LXIV.-The Solubility of Isomeric Organic Compounds and of lllixtures the Relatiom of Solubility to of Sodium and Potassium Nitrates, Fusibility. By THCIS. CARNELLEY D.Sc. and ANDREW THOMSON D.Sc. M.A., University College Dundee. NEARLY seven years ago (1881) one of us showed in an opening address given to the Owens College Chemical Society and subse-quently published in the PhilosophicaZ Magazii2e [5] 13 180 that solubility and fusibility were very closely related t o one another so that of two or more isomeric bodies that dissolves the most easily which has the lowest melting point and in which the atomic arrange-ment is the least symmetrical. Very shortly after Dr. Tilden in a paper in the Philosophical Trafisactions 1884 Part I p. 23 also this Journal 45 266 proved that there was a similar connection between the solubility and fusibility of inorganic compounds.At that time the investigation had included only a comparatively small number (58) of compounds taken at random from Watts’s Dictionary of Chemistry and involved no original experiments made with special reference to the subject. We have now extended the investigation so as to include all isomeric sets of carbon compounds whatever. Our data have been obtained partly from literature and partly as the results of our own determinations. Before giving the results of our work it will be necessary to describe the method we have employed for the estimation of the solubilities of any set of isomeric organic compounds and for the other cases referred to in the present paper.For this purpose it WRS absolutely necessary that for each set of isomerides the determinations should be carried out under identically the same conditions and t o attain this we have in all cases made the determinations for the several members of a set of isomerides simultaneously. The results, however are not only strictly comparable but are likewise a8 absolutely correct as we could make them. A set of long (110 mm.) and narrow (14 mm. outside diameter) glass phials accurately stoppered were employed to hold the solvent and various isomerides under investigation. These bottles were about two-thirds filled with the solvent and such a quantity of each of the isomerides respectively as would be more than sufljcient to form a saturated solution at a temperature 10’ above that at which the solubility was to be determined.The stoppers were then firmly inserted and the bottles Gxed in suitable clips attached to an axi THE SOLUBILITY OF ISOMERIC ORGANIC COMPOUNDS. 783 passing through the centre of a bath filled with water and so arranged that the phials were completely immersed. The axis could be kept in regular rotation any length of time by means of a small water motor whilst the bath was maintained at a constant tempera-ture by a thermo-regulator. The following sketch will represent the arrangement. ABCU is a bath made of tin-plate and styanding on a tripod. To the axis E are soldered six bras8 clips by means of which as many bottles may be fixed on the axis at the same time. FF re-presents six such phials containing the solvent and substances under investigation.Six determinations could thus be made simultaneously. GG and H is a pair of pulleys connected by a cord H being fixed on About +th the actual size. the axis of the small water-motor for driving the apparatus. L is the thermo-regulator to be used in connection with the lamp for heating the bath and K the thermometer for indicating the tem-perature. All our determinations contained in the present paper, were made at 20" C. In order therefore to obtain thoroughly saturated solutions the bath was heated to about 30° and the phials kept immersed aud in continual motion for about two hours matters being so arranged that some at least of the compounds remained still undissolved at 30° in order to avoid supersaturation.At the end of that time the lamp was turned out and the bath allowed to cool dow 784 CARSELLEY AND THONSON SOLUBILITY OF to exactly 20" C. the phials being kept in continual motion through-out. After standing for a few minutes so as t o allow the still undissolved portions to settle