138 X.-On the Basicity of Tartaric Acid. By MT. H. PERKIN, F.R.S. TEEtetratomic character of tartaric acid has been fully shown by its artificial formation &om auccinic acid. Reasoning upon this fact several chemists have been induced to regard it also as tetrabasic and in confirmation of their hypothesis have ob-tained several metallic combinations having the formula of tetra-basic tartrates. Thus Hugo 8chiff found that by heating the basic tartrate of lead to 130"C. it lost an equivalent of water yielding a body which he regards a8 the tetrabasic lead-salt* thus :-C4R4Pb",0,.Pb",0 = H,O + c4H,Pb",06. Fr isc h also obtained a remarkable zinc-compound by boiling metallic zinc with potash and tartaric acid and t,hen carefully neutraliskg the solution with dilute nitric acid.The product has the composition C,H,Zn",06 + +H,O; but if we consider the curious methods by which these sub-stances have been produced and the high temperatures at which Home of them have been dried and also the fact that tartaric acid will easily lose and again take up an equivalent of water it becomes very difficult to judge of the true constitution of these bodies especially as they are insoluble and cannot be crystallised. I have therefore thought it worth while to study the replaceable hydrogen in a different manner and for this purpose have examined principally new derivatives obtained by the action of the chlorides of the acid radicals on tartaric ether. Action of CJiloride of Benzoyl on Taratarie and Paratartaric Bthers.Chloride of benzoyl when mixed with tartaric ether does not react in the cold but upon. the application of heat hydrochloric * Ann. Ch. Pharm. lxix 2'12 PERKIN ON THE BASICITY OF TARTARIC ACID. 139 acid is abundantly given OE A mixture of these two bodies in about equivalent proportions (with however a slight excess of the ether) was heated in a water-bath for two or three hours. The product was well agitated at intervals with a solution of carbonate of sodium for the purpose of decomposing and re-moving any chloride of benzoyl or tartaric ether that might still be unacted upon. The alkaline solution was then separated fkom the new compound which presented itself as a very viscid oil This oil was then washed with water dissolved in ether and agitated with dry carbonate of sodium to remove any water or acid that might still be remaining.The ethereal solution was then filtered and evaporated over the water-bath. By this means the new body was left as a very viscid but transparent oil. Two combustions of this substance in oxygen gave the following numbers :-I.-.2270 of substance gave 4799 of C02and -1205 of H,O. 11.-2315 of substance gave -4913 of CO and *1217 of R,O. These numbers show the new product to be tartaric ether with one equivalent of hydrogen replaced by benzoyl as the following comparison will show :-Experiment. Theory. I. IT.\ -c, ...... 180 58-06 57.64 57.86 H, ...... 18 5.80 5.90 5.83 ~-0 ........ 112 36-14 .... 310 100.00 This product after standing especially under water for two or three weeks is almost entirely converted into a maas of colourless prismatic crystals which on being well pressed between bibulous paper under a very powerful screw-press until perfectly fkee from oil appears like a cake of white was. 140 PERKIN ON THE BASICITY OF TARTARIC ACID. This on being heated easily fuses and on cooling gradually solidifies into a beautifully radiated orystalline mass. A por-tion of this substance burnt in oxygen gave the following numbers :--2470 of substance gave -5256 of GO, and -1330 of H,O. These furnish the following percentages :-Experiment. Theory. Carbon ...... .... 58-03 58-06 Hydrogen ,.... . 5-98 5-80 These numbers it will be observed correspond to those ob-tained on burning the non-crystalline product.When I first obtained this solid body I was inclined to believe that it was isomeric with the oily product ;but on further consideration I am induced to regard it simply as a purer substance and this n70uld appear to be the case from the analysis of the solid body yielding the best numbers. I have had an oily product stand- ing for nearly three months perfectly clear but on rubbing the sides of the bottle with a glass rod it gradually began to crys tallise. This body I propose to call benxotartaric ether. When pure it is a white inodorous solid and may easily be powdered. It melts at 64O C. and cools to a viscid oil which does not solidify until it has stood for some considerable time but the introduction of a minute piece of the solid product im-mediately causes it to commence crystallising ; and this takes place with a very considerable elevation of temperature.It crystallises in transparent prisms. If strongly heated it distils with decomposition. It is soluble in all proportions in alcohol and ether; slightly soluble in boiling water the solution becoming milky and depositing oily drops on cooling ; after-warda beautiful but small prismakic crystals