIN GROUP. 1063 cipitated, was divided into three fractions: Bl, Bz, and B,, the first bring the least soluble. B,. The free acid and barium and calcium salts of this fraction were examined. 3.596 grams free acid, dried as usual, gave 100 C.C. solution, sp. gr. 1.01318, and this solution, in a 200-ram. tube, gave a rotation a, = +7*03".0.899 gram dry acid gave 0.0015 gram ash.* 0.649 gram dry barium salt gave 0.0394 gram BaS04. 2.596 ,, 7, ,, 100 C.C. solution, sp. gr. 1.01034, 3.274 ,, 9 7 ,, 100 C.C. solution, sp. gr. 1.01248, arid this solution, i n a 200-mm. tube, gave a rotation aD = +4*9gn. 0.818 gram dry calcium salt gave 0.020 gram CaCO,. arid this solution, in a 200-mm. tube, gave a rot t' ion + 6.63". = &. This fraction was examined as calcium salt. 0.732 gram dry calcium salt gave 0.018 gram CaC03. 2.928 grams 7, ,, 100 O.C. solution, sp. gr. 1.01132, and this solution, in a 200-mm. tube, gave a rotation aD = +5-82". B,. This was also examined as calcium salt. 0.717 gram dry calcium salt gave 0.018 gram CaC03. 2.870 grams 7 7 ,, 100 C.C. solution, sp.gr. 1.01110, and this solution gave a rotation = + 5-70" in a 200-mm. tube. From these data we have the following factors :- Free acid, Barium salt, ,, [a]D$ = +99*3"; D = 3.98; Ba0 = 3-99s ,, Calcium salt, ,, [.]Dl = +98*9' ; D = 3.81 ; CaO = 1.37 ,, Clalcium salt, B2. [a]Dz = +100*8" ; D = 3-87 ; CaO = 1.38 ,, Calcium salt, Bs. [.ID,$ = +100.8' ; D = 3.87 ; CaO = 1-41 ,, B1. [a]~) = + 97.7" ; D = 3.67 ; ash = 0'17per cent.? The specific rotatory power and density of the free acid are low, ant1 I may say that this is a general rule when these acids are examined iii the free state ; I have 1-eason to believe that this is due to the reten- tion of alcohol by the free acid, probably as an ether. The true divisor, D, of the free acid I estimate at, 3.75, and if the optical activity ob- served be corrected by this factor, the specific rotatory power of the acid will be [a]D = +loo".All the factors agree so well, that there can be no doubt that the B fraction is a homogeneous substance. * Proved to be calcium carbonate. f- Calcium carbonate. $ The specific rotatory power is calculated for the organic portion of the salt. 4 Corrected for ash (calcium carbonate).1064 O'SULLIVAS : RESEARCHES ON THE C fraction.-All the pieces with a specific rotatory power [&ID = +go" (about) were mixed, small fractions taken out above and below as in the preceding case, and the large, middle portion, after dialysis and repeated precipitations, was divided into three fractions, C,, C p , and C3, the latter being the most soluble. The calcium salt of each of these was examined.GI. 0.826 gram dry calcium salt gave 0.025 gram CnCO,. 3.306 grams 9 ,, 100 C.C. solution, sp. gr. 1.01278, and this solution, in a 200-mm. tnbe, gave a rotation aD = + 5.85'. C,. 0.766 gram dry calcium salt' gave 0.023 gram CaC0,. 3.066 grams 9 , ,, 100 C.C. solution, sp. gr. 1.01186, and this solution, in a 200-mm. tube, gave a rotation aD = +5*45". C3. 0.794 gram dry calcium salt gave 0.022 gram CaC03. 3.178 grams ?, ), 109 C.C. solution, sp. gr. 1 *01230, and this solution, in a 200-mm. tube, gave a rotation aD = +5*72". These data lead t o the following factors :- cl. [a]D = +90.0" ;* calcium salt, D = 3.86 ; CaO = 1.70 per cent. c:z. [%ID = +90'5"; ,, D = 3.87; CaO = 1.70 ,, c,. [a], = +91*1'; ), I> = 3.87; CaO = 1.60 ,, These numbers prove conclusively that the C fraction is a homo- geneous substance. Other fractionations of other preparations of this fraction .were made with results agreeing well wit,h those.I may give another set of determinations. A free acid, [aID = 90°, was divided into two fractions, Ca and C5, the former being the least soluble, and of these calcium salts were prepared and analysed. C4. 0.778 gram dry calcium salt gave 0.024 gram CaCO,. 3.113 grams )) 9 7 100 C.C. solution, sp. gr. 1.01214, = + 5.55" in a 200-mm. and this solution gave a rotation tube. C5, 0*575 gram dry calcium salt, gave 0.017 gram CaC03. 2.300 grams ,, ,, 100 C.C. solution, sp.gr. 1.0090, and this solution gave a rotation a, = +4*15' in a, 200-mm. tube. From these data the following factors are calculated :- C4.[a]D* = +90.7 ; calcium salt, D = 3.90 ; CaO = 1-73 per cent. C5. [a]= = $91.8; 93 D = 3-91; CaO = 1.66 ,, * The specific rotatory power is calculated for the organic portion of she salt.GUMS OF THE ARABIN GROUP. 1065 The agreement of these numbers amongst themselves and with those obxained in the previous experiments is perfectly satisfactory. These two fractions were again mixed, dissolved in a little water and hydrochloric acid, and then alcohol added ; the precipitate was redissolved and reprecipitated until it was free from hydrochloric acid. A barium salt of a portion of this was prepared, and the free acid itself was dried ; both preparations were examined. 12.783 grams dry free acid contaiaing only traces of ash, gave 100 C.C.solution, which, in a 200-mm. tube, gave a rotation aD = +5.0". 0.733 gram dry barium salt gave 0.047 gram BaS04. 2.934 grams ,? ,, 100 C.C. solution, sp. gr. 1.01186, and this solution, in a 200 mm.-tube, qave a rotation aD = + 5.05". From these data- Free acid, [a]D = +89*8"; Barium salt, [alD = +89*9" ;* D = 4.04; BaO = 4.21 per cent. JI fraction was, as I have said, small, and, taking into considera- tion the fact that its specific rotatory power holds pretty well the same relation to that of the C fraction as the latter holds to B and B to A, it seemed to me fair t o conclude that this acid is one of a series, and if I can establish a relationship between A and €3 and R and C, it will not be unreasoiiable to conclude that the same relationship holds between C and I>.The optical activity of this fraction may be hken as [ a ] D = +80-82". On comparing the factors of the members of this series with those of the members of the A sample of gum, we find that, ap- parently, these acids are higher members of the A sample series, the acid with highest optical activity [a]= = +59", and with a calcium salt containing 2.1 per cent. lime, tetrarabinantrigalactan- geddic, is separated seemingly by an absent acid [aID = +70" (about),from the lowest member, [a]D = +80--83", of the l3 sample group ; in other words, a pentarabinnntrigalactangeddic acid would appear to be absent, and a hexarsbinant1,igalactaiigeddic appears as the first and lowest member of the B sample series ; the next highest member, [a]D = +goo, being heptarabinantrigalactangeddic acid, and so on for the [%ID = 4-100" and the [a]D = +110" acids.This would seem to be borne out, too, by the decreasing amount of lime in the calcium salm, and, also, by the increasing solubility in alcohol in the higher acids. If this is the relationship that exists, the first resting stage in the action of sulphuric acid on the wids of the B series * The specific rotatory power is calculated for the organic portion of the salt.1066 O'SULLIVAN : RESEARCHES ON THE should be the same acid as that which was found in dealing with the A series? namely, trigalact angeddic acid. A portion of the B fraction above described was acted upon with sulphuric acid f o r 20 minutes, in the usual way, and the resultiiig solution treated with alcohol after the manner already described. The portion insoluble in alcohol was purified from sugar by further treat- ment with alcohol, then dialysed with hydrochloric acid until free from ash, and further puritied by removing a slight, most soluble, and least soluble portion with alcohol.The middle, and by far t h e greatest portion, was then divided into two fractions, a: and ,!3 ; these were converted into barium salts. Ba. 0.669 gram barium salt, dried as usual, gave 0.0'71 gram Bas 04. 2.678 grams gave 100 C.C. solution, sp. gr. 1.01182, and thih solution, in a 200-mm. tube, gave a rotation aD = +1*0". Bp. 0.770 gram barium salt, dried as usual, gave 0.084 gram Bas 0,. 3.030 grams gave 100 C.C. solution, sp. gr. 1.01370, and thi:, solution, in a 200-nim.tube, gave a rotatioii [ a ] D = +l.l". Hence, BZ. [a]D = +20*1", R/3. [a]D = +19*2". Ba: = 6.97 per cent. BaO; D = 4.41, BP = 7-17 ,, BaO; D = 4-45. The free acid, [a], = f19.2, was again acted upon for 10 minutes with sulphuric acid, and the resulting solution treated as usual with :dcohol. Very little action had taken place, as the alcohol contained only a very small quantity of sugar. The precipitate was washed and purified, and a barium salt prepared of it. 0.658 gram dry barium salt gave 0.070 gram BaS04. 2.632 grams 37 ,, 100 C.C. solution, sp. gr. 1.01186, and this solution, in a 200-mm. tube, gnve a rotation aD = + 0.95". Hence 6.99 per cent. BaO ; [aID = +19.4" ; I) = 4.50. Fraction C, [a]= = +go", was acted upon in the same way witli exactly the same results.Many other transformations were made without any material alteration in the character of the product. In. all cases, the acid obtained had an optical activity [cz]D = +20° (about), and gave a barium salt which yields 7.2-7-4 per cent. BaO. It is therefore certain that this acid is a resting stage in the action of sulphuric acid on the acids of the B fraction of the B sample of gedda gum. Before proceeding further, we may emmine the alcoholic solution,GUMS OF THE ARABIN GROUP. 1067 and determine if arabinon and arabinose are the only other products of the degradation. It was neutralised with baryta-water, the barium sulphate removed by filtration, and the filtrate evaporated to a syrup in ib vacuum. This was treated with strong alcohol, which, in all cases, dissolved it all except a small portion of the barium salt of a gum acid, a portion of which was also thrown down with the bariuni sulphate on neutralising the alcoholic solution ; this was separated by treating the sulphate with boiling water and filtering; it will be returned to later on.The alcoholic solution was again evaporated t o a syrup which, on standing a few days, yielded a large crop of crystals (a), a second ( b ) , third (c), and fourth ( d ) crop were obtained, and, finally, a small portion of mother liquid (e) was left. An optical activity determination was made of all these crops, (a.) [a]n = +108-9" c = 3.122, (c.) [a]u = +102.5O c = 3.636, ( h . ) [a]= = +104*6" c = 4.062, (d.) [ a ] , = +102*0" c = 9.010, (e.14 [a]D = +100*0". (u) contained arabinon, ( b ) is pure arabinose, so are ( c ) and ( d ) practically, and (e) is contaminated with the products of the action of siilphuric acid on arabinose ; it also coiitaim a trace of galactose, yielding, on t'reatment with nitric: acid, a trace of mucic acid ; a trace of barium salt is also present ; when we take that into acconnt, and consider how the specific rotatory power is arrived at, we can easily see that the amount of matter of any kind, other than arabinose, is very small.The galactose and the more soluble gum acid, the barium salt of which is mentioned above as being precipitated with the barium sulphate and as being insoluble in strong alcohol on treat- ment of the syrup, are products of the further degradation of a small portion of the resting-stage acid.If we now compare the numbers obtained for the optical activity of this acid and f o r the percentage of lnaryta in the barium salt wit11 the same numbers obtained for either geddinosic acid o r trigalactangeddic acid, we shall find thai, although the sp. rot. power does not differ very much from t,hat of the latter, being 011 the whole somewhat less, the percentage of baryta in the barium salt is materially less than that in barium trigalactangeddate, or in barium geddinosate, in other words, the resting-stnge acid of the acids of the B sample is of greater molecular weight than that of the A sample of acids. The question which then presented itself is, to what is this * In this case the solid matter in solution was inferred from the sp.gr. ; in all the other cases the crystals were dried at 100" and weighed ; they contained no water of crystallisation.1068 O'SULLIVAN : RES EARCIIES ox THE due? Is it that the resting stage of the I3 sample contains morc galnctan groups, or is it that the substances are altogether diffe- rently constituted ? If the former supposition be correct, the resting- stage acid of the B sample when acted upon by sulphuric acid should yield final acids identical with those obtained from the A sample. A portion of the resting-stage acid, [a]D = +go", and 7-52 per cent. baryta in the barium salt, was acted upon at the boiling temperature with 2 per cent. sulphuric acid solution for If hours. The resulting liquid was cooled, neutralised with baryta-water, filtered, and con- centrated in a vacuum.The syrup was repeatedly extracted with alcohol, the alcoholic solutions concentraA,ed, and the residue allowed to crystallise. A determination of the optical activity of the crystals gave On treatment with nitric acid, they yielded mucic acid. From these facts, and its solubility, crystalline habit, &c., the sugar is undoubt- edly galactose. The solid matter in the mother liquid gave an optical activity but ii; contained a little barium salt, hence it is clear, the alcoholic solution contained no other sugar except galactose, the colonred sub- stances produced by the action of the acid on galactose, and a little barium salt of a gum acid, to be described later on. The transformed products insoluble in alcohol, that is, the gum acids, were completely freed from sulphuric acid and sugar by treat- ment with alcohol, and then divided into two fractions, R, and R2, and barium salts of them prepared.[RID = +81.0". [&ID = +71*1", 0.653 gram R,-salt, in dry state, gave 0.137 gram BaS04. 2.614 ,, R1-salt ,, ,, 100 C.C. solution, sp. gr. 1.01316, and this, in a 200-mm. tube, gave a rotation aD = 2.800 grams R,-salt gave 100 C.C. solution, sp. gr. 1.01412, and this, 0.700 ,, R,-sttlt ,, 9 7 0.146 7 7 + 2.72". in zt 200-mm. tube, gave a rotation a, = +2%5"* Barium salt of R, = 13.8 per cent. BaO ; D = 5.04, Free acid, R,. [RID = +60.2", 7 7 R, = 13.7 ,, ,, ; D = 5.04. ,, R,. [ a ] , = +58*8". There is no doubt we have here a fairly pure substance; it is, in fact, monogalac tangeddic azid-compare optical activity and percent- age of baryta in the barium salt.The more soluble portion of the gum acids obtained in the aboveGUMS O F THE ARABIN GROUP. 1069 (1egrad:ttions of the resting-stage acid was purified as usual and divided into two parts, G1 and Gz, and of these baryta salts were pre- pared. 0.255 gram Gl salt, dry, gave 0.051 grani BaC03. 0.771 ,, G, 7 7 0.173 ,, BaS04. 5.436 granis GI 7 9 100 C.C. solution, which, in a 200-mm. tube, gave a rotation [a]= = +6.2So. tube, gave a rotation aD = t 4.25". 3.856 grams G, salt, dry, gave 100 C.C. solution, which, in a 20O-rr1m. Barium salt, Gl = 15.5 per cent. BaO, Free acid, Gl. [cx]D = +67.6", ,, Gz = 14.7 ), 7 9 7 ) G,. C a ] D = +64*7O. 02817 gram G1 dry free acid gave H,O 0,149 gram, CO, 0.440 gram.B-Geddinosic acid. Theory for Found. 7 c29H48027* C ........ 42.60 42.48 42-54 48-03 H . . ...... 5.88 5-89 5.93 5.77 BaO.. .... 15.5 15.7 15.7 15.6 Cz9H4,0,;Ba0. A comparison of these numbers with those obtained f o r P-geddinosic acid shows clearly that the acids are identical ; hence p-geddinosic acid is the product of the degradation of the gum acids of both gums. Further, when this acid was digested for an hour longer with sulphuric acid, an acid was obtained containing increased percentage of BaO, and I have no doubt, if the digestion be continued sufficiently long, an acid can be obtained the optical activity of which is [ a ] ~ = f71", and the barium salt of which contains 18-76 per cent. BaO, hence geddic acid, CZ3H3,O,. Let us now inquire what relationship does the resting-stage acid hold to this one ; C23H3~022 + 4cd&ooio - 4HZO = ~ C ~ & ~ O O ~ I ? , C Z ~ H ~ O O ~ ~ represents tetmgalactnngeddic acid.yielded the following results :- An elementary analysis of the barium salt of the resting-stage acid 0.3125 gram dry barium salt of purified resting-stage acid gave HzO 0.157 gram, CO, 0.471 gram, BaCO, 0.0299 gram,SO70 O'SULLIVAN : HESEARCHES ON Y HE Calculated for Found. 