to the bottom of the phiais a quantity of the saturated liquid from each was withdrawn into a small tared glass bottle and weighed. This gave the weight of the solvent and substance dissolved. To determine the quantity of the latter the solution was evaporated to dryness and the residue weighed. If however the substance was volatile when evaporated this method could not be employed and in some cases very great dificulty has been experienced in making an exact deter-mination.But when the compound was coloured the quantity in solution coiild be easily estimated colorimetrically by titrating it against a standard solution of the same substance in the same solvent. In the case of acids or alkalis also the amount could be determined by means of a standard acid or alkali certain modifications however, being necessary in the presence of some of the solvents. The differ-ence between the total weight of the solution and that of the sub-stance dissolved gave the weight of the solvent which had been required to dissolve that weight of substance at ihe given temperature. By the above method results are obtained which are strictly com-parable because each set & isonerides can be worked simultaneously, whilst complete saturation is ensured and supersaturation avoided.Special tests made for the purpose showed that the phials were thoroughly tight at the stoppers even when such a volatile solvent as ether was employed. Altogether we have found the arrangement to answer admirably. This usually took about two hours. I. ( a . ) Determination of the Solubilities of 1 3 and 1 4 Nitrawiliiies (m. p. 114" and 147"). The qunnt'ity of these compounds dissolved by a given solvent could not be determined by evaporating a weighed quantity of the saturated solution to dryness on the water-bath as both of them werc slightly volatile under these circumstances. Thus 0.7015 gram of the 1.3 compound after solution in water and evaporation to dryness, weighed only 0.6280 gram or had lost nearly 15.0 per cent.in weight, while the 1*4-compound though not nearly so volatile still suffered a perceptible loss. Therefore as both met%- and para-nitraniline form coloured solutions the quantity of these compounds dissolved by a given solvent was estimated colorimetrically by titration against a standard solution of the nitraniline in the solvent under investigation. That this method gives good results was shown by the following test determinations of the nitraniline dissolved in solutions containing known quantities : ISOMERIC ORGANIC COMPOUNDS. i 8 5 Weight of nitrani- Weight found by line taken. experiment. Experiment I 0,0162 gram 0.0165 gram. . 9 11. 0.0030 , 0-0029 ,, 9 111 0-0075 , 0.0075 ,, IV 0.0099 , 0~0100 , 9 , As the colour of the solution pales somewhat when kept for more than a day especially on exposure to light it is necessary that the determinations should be made as soon as possible and that the standards should be freshly prepared.By the above method we have determined the solubility of the meta- and para-nitranilines in 13 different solvents with the results given in Table I p. 786. All the determinations refer to 20" C. Column I gives the solvent ; Column I1 the modification of nitraniline referred to ; Column 111 the number of the experiment, each of which is the result of an entirely independent determination. The determinations in a given solvent to which the same number is attached were always made simultaneously and in all respects under the same conditions. Column IV gives the weight of the solution taken saturated at 20" C ; Column V the strength of the standard employed ; Column VI the number of C.C.of this standard required to produce a colour in 50 C.C. of the solvent equal to that of tho weight of liquid in Column IV also made up to 50 C.C. ; Column VII the solubility represented in parts by weight dissolved in 100 parts by weight of the solvent ; and Column VIII the meau of the results in Column VIT. ( b . ) Determination o f the Solubility of Mixtures of Xodium and Potassium Nitrates in Water. The resnlts were obtained with the apparatus above described and in the manner