separate. Its aqueous solution has a slightly bitter taste. Its solutions do not change the colour of litmus paper. Aqueous potash ap- pears to be without action upon it. With sodium it liberates hydrogen and appears to form a sodium-compound.It is heavier than water Benzotnrtaric ether when heated in a sealed tube to 100" C. with alcoholic ammonia slowly decomposes and becomes of a PERKIN ON THE BASICITY OF TARTARIC ACID. 141 pale brownish. yellow colour. On evaporating the product to dryness and then adding water a few oily drops of the un-changed ether appear; these may be separated by filtering through wet paper. The filtrate on the addition of hydrochloric acid becomes filled with crystals of benzoic acid and upon re- moving these and evaporating the solution to dryness the residue presents a gummy appearance and is very soluble in water. After boiling this with strong potash (which causes the evolution of a considerable quantity of ammonia) the addition of hydrochloric acid again causes a very large quantity of benzoic acid to separate showing that in this reaction an amide is produced containing benzoyl ; no benzumide however is formed.The products of this reaction are probably benzoic acid tartramide benzotartramide and alcohol. I have made several attempts to replace a second equivalent of hydrogen in tartaric ether by benzoyl but at present have not succeeded. The result of heating tartaric ether with two equivalents of chloride of benzoyl or of heating benzo-tartaric ether with one equivalent of the reagent is the production of an oil the ethereal solution of which becomes gelatinous when evaporated. I have made several combustions of diEerent pre- parations of this product which were purified in various manners the last one having been heated aa high as 180"C.with the chloride of benzoyl. I have not however obtained any useful results. The carbon determinations gave numbers varying from 60.24 p. c. to 61 the hydrogen being about 5-4p. c. This increase of carbon I do not tlxink is due to the introduction of benzoyl as even at a temperature as high as 240"C. a mix-ture of benzotartaric acid and chloride of benzoyl evolves scarcely any hydrochloric acid gas. Chloride of benzoyl acts on paratartaric ether in precisely the same manner as upon tartaric ether one equivalent of hydrogen being replaced. The product I propose to call henzoparutartaric ether as it is isomeric and not identical with the preceding body. It melts at a considerably lower temperature viz.,57" C.and does not crystallise so easily as that substance. Action of Alcoholic Potash on Benzotarturic Ether. Eenzotartaric ether is easily decomposed by alcoholic potaah yielding if in excess benzoic and tartaric acids; but the de- 142 PERKIN ON THE BASICITY OF TARTARIC ACID. composition may be moderated and intermediate products obtained if a dilute alcoholic solution of this ether and a weak aolution of alcoholic potash are employed the potash solution containing an insufficient quantity of alkali to decompose all the ether. A quantity of benzotartaric ether treated in this manner was gently heated until the excess of alcohol had eva- porated; water was then added to cause the separation of any oily products.After filtration through a wet filter the solution was acidified with hydrochloric acid ; this caused the separation of an oil which was again separated and the clear liquid placed over sulphuric acid under the bell-jar of the air-pump. After evaporating in this manner for a few days beautiful tufts of crystals separated which on being well washed with water were found to be pure. A specimen dried at 100°C. and burnt in oxygen gave the follawing numbers :-,1474 of substance gave 02964of CO, and -0675of H,O. These numbers agree with the formula as will be seen by the following comparison :-Theory. Experiment. 4 A \ C, ........ 156 55.32 5484 HI,. ....... 14 4.9 G 5-08 0,........ 112 39-72 .. 282 100*00 This iB therefore ethylbenzotartaric acid or the benmtartrate of ethyl and hydrogen.This acid is a beautiful product crys- tallising in tufts of hard needles difficultly soluble in water but excessively soluble in alcohol and ether. On evaporating its alcoholic or ethereal solution it is deposited in fan-shaped masses of crystals. Its aqiieous solution reddens litmus paper. It is easily decomposed by potash and its salts appear to be rather unstable. Owing to the small quantity I have obtained of this acid I have been unable to examine it more fully. The mother-liquors from ethylbenzotartaric acid contain con- PERKIN ON THE BASICITY OF TARTARIC ACID. 143 aiderable quantities of Dessaigne’s benzotartaric acid.* The oily liquid previously mentioned as being thrown down on acidi-fying the crude product consists of a mixture of benzoic and ethylbenzoic acids with a little neutral oil having the odour of benzoic ether.Therefore by hydrating benzotartaric ether with alcoholic potash the following reactions take place :-I. c2H5} C4H3(CrH50)06 + HzO = C~Htio1. gH5} C4H3(C7H5O)Ofj-C2H5 Benzotartaric ether. Alcohol. Ethylbenzotartaric acid. 11. zz}C4H3(C;B,0)06 + 2H20= 2C2H60 +-i} C4H3(C7H50)Os. Benzotartaric acid. 111. ~$}C4H3(CiH50)06 + 3H20 = 2C2H60 + C4H406+ C,H,O,. 