4C12H20010,C23H3,01~,Ba0. C .............. 41.72 41.7 I3 .............. 5.58 5.4 BaO ............ 7.43 7.49 Hence, the resting-stage acid for the gum acids of the B sample of gum is tetragalzctangeddic acid, in the place of trigalactangeddic acicl for the A sample. I have shown that arabinon and arabinose and tetragalactaiigeddic acid were the only products of the action of sulphuric acid on the original gum acid.Let us see what is the composition of these acids. The molecular weight of the acid [aID = +go", with calcium salt containing 1.7 per cent. CaO, is 5600 -+ 1.7 - 56 = 3238, aiicl the molecular weight of .tetragalactangeddic acid is 1890 ; hence the molecular weight, of the matter eliminated is 3238 - 1890 = 1348. The molecular weight of arabinan is 264; hence 1348 4 264 = 5.1, say 5, the number of molecules of arabinan hydrated. The acid under consideration is, therefore, pentarabinantetragalactangeddic acid, 5C10H1,0*,4C,~H~0010, Ca3H30O18. Several combustions of the free acid and of the calcium salt were I. 0,2987 gram dry, free acid, gave H,O = 0.167 gram, CO, = 11.0.3568 gram dry, free acid, gave H20 = 0.200 gram, CO, = 111. 0.2640 gram dry, free acid, gave H20 = 0.207 gram, CO, = IV. 0.5628 gram dry calcium salt gave H20 = 0.194 gram, CO, = V. 0-2685 gram dry calcium salt gave H,O = 0.145 gram, CO, = TI. 0.2946 gram dry ca,lcium salt gave H,O = 0.163 gram, GO, = made, different preparations being employed. 0.488 gram ; ash, a trace. 0.587 gram ; ash, a trace. 0.602 gram ; ash, a trace. 0*585* gram, and CaC03 = 0.0099 gram. 0*436* gram, and CaCO, = 0.0086 gram. 0*476* gram, and CaC03 = 0.0099 gram. Found. Theory for f------,---- 5ClOH,,O8,4C,,I-I,OO,,, I. 11. 111. 1v. v. TI. C23H30G18- i: .... 44 -65 44 -87 45 '10 44.65.f. 45 -09.f. 44 -9Of 45 '23 i i . . .. 6-22 6.23 6'32 6'03.f. 6 . l l . f . 6.26.f. 5 *92 5CioHi,0,,4C,2H,o0io, C23H300 18, CaO * ca0..- - - 1.52 1-7 1'8 1 -71 * This includes the C02 in the ash in boat, determined by weight of boat before f These percentages are calculated on the substance free from ash. reatment with ammonium carbonate.GUMS O F THE ARABIN GROUP. 1071 These figures leave no doubt that the theory advanced for the con- stitution of the gum acid k the correct one. The acid of the B fraction, [a]= = +100-101", gave a calcium salt containing 1.4 per cent. lime ; heptarabinantetrngula~~angeddate of calcium 7CloH160s,4C1~H,,010,C23Hw018,Ca0, requires 1.48 per cent. CaO. Of this acid and its calcium salt several combustions were made, different preparations being employed. I. 0.2907 gram dry free acid gave H,O = 0.163 gram, CO, = TI. 0.2873 gram dry fkee wid gave HzO = 0.159 gram, CO, = 111.0.2284 gram dry free acid gave H,O = 0.128 gram, CO, = TV. 0.2886 gram dry cnlcinm salt gave H,O = 0.15s gramj GO, = V. 0.2096 gram dry calcium salt gave H20 = 0.115 gram, CO, = 0.480 gram, ar,d a trace of ash. 0.476 gram, and a, trace of ash. 0.380 gram, and a, trace of ash. 0 474 gram, and CaCO, = 0.0075 gmm. 0.346 gram, and CaCO, = 0.0050 gram. Pound. Theory for r----------- 7 ~ C ~ ~ I I : , ~ , , ~ C ~ : H ~ ~ O I * , T. 11. 111. IV. Q. C'23I-T3,0 1% C.. . . 45.03 45-19 45.357 45.45* 45.63* 45.26 H . . . 6-23 6.15 6-23 6.17s 6.18% 5.94 ~CloH160,,4C,3H2,0,,, C23H3001SCa0. cao . - - - 1-48 1.34 1.48 The higliest rotating acid, [a]= = +108-110", gave a calciuni salt, which yielded CaO = 1.1 and 1.3 per cent. ; calcium n,onarabinantetl.a- gdactangeddate, 9 C ~ o H 1 ~ 0 s , 4 C 1 ~ ~ ~ ~ 0 , 0 , ~ ~ ~ H ~ ~ O l ~ , C a 0 7 requires CaO = 1.29 per cent.No serious attempt was made to purify the low rotzting acid; it was exceedingly difficult to free any portion of it from nitrogen, but i t was found that the fractior, with opt'ical activity [&ID = +8+433" yielded a lime salt containing CaO = 1.9-2.0 per cent. ; calcium t r i - n~nbinaiztetragnZactangeddate, 3 C ~ , ~ ~ ~ ~ O ~ , ~ C ~ ~ ~ ~ ~ O ~ ~ , C ~ ~ re- quires CaO = 1.98 per cent. It is, therefore, satisfactorily proved that the portion of the gum free from nitrogen consists of the c a l c i u ~ , magnesium, and potassium salts of four acids, viz. :- I. Nonarabiiiantetragalactangeddic acid, SCIOII,,OS,~C,,II,,O,~,C,:,H,,O~,. 2.Heptar~binantetragslactangedclic acid, 7 C 1 0 € I l ( i 0 , , 4 C , 2 € ~ ~ ~ ~ ~ , , , ~ ~ ~ ~ ~ 0 ~ 1 ~ . 3. Pentar~binantetragalactttrigeddic acid, ~ C ~ O H ~ ~ ~ , , ~ C , , H ~ ~ O ~ ~ , C ~ ~ 4. TriarabinantetragsLlactangcddic acid, 3C1,Hl60,,4C,2II,"Olo,CzaHs~O1,. * CLtlculatecl on substance free from ash.1072 O'SULLIVAK : RESEARCHES ON TRE The heptarabinan acid exists in by far the largest quantity. I n purifying these acids, there was some evidence, as I have said, of the presence of intermediate acids; I have, however, satisfied myself that such substances do not exist in any quantity in the gum under examination. There are, no doubt, places for them, as there would appear to be no reason why we should not have octa-, hexa-, and tetra-arabinan acids; it is curious, indeed, that onl? tthe odd number arabinan acids come to the surface, so t o say, i n this gum.A comparison of the composition of the series of acids in the two samples of gum shows clearly the difference between them, and the close relation they bear one to the other ; I need not dwell further on the subject. Besides the acids above described, the B sample of gum contains, ns I have already said, a large quantity of substance which I found it impossible t o free from nitrogen. I must reserve f o r a future occa- sion a fuller account of this sustance ; its general characteristics are given above. The low percentage of ash in the B samples of gum is no doubt in part due to the presence of this substance, but the chief cause of i t is the greater molecular weight, and consequently lowel.neutralising power, of the acids of tlhis gum. Although the evidence adduced for the conclusions arrived a t i n this work leaves little to be desired, I could not help feeling, notwith- standing many repetitions and confirmations, that there were a few points I should like t o see more completely worked out. I could not do this i n consequence of the difficulty in working with the B sample, due t o the presence of the nitrogenous substance and to the high solubility of the gum acids in alcohol. I tried several other samples of commercial gedda gum, but could not find any one 30 easily work- able as the A sample, the great bulk of which I had unfortxmately used up. I: have, therefore, t o content myself with leaving the work as it is, for the present.In time, I hope t o be able t o describe more fully and accurately the different acids, and more especially to give a complete account of the final acid, geddic acid, C23H38022, [a]D = +71-'72", and to determine in what it agrees with and differs from fhe similar andisomeric acid of the arabin series. This acid we should now call arabin acid, the resting-stage acid in the arabin series being tetragalactanarabic acid (Trans., 1884,45,55), the acid of the barium salt of which yielded '7.36 per cent. Ba0 the result of the action of sulphuric acid for 15 minutes on arabin : 4C,aH,o0,,,C23H300,S,Ba0 requires 7.49 percent. BaO. This formulais C71H11005S, against C71H1120z9, or a molecule of water less than that gi-ren in the place quoted.Arabin, or the then arabic acid, would be diarabinantetragalacta~iarabic acid 2CloHl~08,4C~~H2,0,0,C2~H~01~ = C91H1&4, against CSJL2O7~, given,GUMS OF THE ARABIN GROUP. 1073 in the paper quoted. I n the third part, of this work, I shall, I hope, be able to make these points clearer. Finally, I cannot conclude this paper without a brief account of the results obtained in an examination of a third sample of gedda gum. Sample of Gedda Gum C. This sample contained less nitrogenous matter than the B sample, but considerably more than the A one. The free acids were still more soluble in alcohol than those of the B sample; they had, however, the same syrupy character when precipitated by alcohol. I n cnnse- quence of the high solubility of the gum acids in alcohol, I fonnd it impossible to effect a clean fractionation.Sufficient, however, was done to show that the sample consisted of mixed gums, that, i n fact, it contained a t least two series of acids, neither of which was the same as the acids of the series of either the A or B samples. The nitrogenous matter was again the portion least soluble i n alcohol, but it was exceedingly difficult to free the gum acids from it. The portion rendered as free as possible from it was divided into a, b, and c fractions, a being the least soluble, and these, after being dialysed, were again divided into al and 1x2; b,, b2, bB, b,; and cl, 6, c3, and c4. The barium salts of these fractions were prepared and analysed. It would serve no useful purpose to give the analytical data.I may, however, record the results :- a,. [a]D = +64*9'; BaO = * per cent. a2. [a]= = f77.1" ; BaO = * 7 9 bl. [aj, = +83.2" ; BaO = 360 ,, b?. [E]D = +82*1" ; BaO = 3-54 ,, b,. [ R I D = 4-83.9" ; BaO = 4.03 ,, b4. [a],, = +84.2"; BaO = 3.77 ,, cl. [a]= = +85.6"; BaO = 3-91 ,, c3. [a]= = +89*9"; BaO = 3.66 ,, CZ. [a]= = +88.1"; BaO = 3.93 ,, cq. [ a ] ~ = +92'3"; BaO = 3-53 ,, The fraction 6, was acted upon with sulpharic acid for 30 minutes. The sugars formed consisted chiefly of arabinose with a little arabinon, and only a trace of galactose. I n this case, the arabinose crgstallised from the concentrated alcoholic solution in beautiful, warty aggrega- tions.* The gum acid insoluble in alcohol was freed from sugar and sulphuric acid in the usual way and then divided into three fractionp, Z, y, Z. Of these, barium salts were prepared and analysed :- * I may here remark that the crystalline habit of arabinose varies with the source whence it is obtained.1074 O'SULLIVAN : RESEARCHES ON THE x.[a]= = +32.2" ; BaO = 6.80 per cent. y. [a]= = +30-0"; BaO = 6.77 ,, 2. [ a ] , = +31*4"; BaO = 6.84 ,, Barinm pentragalactangeddate, 5Cl,B,,010,C,,H,,017,Ba0, requires 6.51 per cent. BaO. The fractions b4 to c4 are, no doubt, members of the same series. c4 is probably heptai~ahinnn~entugalucta~~gedtlic acid, 7CloH160s,5 C12Hzo010,C,3H2,01i,Ba0, requires 3.64 per cent. BaO, and cl, h e.carabinananpentaga1actangeddic acid, requires 3.88 per cent. BaO, the intermediate substances being mix- tures of these two acids.ccl does not belong to this series, for although it is the least soluble and its optical activity is lowest, the percentage of barium in the barium salt excludes it from the group ; it evidently belongs to a distinct gum, the sample being, no doubt, a mixed one. a2, bl, b,, and h:, are obviously mixtures of this series and the pentagalactan series. The nitrogenous snbstance in the C sample behaves in the same way to reagents as that, of the B sample; it is, however, devoid of diastatic action. Although the work described in t,his paper shows clearly the con- stitution of these dextrorotatory gums, and, indeed, of the whole group of gums, in a broad way, there are many points of interest which I should like to see more definitely settled. For example, the acids of the A sample differ one from the other by a molecule of arabinan; there is a difference in specific rotatory power between the mon- and di-acids of 6", and between the di- and tri-acids of the same amount, whilst the difference between the tri- and tetra-acids appears t o be 10" ; it would be interesting t o have these differences accurately determined, and to ascertain if they are constant factors. Of course, the amount of time requisite to purify the number of substances with which I had to deal, and t o make accurate determina- tions of t,hem was far greater than I could devote to the work. I had, therefore, to be satisfied with such factors as enabled me to identify the compounds and t o decide that they were sufficiently pure to warrant the inferences drawn from the analyses of them. Again, we see the four acids of the B sample, which differ one from the other by two arabinan molecules, differ in specific rotatory power by about loo, but when the final three arabinaa molecules are eliminated, that is, when triarabinantetragslacfxngeddic acid is converted into tetraga- lactangeddic acid, there is a fall of 50"; the elimination of the one final molecule of arabinan in the A sample acids, the degradation of rnonnrabinantrigalactanfreddic acid to trigalactangeddic acid is accom-GUMS OF THE AIEABIN GROUP. 1075 panied by a frill of 15-18". In the acids of the C sample of gum the difference of one molecule of arabinan makes, a s in the A sample series, apparently a difference of 6-7" in the specific rotatory power of the acids, and the elimination of the six final molecules makes a difference of SO", or about 8" for every molecule of arabinan. Yet again, turninc +o the elimination of the galactan groups ; be- tween tFgalactangeddin ac;d and the tetrag.alactnn one there appears to be a difference in optical artivit8v of only 2". I believe, however, this might very well be increased to 7" in consequence of the nature of t h e impurities with which the preparations of both acids are likely to be con taminated. The specific rotatory power of digalactangeddic acid I believe to be about [a]= = +]lo, thus differing from the trigalactan body bv 7-43' ; but there can be no doubt the mon-acid is close upon [alD = 60°, whereas the difference between this and the final acid is not more than 11". We see that with the elimination of the galactan proups the optical activity goes rapidly u p after we pass the di-acid ; the presence of a small quantity of the low acids in the preparation of the di- and tri-acids wonld materially increase their optical activitv (compme some of the fractionations of the tri-acid, A sample). It is pretty clear, in a broad way, that the optical activities and the com- position are closely related ; I shwld like to be ahle to indicate the relationship more closely than is shown above. This couId only be donP hv very careful pnrification of the snbstances and accurate determinations of the factors; it is obvious these things are beyond the scope of this papcr. I hope, however, to be able to return to them later on, hut. after all, t'he questions are more interesting from a phvsI'ca1 point of view than a chemical one. There is one point more to which I should like to refer, the baryta in the barium salts; when the acids are dry, water appears not t o take the place of baryta, the dry acids are anhydrides. This, I say, appears to be the case, as a result of a general analysis, brit it would Fe well to have special evidence on the subject. This, I believe, I shall he able to obtain by employing more easily workable gums. In the immediate future, however, I intend to continue this work with an investigation into the constitution of an optically inactive gum, of which I have an eyample in a sample of Auptralian gum, and to con- clude it with a, full examination of the final acids, geddic acid, arabic acid, &c., C23H3R012, in all cams. Finally, I need hardly say that the amount of labour involved in this work, especially when it is remembered that what I have recorded here is only a fraction of the work upon which the results are based and which led to them, could not be performed without much assistance. For that assistance, given in a helpful and thoughtful manner,I have to express my best thanks to my assistant, Mr. A. L. Stern, B.Sc. VOL. LIX. 4 E