already detailed except that the mixed salts were exposed to the action of the solvent for four instead of two hours. Those experiments to which the same number is attached were made simultaneously.The composition of the dissolved portion of the mixed salts was ascertained :-(a.) By finding the weight of mixed sulphates obtained from a known weight of the mixed nitrate residue left on evaporating the several saturated solutions to dryness ; (b.) By determining the melting and solidifying points of the several mixed nitrate residues. R,eference to the curves in Diagram I then gave the composition corresponding to a given melting or solidifying point. The melting and solidifying points of the original mixtures before solution and of the dissolved portion after solution and evaporation, (See Table 11 p. 789.) VOL. LIII. 3 TABLE I. I. Solrent. --Water H20 . . . . . . , -___.-Methyl alcohol CH,.OH'.. -Ethyl alcohol C,H,.OH . . Propyl alcohol C,H,.OII. , 11. hbstance :-Nitraniliii e. --meta 7 J J J 9 9 para ) J 9 9 7 1 meta pars J ) J J meta J J ) J para J J J 1 111. No. of experiment. --1 2 3 4 1 2 3 4 1 2 1 2 1 2 3 1 2 3 1 2 1 2 ----IV. Weight of solu-tion taken, saturated at 2c)" c. 4 -839 4 *716 3 -605 5 *022 4 -859 4 *S70 4 *521 4.959 2.592 2.412 2.522 2 '069 2 *393 2 -566 2.595 3 *929 2 -540 2 -388 2 '472 2 -662 2 -795 2 -840 ~ v. 1 C.C. of standard solution con-tained of nitraniline. ~ gram. 0.00081 0 *00068 0 *00098 0 * 00098 0 *00080 0.00114 0 *00062 0 *00062 0 *02208 0 *01585 0 -02789 0 -01338 --0 *0220 T A 13 L E I-con t inucd.I. Solvent. Isoarnyl alcohol CJ11~*OII Ethyl Ether (C2H,)J0 . . -Benzene CGHG . . . * . . . I w *---KJ Toluene C7Hs . . I . . . . . . . 11. Substance :-Nitraniline. --ineta para 9 , 9 -mela para ,> >> -me ta para 9 ) 9 , meta para 9 9 9 -me ta para 9 , ;> 111. No. of experiment. --1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 -_ _ . ~ --IV. Weight of solu-tion taken, saturated at 2OC c. 2.685 2 '378 2,414 2.288 ~ 3,399 2 *715 3 '059 2 -680 2 *426 2 *606 2 -930 2.365 2 *705 3 -726 2.413 3 *700 3 -052 3 -212 2 962 3 *417 v. 1 C.C. of standurd solution con-tained of nit raniline.gram. 0 * 02000 0 *di356 9 , 0.01587 0 -$1278 I > 0 *015'76 0 *do698 0 '01576 0 &698 > > I__-$ TABLE I-continued. 11. Substance:-Nitraniline. -meta para 39 Y Y meta para 9 , IIT. IV. v. experiment. saturated at tained of 20° c. nitraniline. gram. Weight of solu- 1 C.C. of standard No. of tiou taken solution con------1 3.119 0 *01660 2 3 -086 2 3,120 39 1 5 -487 0 -01660 2 3 * 846 1 3 -520 0- 6i 158 --------1 4 *329 O.d)1158 I. Sol \-en t. 2 1 2 1 2 Cumene C,H, ,, 4.790 9 9 ___-____- ---5 *178 0.01660 5 -992 4 '878 0&15S 3 *450 9 , -_e__--Chlorsform CHCI,. . . . . . , Carbon tetrachloride CCI,. --Carbon disulpliide CS2.1 2 1 2 4 -854 0 -01660 5.138 2.906 0 *di158 3 '416 Y9 meta para 9 ) 9 I. No. of experi-ment. 11. Composition of mixed salts sub-mitted to the action of the solvent. 1x1. Weight of solu-tion taken saturated at 20". IT. Weight of residue obtained. 1 4 2 *372 3.037 5 '036 6 *614 1 2 4 '509 4 -807 4.407 9 * 6005 8 *083 5.778 4.543 6 -019 3 -351 2 *817 -----2 *339 2 -537 1 3 4 5 ~ ~~ SO p. c. NaN03 + 20 p. c. KNO, 9 9 7 7 9 9 9 7 7 7 7 7 7 7 17 1 9 2 -523 5 *566 4 *629 3 -353 2 '574 3 *5095 1.939 1.635 2,868 3 *268 2.530 2 -001 2 *254 ~ ----1 3 4 5 70 p. c. NaNO + 30 1). c. KNO, 11 9s 91 9 91 9 9 97 9 7 9 9 1 3 4 GO p.c . NaNO,+tEO p. c. KNO, Y t 19 9 9 9 9 7 9 9 ) ~ ~~ 4 *978 5 a 573 4 '354 3.898 4.364 --1 2 ~~ 50 p. c. NaN03+ 59 p. c . KNO, 9 9 9 9 9 9 1 2 45-7 p. c. NaNO + 54.3 p.c. KXO, = molecular proportions 2) 79 9 7 5.727 4 -167 2 *669 1 -95 11. I. I No of experi. nzent. Composition of mixed salts sub-niittcd to the action of the solvent. ! I----- --1 2 1 2 103 1'. c. KNO, 7 ) TABLE 11-continzhed. 111. Weight of solu-tion take11 saturated at 20". IT. Weight of residue obtained. I-- --ISOlMERIC ORGANIC COMPOUNDS. 