3 Tartaric acid. Benzoic acid. Action of Chloride of Strccinyl on Tartaric Ether. Chloride of succinyl and tartaric ether when heated together evolve hydrochloric gas in considerable quantity and produce a neutral oily body. In preparing this substance I have gene- rally employed the tartaric ether and chloride of succinyl in the proportions of two equivalents of the former to one equivalent of the latter.These bodies were heated together in a large test-tube in the water-bath until hydrochloric acid ceased to be evolved. The product after having been frequently agitated with water for several hours for the purpose of decomposing any free chloride of succinyl and removing any tartaric ether that might not have been acted upon was purified in the same manner as benzotartaric ether. Two combustions in oxygeii gave the following numbers :-I. -2564 of substance gave -4487 of CO, and 01343of H,O. 11. -2723 of substance gave -4772 of CO, and -1426 of H,O. # J. Pharm.[3] xxxii 47. 144 PERKIN ON THE BASICITY OF TARTARIC ACID. Theae numbers give percentages which nearly agree With those required by the formula This represents two equivalents of tartaric ether bound together by*the replacement of two equivalents of hydrogen by diatomic succinyl. The following is a comparison of the theoretical and experi- mental numbers :-Experiment. /. Theory. /- I. 11. ‘ c2,....... 7-240 48-58 47.72 47-79 HSw....... 30 6-07 5-81 5-81 O, ........ 224 45-35 8. .. 494 100*00 The carbon and hydrogen in these analyses are rather low. This originates I believe &om the presence of a small quantity of a neutral chloriuated oil which is generally found in the chloride of succinyl; and on account of the nature of the new product I was unable to purify it further.A compound repre senting one equivalent of tartaric ether with two equivalents of hydrogen replaced by succinyl would require a very much higher percentage of carbon viz. 53.7. This product I propose to call succinotartaric ether. It is an extremely thick oil generally of a pale yellow colour but I believe would be colourless if perfectly pure. It is soluble In alcohol and ether in all proportions producing solutions which are neutral to litmus paper. If heated with alcoholic potash it is decomposed. It cannot be distilled without undergoing decom- position. Action of Chloride of Acetyl on Benzotartaric Ether. Not succeeding in replacing a second equivalent of hydrogen In tartaric ether by treating it with chloride of benzoyl I thought it well to employ a more active chloride and selected the chloride of acetyl.PERKIN ON THE BASICITY OF TARTARlC ACID. 145 A mixture of benzotartaric ether and chloride of acetyl in about equivalent proportions using a slight excess of the chlo-ride was heated in a sealed tube to 140' or 150"C. for three or four hours. On opening the tube large quantities of hydrochloric acid escaped. The oily product was well agitated with water and dissolved in ether. The ethereal solution was then agitated with dry carbonate of sodium filtered and evaporated to dryness over the water-bath. A specimen of this product burnt in oxygen gave the following numbers :--2332 of substance gave -4965 of CO and *1219of H,O.These numbers give percentages which agree very closely with those required by the formula which represents benzo-tartaric ether in which one equivalent of hydrogen is replaced by acetyl. Theory. Experiment. n C, ........ '204 57.93 38.02 H2,........ 20 5.68 5.80 0 ........ 128 36-37 .. 352 100*00 This product which I propose to call Acetobenzo-tartaric ether is an extremely thick colourless oil heavier than water. It does not show any tendency to solidify as specimens which have been kept for months remain perfectly clear. It is very soluble in alcohol and ether and is quite neutral to test-paper. Heated with alcoholic potash it is entirely decomposed into alcohol acetic benzoic and tartaric acids thus ~c2H~~~c~H2~c~H~0~~c2H~0~06= + 4H'20 Acetobenzotnrtaric ether.2C,H,O -l-C2H4O + C,H,O + C4H,O6* Alcohol. Acegic acid. Benzoic acid. . Tartaric acid Action of Chloride of Acetyl on Tartaric Ether. Tartaric ether is freely attacked by chloride of acetyl even at 146 PERKTN ON THE BASICITY OF TARTARIC ACID. the ordinary temperature large volumes of hydrochloric acid gas being evolved and the mixture becoming quite hot. If equivalent proportions of these two substances are employed an oily body is produced. This product may be purified in the game manner as the aceto-benzo-tartaric ether. Combustion of this oil in oxygen gave the following numbers :--2639 of substance gave -4662 of CO and *1574of H,O these numbers give percentages agreeing with the formula which represents tartaric ether with one equivalent of hydrogen replaced by acetyl as the following comparison will show :-Theory.ExDeriment. I?,, ........ -120 48-38-48-17 HI6........ 16 6-45 6.62 0 ........ 112 45-17 - 248 100.00 Thissubstance which I propose to call Aceto-tartaric ether is n colourless oil not nearly so viscid as those already described being of about the consistency of olive oil. On heating it in a retort placed in an oil-bath it begins to decompose after the temperature has been considerably raised acetic acid condensing in the neck of the retort ; and at about 287O.C. an oil distils over leaving a residue of carbon. Aceto-tartaric ether is heavier than water and slightly soluble in that menstruum.It may be separated from its aqueous solu- tion by the addition of saline solutions as that of common salt. It is perfectly neutral to test-paper and has a rather bitter taste. With boiling aqueous ammonia aceto-tartaric ether decom- poses and on evaporating the solution it yellowish syrupy product is obtained having a bitter and slightly burning taste. Aceto-tartaric ether when heated with chloride of benzoyl evolves hydrochloric acid forming a thick colourless oil probably benzoacetotartaric ether. Sodium acts rapidly upon this ether with evolution of hydro- PERKIN ON THE BASICITY OF TARTARIC ACLD. 147 gen gas. The reaction is facilitated by the addition of benzole which renders the ether more fluid. The 'resulting product of this reaction is a transparent gum-like substance.This is pro- bably Sodacetotartaric etl~er. On treating tartaric ether with two equivalents of chloride of acetyl arid after the reaction has abated heating the product in a sealed tube to 100" for a short time a second derivative is obtained which on being purified like the preceding solidifies after standing into a beautiful crystalline mass wlkh may be separated from a small quantity of oily aceto-tartaric ether by pressure between bibulous paper under a powerM press and then crystallised from water. Specimens dried by fusion and burnt in oxygen gave the following numbers :-I. 02606of substance gave 04763of GO and -1500of H,O. fL 02604of substance gave *4763of C02 and 01507of H,O.These numbers give percentages which agree closely with those required by the formula- which represents tartaric ether in which two equivalents of hydrogen are replaced by acetyl. The following is a comparison of the theoretical and experi- mental numbers :-Theory. Experiment. 1. 11. C,2.. ...... 144 49.65 49.84 49-88 HI*........ 18 6.20 8-39 6-43 I 0 ........ 128 44-15 .- 290 100.00 This substance is therefore Diacetoturtaric ether. It is mluble in alcohol and ether in all proportions but crystallises from its alcoholic solution on dilution with water. If boiled with water B considerable quantity dissolves and the solution on cooling deposits it in splendid prismakic crystals more than an inch and 148 PERKIN ON THE BASICITY OF TARTARIC ACID.a half in length. It is slightly soluble in cold water but on the addition of a strong solution of chloride of sodium this solution becomes cloudy and the diacetotartaric ether crystallises out on standing. Diacetotartaric ether melts to a colourless oil at 67' C. and cools without resolidifying but as soon as a small particle of the solid product is thrown into it it begins to crystallise in tufts of needles and in a few moments becomes perfectly solid. During its crystallisation it evolves heat sufficient to become quite hot to the hand. If strongly heated this ether distils with only slight decom- position. Its boiling point is between 294" and 298' C. The second combustion was made with a product which had been distilled.I believe this to be the only derivative of tartaric acid which is known to bear distillation without being entirely or at any rate very much decomposed. Diacetotartaric ether is not quickly decomposed with aqueous potash and if it be dissolved in cold alcoholic ammonia the rsolution after having been kept for several days will be found to contain a very large quantity of unchanged product. Fused diacetotartaric ether evolves only a very small quantity of hydrogen when brought in contact with sodium-very much less than the monoacetotartaric ether-in fact its solution in benzole scarcely evolves a trace of hydrogen and on the evaporation of the benzole t'he product is found unchanged. This would tend to show that all the typical hydrogen in tartaric ether is substituted in this compound.Paratartaric ether yields two new bodies when treated with chloride of acetyl viz. acetoparntar.taric and diacetoparatar-taric ether. They are prepared precisely in the same manner as the two preceding bodies. The monacetoparat'artaric ether I have not closely examined. It is a colourless oil. The diaceto-paratartaric ether is a solid body melting at 5005~ C. It boils at about 298' C. and distils with slight decomposition. It is soluble in alcohol or ether in all proportions. It is