ISSN:0368-1645    

DOI:10.1039/CT8915901029

出版商:RSC    

年代:1891

数据来源: RSC

摘要:

1076 XCI1.-Dissociation of Lipid Nitrogen Peroxide. By J. TUDOR CUNDALL, B.Sc., Lecturer on Chemistry in the Edinburgh Academy. WHILST examining nitrogen trioxide with Professor W. Ramsay (Trans., 1885, 47, 1137 and 672), we were led to expect from the appearance of the substance t-hat it was to a certain extent dissociated while yet in the liquid state. I t was suggested that I should make some colorimetric experiments to ascertain whether this was the case, commencing with nitrogen peroxide as a substance for which experimental methods might be more easily devised than for the trioxide. It was assumed that liquid nitrogen peroxide at low temperatures was a colourless Substance, consisting of nitrogen tetroxide, Nz04, but as the temperahre was raised i t became yellow and then red, owing t o its partial dissociation into NO,.N,O* = 2N0, ; NO, being assumed to be an intensely red substance, resembling bromine iu appearance. Judging, too, from the behaviour of the snbstance in the gaseous state, where the change of colour is accompanied by decrease in the density, these assumptions seem not unreasonable. At the time, I made some experiments in which chloroform solu- tions of the peroxide were matched against different coloured liquids. However, although the results seemed promising, the difficulties of obtaining a good match for the peroxide, and of keeping its solution, during the experiment, without loss of strength, and from the action of atmospheric moisture, proved insuperable with the crude apparatus I had then devised. The pressure of other work prevented any de- velopment of the idea until 1889, when t'he present research was commenced.The object in view was, firstly, to decide whether liquid nitrogen peroxide dissociates when it is diluted; and, secondly, to measure its dissociation when heated. With regard t o the first point, Ramsay has shown (Trans., 1888,53,621), from the depression of the freezing point of acetic acid by varying amounts of peroxide, that no large amount of dissociation takes place on dilution. That alteration of temperature produces dissociation, as indicated by the colour, is obvious: and it has been shown by 3. J. Boguski (Compt. rend., 109, 80&806), that the electrical resistance of slightly impure peroxide rises with the temperature.DISSOCIATION OF LIQUID NITROGEN PEROXIDE.107 7 Preparation of the Materiuls. Ramsay (Trans., 1890, 57. 590) suggests that the best method of preparing nitrogen peroxide is to act on arsenic trioxide with nitric acid, and so obtain a mixture of trioxide and peroxide. This mixture is purified by oxidising the trioxide to peroxide, by treatment wit11 nitzogen pentioxide, and subsequent distillation. He deprecates the use of sulphuric acid to prevent the formation of the trioxide on account of the diminished yield, but that objection can be obviated to a large extent by the use of fuming nitric acid, so that the amount of sulphnric acid required is kept low. The plan I have found most effective is to mix 315 grams of fuming nitric acid (sp. gr. 1.5) with 150 grams of concentrated sulphuric acid and 250 grams of moderately finely powdered arsenic tri- oxide.The proportions are roughly those required for the reaction, 4HN03 + As,O, = 2N20, + 2HAs0, + H,O. The arsenic trioxide is purposely in excess, and there is enough sulphuric acid t o allow one formula-weight to each formula-weight of water present in the nitric acid formed in t'he rittaction. The mixture was made in a flask of about 1 litre capacity, fitted by a cork with an upright, thin-walled tube 1 cm. in diameter. This tube passed through the cork nearly to the bottom of the neck of the flask (an expedient which greatly protects the cork from the fumes), and projected about 30 cm. The projecting part was surrounded by a vessel of water, kept at 20-25", to condense any nitric acid that might be carried over.The gas issning from the top of the upright tube was then passed through a long U-tube containiiig phosphorus pentoxide, and finally into a bulb surrounded by ice and hydrochloric acid, where it was condensed. The action was started by gently heating the mixture ; it then proceeded quietly, with occa- Eiond heating, until the mixture began to turn green, and the character of the effervescence to change, when it was stopped. Tho crude product was then distilled at as low a temperature as possible. As it was feared that it might contain traces of anhydrous nitric acid, which is soluble in the peroxide, the vapour was passed through a tube of heated arsenic trioxide, and again through phos- phorus pentoxide before being condensed as above.It wab: condensed in three fractions, the first weighing 97.5 grams, the second 56.0 grams, a,nd the third 16.5 grams. The last fraction, as well as the residue in the flask, was not used. The portions collected are about 82 per cent. of the theoretical yield if the nitric acid of sp. gr. 1.5 contained 90 per cent. of HNOs. As in the subsequent experiments, it would be necessary to find the volume of small quantities of the peroxide accurately, and as the 4 E 21075 CUNDALL : DISSOCIATION OF substance examined by Thorpe (Trans., 1880, 37, 141) was prepared in quite another way, the sp. gr. and expansion were carefully deter- mined by a method similar to that employed by Thorpe. A dilato- meter, whose bulb held about 20 c.c., was used, and the following results were obtained (weight of peroxide taken 25,6225 grams) :- t.0" 6.7" 11.8" Sp. gr. compared with water at do. 1.4880 1.4734 1.4649 Thorpe found the sp. gr. at 0" = 1.4903. prepared by Messrs. Duncan and Flockhart from methylated spirit. I t was dried with fused calcium chloride and distilled, the middle portion was then repeatedly treated with phosphorus pentoxide and distilled again, that portion only being employed which was of constant boiling point.. The solutions of nitrogen peroxide in chloroform were usually prp- pared by weighing i n sealed bulbs, but were also prepared by measure- ment of the nitrogen peroxide in a, special narrow graduated tube, and of the chloroform in a stoppered burette. The temperature was noted and correction made for it in cnlciilating the strengths of the solu- tions, which are stated in percentages of nitrogen peroxide by volume at 0".In almost all cases, the calculated strengths of the solutions were confirmed by analysis. The method adopted was that of shaking in a small stoppered bottle a suitable rolume of the solution with a meamred quantity (usually 20 c.c.) of normal caustic soda, and titrating with semi-normal sulph- uric acid, phenolphthale'in being used as an indicator. This method was proved to give accurate results by check analyses i n which weighed bulbs of the peroxide were broken in chloroform and then stnalpsed. The method of analysis by the nitrometer was also em- ployed with similar, but, not as satisfactory, results. The solutions were prepared just before they were examined, except when they were sealed up in the colorimeter itself, and so unable to alter in strength.The chloroform used for dilution was that sold as "pure," and* Apparatus used. It was obvious that some special form of colorimeter must be ern- ployecl, in which to make the comparisons of tint, as the free access of air which obtains in the common forms was out of the question, and also the observations had fo be made under known and steady condi- tions of temperature.LIQUID NITROGEN PEKOXIDE. 1079 One of the most suitable forms det-ised waB a modification of Mill’s colorimeter. Of this pattern, two, called A and B, were con- Rtructed of different lengths for use with pale and darker solutions ; they were alike in all other particulars. Each consisted of a tube, a, Fig.1, about 22 mm. in diameter, in A 60 mm. long, and in B 100 FIG. 1. C mm. long, closed at the top by an end that was blown as flat and as thin as possible. At the other end, a narrow tube b, 7 mm. in diameter, and 200 mm. long, was sealed axially, and at the side a bent tube c with a small bulb was attached. Before the top of a was closed, the disc d and its float e was inserted. The disc was secured by a, clip of platinum sealed to the top of the bulb and stalk of which the float consisted. The stalk was weighted with a piece of stout irox wire fixed inside wititi a plug of asbestos, so that the whole float and disc just floated in chloroform. The end of the stalk was then sealed. When the apparatus was constructed, n was graduated in mm., sttart- ing as nearly as possible from the inside of the flat end.It was filled with the required solution through the open end of b, and when nearly full was sealed. The bubble was then transferred to c , where it did not interfere, and also allowed the liquid t o expand. The disc with its float was moved up and down, or fixed by the1080 CUNDALL : DISSOCIATION OF action cf a small soleno'id encircling the jacketing tube in which b was supported. It was thus possible to make colour comparisons i n an entirely-sealed apparatus. For shorter columns, and for comparisons in which it was Eecessary t o dilute during the progress of the experiment, two colorimeters of another form, called E and I?, were employed. Each apparatus con- sisted of a straight tube y, b'ig.2, open at the top and closed at the Bottom by a flat end, with a reservoir h sealed on near the bottom. A narrow inlet tube, provided with a slight funnel expansion, was sealed to the top of h for introducing the substance to be examined. In E, g was 100 mm. long; in F, 120 mm. long; and in both 22 inm. wide. Closely fitting into y was a tube i closed a t the bottom with a flat end, In order to make the end of i of uniform thickness, it was ground and polished on the outside, whilst a, microscope cover-glass was cemented on the inside with Canada balsam. This device provided a transparent end to the tube, which did not distort the disc k lying on the bottom of g when seen through it, and, above all, was not affected by the liquid in which it was immersed. The tube g was graduated -" 2.in mm., and as the levels of the top of the disc and of the bottom of the tube i could be accurately read, the length of the column between them could be determined with a limit of error of 0.2 mm. The tops of g and i were held in brass clamps I and m, which could be separated and approached by a rack and pinion working in a brass pillar n. This piIIar fitted in a socket OP the bottom of the constantLIQUlD NITROGEN PEROXIDE,. 1081 temperature-bath, of which the level of the water rose to about 0, whilst the level of the liquid in the colorimeter was never higher than p . The rack and pinion was actuated from the outside by a kq-, which passed through the wall of the bath. The only access of air was through the inlet tube and the narrow annular space between q and i.The effect of any deterioration through these openings was so miiiimised by the large bulk of liquid in h as to be negligible. In order to obtain the necessary conditions of temperature and illumination, the colorirneters were enclosed in a wooden tank, Fig. 3, FIG. 3. * 26 cm. long by 13 cm. wide by 23 cm. deep, provided with a double glass f.ront. The tank was divided into two equal parts, r and s, by a thick, watertight, wooden partition, and the whole was painted black. Both compartments were provided with drain cocks t , t, and with over- pipes TJ t~ A supply of water could be allowed to flow in from the main- by taps u, u. I n the side r was fixed a metal pipe TO, which, when heated at its upper part, caused a, circulation of hot water, and permitted the temperature of the bath to be raised.Tubes x, x, x, on which the solenids slid stiffly, wcre provided in both compartments as sockets for the colorirneters A and B, whilst n, socket for the recep- tion of the stand of colorirneters E and l? was fixed in r. Uniformity of temperature was secured by an Archimedean screw y, fixed at an angle and turned rapidly by an electromotor. This stirred so effec- tirely that when ice and water were in the bath, the temperature remained absolutely constant i n every part until almost the last tracelo82 CUNDALL : DISSOCIATION OF of ice was melted. A similar stirrer was provided in the other cool- partnient, but it was not much used as the temperature generally required there was that of the tap wat,er, which was allowed to flow sufficiently rapidly to keep up a good circulation, and whose tempera- ture remained cons tan t during long periods.Higher temperatures i n r were produced by regulating the sizc of the flame under w, and these, owing to the large mass of water and constant stirring, were easily kept uniform. The measuring apparatus employed was checked either by weighing with water or against measures of knoyn accuracy. Method of Comparison and Calculation. An attempt was made i n the earlier experiments to compare the colour of the peroxide with the solution of some colouring matter such as chromic acid or Bismarck brown. As this plan proved un- satisfactory, it was given up in favour of a mc>thod in which different lengths of some standard solution of the peroxide were compared with the solution that was being diluted or raised in temperature.The method of procedure, initially, was to take some pure peroxide in colorirneter E at 0" and to compare i t with a standard solution ii: colorinieter A G r B either at the same or another temperature. The comparison was effected by fixing the disc in A or B amnd then match- ing the colour of that particular column of solution by moving the inner tube i of E or Ik' u p or down by the rack and piniun t o increase or diminish the colour. In order to make the operation easier, the columns to be compared were put close side by side, and the whole of the bath and apparatus was shut off from view by a screen of card with two holes directly over the columns to be matched.The discs were illuminated by two gas jets so placed as to equally light them, and whose heat rays were cut off from the bath by a double screen of glass. The flames themselves were hidden from the eye by shades. When the colours in the two coiorimeters were matched, the heights of the columns of equal tint were read and recorded. The solenoid was then moved and a new match made, and so on until five or mom comparisons had been taken, when the liquid in E: WAS diluted or its temperature raised, and a new match made. Great care was taken that the color~inieters and tlieir content,s were a t the temperature of the bath by waiting a sufficient time and by stirring the liquids in them by the movement of the disc or tube. Examples taken a t random oE these comparisons are given in order t o show the amount of accuracy possible.It will be seen i n the sequel that the errors that do occur-and shades of yellow and orange are notoriously difficult to match-are of a magnitude which do notLIQUID NITROGEX PEROXIDE, 1083 affect the general result. A column of the peroxide 10 mm. in length a t 0" was taken to be of unit depth of colour, and all the lengths given are those of columns matching it. Pure peroxide at 0' was in E, and a standard solufion called A/3 of 8.9 per cent. by volume was i n A also at 0". Hence, a t O", 5i.1 mm. of the solution A/3 match 10 mm. of pure Again a solution of 3.0 per cent. at 0" in E' was compared with a peroxide. standard called Bi3 of 1.44 per cent. by volume at 132" in B.F. R. 22.0 mm. match 14.8 mrn. 38.0 ,, ,, 23.2 ,, 25.0 ,, 9 , 15.0 Y, 25.0 ,, > 7 14.0 9, 28.4 ,, ,, 19.3 >> - - 138.4 86.3 Now 63.9 mm. of B/3 a t 13.2" had been found to match 10 mm. of pure peroxide at Oo7 so at O", 63'9 138*4= 102.5 mm. of a 3 per 86 3 cent,. solution have the same tint a s 10 mm. of pure peroxide. The first column in Table I (p. 1084) gives the length of different solutions at 0" which have the same colour as 10 mm. of pure per- oxide. The corresponding number in the second column is the per- centage strength of the solution. Now, if the colouring matter in the pure peroxide remains un- changed in amount by dilution, then the product of the 1engt)h of the columns that match into their strengthfi should be constant, for the lengths should vary inversely as the percentage strength by volume, that is, if you halve the strength of a solution it ought to require a, column of double length to give the same cnlour.The third column gives the product of length by strength, and i t is seen that it is not a constaut, but diminishes with increased dilution. The colouring matter then increases with the dilution, and its rela,tive quantity is108.1 . CUXDALL: DISSOCIATION OF as the reciprocal of' the product length by strength, for example, if at 0" a 10-mm. column of pure peroxide (100 per cent.) matches a 40.6 mm. column of 10.1 per cent., then we have a column containing 100 x 10 = 1000 pal-ts of peroxide matching a column containing 40.6 x 10.1 = 410 parts. Therefore the relative amounts of colouring matter (NO,) in the 10.1 per cent.solution to that in the pure per- oxide are as x+5 : i5k5, or as 2.44 : 1. This recipi-ocal multiplied by 1000, which expresses the relative amorint of NO, present, is stated in the fourth colnmn. It is multiplied by 1000, so that the amount present in the pure peroxide a t 0" is represented by one unit. TABLE I.-Comparison at 0" of the Colozir of Pure Nitrogen Perozicle with that of its Solutions. Columns of equal tint. mm. 10 .o 11 .o 12 *5 1 2 -7 13 .7 13 -7 15 -0 15 -8 15 -2 18 '0 21 '0 25-0 34 -6 39 -2 37 '3 37 '3 40.6 57 '3 56 '0 50 -5 63 -1 63 .1 80 -7 102.5 136 -0 118 -8 Per cent. strength by volume. 100 .o 88 -3 7s -4 .67 -6 63 -0 62 -6 55 -0 48 - 2 46 -8 37 -5 25 '8 25 -1 14 -3 13-0 10 '4 10 '37 10.1 8 - 9 7 6 7 - 5 7-1 5 '3 4.0 3 -0 1 -44 1 .G Per cent. strength x lcngth. 1000 97 1 942 860 865 858 825 76 1 71 L 675 541 627 495 5 10 388 387 410 510 426 379 4'6 334 323 307 196 171 Relative amount of NO, present. 1 .oo 1 '03 1 -06 1'16 1 .20 1.17 1-21 1.31 1-91 1-48 1 '85 1 -59 2 '02 1 -97 2 -60 2 -58 2 '44 1 -95 2 -35 2 '64 2.46 2.99 3 -1 8 -25 5 ' 1 5 85 The last column is plotted as Curve I (p. lOS!;) against percentage strength by volume. It will be seen that the rate of dissociation is extremely small, until the peroxide is about 20 times diluted, after which i t becomes much greater. It shonld benoted that these figures are purely relative, their absolute value being found later.LIQUID NITROGEN PEROXIDE. 1085 Measurement of Relative Dissocirttion on Rise (,f Temperaiure.Having thus established that dilution produces dissociation, it became necessary to compare this dissociation with that caused by rise of temperature. The comparisons were made, and the results calculated in precisely the same way as be€ore; a, 20 mm. length of pure peroxide at 0" being again taken as containing unit quant,ity of peroxide. The relative dissociation is plotted nqainst temperature for eight different dilutions, as Curves VIII-XIV, in Plate I1 (p. 1086); the amount at 0" being cori-ected by reference t o the smoothed curve representing the dilution at 0". TABLE II.-Dissociation caused by Bise of Temperature at Diferent Dilutions. Columns Of equal tint. mm. 10 .o 5 -4 4'1 15 -2 11 '1. 7 0 6 -8 21 -0 15 -3 9 -4 6 . 1 25 .O 14.6 6 . 5 34 *6 26 -0 18.0 12.5 8.1 37 -4 30 -1 21 -3 14 -4 11 -0 -- 6 -96 Tempera- ture.-- 0 % 7 ?A 11 -2 0 -0 4 *2 9 . 2 15 -2 0 *o 5 *o 11 -6 17.6 0.0 7 -2 17 '3 0.0 5 *7 10 -7 15 7 20 *2 0.0 5 *4 11 '2 15 -7 20 '2 27 -0 Per cent. strength by volume. 100 -0 loo -0 100 -0 46 -8 46 -8 46 -8 46 *8 25 * 8* 25 -8 25 -8 25.8 25 '1" 25 -1 25 -1 14 -8 14'3 14 3 14 *3 14 '3 10 *1 10'1 10 -1 10.1 10 -1 10 -1 Per cent. strength x length. 1000 640 410 711 519 328 248 5442 395 243 157 628 3titi 158 495 3'72 257 179 116 378 3( 4 215 145 111 70 Relative amount of NOz present. 1 -00 1-85 2 -40 1 -41 1-92 3-05 4.03 1 -85 2 -53 4 '13 6 -36 1 59 2 -73 6-15 2 *02 2 -69 3 -87 5 *Fig 8 -66 2 -64 3 '29 4 -67 6 -89 9 -06 14 -2 * These two percentages m e combined in the ,same curve, No. X.1086 Columns of equal tint.Columns O f e q u d tint. Per cent. strength Per cent. strength by volume. x length. mm. 57 -2 41 -1 29 -3 16 -9 13.6 8.8 102 *5 69 -3 48 -0 41 -6 29 '2 21.4 118 *8 69 -8 63 -9 62 -8 37 -7 26 -6 CUNDALL : DISSOCIATION OF Tempera- rature. --- 090 5-1 9 -9 15 -9 20 *2 %5 -0 0 -0 5 . 1 10 *9 15 .O 20 -0 25 *O 0 .o 11 -9 13.2 15.1 19 - 9 25 *O TAHLE II-~ontin~ed. Per cent. strength by rolume. 7 -1 7.1 7 . 1 7 . 1 7 . 1 7 . 1 3 -0 3-0 3 -0 3 .o 3 -0 3 -0 1 -44 1 *44 1 -44 1 '44 1 *44 1 -44 Per cent. strength x length. 406 291 207 117 97 62 307 208 147 125 88 64 I71 101 92 90 54 38 Relative amount of NO,, present. 2 *443 8 -43 4 -82 8 *51 10.33 16 -06 3 *25 4 -81 6 -79 8 -02 11.4 15 -5 5 *85 9 -9 10 -9 11 -2 18 *4 26 -1 The amount of dissociation at every 5" for eash of these strengths was then read off the smooOhed rise of temperature curves, VIII-XV (Plate II), and plotted as dilution curves, 11-VI (Plate I). These isothermals show that dissociation at temperatures above 0" takes place i n the same manner as at O", only at a greatel.rate. The form of the curve below 10 per cent. is confirmed by a dilutiou experiment that was performed at, 82", and of which the results are stated in Table 111. It is plotted as Curve V I I (Plate I). TABLE 111. mm. 26 *5 37 *6 4 5 -8 48 *8 61 -5 71 -6 '77 -5 82 -1 111 -5 9.93 6 *O 4'8 3 -6 2 -5 2 .o 1.6 1 *4 0 -8 263 266 220 176 154 143 124 115 89UX&S of DissociatLo~~. I I I l j . , , I I 1 ' . 1 : I I I I I 1 I 1 I 1 I 1 I I I I I I I 1 I 1 c ul I u) 0 d 2 VI K K I$ 4HARTLEY JouLm.C?zm. Soc. A h . /&9!L ABSORPTION CURVE OF BLUE CHLOROPHYLL ALCOHOLIC SOLUTION The dotted Line indicates the cwve o f y e h w ch2orophyl.L. ScaLe 5 6 7LIQUID KITROGEN PEROXIDE. 1087 Determination of the Absolute Amount of NO, present. I n order to arrive at the absolute amount of the peroxide dissoci- ated, a cornparison was made between the colour of the pare liquid peroxide at 0" with that of the gaseons peroxide at 30". It is true that the substances compared are in different physical states, but it will be seen that the NO, present in the liquid peroxide and in the solutions is never in a state of aggregation differing widely from that obtaiiiing in the gas, and SO might be expected to have the same properties as regards colour as when actually in the gaseous state.That this assumption is legitimate is borne out by the fact tbat liquid and gaseous nitrogen peroxide have both the same absorption spectruiu, that, of the pure liquid being fainter than that of the gas ; on dilution and rise of temperature, however, i t becomes much more distinct. Again, another property of the molecule, namely, its ac.tion on polarised light, is the same in the l i q ~ i d and gaseous state, pro- vided equal numbers of molecules are traversed by the ray. The comparison was made by means of a solution A/3 of 8.9 per cent. by volume, 22.5 mm. of which at 11.8" matched 10 mm. of the pure liquid peroxide at 0". This solution A/3 in A was compared with the gas in F ; F was filled with t,he gas by allowing aborit I C.C.of the liquid peroxide to run in at the inlet tube and cover the disc, where it soon evaporated, the excess escaping, and leaving the apparatus qiiite full of the gas. The temperature was allowed to become steady, and then matches were made in just the same way as with the liquid. After five matches had been made, a little more liquid peroxide was added, and when it had evaporated, and the temperature had become steady, comparisons were again made. The results of the first series of comparisons are given in Table IV, m-here the gas' in colorimeter F at 29.8" was compared with standard solution A/3 (of 8.9 per cent. by vol.) in colorimeter A at 12.0". TABLE IV. 30.8 mm. match 12 mm. 19.5 ,, 7 7 7 7 , 11.0 ,, 7 , 5 7, 21.0 ,, 9 9 9 9 7 13.0 ,, ? > 6 3 , 95.3 39 A.F. - - Now, 22.5 mm. of A/3 at 12.0" matched 10 mm. of pure liquid per- oxide at O", EO at 29*8", ~~- = 9.2 mm. of gaseous nitrogen 39.0 x 22.5 95.3I088 CUMDALL : DISSOCIATION OF peroxide have the same depth of colour as 10 mm. of liquid peroxide at 0". The experiment was repeated on another day with gas from a different sample of peroxide, again comparing with standard solu- tion A/3 at 11.8'. The temperature of the gas during the first five matches was 19*6", but during the remaining 15 it only varied between 29.9"and 30". The results are given in Table V. TABLE V. Hence, calculating its before, 9.3 mm. of the gas at 30" match 10 mm. of the pure liquid peroxide at 0". Now, by J. W. Gibbs' formula (Amer. J . Sci., 18, 1879), the density of nitrogen peroxide gas compared with air at 30" and 755.8 mm.(the pressure at which the matches were made) is 2.618, which agrees sufficiently well a-ith Deville and Troost's experiments (Compt. rend., 64, 237) under similar conditions. Now, a density of 2.618 requires that the gas shouId contain 35.2 per cent. of NO, and 64.8 per cent. of N204. But .there is the same amount of NO, in a column 9.3 mm. long of the gas at 30" and 755.8 mm. as there is i n one of 10 mm. of pure liquid per- oxide. Assuming that a column with a base of 1 sq. CTT-. is beingLIQUID NITROGEN PEROXIDE. 1089 looked at, 9.3 c.c, of the gas has as much NO, in it as 10 C.C. of pure peroxide at 0". 9-13 C.C. at 50" and 755.8 mm. contain 35.2 per 273 755.8 46 303 160 22320 cent. of NO, = 3.78 c.c., which weigh 3.78 x T-- X -7- x = 0.00698 gram of NO,. Now, 10 C.C. of pure peroxide at 0" weigh 0.00698 14903 14.903 grams ;. therefore iiitrogen peroxide at Qo contains x 100 = 0.0468 per cent. by weight of NO,. The weight per cent. of NO, present in the peroxide in any of the solutions at any temperature examined can then be found tiy multi- plying the number expressing the relative dissociation by 0,0468, as there is this amount i n the pure peroxide atl 0" which is taken as being dissociatsd to the extent of 1 unit. Conclusions. 1. Liquid nitrogen peroxide dissociates when diluted with chloro- form, just as the gas dissociates when diluted with an indifferent gas, or when the pressure on it is lowered. The dissociation is of small amount, and its rate of increase is very small until the liquid is diluted about twenty times, when i t becomes much grmter. 2. Liquid peroxide of nitrogen and its solutions dissociate when heated, but its solutions do so at a much greater rate as they me diluted. I n the most dissociated solution yet examined, that is, one containing 1.44 per cent. by vol. of nitrogen peroxide at 25", there is only 1-22 per cent. of NO, present in the peroxide, It is proposed to examine more dilute solutions at higher tempera- tiires with a, view t o obserring whether t.he liquid peroxide under those circumstances behaves like the gas, but as this will involve working at pressures greater than atmospheric, some considerable time must, elapse before results can be obtained. Edinburgh Academy.