791 were determined in the ordinary way by heating small quantities of the substance in somewhat large capillary glass tubes attached to a thermometer immersed in a bath of strong sulphuric acid.I n the case of the pure nitrates for which an ordinary thermometer cannot well be used the specific heat method formerly described (this Journal, 29 489) by one of us was employed. The results obtained are given in Table I11 (p. 792). The relation between the composition of various mixtures of sodium and potassium nitrates and the corresponding melting or solidifying points is shown in Diagram I the thick broken curve being for the melting points and the thin broken curve for the solidifying points. By reference to these curves the composition of any mixture of the nitrates may be determined approximately if the melting o r solidifying point of the mixture be known. The form of the curves however, shows that in nearly all cases there are two different mixtures which hare the same melting and solidifying points while mixtures contain-ing from 40 to 50 per cent.of NaNO all have the same melting and solidifying points these points not being affected by alterations in composition between those limits. 11. .Eelation of t h e Solubility to the Fusiliility of Isomeric Organic Co Ynp ou nds. Raoult Pictet has shown (Compt. Yettd. 88 S55) j i r s t that the lower the melting point of a solid the longer are the oscilletions of its molecules ; and second that the melting poifits of solids correspond to equal lengths of oscillation so that t x zv = c where f is the melting point measured from the absolute zero zu the length of oscillation and c a constant. Of two isomeric bodies therefore the one with the lower melting point mill at any given temperature below the melting point have its molecules moving with oscillations of greater amplitude than the one with the higher melting point.The molecular weights being equal, the force of restitution will be less in the case of the more fusible compound and hence its molecules will be in a less stable condition, and be the more readily separated from their fellows than those of the less fusible compound. Now in order that a solid may dissolve in any liquid it is necessary that its molecules should undergo a sort of unloosening process and we should therefore conclude that of two isomeric compounds that wonld dissolve the more easily in which the attraction or the force of restitution to the mean position of oscilla-tion was the least i e .the one which is the more easily fusible. This argument shows :-Rule (1.) That for any series of isomeric organic compounds the orde TABLE 111 Per cent. NaNO in original mixture before solution. --100 90 80 70 60 50 45 *7 40 30 20 10 0 Mixed nitrates before solution. Uncorrected. Melting point. 303" 288 274 260 238 225 225 225 235 27 5 29 5 -3olidify ing point. --286 266 255 23 0 220 220 219 230 275 290 -Corrected. Melting point. 316t 298 283 268 242 231 23 1 231 242 284 306 339:: Solidifying point. --296 275 263 237 226 226 226 237 284 300 -Mixed nitrates after solution and evaporation to dryness.Uncorrcct.ed. Helting poiii t. 303 284 259 250 241 235 231 225 269 --I Jolidifying point. ---282 255 2 45 236 230 226 221 266 ---Corrected. Melting point, L 316 294 267 256 247 242 838 231 278 339 --golidifyin! point. ---292 263 251 242 237 233 227 275 ---* By t8herniometer. t Bv specific heat method or 319" bv thermometer. 8 I n taking thesc averages no greater &;gllt has been given' to the reiults from analysis solidifying points because the method of analysis employed does not admit of very great exactness, of mixed sulphates obtaincd makes a rather large error in the final result. Note.-M. p. of pure Itu'aNO = 3L43 (Braun po*qLq. Ann. 154 190) ; = 310" (Person, (Carnelley Jour.Chem. Xoc. 33 276). (Brann Zoc. cit.) ; 339" (Carnclley Zoc. cit.) ; 339" (Person Zoc. cit.). S. p. of puye KNO, (Phil. Afczg. [ S ] 18,114). Sclisffgotscli (Pogg. Ann. 102 293) gives the following as the sodium and potassium nitrates :-I00 per cent. NaNO = 313" ; 90 per cent. = 298"; 80 per cent. = 244"; EO per cent. = 229"; 45.7 per cent. (molec. proportion) = 225.6"; 40 per cent. = 280"; 10 per cent. = 311"; pure KNO = 335'3". 5 . p of pure NaRTOJ = 295" (Maurnen& Compt TABLE IV. Order of solubility. 