deposited from its boiling aqueous solution on cooling in small tufts of needles and after long standing sometimes forms short but very brilliant prisms on the sides of the vessel. It differs from the ordinary tartaric derivative in its melting point which is 16.5 degrees lover and alao in the different PERKIN OF THE BASICITY OF TARTARIC ACID.149 manner in which it crystallises from iks aqueous solutions. It appears to be rather more soluble in water but does not cry+ tallise pearly so freely as its isomer. arid when fused it takee much longer to aolidify and then does not produce so well crystallised a mass. A portion of this product gave the follow- ing numbers :-*2063of substance gave 03769of CO and 01169 of H,O. Percentagu composition. Theory Carbon.. ,..... 49-82 49-65 Hydrogen ... 6-29 6-20 These results show it to possess exactly the same composition as diacetotartaric ether. I was anxious to take the vapour-densities of these two diaceto-ethers because if paratartaric acid be composed of right- and left-handed tartaric acids its formula should be doubled and this diffsreiice a vapour-density would show directly; but although these bodies are volatile they would decompose too much at the temperature required by such ex- periments to give trustworthy results ; but the boiling points being nearly identical affords strong eyidence that the formula must not be altered because if it were double it is certain that the boiling.point would be immensely raised. I may here meu- tion that diacetoparatartaric ether after having been distilled has exactly its arigirial characters arid fusing point. It is very curious to find that paratartaric acid which yields right- and left-handed tartaric acids and is also formed agair when their solutions are mixed even with evolution of heat does not possess a formula equal to two equivalents of tartaric acid.It would appear therefore not to consist of these two acids but to change into them when converted into certain salts. Action of Chloride of Acetyl on Tartaric Acid. If dry powdered tartaric acid be digested with about three times its weight of chloride of acetyl very little change ap- pears to take place at first but after continuing the heat for several hours it gadually disappears leaving a syrupy liquid VOL. xx. M 150 PERKIN ON THE BASICITY OF TARTARIC ACID. which on cooling generally solidifies to a crystalline mase; should this not take place more chloride of acetyl must be added.The crystalline product after being strongly pressed between dry thick bibulous paper may be rendered perfectly pixre by hsion in an open dish so as to volatilize adhering chloride of acetyl &c. On cooling it solidifies to a splendid white crystalline body. Specimens of this substance burnt in oxygen gave the fol- lowing numbers :-I. *2539 of substance gave *4122 of CO, and *0907 of H,O. TI. a2438 of substance gave 93974 of CO, aad *0862 of H,O. These numbers give percentages which agree with those required by the formula- representing diacetotartaric anhydride. The following is a comparison of the numbers :-Theory. Experiment. r ,. 1 --1. 11. C8 ........ 96 44-44 44-27 44.45 H ........8 3.70 3.96 3.93 0 ........ I12 51-86 - 216 100.00 Diacetotartaric anhydride is a tough orystalline solid melting att 126O-127" C. When distilled it undergoes a considerable amount of decomposition especially if the distillation be carried anslowly. It boils above 250" C. but no fixed point can be ob-tained acetic anhydride coming over during the distillation as well as other products some of which affect the eyes like acrdene; a residue of carbon is left in the retort. If heated gently it sublimes in beautiful but small prisms. It is slightly soluble in benzole and crystallises fkom this solvent in slender white needles. It also crystallises from acetic anhydride. Dried paratartaric acid when submitted to the action of PERKIN ON TIW BASICITY OF TARTARIC ACID.151 chloiide of acetyl undergoes precisely the ~amechange as ordinary tartaric acid but the reaction goes on rather more slowly. The diacetoparatartaric anhydride is a beautiful crystal- line body resembling its isomer in its characters. Its melting-point is also the same viz. 126O C. The following is a combustion of a specimen of diacetotar-taric anhydride prepared from paratartaric acid :--3012 of substance gave -4873of CO, and *lo56of H20. Percentage composition- Theory. Carbon . . . . . . . . 44.12 46.44 Hydrogen .. . . 