ISSN:0368-1645    

DOI:10.1039/CT8915901076

出版商:RSC    

年代:1891

数据来源: RSC

摘要:

XCIII. --On Irort Caybonyls. By LUDWIG MOND, F.R.S., and CARL LANGER, Ph.D. THE existence of a volatile componnd of iron and carbon monoxide was made known to t'he Societ;y last session in a communication by one of UR and Dr. F. Quincke (Trans., 1891, 59, 604). I n a paper read before Section B of the British Associatinn in August last,* it was announced t h a t we had succeeded in obtaining this compound in the form of an amber-coloured liquid. which boils a t lOe0, solidifies below -21°, and deposits tabular crystals of a darker colour on stand- ing. We have since found that these crystals are only obtained when the liquid is exposed to lizht, and that, their formation is accom- panied hv the evolution of carbon monoxide. This liqnid compound of iiwn with carbon monoxide i s preparrd in the following manner :-Ferrous oxalate, precipitated from a hot solution o f ferrous siilphate by adding to it a slight excess of potassic oxa'ate, is washed by repented decantRtion with wnter and dried at 120".The dry powder is introdnced into a combustion tube, and heated in a gentle ciirrent of hydrogen, the temperatnre being praduall-y rRised until the oxalate bas turned b l ~ c k , and is then kept stationary until, on nrrestinq the current, of hydroqen, no more gas escapes. The finely-divided iron thus obtained is allowed to cool to the ordinary temperature, and is then put into water without allowing it to come into contact with the air, boiled several t,irnes with water until all sulphate is removed, dried quickly on plates of gypsum, and then returned to the combiistion tube and slowly heated in a current of hydrogen to nbont 300" to drive off d l the water.After allowing i t tocool again, the hydrogen is displpced bycarbon mon- oxide and the tube is then closed a t one end, the other remaining con- nected with the gasholder containing carbon monoxide, which is slowly absorbed b-y the iron. After the lapse of 24 hours, the tube is heated to about 120", when the iron carbony1 formed distils over. The yield is somewhat increased when this distillation takes place in a, slow current of carbon monoxide and the issuing p s e s are pRssed through a tube cooled to -20'. When no more iron carbonyl comes over, the tube is allowed t o cool, and put into communication again with the gasholder containing the carbon monnxide, which is absorbed as before.This operation can be repeated during several weeks; the daily yield, however, is always small, and rarely exceeds 1 gram from * '' On Nickel Carbon Oxide and its Application in Arts and Manufactures," by Ludwig Nond.MOND AND LAXGER ON IRON CARBONYLS. 1091 100 grams of iron. We have tried t o increase this yield by ex- posing the iron t o carbon monoxide under a pressure of 10 atmo- spheres, but have obtained no material augmentation. The iron carbonyl thus obtained is a somewhat viscous liquid of a pale-yellow colour. Its sp. gr. at 18" is 1.4664 compared with water a t 18". It distils completely without decomposition at 102.8" iicder a pressure of 749 mm. It solidifies below -21" into a mass of yellow- ish, needle-shaped crystals. It is soluble in many organic liquids such as alcohol, ether, benzene, mineral oils, &c.On exposure to the air, it is slowly decomposed with formation of a brown precipi- tate consisting mainly of hydrated ferric oxide. On heating the vapour to 180", the compound is completely decomposed into iron and carbon monoxide. The vspour density has been determined in a Victor Meyer apparatus filled with hydrogen and heated in a bath of xylene. 0.191'2 gram substance displaced 25.0 C.C. hydrogen at 17.7" ; air 0,1249 gram substance displaced 16.4 C.C. hydrogen at 17.7"; air This corresponds very nearly to the calculated density of Fe(CO), = 6.7, whilst Pe(CU)* would require 5.7, and Fe(CO), 7.25. The formula Fe(CO)5 is also corroborated by analysis of the substance. We determined the iron by heating a weighed quantity with chlorine-water in a sealed tubc and subsequently precipitating and weighing the ferric oxide ; and the carbon monoxide by burning a weighed quantity in a, current of air in a combustion tube pa,rtly filled with copper oxide, and weighing the carbon dioxide formed.The following figures were obtained :- I. 0.3014 gram substarice yielded 0.1265 gram Fe,O, = 0.08855 11. 0.1318 gram substance yielded 0,05523 gram Fe,O, = 0.03867 111. 0.2463 gram substance yielded 0.2702 gram CO, = 0.17195 IV. 0.2838 gram substance yielded 0.3172 gram CO, = 0.2022 Found. r--------- 7 I. 11. 111. IV. F e (0 0) 5 . We obtained the following figures :- pressure, 751 mm. ; density, 6.5. pressure, 759 mm.; density, 6.4. gram Fe. gram Fe. gram GO. gram CO. Theory. Pe ......... 29.37 29.34 - - 29.08 co ........ - - 69.88 71.14 70-92 We propose to call this cornpound ferropentacarbonyl. It is iso- meric with ferrous croconate, C5Feo5. VOTJ. LTX. 4 F1092 MOND AKD LANGER ON IRON CARBONYLS. Ferropentacarbonyl is not acted upon by dilute sulphuric, hydro- chloric, or nitric acid at the ordinary temperature. Concentrated nitric acid, chlorine-water, and bromine-water, however, act readily, forming ferric nitrate, chloride, or bromide respectively. Alcoholic solutions of sodium and potassium hydroxide absorb the vapour rapidly, and also dissolve the liquid without evolution of gas. After a while, a greenish precipitate is formed, which contains chiefly hydrated ferrous oxide, and the solution becomes brown. On ex- posing it to the air, it takes up oxygen, and the colour changes to a dark-red, whilst hydrated ferric oxide separates out.On account of this rapid change of the solution in contact with air, and the small quantity at our disposal, we have not yet been able to isolate any well-defined compound suitable for analysis from it. On mixing alcoholic solutions of ferropentacarbonyl and mercury chloride, a slight evolution of carbon monoxide is observed, and a yellowish, crystalline precipitate is formed containing iron, mercury, chlorine, and carbon monoxide. The analpes, however, did not give figures from which a definite formula could be deduced. The liquid Ferropentacarbonyl undergoes no change when kept in the dark, but when it is exposed to the light for Eeveral hours in a sealed tube, gold-coloured, tabular crystals are formed, and the pressure in the tube rises very high.When dry, these crystals have a metallic lustre, and resemble flakes of gold, On exposure to the air, they are gradually decomposed and coloured brown. We have prepared this sub,stance for analysis by collecting i t on a filter, washing with ether, and drying over sulphuric acid for about 30 minutes. A weighed quantity was treated with bromine-water, which dissolves the sub- stance with evolution of gas. The iron was precipitated from tthe solution and weighed as ferric oxide. 0.0800 gram subst’ance yielded 0.0412 gram FeaOs = 0.02884 Fe, 0.0’742 gram substance yielded 0.0381 gram Fe,O, == 0.02662 Fe, or 36.05 per cent.Fe. or 35.90 per cent. Fe. These figures agree sufficiently with tlhe formula Pez(C0)7 (diferro- heptacarbonyl), which requires 36.36 Fe. We have, ~ro far, not succeeded in obtaining a sufficient quantity of this substance to con- trol its composition by determining the carbon monoxide, nor has it been possible to ascertain its molecular weight either by the vapour density or by Raoult’s method, as the substance is neither volatile without decomposition nor sufficiently soluble in sui tabIe solvents. Chlorine, bromine, and nitric acid decompose the crystals ; sulph- uric acid and hydrochloric acid do not act upon them at the ordinary femperat ure. Alcoholic potash dissolves them, forming a solutionCOMPOUNDS OF OXIDES OF SILVER AND LEAD. 1093 similar in appearance and behaviour to that obtained with the liquid iron carbonyl. The solution of iron carbonyl in heavy mineral oil investigated by Moad and Quincke, which gave figures for the proportion of iron to carbon monoxide varying from 1 : 4.03 t o 1 : 4.26 undoubtedly contained mixtures of the two substances we have described. We are continuing the investigation of these compounds, and hope shortly to make a further communication to the Society.

ISSN:0368-1645    

DOI:10.1039/CT8915901090

出版商:RSC    

年代:1891

数据来源: RSC

摘要:

COMPOUNDS OF OXIDES OF SILVER AND LEAD. 1093 XCIV.-Note o n some Conzpoun.ds of the Oxides of Silver a d Lead. By ENTLY ASTON, B.Sc. (Lond.), Chemical Department, University College, Gower Street. ACCORDING to Wohler (Ann. Phys. Chem., 1837, 41, 344) the yellow precipitate obtained by the addition of caustic potash to a solutioii of a lead and silver salt is a compound, containing 1 mol. of silver oxide to 2 mols. of lead oxide (Ag2O,2PbO). It is insoluble in caustic potash, and so, by digestion with it, may be separated from the free lead oxide, which is simultaneously precipitated. On analysis, Wohler obtained the numbers 34.23 pcr cent. AgaO and 65.77 per cent. PbO. At Professor Ramsay’s suggestion, I have made the following attempts to obtain Wohler’s compound, but at present without success.First Method of Preparation.-The first sample prepared was a, yellow precipitate, obtained by the addition of a strong solution of caustic potash to a solution of lead acetate and silver nitrate, in which the ratio of lead oxide to silver oxide was appmximately 3 : 1 ; the mean of two analyses gave 22.79 per cent. Ag20, and 77.18 per cent. PbO. The nearest formula is 7Pb0,24g20, which corresponds to 22.91 per cent. Ago. This sample was then treated with a strong solution of caustic soda, and the residue, on analysis, yielded 48.19 per cent. Ag20, showing that a large amount of lead oxide had been dis- solved. The Pormula PbO,Ag,O corresponds to 50.98 per cent. AgzO. A second yellow precipitate was obtained by the addition of potash to a solution of lead and silver nitrates, containing approximately 2Pb0 to lAgzO ; the mean of two analyses gave 40.68 per cent.AgzO. This number approaches 40.94 per cent. AgzO, corresponding to the formula 3P bO, 2A g 2 0 . A third sample, somewhat darker in colour, was obtained by adding a very large excess of caustic potash to a solution of lead and silver1094 ASTON ON SOME COMPOUNDS nitrates containing approximately lPbO to lAg?O ; on analysis, it gave 49.01 per cent. AgzO as the mean of two estimations, a number not very far from that obtained with the first sample after extraction with soda (48.19 per cent.). This sample mas extracted with ammonia eo dissolve any free silver oxide (for it stained the skin), and then with soda. After extraction, it contained a still higher percentage of silver oxide, viz., 57.54.Second Method of Preparation.-To ensure the presence of excess of caustic alkali, three more samples were prepared by the addition, of silver nitrate to a solution of lead nitrate to which excess of pure caustic soda had been previously added ; the colour of the precipitate obtained was yellow a t first, but quickly turned t o green in the case of samples 4 and 5 ; in the case of sample 6, it did not change from yellow to green for s, few days. In common with other silver salts, the mixed lead and silver oxide is darkened by the action of light, so that the colour is apt to vary, unless the substance is kept in the dark. The fourth sample was made by adding silver nitrate to lead nitrate in excess of soda, in such amount that the ratio of lead oxide to silver oxide was about 1.5 to 1; the precipitate obtained gave 56.81 per cent.Ag20. The fifth sample was made by adding silver nitrate t o lead nitrate in excess of soda, in such amount that the ratio of lead oxide to silver oxide was nearly 5 to 1 ; the precipitate, on analysis, gave 49.01 per cent. Ag,O, a number agreeing with that obtained from sample 3 (49.01), and also very near sample 1, after extraction with soda (48.19). The sixth sample was prepared with a still larger proportion of lead oxide, 9Pb0 to 1AeO ; the precipitate obtained gave on analysis 47.46 per cent. of AgzO. Third Method of Srepuratiom-A sample was then prepared accord- ing to Wiihler’s method, by the addition of caustic soda to a solutior, of a lead and silver salt, and the precipitate repeatedly extracted with soda, till no more lead oxide was dissolved; the precipitate, which was originally yellow, turned first green and then black.On analysis, it yielded 66.29 per cent. AgzO; the nearest formula, 2Agz0,Pb0, corresponds t o 67.53 per cent. AgzO. Fourth Method of Preparation.-Another way was then tried. Lead hydrate and silver hydrate were mixed together, and allowed to stand for some weeks with a solution of soda; the mixture turned dark- p e e n ; it was then repeatedly extracted with soda till no more lead oxide was dissolved. On analysis, it gave 65.39 per cent. AgaO, a number not very far removed from that required by the formula 2Agz0,Pb0, viz., 67.53 per cent. Ag20.OF THE OXIDES OF SILVER .AND LEAD.1095 Tabular Xtatement of Besults. Method of preparation. (i.) Mixture of lead and silver salts, preci- pitated by caustic potash. .......... (ii.) Addition of silver ni- trate to lead hydrate dissolved in cxccss of caustic soda.. .. (iii,) Method same as (i), but repeatedly ex- tracted with soda.. (iv.) From lead hydrate and silver hydrate by addition of soda solution. ......... Found k 2 O . Sample. I-- -- 22 *79 !- 1 40.68 56 -81 49 -01 47 -46 Ag20 calculated for forinula. 22 -91 7Pb0,2Ag20. 40-94 3Pb0,2Ag20. 50 ’98 YbO,Ag2O. 6’7 -53 Pb0,2Ag20. These results seem to show :- 1. Under the conditions given by Wohler, the substance formed varies in composition, the highest percentage of silver obtainable being 49.01, whereas the amount required for Wohler’s compound is 34.23.2. When a mixture of lead and silver hydrates is allowed to stand in presence of soda, a change of colour occurs, and after repeated extraction with soda, a substance is left having a composition nearly that required by the formula 2Agz0,Pb0. A compound of similar composition is obtained by precipitating the mixed nitrates of lead and silver, and exhaustively extracting with soda. I also tried to obtain compounds of lead oxide with the oxides of barium, strontium, sodium, and potassium respectively, but without success. When a solution of a lead salt is added to a strong boiling solution of barium o r strontium hydrate, crystalline scales of litharge of a yellowish-grey tinge are deposited on slight cooling. In the case of sodium and potassium hydrates, the solution had to be evaporated to a small bulk, and the crystals were cot so well defined. Some attempts were made by evaporating a mixture of lead acetate and the hydrate over sulphuric acid, but without success ; the result was in some cases a crystalline deposit OP litharge, in others, crystals of the hydrate containing a little lead separated out, but of no definite composition. University College, London.