1 2 2 4 One part dissolved by parts solvent. --___--- --1 2 1 2 1 2 3 4 39 water at 100" 155 I J ----692 water at 19' 3500 , 17 -__----- - -1 2 3 1 1 1 2 3 4 1 2 20 water at 18' 49 , 14.5' -__----.-- - - -- -----4'5 alcohol at 20° 26'7 1J 11 -a06 245 158 226 165 179 189 --, i Formula. Subst,ance. Constitution. Solvents. Authority. ---Dinaphthyl Dinaphthyl . Phenylanthraceno . Dinaphthyl 76' 154 152 187 94 192 61 82 131 51 130 180 -----a ; p p ; p a ; a -Smith Chem. SOC. Jouc. 32 562. Schellinger A&alen $02 61. Smith Chem. Soc. Jour. 32 562. Zincke Ber. 10 999. Xigatti Gazzetta 11 357. Guareachi 1 1 Annalen 222 262. 5, -____--9 9 1 ~___-___--J I , p3tilbene dichloride. a- J) . Dibromonaphthalene . 11 , Ph*C,H,Cl,*Ph . ,> . 1 all proportions in alcohol 10 of alcohol alcohol ,, ~~ ~ CloH,Br2.. alcohol 9 % 9 - -I--I- -water alcohol ether water water alcohol ether 2 1 16-5 alcohol at 56" 3 50.0 C,H60 Dimethyloxalate Isosuccinic acid . Succinic acid . (COZCHJ2 . COZH'CH (CH,) . CO,H. C02H.(CH2)2*COZH . Watts' Dict. 4 272. Wichelhaus 2eit.f. Ch. (2) 3 247. 5.5 cold water I 20'0 ) I I J, ---_.--Watts' Uict. 5 353 ; 7,1093. , 5 1091. , 3 823. , 2 349. Sorbitol . Isodulcitol . Mannitol Dulcitol 110 110d 162 182 1 1 0.6 water water ; alcohol 2, 1 , ,9 --water alcohol ether chloroform ,, 1 9 ,I 4 - I 2'1 >> 6'5 21 31.0 ), -_I-_.-_ 1500 water at 25" - -2600 water at 25" __ --~--CO,H OH COH = 1 . 2 3 . J I = 1 3 4 . $1 = 1:4:5 . I = 1 2 5 . . . . . . . . . -_.__--C8H604 .166 234 244 . 249 207 250 106 133 141 217 261 278 82 118 199 67 81 103 156 176 192 59 114 68 125 --------__ --Aldehydohydroxybenzoic acid. I ,, ,J 9 9 . ,J ,J . . 4 Tiemann Ber. 10 1562. , Ber. 12 1334. , Ber. 9 1268. , Ber. 10 1562. Matsmoto Ber. 11 122. Watts' Dict. 6 249. Schall Ber. 12 829. I h i d . ; Patern? Gazzetta 9 485. Watts' Dict. 6 715. , 6 249. Jacobsen Ber. 12 434. Schall Ber. 12 825. Gekhten B:raJ 11 1587. Schall Ber. 12 825. Swarts Be?. 15 1662. Perkin Chem. SOC. Jour. 39,426. _. -, 1 -_ -9 J J ~-2 9 I , ,J ,I --~ , J, 3 J > ---1 J, -_I__.__-,J ,J ,J water ,, CO,H OH OMe = 1 4 5 . . . . , , = 1 3 4 . -- -Ph.C(C02TI):CH .Ph.CH:CH*CO,H --~--(CO,H) :OMe = 1 2 4 , = 1:3:2 1 = 1:3:4 , = 1 4 5 . . Methoxyhydroxybenzoic acid . I I . 1 I. .~. 1 Lry.u -;id Cinnamic acid Methoxyphthalic acid Methoxyisophthalic acid . Metboxyterephthalic acid. _- ---,J ,> . CyHsOd . C,E<c!' ;. . . . . . . ----CgHsO water ,) water ether 2, I 9 , Hydrocoumaric acid Tropic acid. Hydroxyxylic acid -______-__---Dfelhoxytoluic acid . , I 1 1 I OH (CH~QCH~GO~H) = 1 2. e . . . HO*CH,*CHPh*COzH. Me,:OH:CO,H = 1 2 4 5 . . . . . . . . OMe :Me COzH = 1 :4 6 I1 = 1:2:6 2 = 1:3:6 I = 1 2 5 . . I = 1 3 4 . . = 1 2 4 -___ -----__- - -vater ,, > J -C,HI0O3 . water I I I J , ~ ~~ a-Dibromocamphor 8-Methyl-p-methoxyphenyldibromopropionate .XethJ-1-a-methoxyphonyldibromopropionate . . . ,> . -----~-alcohol ,, MeO~C6H,.C2H,Br,*C02Me = 1 2,. . I 9 = 1 2 . . , carbon disulphide; alcohol I 3 Dinitrobenzene . I I . (KO,) = 1 3 , - - 1 a )) = 1 4 90 118 172 16'9 alcohol at 24%' 26'3 , 24.8 1 - 3 __---_I_-alcohol chloroform ether benzene water ,? ,, Kiirner Gazzetta 4 305. Rinne Ber. 7 869. 2 2 Dabney Amer. Chem. J. 5 20. 9 ,> --> I , ,, I 9 ) ,, ~ ~~~~ Amidodinitrophenol . 1 ) 9 , . . ---- --Amidocaproic acid . > . water; alcohol ,) ,> OH:KH* (NO,),= l 2 4 6 + - . a s . -I = 1 4 2 6 . NH (SOz) = consec . 714 water at 22' 1 2 C6H1302N water 3 Kencki Watts' Dict.J. pr. 3 Chem. 581. [21J 15 390 27 cold water 2 43% water a t 14'5' 170 ubl. 210 120 151 140 147 238 86 116 ___--_. -___ NH2*CO*NH.CH2*Bu. . KH,*CO.NH*CHhfePr ------Watts' Diet. 6 1116. ?Yatts' Dict. 1 555. , 6 314. J J , ~ - -, 7 947. Amylcarbamide . Nitrobenzoic acid 1 . -- --_- I 3 1 28 water at 27" 79'3 , I _ _ - ~ - _ _ _ _ ,i COzH NO2 = 1 3 . , = l 2 . I ' = 1 4 . . . 400 water at 10' 2' I 400 , 22 3 1327 , 16 ~~ ORfe (NO,) = 1 2 4 . . , = 1 2 6 -_. __-alcohol Sillliowski Ber. 7 370. I , DinitranisoPl . Ethamidobenzoic acid . Phenyl-P-amidopropionic acid Phenyl-a-amidopropionic acid x-Amidohydratropic acid . Phenylmethamidoacetic acid x-DinitrodesoxybenzoYn 3- 3inchonidine .hchonine . o . e . . , . - - - - ~ - ~ _ _ ---- ~ - _ _ -J ---________-- ---,> - -alcohol ,> ,> t ,, S H E t C02H = 1 3 Ph*CH(NH2).CH2.CO$I. Ph*CHZ.CH( NHB)*COgH . . . . . . . . . . . . CH,*CPh(NB2)*C02H . NHMeGHPh.C02H Griess Ber. 5 1038. Posen Annalen 195 143. Schulze Bw. 14 1785. Tiemann Ber. 14 1976. , Ber. 14 1982. Golubeff Bull. SOC. Chim. [2] 34 345. Hesse Watts' Dict. 6 463. Ebert Ber. 9 598. 