3-89 3-70 Action of Water on Diacetotartaric Anhydride. If exposed to the air this anhydride quickly absorbs moisture or if placed in contact with warm water it gradually dissolves producing a powerfully acid solution.This contains an acid which I propose to call diacetotartaric acid. Its formation may be explained thus :-C,H,(C,H,%O + H,O = C4H,(C,H,Q),OB* Diacetotar taric anhydride. Diacetotartaric acid. Diacetotartaric acid obtained by evaporating its aqueous solution under the bell-jar of an air-pump generally presents itself as a transparent gum-like substance. It is very deli- quescent and possesses a powerfully acid taste. If strongly heated it decomposes without formation bf its anhydride. When heated with a solution of potash or soda it is decom-posed according to the followhg equation :-C,H,(C,H,O),O + 2H,O = C4H606+ 2C2H,0,. Diacetotartaric acid. Tartaric acid. Acetic acid. A quastity of diacetotartaric acid which had been prepared from ordinary tartaric acid was decomposed by potash and after neutralising with an acid the tartaric acid was precipitated as a calcium-salt which after being well washed was decomposed with dilute sulghuric acid and filtered.The filtrate on being con- 31 2 152 PFAlCIJY ON THE BASICITY OF TARTARIC ACID. centrated deposited large crystals of tartaric acid which were washed and recrystallised. The acid thus obtained does not contain water of crystallisa-tion nor does it precipitate solutions of chloride or nitrate of calcium; it would therefore appear to be ordinary tartaric acid but it appears to crystallise in a somewhat different manner. As I have obtained it in square tables about a quarter of an inch in diameter I hope to again examine this product.A portion of the acid was converted into the acid potasaium- salt. A potassium determination gave the following num-bers :-*1219 of substance gave 00357of sulphate of potassium = 20.4'7 p. c. of potassium. Theory requires 20.74 p. c. Diacet0tartrates.-With bases diacetotartaric acid forms salts containing one and two equivalents of metal,-it is therefore bibasic. These compounds are rather difficult to obtain pure and are remarkable for their great solubility. Sodiurn-salt.-This is obtained by carefully neutralising a solution of the acid with carbonate of sodium concentrating at a very gentle heat and then finally evaporating in vacuo. Thus obtained it is a crystalline solid remarkably soluble in water and very deliquescent.Tlzepotassium-salt is obtained in the same manner as the above substituting carbonate of potassium for carbonitte of sodium. It is a crystalline salt very soluble in water and deliquescent. Aeid potassium-saIt C,H,O,.-To prepare this salt two J equal quautities of an aqueous solution of the acid are taken one portion is neutralised with carbonate of potassium and then mixed with the other. The new salt crystallises out on concentrating the solution at a gentle heat or in vacuo; it is then separated from the mother-liquor by being strongly pressed between bibulous paper and purified by a second crystallisation. Thus obtained it forms a crystalline powder very soluble in water but not deliquescent.It reddens litmus powerfully and poswsses an acid taste. A specimen dried at 100" C. gave the following numbers :-0-3415 of mbstance gave 091080 of aulphate of potash ps= 14-17 p. c of potassium. Theory requires 14.34 p. c. PERKIN ON THE BASICITY OF TARTARIC ACID. 153 Calcium-salt Ca”C,H,O,.-T2lis salt is obtained by neutralis- ing a solution of the acid with carbonate of calcium filtering and concentrating first at a low heat and then in vacuo. I have not succeeded in obtaining it in a crystalline condition. Its solution concentrates to a syrup and then dries up to an opaque friable mass. It is deliquescent. A specimen dried at 100”C. gave the following numbers :--1871 of substance gave -0951 of sulphate of calcium = 14.94 p.c. of calcium. Theory requires 14-70p. c. of calcium. Barium-salt Ba”C,H,O,.-This is obtained in the same manner as the calcium-salt. Its solution after being concen- trated to a syrup on standing for a day or two deposits fine needles of this nev salt sometimes half an inch in length. It is excessively soluble in water and deliquescent. Specimem dried at 100”C. gave the following numbers :-I. -2065 of substance gave -1297 of sulphate of barium. 11. -1377 of substance gave *1307of CO,,. and -0300 of H,O. The following is a comparison of the theoretical and experi- mental numbers :-Experiment. A Theory. I. IT. -c8 ..* ..... 96 26.01 25.88 L H ........ 8 2.17 -2-34 ..... 137 37-12 36-93 -Ba”. .. 0 ........ 128 34-70 -369 100*00 Copper-salt Cu”C,H80B.-~Obtained by saturating a solution of the acid with carbonate of copper and concentrating the solu- tion at a low temperature.It is a blue crystalline salt very hble in water. Specimens dried at 100”C. gave the following numbers :-I. -3527 of substance gave 00945 of Cu”0 = 21-40 p. c. of copper. 154 PERKIN ON THE BASICITY OF TARTARIC ACID. 11. ,2268 of substance gave 00603of CuO = 21.22 p.c. of copper. Theory requires 21.48 p. c. of copper. Silvel-sa zt gac8H,08.-Diace t0tartaric acid is difficult to saturate with carbonate of silver. To obtain this salt it is best to agitate a solution of the acid with freshly precipitated car-bonate of silver filter and concentrate. After a time the silver- salt will then separate in the form of an albuminous magma of very minute silky needles which must be separated fiom the mother-liquor by pressure between bibulous paper and then recrystalliaed.Thus obtained it dries to a white crystalline mass extremely soluble in water and very slowly acted upon by light. Mercuq/-saZt.