ISSN:0368-1645    

DOI:10.1039/CT8915901093

出版商:RSC    

年代:1891

数据来源: RSC

摘要:

1096 XCV.-A New Nethod of Preparing p-DinaphtJLylene Oxide, C,,H,,O, and the Constitutioia of its Tetrasulphonic Acid. By W. R. HODUKINSON and L. LIXPACH. THIS substance has already been prepared in a variety of ways: by heating naphthol with lead oxide (Graebe, Kiiecht, and Unzeitig, Amzalen, 209, 138), by long boiling of naphthol in contact with air (Merz and Weitli, Uer., 14, 200), by heating naphthol with phosphoric anhydride, and by t,he dry distillati011 of calcium naphtbyl- ate (Ber., 15, 1122). Our new method of formatioil is to heat the sodium salt of Schaeffer's p-naphtholmonosulphonic acid in a metal retort, platinum o r copper, to low redness. The distillate condenses almost imme- diately to a yellowish solid. The total yield from t'ne dry sodium salt is considerable, but was not determined quantitatively ; the main portion of this distillate was found to consist of I, /%naphthol ; 11, p-dinaphthyl ether ; and 111, dinaphthylene oxide.To separate these three compounds, the distillate was first treated with dilute sodium hydrate solution, in which about 50 per cent. dissolved, and on acidify- ing the alkaline solution, a very pure p-naphthol, melting at 122", was precipitated. Little else appeared t o have been dissolved. The psriion insoluble in soda was then boiled with glacial acetic acid, in which it dissolved completely, and, on slightly cooling, a yellow, fluorescent substance crystallised out , which, on recry stallisation from acetic acid, formed very fine, rhombic plates melting at 153". Analysis :- 0.1850 gram substance gave 0.6080 gram CO,, and 0.0755 gram HZO.Found. Theoyy. C2". ....... 89.62 per cent. 89.55 per cent. H12. ....... 4.53 ,, 4.48 ,, 0 ......... - 9 , 5.97 y, With picric acid in alcoholic solution, a compound is formed which melts atl 134". The acetic acid solution evaporated to some extent, on cooling, gave colourless crystals of a substance of low melting point, which, on recrystallisation from dilute alcohol, was obtained in bright, silky plates melting at 105". The analysis of this substance and its physical properties show it to be dinaphthyl ether. It forms groups of dark-red needles.METHOD OF PREPARING P-DINAPHTHYLENE OXIDE. 1097 0.2010 gram substance gave 0.8536 gram CO, and 0.0350 gram H,O. Found. Theory. C z o . . . . . . . . HI4... . . . . 5-25 ,, 5.18 ,, 9 9 5.94 ,, 0 . . . * . * . . . 88.67 per cent. 88.88 per cent. - The yield of dinaphthylether t h u s obtained is between 4 and 5 per cent., and that of dinaphthylene oxide about 25 per cent. of the total crude distillate. The dinaphthylene oxide agrees in all its properties with the com- pound described by Knecht and Uneeitig ( B e y . , 13, 1724, and (Anlza'leia, 209, 138). I t melts at 153", and boils without any decom- position at a temperature close upon, i f not above, 500". This method of formation of dinaphthglene oxide from Schaeffer's p-naphtholmonosulphonic acid points quite clearly to its constitution. /\/\OH H I . Schaeffer's acid has the constitution \/\/ HSO, The sodium salt of this acid when heated 'gives sodium siilphite.The two residues unite t o form dinaphthol and diiiaphthylene +oxide respectively. As a matter of fact, dinaphthol does exist amongst the products of the distillation, but only in very small quantity, for, as might be expected, the high temperature at which the reaction takes place causes the $elimination of water and forniation of dinaphthylene oxide. Dinaphthol. Dinsphthylene oxide. Dinaphthylene oxide is easily sulphonated, either by warming with ordinary strong sulphuric acid, or by dissolving in chloroforni and treating with sulphuric acid mixed witli SO per cent. of the anhydride ; or by the action of sulphonic monochloride, S0,HCl. As the sulphona- tion proceeds, the colour changes from blue t o brown, and the solu- tion of the acid in water in also brown.The same stage of sulphona- tion is obtained by all the above mentioned methods, namely, a tetra- sulphonic acid.1098 HODGKINSUN AND LIMPACH: h NEW METHOD The barium salt of this acid is very soluble in water when first made, and the solution when filtered through animal charcoal is not fluorescent. After evaporating to dryness, the salt is not quite so soluble in water. When slowly crgstallised from the original solution, it contains 10 mols. HzO, which can be driven off at 120". The less solnble salt conta-ins but 2 mols. H,O. 0.1229 salt lost 0-0218 = 17.8 per cent. water at 120". 0.1056 salt lost 0.0186 = 17.6 ,, ,, ,, (barium lost). 0.2402 ,, 0.0412 = 17.1 ,, ,, 7, and gave 0.1070 Bas04 = 26.2 Ba per cent. C,,H,(S03)4Ba2,10H20 requires 17.6 per cent.water and 26.8 barium. The ammonium and other salts of this acid have been made, but no particular interest attaches t o them. This sulphonic acid is of interest, because it can be obtained directly from ,&naphthol. When P-naphthol is heated with sulphuric acid to a high temperature (120-150"), there is always produced along with the /.3-naphtholsulphonic acid a certain quantity of slxlphonic acid, which does not react with diazo-compounds, and con- tains therefore no hydroxyl group. The amount of this acid depends on the extent of heating with the sulphuric acid. That these sulph- onic acids are derivatives of dinaphthylene oxide can be shown in the following manner :- The mixture of naphtholsulphonic acids obtained by sulphonating naphthol at a high temperature is made into an ~lkaline salt, and after the removal of the excess of sulphuric acid with barium carb- onate, treated with a diazo-compound.For this purpose diazo- benzidine is very convenient, as it forms coloured compounds with all the isomeric iiaphtholsulphonic acids. These compounds can be precipitated by the addition of salt to their aqueous solutions. The filtrate, which should give no colour reaction on the addition of a little more of the diazo-compound, is somewhat concentrated by evaporation, mixed with an equal bulk of dilute sulphuric acid, and heated in a sealed tube or other closed vessel at a moderately high temperature, 150" or thereabout. By this means all the sulphonic groups are eliminated, and the dinaphthylene oxide separated in a crystalline form.After one or two crystallisations from acetic acid, it is obtained quite pure of m. p. 153", and is otherwise also identical with the product from Schaeff er's acid above mentioned. This process can even be watched, as the /3-naphtholtrisulphonic acid is gradually converted into dinaphthylene oxide tetrasulphonic acid :- 1. By analysis at different stages of heating, when it is observedOF PR~PAHIKG P-DINSPHTHYLENE OXIDE. 1099 that both the sulphur and the barium (in the barium salt) diminish in amount. 2, By the difference in the colour of the fluorescence of the acid solution, the green fluorescence of the trisulphonic acids passing into blue as the dinaphthylene oxide acids are formed. 3. In a more definite manner by heating naphthol with five times its weight of slightly fuming sulphuric acid quickly t o 120-139", and keeping it a t that temperature for one hour.It is then converted completely into the trisulphoiic acid. Small weighed quantities of this are taken, rendered alkaline, and titrated with a solution of diazobenzidine of known strength, as long as the colouring matter is fdrmed. From the amount of diazobenzidine solution used, the quantit,y of the trisulphonic acid can be calculated. If the heating be continued past this stage, sulphur dioxide is formed, and the amount of trisulphonic acid regularly diminishes. This can also be watched by taking analyses a t different stages. An analysis of the barium salt of a portion immediately after the stage of formation of the trisulphonic acid gave Ba 34.85 per cent.and S 16.15 per cent. Barium p-naphtholtrisulphonate requires Ba 35.04 per cent. and S 16.37 per cent. After several hours continued heating a t 140-150°, an amalysis of a barium salt gave considerably less barium : Ba 32.6 per cent. and S 14% per cent. Barium di nap hthy lene oxide te trasul phonate, C,,H, ( S 0,) 4B %, dried at 120°, requires Ba 32.54 per cent. and S 15.2 per cent. It would appear, therefore, thak in this action, as in the distillation of the sodium salt first mentioned, the SO, groups in the para-posi- tion relatively to the hydroxgl are successively removed with simul- taneous joining up of the two naphthol rings. The constitution of this tetrasulphonic acid is also clear from the fact that all the isomeric mono- or di-sulphonic acids of P-naphthol give one and the same trisulphonic acid on further, somewhat drastic, treatment with sulphuric acid; (Timpach, Rer., 16, $26) ; see also Nietzki (Chewz.Zeit., 15, 296 ; also Jour. Chem. Id., 10, 6, ,536). As this acid must have the constitution the only pozsible constitution for dinaphthylene oxide tetrasulphonic acid is YOL. LIB. 4 G1 LOO METHOD OF PREPARING P-DINAPHTHTLENE OXIDE. Dinaphthylene oxide is brominated tolerably easily either i n acetic or chloi*oforrn solution ; with a large excess of bromine, a product was obtained crystallising in very slender needles, and almost colourless ; i t melts at 231". On analysis, 0.2578 gram substance gave 0,3353 gram AgBr, or The formula CloH,5r40 requires 55-17 Br. It is therefore a tetra- hromo- derivative.When less bromine is employed, evidently some lower brominated derivatives are formed ; these have lower melting poir-ts than the one mentioned, but were not examined further. Chlorine in large excess acting on dinaphth-ylenc oxide in chloroform Polution gives a chlorine derivative of similar appearance t o the bromine one. Dinaph thylene oxide dissolves in glacial acetic acid very easily, and on adding strong nitric acid to this solution without heating, a nitro-derivative is precipitated in fine, red needles, melting at 185". 0.2658 gram substance gave 10.3 C.C. N at 763 mm. and 17.4" = 4.47 per cent. N ; C,,H,,NO, requires 4.47 per cent. of nitrogen. Attempts to reduce this substance have not as yet been siiccessful. On boiling with strong nitric acid for a little time the nitro-derivative dissolved, and on cooling separated in crystals not very distinct in shape. It was washed with hot acetic acid, dried, and analysed. 0.1264 gram substance gave 13.9 C.C. N a t 755.2 mm. and 18", or 0.1036 gram substance gave 11.4 C.C. N at 754 mm. and 17", or 12.66 C,H,N,O, requires 12.5 per cent. nitrogen. It is therefore a tetranitro-derivative. This compound begins to melt at about 250", but decomposes at the same time. The mononitro-derivative, when treaked with excess of bromine, gives a yellow bromonitro-derivative forming rhombic prisms, some- what resembling nailhead spar in appearance. 55.35 per cent. of bromine. It melts at 220". 12.5 per cent. nitrogen. per cent. nitrogen. It melts at 295".

ISSN:0368-1645    

DOI:10.1039/CT8915901096

出版商:RSC    

年代:1891

数据来源: RSC

摘要:

INDEX OF A UTHO-RS’ N-4 M E S. T R A N S A C T I O N S . 1891. And also t o such papers as appeared in the Proceedings during tElc Session 1S90--1891 (Nos. 87-100, November, 1890, t o Jnly, lSSl), but not in t h e Transactions (marked PROC.). A. Ad i e. R. H., a direct comparison of the physical constants involved in the determination of m o l e d a r wcights hy R:toult’s method, PROC., 1891, 26. .- compoiinds of the oxides of plios- phorus with eulphuric anhjdride, 230. - osmotic pressure of salts in solu- tion, 344. A r m s tro n g , H. E , terpenes and ailied compounds : the nature of turpentinc oils, including that obtained from Pim~rs khnsyanu, 311. __ the formation of Fit1ts-a contri- hiition t,o the theory of clectrolysis and of the nature of chemical change in the case of non-electrolytes, PROC., 1891,118.- the function of chlorine in acid chlorides as exemplified by sulphuryl chloride, PROC , 1891, 60. A r m s t r o n g , H. E., and W. J. P o p e , terpenes and allied compounds j sol,reroI, a product of the oxidation of terebenthene (oil of turpentine) in sunlight, 315. A r m s t r o n g , H. E., and E. C. Ros- si t e r , a new method of preparing nitro drrivatires : the use of nitrogen dioxide ns a nitrating agent, PROC., 1891, 91. -- brorno-derivatives of p-naph- thol, PRO~., 1891, 8’7. -- cliloro- and bromo-deriva- tives of naphthol and naphthylamine, YROC., 1891, 32. -- the action of nitric acid on 11 aphtliol-de1,irntivc.s as indicative of the manner in which nitration is effected in the case of benzenoi’d com- pounds genernily ; the formation of nitro-keto-compounds, PROC., 1891, 89.A r ni s t r o n g , H. E . , a nd W. P. W y n n D , studies on the constitution df tri- dcrivntivcs of naplit,hnlenc. No. 9. Andresen’s nuplitliyinrr~inediaulpll- onic acid, PROC., 1891, 27, B. Bassekt, II., eulyte and dyslyte (‘I B e v a n , E. J. B i s h o p , A. W., and W. H. P e r k i n , juii., the action of ethyl dichlor- acetate on the sodium derivative of ethyl nialonatr, PROC., 1891, 41. Rrsclr, H. R., obitnary notice of, 452. B r a u n e r , B., volumetric cstimatioii or“ B u r t o n , C. I., obituary notice of, 453. correction), 978. See C. F. Cross. telluriuni, 58, 2.38. C. C a i n , J. C., and J. R. Cohen, action of ncetic acid on plienylthiocarbamide, 327. C a r 11 e 11 e y, T., obitnary notice of, 455.C a r n e n t e r , W. L., obituary notice of, 461. Chapman, A. C., compounds of dex- trose u-ith the oxides of nickel, chrom- ium, and iron, 323. Cleve, P. T., the formation of an explosive substance from ether, PROC., 1891, 1.5. C o h en, J. B., dibenzanilide, 67. C o h e n , J. B. See also J. C. Cain. Colefnx, A., phenuvic acid: its con- stitution and relationship with Paal’9 plicnylniethylfurfurancarboxylic acid, 190. C o l l i e , J. N., action of heat on ethyl P-arnidocrotnnatc, 172. 4 6 21102 INDEX OF AUTHORS. C o l l i e , J. N., constitution of dehydr- - lactone of triacetic acid, 607. - reziotions of dehydracetic acid, 61'7. C o u s i n s , H. H. Cross, C. F., and E. J. B e r a n , the srrtion of nitric acid on the lignocel- luloses, PROC., 1891, 61.C u n d a l l , J. T., the dissociation of liquid nitrogen peroxide, 1076. acetic acid, 179. See J. E. Marsh. D. Day, T. C., the inflnence of tempera- ture on germinating barlcj, 664. D i x on, A . E., new benzyl derivatives of thiocarbamide, 551. D iifton, 8. F., orthoquinolinehydr- srzirie, 756. D i i f t o n , S. F. See also S. R u h e - mann. D u n s t a n , W. R., and T. S. Dyinond, action of alkalis on the nit,ro-com- pounds of the paraffin series. Forui- :ition of isoxazoIes, 410. D u n s t a n , W. R., and W.H. I n c e , aconite alkaloi'ds. Part I. Cryst al- line alkalo'id of A conilum napellus, 271. Dyniond,T. S. See W. R. D u n s t a n . E. E n s t e r f i e l d , T. H., oxidation of man- nitol by nitric acid. d.-Manncsac- clim*ic acid, 306.- phenylbroinacetic acid, an apl)a- j*ent exception to the Lt: Bel-Van't Hoff hypothesie, '71. F. See R. Meldola. F o r s t e r , M. 0. Fisankland, P. P., and W, g r e w , optically active glyccric acid, 96. -- - fermentation of calcium gly- cerate by Bacillus ethaceticus, 81. P r a n k l a n d , P. F., A. S t a n l e y , and W. F r ew, fermentations induced by the Pnc umococcus of Ibiedllnder, 253. F r e w , W. See P. F. F i - a n k l a n d . G. G a r d n e r . J. A. P u l i i n g e r. Fee J. E.Marsh, F. G l a d s t o n e , J. H., molecular refrac- tion and dispersion of various sub- stallces, 290. - molecular refraction and dispersion of various substances in solution, 589. G o r d o n , H., studies on the formation of substitution deriratires, PROC., 1891, 62.Grecii, -4. G., and T. A. Lawsoii, the ortho- and para-nitro-deri~atives of paratoluidine, 1013. H. H a y r o w , G., rapid method of estimat- ing nitrates in potable n aters, 320. H a r t l c y , W. N., spwtra of blue and yellow chlorophyll, wit11 some ob- servations on lcaf-green, 106. Henderson, G. G., on diphenyli~o- succiiiic acid and diphenylpropionic acid, 731. H e wi t t, J. T., chlorinated phenyl- hydrazines, 209. - cit raconfluoresce'in, 303. Hegcock, C. T., and F. H. N e v i l l e , on the freezing points of triple alloys of gold, eadniiuni, and tin, 936. H o d g k i n s o n , W. H., and L. L i m - pacli, a new method of preparing dinaphthylene oxide, C20H120 ; siid the constitution of its tetrasulphoiiic acid, 1096. H o r i , E., and H.F. M o r l e y , normal and iso-l)rol~~lpai"i~tol~~i(lines, 33. ITuglies, F. See It. Meldola. H u g h e s , R. E., the action of alumin- ium chloride on EenzeiioYd acid chlorides, PYOC., 1891, 70. H u n t l y , G. N., action of phosphorjl chloride 011 phosphorus pentoxide, 202. H u t c h e s o n , J. E., obituary notice of, 4463. I. I n c e , W. H. See W. R. D u n s t a n . J. J a p p , F. R., the gravivolumeter, an instrument by iueans of which t,tie observed rolume of any single gas gires directly the weight of the g a e , 894. J a y p , 3'. R., and F. K l i n g e m a n n , etllyl pherianthroxyleneacetoacetate, 1.ISDES OF AUTHORS. 1103 K. K i p p i n g , F. S., and J . E. Mackenzie, ethyl a a'-dimeth yl-o a’. diucet yl pirn el- ate and its decomposition products, 569.K i p p i n g , F. S., and ?V. H. P e r k i n , jun., action of reducing agents on aa’-diacetylpentane : synthesis of dimethyldihydroxyheptamethy lene, 214. K l i n g e m a n n , F., and W. F. rJay- cock, action of ammonia and of mcthylaniine on the oxylepidens, 140. K l i n g e m a n n , F. See also F. R. Japp. L. L a n g e r , C. See L. Mond. Lawson, T. A. Laycock, W. F. See F. K l i n g e - L i r n p a c h , L. See W. R. Hodgkin- See A. G. Green. i n a n n . so 11. M. Mc Gowan, G., iodometric estimation of nitric acid in nitrates, 530. Mackenzie, J. E. See J?. 8. K i p - p i n g . Marsh, J. E., and H. I€. Coilsins, the sulphonic deriratives of camphor, 966. -- researches on the tcrpenes. I T . On turpmtine, 725. __- researches 011 the terpenes.On camphene, G4S. M a r s h a l l , W., oxidation of cobalt salts by electrolysis. 760. - the peraulphatep, 771. M a r s h a l l , T. R., m d W. H. P e r k i n , jun., the sjnthetical forniation of closed carbon c*hains. Part 9 (cont.). The action of ethylene brouiide on the sodium compounds of ethyl aceto- acetate and ethyl benzoylacetoacetate, 858. M a r s h u l l , W. See T. Purclie. Masson, O., and U. T. $1. W i l s - more, does magnesium form com- poiinds with hydrocarbon radicles ?, PROC., 1891, 16. Mat thews, F. E., the a- and p-mndifi- cations of benzene hexacliloride, 165. M r l d o l a , R., and F. Hnghes, RZO- derivatives of P-naplithglamine, 378. M e l d c l a , R., and M. 0. F o r s t e r , researches on the triazine series, 678. M i l l a r , J.H. See J. J. Sud- b o r o 11 g h. M o n cl, L.. a d C. L n 11 g e r, iron mrb- onjls, 1090. Moiid, L., and F. Q u i n c k e , volatile compound of iron with csrbonic oxide, 604. Moody, G. T., combustion of mag- nesium in water wpour, Pnoc., 1891, 20. Morley, H. F. See E. Hori. M o r r e l l , R. S. See S. R u h e m a n n . N. N e v i l l e , F. I€. N i c h o 1 son, E. C., obituary notice of, See C. T. Heycock. 464. 0. 0 8 t w a1 d, W., magnet,ic rotation, 198. O ’ S u l l i v a n , C., researches on tlie Part 11. O ’ S u l l i ~ a n , C . , and F. W. Toinp- gillm of the arabin group. Geddic acids, gedda gums, 1029. son, estimation of cane sugar, 46. P. P e r k i n , A. G., action of nitric acid on P e r k i n , P. M., derivatives of pipcr- P e r k i n , TV.H., the niagnetie rotation of saline solutions, PHOC., 1890, 141. - tlie magnetic r0tator.y power o f solutions of animoninm and Sodi iiin salts of some of the fatty acids, 981. - tlie refractive power of certiiin organic compounds a t different teni- peratures, PROC., 1891, 115. P e r k i n , W. H., jun., acetglcarbinol, 786. - action of nieth-ylene: iodiclc on t h o disodium compound of ethyl peiitaiie- tetmcarboxglate. Synthesis of hexil- nietliyiene deriva ti\ es, 798. - formation of anthraquinone from ortliobeiizoplbenzoic acid, 1012. P e r k i n , W. H.,;jun.,andB.Prentic.e, new synthesis of the hexarnethjlenc- dicarboxylic acids, 990. -- synthesis of homologucs of pentanetetr:rc.arboxylic acid and of pimelic acid, 818. P e r k i n , IT. H., Sun., and J. Stcln- h o u s e , benzoylscetic acid and swie of its derivatives.anthracene, 634. onyl, 150. Part V, 990.1104 IXDES OF SUTHORS. P e r k i n , W. II., jun. See also A. W. Bishop, T. R. M a r s h a l l , P. S. K 1 p p i 11 g. P i c k e r i n g, S. U., a recent criticism by Litpton of the conclusions dramn Asom a study of various properties of sulphuric acid solutions, PXOC., 1891, 105. - the nature of solutions as eluci- dated by a study of the densities, hcat of' dissolution, and freezing points of solutioiis of calclurn chloride, PROC., 1891, 105. - theory of dissociation into ions and its consequences, PROC., 1890, 170. Pope, W. J. See H. E. Arnistrong. Yrentice, B. See W. H. P e r k i u , inn. F i l l l i n g e r , F., and J. A. G a r d n e r , the vupour densit) of arnnionium chloride, PXOC., 1891, 2.Y u l l i n g e r , W., volatile platinum (oompounds, 598. Yiirclie, T., and W. M a r s h a l l , addi- tion of the elements of alcohol to the etliereal salts of unsaturated acids, 4b8. Q9 Qnincke, F. See L. Mond. R. R c y n o I d s, J. E., additive compounds 01 " thiocarbamidc " which afford eridence of its constitution, 383. Richardson, A., action of light on pure ether in presence of moist oxy- gen, 51. - - decomposition of silver chloride by light, 536. Eosuiter, E. C. See H. E. Arm- strong. Euhemann, S., and S. F. D u f t o n , inucic acid. I'art IV. Action of phos- phorus pentacliloride on mucic acid, 26. - - contributions to the linom- ledge of rnucic acid, $50. Ituhemann, S., and E. S. M o r r e l l , :;Lct,ion of ammonia on ethereal salts of organic acids, 743.S. S c h r y v e r , S. B., the asyinnietry of nitrogun in Substituted aininonium coinpounds, Pmc., 1891, 39. S hieid s, J., preparation and propertic3 of ethyl hydrogen fuuiarate and eth: 1 hydrogen maleate, 786. S m i t h, H., obituary notice of, 465. S p r u g u e, C. T., ethyl thiacetoucetat c , Stanley, A. Stenliouse, J. See VI'. H. Perkill, j un. Sudborough, J. J., action of nitros!l chloride on metals, 655. Sudborough, J . J.,and J. H. M i l l a r , action of heat on rittrosyl chloride, 329. See P. F. F r a n k l a n d . 73, 270. T. Thorpe, T. E., and A. E. T u t t o n , phosphorous oxide. Part 11, 1019. Tonipson, P. W. See C. O'Sul- l i v a n . T u r p i n , G. S., the action of picric chloride on amiues in presence of alkali, 714. T u t t o i i , A. E., crystalline form of the calcium salt of the new optically active glyceric acid, 233. - crystallographical characters of aconitine from Aconitunz riapelltrs, 288. T u t t o n , A. E. See also T. E. 'I'h o r p e. V. Vernon, H. M., a new modification of phosphorus, PBOC., 1891, 3. W. W a r i n g t o n , R.,nitrification. Part IV, 484. Weriier, E. A., action of acetic an- h j dride on substituted thiocarb- ainides, and an improved methoJ of preparing aroniattc: thiocarbamide?, 396. - interaction of phenylthiocarbimidt. with acetic and propionic acids rc- spectivcly, 544. W i l l , H., obituary notice of, 466. Wilsinore, U. T. M. See 0. TVjnne, W. P. See H. E. Arni- Y. ivl a R B o n. s t r o n g . Young, S., dibenzylketone, 621.INDEX OF AUTHORS. 1105 Young, 8., rr,olecular volumes of the satnritted vapours of benzene and of its halogen derivatives, 125. - new method of determining the specific volumes of liquids arid of their saturated vapourd, 37. - on the vapour pressures and mole- d a r volumes of acetic acid, 903. Young, S., on the vapour pressures and molecular volumes of carbon tetrachloride and of stannic chloride, 911. - vapour pressures of dibeiizyl ko- tone, ti26. - vapour pressures of mercury, 629.

ISSN:0368-1645    

DOI:10.1039/CT8915901101

出版商:RSC    

年代:1891

数据来源: RSC

摘要:

INDEX OF SUBJECTS. T R A N S A C TI 0 N S. 18 9 1. And also t o such papers as appeared in the Proceedings during the Session 1890-1891 (Nos. 87--100, November, 1890, to July, 1891), but not in the Transactions (marked PROC.). A, Acetic acid, vapoui. pressures and mole- Acetone, chlor-, action of ethyl benzoyl- Acetonephenanthraquinone, condensa- Acetopropyl alcohol, oxime of, and its Acetylbenzylthiocarbaniide, 408, 562. Acetylcarbinol, 786, 790. - preparation of, from monochlor- - reduction of, 796. Acetylcarbinyl acetate, 788. Acetyldibenzylthiocarbamide, 406. Acetylene bromide, molecular rcfraction and dispersion of, 295. Acetj ltrimethylene, action of hydrogen bromide on, 876. - hydrolysis of, 85'6. - reduction of 87'4. Acetyltrimeth.Flenecarboxylic acid, . _ - preparation of, 863.-- reduction of, 870. Acetyltrimethyleneoxime, 865. Acids, organic, action of ammonia on the ethereal salts of, 743. - unsaturated, addition of the ele- merits of alcohol to the ethereal salts of, 468. cular voliimes of, 903. sodacetate on, 191, tion of, 105. anhydride, 866. osazone of, 795. - acetone, '794. oxime of, 867. Aconine, 286. Aconitine aurochloride, 2'78. ~ crystalline, properties of 276. - crystallographical cliaracters of, - effect of heat on, 282. - gold cliloride, 279. - specific rotation of, 281. 288. Aconitum napellus, crjstalline alkaloi'i: of, 271. Alcohol, estimation of, by oxidation with potassium dichromate and sulph- uric acid, 93. Alkalokl, crjstalline, of Aconitzcm napel- Eus, 271. Alloys, triple, of gold, cadmium, and tin, freezing points of, 936.Ally1 tribromide, niolecular refraction and dispersion of, 295. Allylacetic acid, molecular rsfraction and dispersion of, 295. AllylanisoYl, ortho-, molecular refrac- tion and dispersion of, 295. Allylbenzylthiocarbaniide, 559. Aluminium, action of nitrosyl chloride on, fi59. - salts, molecular refraction and dis- persioii of, in solution, 595. Amines, action of picric chloride on, in presence of alkali, 714. Ammonia, action of picric chloride on, 715. - molcciilar refraction and dispcr- sion of, in solution, 595. Ammonium acetate, magnetic rotatory power of solutions of, 9S4. - chloride, vapour density of, 1407, PROC., 1891, 2. - formate, magnetic rotatory power of solutions of, 982. - nitroethane, 412. - persulpbate, 777. - propionate, magnetic rotatory power of solutions of, 985.- salts, molecular refraction and dis- persioii of, in solution, 5%. - compounds, substituted, asym- metry of nitrogen in, PEOC., 1891,39. Amp1 formate, molecular refraction and dispersion of, 295.INDEX OF Amylenc, molecular refraction and dis- Anhydro Acetonephenanthraquinone, Anhydroaconitine, formation and pro- - gold chloride, 285. Aniline, action of picric chloride on, Anciversary meeting, 435. Anthracene, action of nitric acid on, - ethyl nitrate, 643. --- action of hydrogen - methyl nitrate, 648. Anthraquinone. formation of, from Anthrone, nitroso-, action of nitric acid - nitrosonitro-, 639. -- action of sodium sulphide on, - pseudonitroso-, 645. Antimony, action of nitrosyl chloride Apoacoititine, formation and propert,ies Arabinantrigalactangeddic acid, 1039.Arabin group, gums of the, 1029. Arsenic, action of nitrosyl chloride on, Australcnc, change of rotation of, 727. - hydrochloride, '728. Azo-derivatives of P-naplithylamine, Azo-orthotoluidine, ortho-, 1016. Azoxyorthotoluidine, para-, 1016. persion of, 295. 105. perties of, 283. 715. 634. iodide on, 647. orthobenzojlbenzoic acid, 1012. on, 641. 640. on, 661. ot, 283. 662. 372. B. Bacillus ethncetirus, fermentation of calcium glycerate by, 81. - fyugi, 501. Balance Sheet of the Chemical Society from March 20, 1890, till March 19, 1891, 445. -- of the Research Fund froni March 20, 1890, till March 11, 1891, 449. Barium persulphate, 779. - salts, molecular refraction and dis- persion of, in solution, 595.Barley, germinating, influence of tempe- rature on, 664. Benzaldehyde, action of, on plienylthio- carbimide, 67. Benzene and its halogen derivativea, molecular volumes of the saturated vapours of, 125. 5UB.JECTS. 1107 Benzene. effect of temperature on the - liexa~*hloridz, n- and B-niodifica- -- molecular refraction and disper- - specific volume of, 44. Benzeneazo-/3-naphthylamine, action of - action of nitric acid on, 379. Benzeneazo-B-naphthylamines, nitro-, acetyl derivatives of, 37% -- formation of pseudazimides from, 378. Benzenold acid chlorides, action of aluminium chloride on, PROC., 1891, 70. - derivatives, nitrdtion of, PROC., 1891. 89. Benzoylacetic acid and its derivatives, 996. Benzoylbenzoic acid, anthraquinone from, 1012. Benzoylphenylhydrazine, PROC., 1891, 42.Benzoylpropyl alcohol, 886. Penzoyltrimethylene, reduction of, 885. Benzoyltriinet h ylenccnrboxylic acid, -- reduction of, 884. Benzoyltriphenylpropiomethylamide, - distillation of, 148. Benzylacctophenone, 1007. - oxiine of, 1008. - reduction of, 1008. Benzylammonium thiocyanate, 553. Benzylsniline, molecular refraction and Bcnzylethylphenylthiocarbamide, 564. Benzylmetatolplthiocarbarnide, 556. Benzplmetaxylylt hiocarbamide, 557. Benzylrnethylphen ylthiocurbamide, 562. Reiizyvl-a-naphthy1tliiocarbamide, 558. Benzyl-P-naphthylthiocarbamide, 559. B enzyl ort hotolylthiocarbamide, 555. Benzylp~wat~olvlthiocarbamide, 557. Benzylphenglbenzylthiocarbamicle, 567. Benzylpimelic acid, attempt to prepare, Benzylpiperidylthiocarbamide, 568. Benzylpropylene-+-thiocarbamide, 560.Benzylthiocarbamide, 552. Benz ylthiocarbimide, preparation of, 407, 552. Bismuth, action of nitrosyl chloride on, 662. -effect of various metals on the freezing point of, PROC., 1891, 159. refraction and dispersion of, 291. tious of, 165. sion of, 295. aldehydes on, 380. orthouitro-, 373. - oxime of, 858. -- oxime of, 8113. 147. dispersion of, 296. 847.1108 INDEX OF SUBJECTS. Bisparatolylmethylpyrazolone, 341. Bisphenylmethylpyrazolone, 339. Boyle’s law applied to salts in solution, Eromoform, molecular refraction and Butenylanis3Ils, para-, molecular refrac- Butenylbenzene, p-, molecular refrac- Bje-laws, alteration in, 451. 351. dispersion of, 295. tion bnd dispersion of, 295. tion and dispersion of, 295, C. Cadmium, action of nitrosyl chloride on, 657.- effect of various metals on the freezing point of, Phoc., 1890, 159. 6- gold and tin, freezing point of triple alloys of, 9%. Calcium chloride, densities of solutions -- freezing points of solutions -- heat of dissolution of, PROC., - glycerate (active), crystalline form -- fermentation of, by the - salts, magnetic rotation of, PEOC., -- inolecular refraction and dis- Caiiiphene, actiori of phosphoric chloride -- oxidation of, 649. - preparation of, 648. Camptioic acid, 649. Cauiphopyric acid, 650. - anhydride, 650. Camphor and its derivatives, molcciilar refraction and dispersion of, in solu- tion, 591. of, Pxoc., 1891, 195. O f , l’ROC., 1891, 105. 1891,105. of, 233. Bacillus ethaceticus, 81. 1890, 142. pereion of, in eolution, 595.on, 658. -- sulplionic derivatives of, 966. Camphors, bromo-, preparation of, 968. - chloro-, preparation of, 976. Camphors ulphonic acid, a- bronio-, and -- P-bromo-, salts of, 975. -- a-chloro-, salts of, 977. -.__ B-chloro-, derivatives of, 978. - chloride, a-bromo-, 974. -- a-chloro-, 978. Carbon bisulphide, effect of temperature on the refraction and dispersion of, 291. - dichloride, molecular refraction and dislersion of, 295. its salts, 971, Carbon, tetrachloride, molecular refrac. -- specific volume of, 43, 45. -- vapour pressures and mole- cular volumes of, 911. Carbon chains, closed, synthetical for- mation of, 853. Carbonic anhydride, influence of tempc- rature on the production of, by gcr- minating barley, 664. Carbonic oxide, compounds of, with iron, 1090.-- volatile compound of iron with, 604. Carbonyl bromoplatinite, 603. - chloroplatinit~s, 598. Cedrene, molecular refraction and dis- persion of, 295. Cerinm chloride, molecular ref ractioii and dispersion of, in solution, 595. Chloric acid, molecular refraction and dispersion of, in solution, 593. Chlorine, function of, in acid chlorides, as exemplified by sulphuryl cliloridc, Pxoc., 1891, 00. -- water, influence of hydrochloric acid on the decomposition of, by light, 539. Cliloroform, molecular refraction atid dispersion of, 295. Chlorophyll, blue and yellow, separation --- spectra of 106. - spectra, comparison of, 113. Chromium dextrosate, 324. - sulphate, molecular refraction and Cinchonine mucate, 754. Cinnaniene, molecular refraction and Cinnarnic clilori:ie, action of atluminiun t Citraconfluoresceh, 301.Cobalt balts, oxidation of, by electrolysis, Cobaltic ammonium oxalste, 769. - nitrate, 770. - sulphate, 768. Copper, action of nitrosyl chloride on, tion and dispersion of, 295. of, 109. dispersion of, in solution, 595. dibpersion of, 295. chloride on, PROC., 1891, 71. 760. 658. D. Dehydraeetic acid, action of sulpliuric -- constitution of, 179. -- preparation and propert ic-s reactions of, 617. DehSdracetonephenauthraquinone, 105. acid on, 609. Of) 618. --INDEX OF SUBJECTS. 1109 Dextrose, compounds of, with the oxides of nickel, chromium, and iron, 323. Dextrosobrerol, 3 17. Uextroterebenthene, 313. 1 )iacetylpentane, ual-, action of reducing Uialiylacetic acid, molecular refraction Diamylene, molecular refraction and Diarabinantrigalactangeddic acid, 1038.Uibenzalpimehc acid, 850. Uibenzanilide, 67. Dibenzoylacetic acid, action of Bydr- v- reduction of, 1001. Dibenzoylstilbene, action of alcoholic L_ action of methylamine on, 146. Dibenzoylstilbenimide, 104. Uibenzyl ketone, 621. -- vapour pressures of, 626. - molecular refraction and disper- Uibenzylpentanetetracarboxylic acid, 1)ibenzylpimelic acid, 846. Dibenzylthiocarbatnide, action of acetic anhydride on, 40. Uidymiuni salts, molecular refraction and dispervion of, in solution, 595. Diethylamine, molecular refraction and dispersion of, 895. Dietliylainmonium bromide, coinpound of thiocarbaniide with, 389. Uiethylpentanetetracarboxylic acid, 833. Uiethylpimalic acid, 835. Diet,hylthiocartrbamide, action of acetic X)iferrolieptacarbonyl, 1092.bigalactangeddic acid, 1057. ~~iliyc~roxybenzoylacetic acid, PROC., Dihydroxydibenzylacetic acid, 1001. L)iisobutylpilnclic acid, 843. Uiisopropylpimelic acid, 840. Diniethyl clipropyl glycol, 875. l h n e t h j lacetylcaproic acid, aa-', 570, l)iiuethyldiacetylpentpne, aa-', 570, 587. - dioxiine of, 588. L)imethyldihrouioheptametliylene, 223. l~irnethyldihydroxyheptaniethvlen~, ac- - action of phenylhydrazine on, - condensation product of, 228. - constitution of, 221. - preparation of, 217. - sod! uni derivative of, 220. - synthebis of, 614. agents on, %14. and dispersion of, 895. dispersion of, 895. oxylainine on, 1004. animonia on, 142. tion of, in solution, 591. 844. anhydride on, 409. 1891, 43. 584. tion of hjdroxylamine on, 221 221.Dimethylheptamethylene, 22'7. - diacetate, 225. - glycol, 217. Dimethyli ydroxyiodoheptaniethylene, Dimethyloximidocaproic acid, 536. Dimethylperitanetetracarboxylic acid, DimethyIpimelic acid, aa-', 570, 577, Dimathylpyridine, aa'-, 177. - oxidation of, 178. Ihnethylpyridone, a d - , 177. Dinaphthylene oxide, p-, new method of preparation of, 1096. -- njtro-, 1100. -- tetrabromo-, 1100. - -- tetranitro-, 1100. Dinaphtliylene-oxide-tetras ulphonic Diphenylamine, action of picric chloride Diphenylisosuccinic acid, 731. .__- preparation of, 732. Diphenylmethanc, brorno-, preparation Diplienylnaphthotriazine, and its deri-ia- Diphenyl- % : 5-phenylpyrrholidone, 3-, Dipheny l-4 : 5 -pheny lpjrrholone, 3 -, Diphenylpropionic acid, f?-, 731. --- preparation of, 734.Diplienylpropyl alcohol, 1009. Dipheng.lthiocarbamicle, action of acetic - action of water on, 328. - symmetrical, action of acetic an- Dipropylarnine, molecular refraction and Dipropyipimnlic acid, 838. Dispersion, molecultu, of various sub- stances in solution, 589. - of various carbon compounds, 290. Dispersive power of organic compounds, 290. Dissociation into ions, theory of, a d its consequences, PROC., 1890, 170. - of liquid nitrogen peroxide, 1076. Dif olylthiocarbamidt, meta-, actioii of - Ortho-, action of acetic anhydride Dixgl~lthiocarbaniide, meta-, action of Dyslyte, 978. 224. 830. 587, 836. acid, constitution of, 1099. on, 716. of, 731. tlves, 681, 146. 144. acid on, 329. hydride on, 396. dispersion of, 296. acetic anhjdride on, 4 ~ 3 .on, 402. acetic anhydride on, 404.1110 INDEX OF SUBJECTS. E. Electrolysis, oxidation of cobalt salts by, 760. - theory of, PROC., 1891, 118. Ethane, nitro-, action of alkali carb- onates and hydroxides on, 411. -- action of animonia 09, 412. Ether, action of light on, in presence of nioist oxygen, 51. - formation of an explosive sub- stance from, PROC., 1891, 15. - influence of temperature on the formation of hydrogen peroxide from, 56. Ethyl ncetyltetramethylenecarboxylnte, molecular refraction and dispersion of, 295. - acetyltrimethylenecarboxylate, ac- tion of isoamyl iodide and sodium ethoxide on, 892. -- molecular refraction and dis- persion of. 295. -- acrylate, action of ethyl alcohol on, 475. - allylscetate, action of alcoholic sodium ethoxide on, 452.- allylmethylbenzoylacetate, 999. - amiducrotonate, action of heat on, - amidoethylenedicarboxylate, 747. - sngelnte, action of alcoholic sodium - benzoylacetate, condensation of, - beri zoylsodacetat e and chloracetone, - benzylbenzoylacetate, 1006. - benzyldicarboxyglutaconate, action - bromide, niolecular refraction and - carbamute, action of parachloro- - chlorofumarate, molecular refrac- - chlorop henylhydrazinepyruvate ---_ crotonate, action of ethyl alcohol - dibenzoylacetate, action of phenyl- -- preparation and properties of, - dibenzoylmethylanetate, 1005. - dibenzylpentanetetracarboxylate, - clibronihydromuconate, 752. - dibromopentanetetracarboxylate, 172. ethoxide on, 482. with furfuraldehyde, 1011. l*riiction between, 191. of ammonia on, 748.dispersion of, 295. phenylhydrazine on, 211. tion and dispersion of, 295. (ortho-), 211. on, 478. hydrazine on, 1005. 1000. 843. 827. Ethyl dicarboxyglutnconate, action of animoiiia on, 745. - dichloracetate, action of, on the sodium derivative of etliyl malonate, PROC., 1891, 41. - di ethy lpentane te tracarboxy late, - clicthylpimelate, 834. - diisobutylpentanetetracarboxylate, - diisobutylpimelat e , 842. - diisoprop ylpentanetetracarboxyl- - diisopropvlpirrielate, 8 10. - aa'-diniethyl-aa'-diacdylpimelate, and its decomposition products, 569. dihjdrazone of, 573. 833. 841. ate, 839. -- -- hydrolysis of, 580. - -- preparation of, 571. - dimethylpentanetetracarboxylate, - ad-dimethylpimelate, 59'1, 575, - diphenylisosuccinate, preparation - diphenglpropionate, 735.- dipropylpentanetetracarboxylate, -- dipropylpinielate, 837. - d i sodiopcnt anetctracarboxylnt e, action of benzyl chloride on, 850. -- action of methylene iodide on, '798. -- action of trimethylene brom- ide on, 99%. - ethylcnetricarboxylate, a-, PROC., 1891, 41. ~ fuwaratc, molecular refraction and dispersion of, 295. - furfriralbenzo~laeetate, 1011. - glutaconate, action of ammonia on, - hexameth ylcnet etracarboxglate -. hydrogen fumarate, preparation - -- maleate, preparation and pro- - iodide, action of, on magnesium, -- nioleciilar refraction and dis- - isophenanthroxyleneacetoacetate, -- action of acetic anhydride -- - activn of bromine on, 8. -- action of phenylhydrazine on, 829. 831. of, 731. 836. 745. [1 : 1 : 3 : 31, 803. and properties of, 736.pertiev of, 740. PROC., 1891, 17. persion of, 295. 2, 5. on, 7. k -- hydrolysis of, with caustic alkalis, 11.1111 IXDES OF SUBJECTS. Ethyl isophenanthroxglenencetoacetate, reduction of, with hydriodic acid, 10. -- reduction of, with zinc and hydrochloric acid, 8. - lutidonecarboxylate, 174. - maleate, molecular refraction and dispersion of, 295. - malonate, action of etht-1 dichlor- acetate on the sodium derivative of, -- methacrylnte, action of alcoholic - inethyldibenzoylacetate, PROC., - nitropiperonylacrylate, 156. - pentanetetmcarboxylate, prepars- -- sodium derivative of, PROC., -- spthesis with the aid of, - phenant,liroxyleneacetoacetate, 1. -- action of acetic acid on, -- action of alcoholic hydrogcn -- - action of alcoholic potash on, -- action of ammonia on, 25.-- action of formic acid on, 3. -- action of propionicacid on, I?’. -- - action of sulpliuric acid and - a-phenanthroxylcneisocrotonate, 2. - piniclate, preparation of, 8:25. - propaiietetracarboxylate, prepara- tion of, 991. - propanetricarboxylate, p-, PROC., 1891, 41. - sodacetoacetate, action of ethylene bromide on, 853. - sodethylacetoacetate, action of ethylene bromide on, 893. - sodiobenzylacetate, action of ethyl- ene bromide on, 833. - tetrabromadipat e, 753. - thiacetoacetate, 329. -- action of phenylhydrazine -- condensation of, with para- -- preparation of, 331. - thiocyanate, molecular refraction Ethylamidopiperonyl-w-carbox~lic an- Et8hyl benzylbenzoylacetate, hydrolysis EtEiylbenzylplienglthiocrtrbaxnide, 565. Ethylene bromide, molecular refraction PROC., 1891, 41.sodium ethoxide on, 481. 1891, 43. tion of, 822. 1891, 43. PROC., 1891, 43. 14. chloride on, 22. 24. alcohol on, 18. on, 332. tolylhydrazine, 339. and dispersion of, 296. hgdride, 158. of, 1007. and dispersion of, 295. Ethyl-P-phenylhydroxypropionic acid, a-. 1009. E th~~l-P-~~henyllactic acid, a- ., PROC., Ethylthiocarbimide, molecular refrac- Eulyte, 978. Explosion occasioned hy impurities in 1891, 43. tion and dispersion of, 296. commercial ether, PROC, 1891, 15. F. Fermentation of calcium gljcerate by the Bncillus rtkaceticus, 81. Fermentations induced by the Pneumo- coccus of Friedlgnder, 253. Ferric chloride, molecular refraction and dispcrsion of, in Bolution, 595. - dex t 1.0 sat e , 3 2 5. Ferropent,acarbonyl, 1091.Forrnglparachloro~henylhydrazine, 213. Freezing point of triple alloys of gold, Fumaric chloride, chloro-, niolec.nlar cadmium, and tin, 936. refraction and dispersion of, 295. G. Galactangeddic acid, 1057. Gas, determination of the weight of, from the volume, 894. Oedda gum, prote’id from, 1061. - gunis, 1029. Geddic acids, 1029. Geddinosic acid, 1041. Qeddinosic acid, p-, 1054. Germinating barley, influence of temp- erature on, 664. Glucose, fermentation of, with Fried- lander’s Pnezimococcus, 254. Glutaric acid, preparation of, 993. Glgceric acid, an opticaliy active, 96. -- - optically active, crystalline forin of the calcium salt of, 233. Glycerol, molccular refraction and dis- persion of, 295. Gold, aotlon of nitrosyl chloride on, 662. - cadmium and tin, freezing point of triple alloys of, 936.Gritvivolumeter, 894. Gums, gedda, 1029. -of the arabin group, 1029. H, HaloYd compounds, magnetic rotation of, PRoC., 1890, 142.1112 TSDES OF’ SUBJECTS. Heptane, molecnlar refraction and dis- Heptarabinanpentagalactangeddic acid, Hept A rabin antetragalac tan gedd ic acid, Her, tn rab insnt rigalac tan geddic acid, Hexamcthylene derivatives, synthesis of, Hexamethylenedirarbox~lic acids, new - anhvdride, Pa-, 812. Hexam e th vlen em etadicarboxg lic - rci.;trana, 814. - acids, conversion of one into the Hexameth~lenetetracarboxylic acid -- formation of. 994. Hexarabinanpentagalactangeddic acid, 10’74. Hexatrabinantrigalactangeddic acid, 1065. Hydriodic acid, molecular refraction and dispersion of, in solution, 593. Hpdrobromic acid, molecular refraction and dispersion of, in solution, 593.Hydrochloric arid, molecular refrartion and dispersion of, in solution. 