9 9 9 51 > ), 11 I alcohol I ---ether J, 1 76.4 ether a t loo 1 2118'0 , at 17 2 CloH,C1,0,S2 a-Naphthalenedisulphonyl chloride I P- I ,? 31,H,(SO,Cl) Me C1 NH = 1 4 . ?. . = 1 . 4 ?. . - 1:3:4. . $1 -L_____--9 , 1 7.5 benzene at 14' 2 1 221'o > J > J benzene and other solvents 9 , --I-- - ~~~~ 19'9 water at 17" 35.5 3 38'5 wad'r at ih0 Wroblewsky Annalen 168 147.91 ,I >, 9 I 1 ,J C7HgC1O3N, . Chlorotoluidine nitrate. . ISOMERIC ORGANIC COMPOUXDS. 793 01 sdubility is the same as the order of fusibilitly i.e. the most fusible compound is likewise the most soluble. In order to test this rule we have collected from literature all the statements which have been made as regards the solubility of isomeric compounds (a task which has involved a very large amount of labour and time) and we have found as follows :-Out of 752 cases in which the rule can be applied there are 736 in which the order of fusibility and of solubility is the same and only 16 exceptions (= 2 per cent.) in which the more fusible compound is the less soluble.This is with reierence to one solvent in each case whereas by taking all the solvents into account which bave been tried we have found that out of 1778 cases in which the rule can be applied there were 1755 which agreed with the rule and only 23 exceptions (= 1) per cent.). The following may be taken as examples (see Table IV). This table shows with what 8 great variety of compounds the rule holds good. Not only do compounds haviiig a similar constitution obey the rule but compounds likewise which have nothing in common but their isomerism. The application of the rule however does not stop here for we find that-Rule (2). In any series of isomeric acids n o t only i s the order of solubility of the acids themselves t h e same as the order of fusibility bqd the same order of solubility eztends to all the salts of the several acids, so that the salts of the more soluble a d more fusible acids are also more ensilly soluble than the correspondiny salts of the less fusible aud less soluble acids.We have been able to apply this rule in 143 cases of which 138 agree with the rule so that there are only five exceptions or about 3+ per cent. The following may be given as examples* (see Table V p. 794). The above is probably but part of a very general law viz. that the properties of the corresponding derivatives of two or more isomeric compounds arc related to one another in the same way as those of the primitive is0 mers themselves. Discussion o f Exceptions.-As we have already seen there are only 28 exceptions out of 1921 cases in which we can apply the rule as to the order of solubility of a series of isomers being the same as the order of fusibility.A list of these exceptions is given in Table VI. In this table the statements as to the solubilities are as nearly as possible verbatim with those given in the original papers. Of the above 28 exceptions 12 (viz. Nos. 1 2 3 4 5 6 7 12 13, 14 20 and 21) or nearly one-half refer to only five sets of isomerides. * In some instances the melthg points of the acids are not known in which case we have given those of the corresponding acid chlorides or amides instead CARNELLE'Y AND THOMSON SOLUBILITY 3F - I I . . . . ,. ^ . . 0 u5 u - d 0 3 0 0 0 W 4 I 0 * u -0 0 hl a'.d d I / d h l __ On5 a m dd __ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x 6 u h? % u" ___ . . . . . . f . . . . . . . . * . . . . . . 0 U 0 5 - A - - ; ++ d c uu ~ - . . . . . . . . - - . . . . . . . . . . . . . . . . . - . . . - - -__ . . - . . . . . O . - -+ & u 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . -m m U O w "1 x-'$ Fs 00 u;l' V Y u u z"k A d . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . - . . - . . .& c; ** . . . . . . m m m rid4 I1 I/ II 0 G E r:: 6 u . . . . . . * . . . . . . . . . . . . . . . 