-A solution of mercurous nitrate when added to a solution of a salt of diacetotartaric acid produces a,gela- tinous precipitate soluble in acetic acid.* Diacetoparatartaric anhydride when brought in contact with water yields an acid viz. diaeetoparatarinric acid. So far as I have examined this body it appears to resemble diacetotartaric acid but when decomposed with potash it yields acetic and paratap taric acids. This at once proves that it and its anhydride are isomers of the ordinary tartaric derivatives a fact which it was important to prove on account of diacetotartaric and dia- cetoparatartaric anhydrides having the same melting point.A portion of diacetoparatartaric acid wm converted into a calcium-salt which was dried at 100"; a calcium determination gave the following numbers :-*2240of substance gave *lo92of sulphate of calcium = 14.33 per cent. Theory requiring 14.7 per cent. This salt was uncrystallisable. Action of Sodium on Tartaric Ether. On bringing sodium in contact with tartaric ether hydrogen * Since making the foregeing experiments I find fhat M. Ro chl eder ha0 also studied the behaviour of chloride of acetyl to tartaric acid and evidently obtained Borne of the bodies described above.He did not however make any analysis of his products. He states that he succeeded in obtaining the acid I have described aa diacetotartaric acid in the crystalline state (Chemical Gazette vol. xvii. (1 859) page 61). PERKIN ON THE BASICITY OF TARTARIC ACID 155 is evolved ; but owing to the viscidity of the ether the reaction takes place very slowly. If however it be rendered more fluid by admixture with five or six times its volume of anhydrous benzole the reaction goes on very rapidly the liquid becomhg quike hot and of a pale yellow colour. On separating the excess of sodium and evaporating the liquid a pale yellowish-brown uncrystalline but friable residue is obtained which quickly be- comes sticky fkom absorption of moisture ; mixed With water it produces a strongly aIkaline solution.A weighed quantity of tartaric ether was treated as above in an apparatus so arranged that the hydrogen could be collected. It was found that the reaction went on rapidly for about half an hour until nearly one equivalent of hydrogen was evolved ; it then slackened the solution remaining clear and the sodium quite bright The action then became still less,and a geIatinous product gradually formed on the sodium entirely stopping the reaction. Reasoning from these facts it would appear that the fist and principal product of this reaction is a tartaric ether with one equivalent of hydrogen replaced by sodium or wdiotartaTic ether the gelatinous product being disodiotartaric ether.By heating the &st sodium-product with iodide of ethyl an oil is produced probabIy ethyltartaric ether. The history of the foregoing substances has I think a con- siderable amount of theoretical interest especially with refer- ence to that somewhat difficult question of basicity and atomicity. Tartaric acid is known to be tetratomic; now if it be Iikew&e tetrabasic it is evident that tartaric ether must be a bibasic acid diethyltartaric acid because only two equivalents of hydrogen out of four are replaced by ethyl thus- H4,C4H20G. H,,E,,C,H,qy Tartaric acid Diethyltartaric acid. If this be true then most of the products 1have described would be diethyltartrates. For example the substance I have called diacetotartaric ether would be diethyltartrate of acetyl ; thut+-.H,,E,,C,H,OG. Ac,,E,,C,H,O,. DicthjItartaric acid. Diethyitartrate of acetyl If this be the case it is obvious that this body would be an 156 PERXIN ON THE BASICITY OF TARTARIC ACID. anhydride because anhydrides cohtaining monatomic acid radi- cals are but the salts of acid radicals; acetobenzoic anhydride for example is the acetate of benEoy1. But this substance has not in any respect the characterietics of this class of bodies. Were it an anhydride or salt of acetyl it would eady decom- pose with water and still more so if treated with ammonia; whereas it may be boiled in water and recrystallised without undergoing any change and its alcoholic solution may also be saturated with ammonia and left for days without any appre- ciable amount of decomposition taking place The same facts also hold true with benzotartaric ether ; and moreover when we decompose this product with a limited quantity of alcoholic potash we find that the ethyl is removed much more readily than the benzoyl ethylbenzotartaric acid and then benzotartaric acid being formed.Had it been a diethyl- tartrate of benzoj71 it would immediately have split up into tartaric ether and benzoate of potassium. Benzotartaric ether also if heated with ammonia does not produce beirzamide. The