59.7. JTvdrocinnamanilide, PROC., 1891, 71. Hydrocinnamic chloride, PROC., 1891, 71. c_ -- nction of dumininm chloride on, P R O ~ . , 1891. 71. Hydrocinnamide, P R ~ C . , 1891, ’71. Hydrogen peroxide, influence of temp- erature on the formation of, from ether, 56. Hydroxides of the alkali metals, mag- netic rotation of, PROC , 1890, 143. Hyd rox ybenzyltrimethylenecarboxjlic acid, 884. H y droxyethyltrimetliylenecarboxylic acid, 870. Hpdroxyhgdromuconic acid, bromo-, lactone of, 753. persion of, 295. L074. 1071. 1065. ’798. synthesis of. 990. acid, rcis-, 808. other form of, 813, 816, 817. [l : 1 : 3 : 31, 804. I. Invertase, preservation of, 47.Iodic acid, molecular refraction and dis- persion of, in solution, 591. Iridium chloride, molecular refraction and dispersion of, in solution, 595. Iron, action of nitrosyl chloride on, 660. - carbonic oxide, 604, 1090. - carbonjls, 604, 1090. Iron, VOhtil8 compounds of carbonic oxide with, 604, 1090. Isoaniyl oxide. molecular refraction and dispersion of, 295. I s o h n t ~ l iodide, molecular refraction and dispersion of, 295. Tsocamphopyric acid, 651. Tsophenant8hrox;yleneacetic acid, act ion Tsophenanthroxglenracet~acetic acid, 11. ~- action of potash on, 12. Tsoprolwlparatoluidine, 34. Tsopropylparat olylnitrosamine, 34. Isoxazoles, formation of, 410. of acetic anhydride on, 13. T U. Lfevosohrerol, 317. LE-croterebenthene, 313. Lanthannm salts, molecular refraction and dispersion of, in solution.595. Lead acetate, molecular refraction an11 dispersion of, in solution, 595. - action of nitrosgl chloride on, 658. - and silver oxides, compounda of, 1093. - effect of various metals on the freezing point of, PROC., 1890, 160. - persulphate, ’782. T.eaf-green, observations on, 106. Lecture experiment : Combustion of niagnesium in water Fapour, YROC., 1891, 20. Light, action of, on phosphorous oxidc, 101 9. - action of, on pure ether in prc- sence of moist oxvgen. 51. - action of, on silver chloride. ,536. - decomposition of silver chloride - influence of h?;drochloric acid on the dccornposition of chlorine-water by, 539. Ligno-cellnloscs, action of nitric acid on PROC., 1891, 61. Liquids, new method of determining t h e specific volume of, 37.Lithium salts, magnetic rotation of, PROC., 1890, 142. c_- molecular refraction and dis- persion of, in solution, 595. Luticline, 177. Lutidonecarboxylic acid, 176. by, 536. M. hgnesiwn, action of nitrosyl chloritle - combnstion of, in water vapour, on, 656. PROC., 1891, 20.INrjEX OF SUBJECTS. 1113 Magnesium, compounds of, with hydro- carbon radiclcs, expwiments on the existence of, PROC., 1891, 16. - ethvl, attempts to prepare, PROC., 1891, 17. - salts, magnetic rotation of, PROC., 1890, 142. - - molecular refraction and dis- pepion of, in soliltion, 595. Magnetic rotation, 198. - - of saline solutions, PROC., 1890, 141. - rotatory power of sdutions of am- nlonium and sodium salts of some fatty acids, 981.lhnganese, action of nitrosyl chloride on; 660. Mannitol. fermentat,ion of, with Fried- lander’s Pncumococrzcs, 256. of, in solution. 591. - nioleculnr refraction arid dispersion - oxidation of, by nitric acid, 306. Mannosaccharic acid, d-, 306. Menthol, molecular refraction and dis- persion of, in solution, 591. Bfeyrury, action of nitrosyl chloride on, 659. - vapour pressures of, 629. Metals, action of nitrosyl chloride on, 655. I_ moleciilar condition of, when al- loved with each other, PROC., 1890, 158. MetaphoslJhoric acid, molecular refrac- tion and dispersion of, in solution, 593. Metliane, nitro-, action of alkalis on, 430. 3fethoxyphenyl- rcld-phenylnaphthotri- azine, az-ortho-, 697. MPthoxysuccinamirle, 470. Mcthoxpsuccinic acid, 4’71. AIetliyI acrylnte, action of methyl al- cohcl on, 474.,--- crot,onate, action of methyl alcohol on, 476. - fumarate, action of sodium meth- oxide and methyl alcohol on, 468. -- action of sodium methoxide on, 472. - hexamethylenemetadicarboxylate, 806. - iodide, molecular refraction and dispersion of, 295. - methoxysnccinate, 468. - nitropiperonylacrylate, 156. - sulphate, molecular refraction and Methylammonium salts, compounds of, Methjlaniline, action of picric chloride dispersion of, 296. with tbiocarbamide, 392. on, 716. Methyldehydropentone, 880. - action of water on, 881. Mcthyldehydropentonecarboxylic acid, Met,hgldiphe~qdamine, molecular refrac- Methyl-3-diphenyl-4 : 5-phenylpyrrho1- Ifethylene diiodide, molecular refrac- Methylglq-col, formation of, from acetyl- 878.tion and dispersion of, 296. one, 149. tion and dispersion of, 295. carbinol, 796. M~thrl-P-phen~llactio acid, C I - , PROC., 1891, 43. Methylpropylcarbinol, formation of, 874. Molecular refraction of carbon com- - volumes of acetic acid, 903. -- of carbon tetrachloride and stannic chloride, 911. - -. of the saturated vapours of benzene and its halogen derivatives, 125. - weighte, direct comparison of the physical ronstants inrolved in the determination of, by Ranult’s method, PROC., 1891, 26. -- of metals in solution, PROC., 1890, 159. Mucic acid, 750. -- action of phosphorus penta- -- constitution of, 753. Muconic acid, 750. -- action of bromine on, 750. pounds, 290. chloride on, 26. N. Naphthalene, constitution of the tri- Naphthol, 0-, bromo-derivatives of - 1 : 3-bromochloro-P-, PROC., 1891, - di- and tri-chloro-p-, PROC., 1891, - I : 3-dibromo-P-, PROC., 1891, 33.- tetrabromo-B-, Paoc., 1891, 88. tribromo-8-, PROC., 1891, 87. Kaphthol-derivatives, nitration of, PROC., 1891, 89. Naphthols, chloro- and bromo-, action of nitric acid and oxidising agents on, Naplitliylamine, a-, action of picric chloride on, 716. ru’nphthylnmine, p-, szo-derivatives of, 372. - chloro- and bromo-derivatires of, derivatives of, PROC., 1891, 27. PROC., 189, 8’7. 33. 32, 33. PROC., 1891, 34. PROC., 1891, 32.1114 ISDES OE Naphtliylaminedisulp'nonic acid, p-, Nickel action of nitrosyl chloride on, 860. -_ dextrosate, 328. Nitrates, estimation of, in potable - iodonietric estimation of nitric - magnetic rotation of, PROC., 1890, - production of, in nitrification, 514.Nitration of benzeno'id derivatives, - with nitric peroxide, PROC., 1891, Nitric acid, iodometric estimation of, in -- molecular refraction and dis- -- organism, behaviour of, with am- -- nutrition of, 519. Nitrification, 484. - a purely nitrous agent of, 486. - conditions which determine the formation or separation of a nitrous agent only, 490. - distinction between the production of nitrites and nitrates in, 485. - isolation of the nitrous organisin - nutrition of the nitric organism of, - nutrition of the nitrous organism - production of nitrates in, 514. - theory of', 523. Nitrites, prodmtion of, in nit rificution, Nitro-compounds of the paraffin scries, Nitro-derivatives, new method of pre- Nitrogen peroxide, liquid, dissociation -- preparation of, 1077.-- use of, as a nitrsting agent, Nitro-keto-compounds, formation of, Nitrosyl chloride, action of heat on, 2'71. -- action of, on metals, 655. Nitrous organism of nitrification, isola- - .- - description of, 507. --- nutrition of, 509. --- properties of, 505. Nonarabinatetragalactangeddic acid, Son-electrolytes, nature of chemical Andresen's, PROC., 1891, 2'1. waters, 320. acid in, 530. 144. PROC., 1891, 89. 91. nit,rates, 530. persion in solution, 593. monk, 521. of, 495. 519. of, 509. 486. action of alkalis on, 4 10. paring, PROC., 1891, 31. of, 1076. PRO~., 1891, 91. PROC., 1891, 89. tion of 495. 107 I. change in, PROC., 1891, 118. SUBJECTS. 0. Obituary notices, 452. Oil of turpentine, oxidation of, in sun- Osmotic pressure of salts in solution, 34,4, Oxylepidens, action of ammonia aiid light, 311, 315.methylamine on, 140. P. Pentane, molecular refraction and dis- persion of, 295. Pentauetetracar\)ox ylic acid, preparation and propertieo of, 824. -- synthesis of homoloques of,818. Pentarabinantetragalactangeddic acid, 1070. Percliloric acid, molecular refraction and dispersion of, in solution, 59% Persulphates, 771. Phenazoxime, dinitro-, '723. Phenol, chlorination and bromination of, - diorthonitro-, action of bromine oil, - pnrabromodioi thonitro-, isomeric ~ parachlorodiorthonitro-, PBOC., 1891, 63. Phcnolorthosulphonic acid, orthopara- dichloro-, action of sulphuric acid on, Phenols, amido-, action of picric chlor- - nitro-, sulphonation of, PROC., Phenuvic acid, 190. -~ constitution of, 194.-- relationship with Paal's phenylmethylfurf urancarboxglic acid, 190. Phengl ether, molecular refraction and dispersion of, in solution, 691. - phenyleemithiocarbazide, para- chloro-, 212. Phenylbromacetic acid, an apparent ex- ception to the Le Bel-Van't Hoff hypothesis, 71. Phenjlcarbamide, diparacliloro-, 212. Phenj ldehydropentone, 886. Plienjlbydrazine hydrochloride, ortho- - orthochloro-, 209. - parabanate, parachloro-, 213. ~. paracllioro-, a l l . -- action of chloroform and PhenylhydraziuepTruvic acid, ortho- PROC., 1891, 64. PROC., 1691, 63. change of, YROC., 1891, 63. PROC., 1891, 64. ide on, 71P 1891, 65. chloro-, 209. alcoholic potash on, 213. cLLloro-, 210.INDEX OF SUBJECTS, 1115 Phenylh~draziiies, chlorinated, 209. Yhenylisouazolone, 1005.Phenyl-ald-metnnitrophenylnaphtliotri- -- metanitro-, 693. Phenylnieth ylf urfurancarboxylic acid of Pad, relationship t o phenuvic acid, 190. Phenyl-ald-methylnRphthctriazine, ax- paranitro-, 697. -- reduction of, 712. Yl~en-ylmetliylpyrazolone, action of azine, az-, 700. sulphur dichloride on, 334. --- properties of, 335. Phc nylmethylpyrazoloneazobenzeiie, P1.leiiyl-nZd-pamnitrophenylnaphthotri- -~ pareni tro - . 694. Phenyl-nld-phenrlnaphtliotriaziiie, nz-, rnetanitro-, 684. -- reduction of, 704. __ orthonitro-, 683. -- reduction of, 702. ~ parabromo- and chloro-, 690. - paranitro-, 685. -- reduction of, 702. - pal-asulpho-, 687. 3’henylproprlc,2rbinol, 886. Phenylsemicarbszide, orthochloro-, 210. P h enyltetram ethylene dibromide, 891. Phenylthiocarbiniide, action of acetic -L action of bcnzaldehydfe on, 67.- action of benzoic aciu on, 67. - action of water on, 328. - interaction of, with acetic acid, 544. - interaction of, with propionic acid, - peparation of, 398. Phenylthiosemicarbazide, orthochloro- Phenylurazole, parachloro-, 212. Phosplia tes, magnetic rotation of, PROC., Phosphodichlorotnuconic acid, 27. Phospliodichloromuconyl chloride, 31. Phosphoric anhydride, action of phos- phoryl chloride on, 202. Phovphorons diamide, 1027. - anhydride, action of sulphuric - oxide, 1019. -- action of ammonia on, 1026. -- action of bromine on, 1020. -- action of hydrogen chloride -- action of iodine on, 1021. .- - action of light on, 1019. - thlo-, 332. 336. azine, az-, 699. - gllrcol, 890. acid on, 327. 550. plienyl, 210.1890, 144. anhydride on, 230. on, 1022. TOLL. JrIX. Phosphorous oxide, action of nitric. -- action of phosphorus trichlor- -- - action of phosphorus penta- .-- - action of selenium on, 1026. -- action of sulphur on, 1022. -- action of sulphur cbloride on, 1026. -- - action of sulphuric acid on, 1026. Phopphorus, a new modification of, PROC., 1891, 3. - action of, on sulphuric anhydride, 231. - oxides, compounds of, with sulph- uric anhydride, 230. ___ sulphoxide, 1023. Phoapl~oryl chioride, action of, on phos- phoric anhydride, 202. Picric chloride, action of, on amines in presence of alkali, 714. Pirroaconitine, 272. Picroseptdccjlamine, 715. Piinelic acid, preparation of, 835. -- synthesis of liomologues of, Pinzcs kha.yyana, turpentine oil from,311.Piperid ine, action of picric chloride on, Piperonyl, some derivatives of, 150. Piperonylacrylic acid, 152. -- - amido-, 158. --- from nitropiperonalJ57. -- tetrabromo-, 160. --- action of potassium -- a- and 8-tribromo-, 163. Piperonylethylene, tribromo-, 161. PiperonyloTn, 164. Platinum, action of chlorine and carb- - action of nitrosyl chloride on, 663. - dibromide, action of carbonic oxidc, Platinum-compounds, volatile, 598. Pmeumococcus of Friedliinder, ferment- Potassium persulphate, 772. - salts, magnetic rotation of, PROC., -- inolecular refraction and dis- Propane, nitro-, primary, action of Propionitrile, molecular refraction and peroxide 011, 1028. ide on, 1029. chloride on, 1028. 818. 716. action of nitric acid on, 153. nitro-, 153. -- salts of, 154.- - -- hydroxide on, 160. onic oxide on, 598. on, 603. ation induced by, 253. 1890,142. peraion of, in solution, 596, 596. alkalis on, 431. dispersion of, 296. 4 HI l l 6 INDEX OF SUBJECTS. I’ropjlamine, molecular ref ractio I 1 and dispersion of, 296. Propylparatoluidine, 35. -- normal and iso-, 33. Propylparatolylnitrosamine, 3 5. l’rote’id from gedda gum, 1061. Quinindole-a-carboxylic acid, ortlto-~ Quinine, mucate, 754. (.2 uitiolinehydrazinc, ortlio-, 756. Q iiiiiolinesemicarbuzide, ortho-, 7%. 758. R. Refraction, molecular, of various carbon vompounds, 290. -- of various substances i n solution, 589. Hc-fi tictive power of organic compounds at. different temperatures, PRO(‘., 1891, 115. Rotatory power .of turpentine, change of, 011 keeping, 7%. S. Safrolt, molecular refraction and dis- Salts, formation of, PROC., 1891, 118.- in solution, osmotic pressure of, -. magnetic rotation of soiutions of, Selenic acid, molecular refraction and Selenious acid, molecular refraction and Silicon tetrabromide, molecular refrac- - tetrachloride, molecular refraction Silver, action of nitrosyl chloride on, - and lead oxides, compounds of, _- chloride, action of ?iglit on, -- action of, on water when ex- -- dnrkened, examination of, for -- decomposition of, hy light, - eairs. molecular refraction and dis- persion of, 295. 344. PROC., 18’90, 143. dispersion of, in solution, 593. dispersion of, in solution, 593. tion and dispersion of, 299. and dispersion of, 299. G59. 1093. 536. posed to light, 537. oxygen, 543. 536. perdion of, in solution, .596.Sobrerol, 313, 315. - inacti\-e, 318. - prepaxation of, 315. Sobrerone, 314. Sodium acetate, magnetic rotatory power -- butyrate, magnetic rotatory power - formate, magnetic rotatory power - propionate, magnetic rotatory __ salts, magnetic rotation of, PKOC., -- molecular refraction and dis- Soil, nitrification in, with amrnoniacal Solution, osmotic pressure of salts iii, - the dissociation hypothesis of Sdutions, magiielic rotation of, PROr.? - nuture of, PROC., 1891. 105. Specific rotatory power of nconitint~, 281. - volumes of liquids and their satu- rated vapours, new method of det,er- mining. 37. Spectra of blue and yellow chlorophyll, 106. Stannic chloride, vaponr pressures and molecular rolumes of, 911. - ethide, molecular refraction a11d dispersion of, 296.Strychnine mucate, 754. 8 ubstitution -derivatives, forination of, Paoc., 1891, 62. Sugar, cane-, and milk-, molecular refrac- tion and dispersion of, i n solution, 591. of solutions of, 987. of solutions of, 988. of solutioits of, 986. power of solutions of, 987. 1890, 140. persion of, in solution, 596. solutions, 4*85. 344. Arrhenius, I’xoc., 1891, 105. 1890. 141. estimation of, 46. Sulphanilic acid, action of picric chlor- Sulphates, magnetic rotation of, PROC., Sulphuric acid ~olutionu, properties of, Sulphuric anhydride, action of phos- -- compounds of oxides of phos- Sulphuryl chloride, function of the -- ide on, 717. 1890, 143. PROC., 1891, 105. phorus on, 231. phorus with, 230. chlorine in, PROC., 1891, 60. T. Tellurium, volainetric estimation of, 58, 238.INDEX OF SUBJECTS.1117 Terebentliene hydroclilorides, 728. - oxidation of, in sunlight, 311,315. Terpene hydrate, molecular refraction and dispersion of, in solution, 591. Terpenes, 311, 648, 725. - and allied conipoiinds, 311. Terpilene dihydrochloride, molecular re- fraction and dispersion of, in solution, 591. l'etragaluctungeddic acid, 1069. Tetrahydrophe tiylmetliylf urfuran, pro- Tetraplienylcrotolactone, action of alco- - artion of methylamine on. 147. Tetraphenjlpyrrholone, reduction of, Tetrarabinantrigalactangeddic acid, Tetrathiocarbamidammonium bromide, - chloride, 386. - iodide, 385. Tetrethylammonium bromide and iodide, compounds of, with thiocnrb- amide, 387. 'I%allium, action of nitrosyl chloride on, 657. Thiocxrbamide and methyl- and ethyl- ammonium salts, 391.- beiizyl derivatives of, 551. - compound of triethylainn~onium bromide with, 390. - compound of, with dietliylamino- nium bromide, 389. - compounds of, with ammonium lialo*id salts, 384. - compotmds of, with tetrethylam- iuonium bromide and iodide, 387. - constitution of, 394. - new additive compounds of, 353. Thiocarbarntdes, aromatic, preparation - substituted, action of acetic - substituted, and ammonium brom- Tin, action of nitrosyl chloride on, 661. - gold, and cadmium, freezing point Titanium tetrachloride, molecular re- Toluene, orthochloro-orthonitro-, 1017. Toluidine, ortho-, ortho- and para.nitro- -- orthonitro-ortho-, reduction of, in perties of, 194. holic ammonia on, 14 1. 145. 1035. 384. Of) 196. anhydride on, 396. ide, 386. of triple alloys of, 936. fraction and dispersion of, 299. derivatives of, 1013. alkaline solution. 1015. l'oli iidines, ortlio-,pB,.a- and ortho-nitro-, rediiction of, in acid solution, 1016. Tolylmethylpyrazoloneketoparatolyl- hydrazone, para-, 340. To ylthiocarbamide, para-, action of acetic anhydride on, 4.03. Tolglt,liiocarbiiriide, ineta-, preparation of, 403. - ortho-, preparation of, 402. - para-, preparation of, 4-01,. Triacetic acid, 6-lactone of, 607. - - -- action of bromine on, --- reactions of, 614. Triarabinantet8rugalactangeddic acid, Triarabinantrigalactangeddic acid, 1037. Triazine-derivatives, nomenclature of, Triazines, nitro-, reduction of, '701. - substituted: preparation of, 679. Triazine-series, 678. Triethylamine, molecular refraction and dispersion of, 295. Triethylammonium bromide, compound of thiocarbainide with, 390. Triethylisoxazole, 432. Trigalactungeddic acid, 1043. Trimethylene cyanide, molecnlsr re- - iodide, molecular refraction and Trimethylisoxazolc, 41 3, 42'3. Tripropylamine, niolecirlur refraction and dispersioii of, 296. Turpentine, 725. -. action of hydrogen cliloride on, '728. - hydrochloride, conversion of, into can1 phe ne hytiroclt lorid e, 730. - increase of rotstory power of, on keeping, 726. - oil of, oxidation of, in sunlight, 311, 315. - oils, 311. 612. 1071. 679. fraction and dispersion of, 295. dispersion of, 2%. V. Vapour pressures of acetic acid, 903. -- of carbon tetrachloride and stannic chloride, 911. -- of dibeilzyl ketone, 626. -- of mercury, 629. Vapours, saturated, new method of determining the specific volumes of, 31. - nnsrtnitro-ortho-. reduction of. in W 7 --- I_ - - - - - __ - , - - I __ - - - - __ - - alkaline solution, 1015. Toluidines, nitro-ortho-, displacement of the amido-group in, by chlorine, 131'7. .. . Water, potable, estimation of nitrates in, 320.1118 X. IKDEX OF SUBJECTS. Z. Xylenes, molecular refraction and dis- Zinc, action of nitrosyl chloride on, 656. pereion of. 295. ethyl, molecular refraction and of, 403. Xylyltliiocarbiinide, meta-, prep:m~tion

ISSN:0368-1645    

DOI:10.1039/CT8915901106

出版商:RSC    

年代:1891

数据来源: RSC