0 . . . . - . . . . . . n . . -__ ~ ~ . . . . . . . . . . I . . . . . . . . . . . . . . . . . . . . .Ccm . . . . . . . . . . . . . . . . . . . . . . . . - ~~ . . . . . . . . . . + . . . . . . . . . . . . . . . . . . - * - - I e U a FL Salt. K-toluenesulphonatc > > )) 9 ) )) Be- 9 ) ) 9 . . . . . . . . 9) )) Na-Ag-Cn or Pb-sa.lt . !) ) ) . Ba - p - naplitlialenesulpliinic ac1d 3 ) a- > > ) 1 --Constitution. (so,Iq2 = 1 3 ) = 1:4 ,) = 1:3 = 1 4 = 1:3 = 1 4 Me:SO,H= 1 3 . . . = 1 2 . ,) = 1:4 . ~ ~ ~~~ ~~ - ) ) - 1:3 . )) = 1 2 . = 1 4 . )) = 1:3 . ) = 1 z . . . = 1:4. -___-CI(,T-T;.SO;H . M. p. -of chloride 131 63 131 63 131 Qf amide 108" 136 154 108 136 154 103 136 154 of acid 1 0 5 O 63' --------high temp. --Order of solu-bility.1 2 -Formula of acid. M. 11. of amide 150° 217 150 217 150 217 150 217 158 229 158 229 158 229 ------------CgH,,O,SN Order bility of S 1 2 1 2 1 2 1 2 1 2 1 2 1 2 -.-K-naplitlialenesulplionst e . . Ba- 77 7 3 * . 9 ) , * * 1 , a -158 229 of acid 51" 87 263 2 76 -----cu- )) , * * Pb- 9 , * ' >) 7 ' f ,) ) J * ' -K-naphtlialetiedisulpl~onatc . 9 9 9 Ne- , ,7 Ca- 9 Y 9 Ba- or Pb-salt. > 2 9 9 9 ) 9 9 I 2 1 2 1 2 _I-9 . --Ba-cymenesulphonatc ) . . . . . . . . Ba-sulplinmine mesitylenate . l 91 Constitution. --_--a p p p -_-__-a ---_I__-a a p -_I---1L p n p p -_____-__-CI a p -----Me:Pra:SO,H= 1 4 2 9 = 1 4 3 Me2 (SO,*NH,) C02H = 1 3 4.5 12 = 1 3 2 5 ___--__)-TABLE VI. insoluble in alcohol ether and easily aoluble in alcohol water, water and sparingly in ether Watts' Dict. 7 570 . 2 913. insoluble in water . easily soloble. . sparingly soluble moderately soluhle. extremely soluble -___.-not very soluble in alcohol Osterland Ber. 7 1286 Markownikoff,Annalen 182,62 ,f ,1 ,, _.-- -l'erkin Ch. S O ~ Jour. 30 414 9 , 31 412 , 21 476 267 alcohol at 19' 77 14 1256 19 -- ~ slightly soluble in alcohol. easily , . . . . . . . readily , . . . . . . . -----sparingly soluble in water. Watts' Dict. 8 1157 9 9 9 I 13 --___._ Watts' Dict.6 848 . , . . . . . . . . . . . -_--Lobanoff Ber. 6 1251. 9 9 I . pretty easily soluble in hot distinctly less soluble in hot water water ------Furth Ber. 16 2182 1 9 ~~ slightly soluble in ether . . . . . . . . very 2 - -___-less soluble solvent not,. more } stated --__----Watts' Dict. 6 856 . Wallach Ber. 3 8-1.6 --.--Bechi Ber. 12 321 I [hot water much more easily soluble in ----Ber. 12 604 , ,. Aniw,len 206 167 Ba- and Ca-salts obey the rule. The sym-metry accords a i t h the m. p. but not nitL , { the solubility. Jacobsen -( _-_____ 147 , , . j bilities very close. I From C3HC1Br20 I Ca-P-chlordibromacrylate . , , -a- I 3 CBia2:CC1C02H . . . . . . . . . . . . . . . . . . . CBrCl:CBr.CO,II . . . . . .. . . . . . . . . . . -M. p. -129O 148 -liquid 112O 170 -:onstitution. -Order of solu-bility. 2 1 -3 1 2 2 2 1 2 1 3 --Remarks. Authority. 1 part dissolved by parts solvent. Name. Formula. Three exceptions since they are exceptions as regards all three solvents. The sym-metry accords with tlie m. p. but not with the solubility. Diglycollio acid crystallises with 1 mol. H,O. Would this affect the i solubility ? -- -- -HO*CH,.C'O*O'CH,'C02H + . 1 . O(CH,.CO*OH),. . Blycollic anhydride Diglycollic acid . ----CH,(CO,Me),. CHEt(CO,H) . CMe,(CO,H),. --C5Hs0 . Dimethvl malonate. Dimethylmalonic acid Ethylm"a1onic acid . P-EthoxyphenylacrIlic acid . P-Methoxyphenylcrotonic acid . Butyrocoumaric acid.---PP-Dinaphthylketone aa- 9 , PP- ,, OEt (C*H,*C02H) = 1 2 I . . . , OMe (C,H4.C0,H) = 1 4 . . . . (C,HiO) COgH = 1 2 . -_.__-__-CO(CioH;) 1) , 136 154 174 p.d. 125 135 165 -Cl1H,,O Does the batyrocoumaric acid form a coin-pound with alcohol ? ------The modification melting at 125" is said to be a physical isomeride of that melting a t 165". 8. -CZIHl~O ___-y-Dicyanonaphthalene 8- I ) ) CioH,(CN) , 204 236 2 1 2 1 -}of. Zeitsch. f. Chern. [Z] 5 571. 9. . CIBH6N2 . ~~ ~~~ Iodophenol. I ) . . . . . . . . . . . . . a . ---Iodosalicylic acid. , 0 H I = 1:3 (?) . , = 1:4(?) . CO2H:OH:I = 1:2:5 . , = 1:2:3 . ---_.-.-_.--65 89 197 198 50 63 78 104 161 U.C.181 179 193 -----The symmetry accords with the m. p. but not } with the solubility. } rule. -_.