properties of diacetotartaric anhydride are also again& this view. This anhydride rapidly absorbs water forming diacetotartaric acid and this %id containing two equivalents of acetyl a radical so very easily hydrated when replacing basic hydrogen may be heated up to 100"C.with water with- out undergoing any appreciable amount of decomposition ; moreover it forrns salts which may be dried at 100" C,. without change. The properties of tartaric ether itself are also against this view as it is perfectly neutral to test-paper and does not form compouiids with bases. We have therefore I think strong evidence that although tartaric acid is tetratomic yet the two pair of hydrogen equiva- lents possess very different functions. The nature of this difference may I think be well understood by examining into the particulars of the artificial formation of this acid from succinic acid. Alaxwell Simpson* has lately shown that succinic acid is a derivative of ethylene and may be produced by hcating the cpnide of that hydrocarbon with p0tash.t * Chem.SOC.J. XY 134. 3 The formation of malonic acid from cyanacetic acid is analogsiis to this PEREAT ON THE BASICITY OF TARTAR10 ACID. 157 Cyanide of Succinic acid. ethylene. On treating succinic acid with bromine two equivalents of hydrogen are replaced. This replacement must evidently take place in the ethylene because the resulting acid retains its two original basic hydrogens ; Succinic acid. Bibromosuccinic acid. By boiling the bibromosuccinate of silver with water the two equivalents of bromine are removed and HO substituted. Bibromosuccinic acid. Tartaric acid. From this it will be seen that tartaric acid is succhic acid in which ethylene C,H[ is replaced by C2H2}f/02 repre sents the hydrate of acetylene or acetylene-glycol which in this case has all its atomicity active and holds together the two groups of (COHO).This shows tartaric acid to be a bibasic acid and diatomic alcohol. Therefore the hydrogen which I have yeplaced in tartaric ether by acid radicals is not basic hydrogen but alcoholic hy- drogen. Each of the tarta,ric ether derivatives will then have the double character of an ether and the hydrin of a glycol and those from tartaric a,cid the double character of an anhydride or reactton a8 will be seen if we start with the cyanide of methylene and stop the reaction at an intermediate stage. CHZCYCJ + 2H20 = C2(H,Cy)04 + NH3.Cyanacetic acid. C2(H,Cy)04 + 2HP0 = C3H104+ NH3. Chem. SOC.J. [2i ii 109. 158 PERKIN ON TKE BASICITY OF TARTARIC ACID. an acid and hydrin of a glycol. This will be best Been if I write out their formdae in the following manner* :-Benzotartaric ether ...... COHO COEO ) Ethylbenzotartaric acid . . (C2€€:)" H 0,. * It is worth remarking that tartaric bears to snccinic acid a relation simila~to that of glycollic acid to acetic acid. It is in fact a bibasic glycollic acid. CqH606 -02 = CqRg04. Tartario Succinic aeid. acid. C2H403 -0 z= C&4O2. Olycollic Acetic acid. acid. And 88 glycollic scid represents glycol half oxidized 80 tartaric acid repmsenta a tetratomic alcohol half oxidized.C2H602+ 03 = G2H403+ R20. Qlycol. Qlycollic acid. C4H100j i-04 = C4H6O6 i2H20. Tartaric acid. Thin alcohol would be the =me body as that termed tartaric alcohol in Xolbe's paper on the prognosis of new alcohols and aldehydes (Chern. SOC.J. [2] iv 56). It will be observed that tartaric acid has the composition of oralic acid + aeety-Eem-glycol thus :-Tartaric Oxalic acid. Acetylene acid. glycol. I have Itlready made several experimenta with the view of causing the two groups of COHO to combine and liberate the glycol but as yet have not succeeded. PERKIN ON TFIE BASICITY OF TARTARIC ACID. 159 (COW?/ Acetobenzotartaric ether ..(,'$!}*,) Acetotartaric ether ......(cp2j Diacetotartaric ether.. .... Succinotartaric ether...... Diacetotartaric anhydride.. ( Diacetotartaric acid ...... According to the foregoing reasoning the metallic compounds yepresenting tartaric acid as tetrabasic (if definite bodies) must 'be both salts and alcoholates. The lead-salt might be thus written :- HUNTER ON THE ABSORPTION OF Seeing then the great difference in the properties between the compounds in which basic hydrogen is replaced by acid radicals from those in which alcoholic hydrogen is replaced it appears to me that this points out a very useful method of de-termining the true basicity of an acid as well as its alcoholic nature if it possesses any,-and this is important because the alcohols of some series i. e. the aromatic series are capable of combining with bases ; therefore an acid of such a series if pos- fiessing alcoholic characters would produce saline compounds making its basicity appear greater than it actually is thereby preventing us from classifying it with analogous bodies of other series.I believe this to be the ca,se with salicylic acid and I am now engaged with experiments in this direction which I hope ahortlp to lay before the Society.