__-_._I___. &I. ps. very close. The Ba-salts obey the 10. I C6H,I01. 2 1 more difficultly soluble in Soc. Jour. 41 404. 9 easily , 9, y -Hy droxyvaleramide a-Methoxybutyramide. . a-Hgdroxyisovaleramide Ethyllactamide . CH3*CH(OB)*(CH2),*CO.SH CH,.CH(OEt).CO.XH CH3CH,.CH (OMe)GO.PI'H . CHMe,CH(OH) CO.NH insoluble in ether soluble in ether . sparingly soluble in ether. , 148 water at 25". . . . . . . . . . . . . . . . . . ---54 . 12 ~ C,H1102N & 13, Neugebauer Annalen 227 97 Duvillier Compt. rend. 88,598 Lipp Annalen 205 1 Watts' Dict. 3 452 . Claus Ber. 13 816 . . . . . . . . . . .. 1 " CO,H (NO,) = 1 3 4 . I = 1:2:4 . -__-CO,H O H (CH:NOH) = 1 2 5 . . , = 1:2:3 Dinitrobenzoic acid --Aldoxime salicylic acid . - ------The 1 .2.5-compound crgstallisea with 1 mol. HzO whereas the other is anhydrous. Was the m. p. determinedwith the former in the anhydrous state ? 2 1 -17. (The symmetry accords with the solubility but not with the m. p. The a-compound contains 2H,O and the P-compound +HzO. 1' Were the m. ps. determined in the anhydrous [ state? (CO,H),:Ilfe:NO = 1:3:5:2 9 = 1:3:5:4 a-Nitrouvitic acid. ,) 9- , 227 250 2 1 -2 1 crystallises first from water . Bottinger Be? 9 806 . second . . . . . 1 ,> . CgH70,X. ---Dinitro-a-naphthol. -B- . --___.-OH (NO,) = ~ i a @ i ; . 1 = p i ? ; p1 .18. -19. 138 197 1 The salts obey the rule. (The symmetry accords with the solubility but 1 not with the m. p. The corresponding imides agree with the rule both as to sym- 1 metry and solubility. Have not the m. pa. got transposed in the original paper ? Me (NH~CO*C,H,~CO~NH,) = 1 4. I = 1:2. 2 1 l'olusuccinamide . . . . . . . . . . . . . . . . . . - -3-Ethgldibenzhydroxamate 3- , 7, 148 160 58 63 -99 104 -263 276 d. of acid 78" 131 251 266 168 192 ----142 169 -99 104 ----_. NBz?EtO. 20 & 21. -22. -23. -24. -25. -26. -27. Cl6HI1O3N,. 2 1 -2 1 (cf. Annalen 175 326). M. ps. very close. These are physical isomerides. Ba-salts obey the rule C a - d t s do not (see No.28) ; m. ps. very close. The symmetry accords with the solubility but not wilh the m. p. , . C3HC1Br,0 I . . . CBr2:C01.C02H . CBrC1:CRr*C02H . 3-Chlordibromacrylic acid. z- , . . . . . . . . . ~-Sulphaminemesitylenic acid . . . 3 . * a . 37'1 water a t 20" Illabery Am. Ch. J. 6 157 . 18'4 , I * * . --__- ----Me COpH (SO,.KH,) = 1 3 5 6. I I = 1:3:5:2. C9HllO4SN . --From C12H120, lmmonium a-hydropiperatc. I P- I ? . , less soluble solvent not Bnri Annulen 216 172 /}The acids obey the rule. more , } stated , ~~ Sa-P-phenanthrenecarboxylate . . From ClsHl0O2 ,. . 2 1 -2 1 I . :a-dibrom~~yromucatc 9 I > From C5H2Brp03 ,. 0~CH:CBr.CBr:C~CO2H . ~~ ~~ C4H3MeK.C0,11 1 __-___-I_-From C6Hi0,N .. 2 1 Ph-P-methylcarbopyrollate ,) I . . I ,. n-y- I I di5cultly soluble solrent not 1 Ciami:ian Bey. 14 1057. easily soluble } stated I " solubility of free acids not stated. Luz:$\ ir ~ e n a ~ { ~ ~ ~ i sa<,G u ~ e ~ ~ T -- ~ ______-- ----1 The acids are also exceptions. Ba-salts obey ~ less soluble in n-ater . I M\labc1y2 Am Ch J 6 157 . !{ the rule. The m. ps. are very close (see more , , . ' > * , . . No. 22). The symmetry accords Trith the solubility but not Tith the m. p. 28. -2 1 ISOMERIC ORGANIC CONPOUNDS. 797 The above 28 exceptions may be classi6ed thus :-(1.) Five are only barely exceptions the melting points being very close viz. Nos. 11 20 21 22 and 28. The information given in respect to these is not sufficient to allow one to say whether the solu-bilities are correspondingly close except in No.22 (‘7.v.). (2.) I n No. 25 the solubilities of the compounds compared are very nearly equal whilst the melting points are not very different. (3.) In No. 27 the two Pb salts referred t o were not analysed and one or other of them might have been a basic salt and therefore not really isomeric and comparable with the other one. (4.) Seven though exceptions themselves yield salts which obey the rule (viz. Nos. 11 18 19 22 23 25 28). (5.) Three salts which are exceptions are derived from acids which obey the rule (viz. Nos. 24 25 26). (6.) Three (viz. Nos. 8 20 21) are physical and not chemical isomerides in the ordinary sense and our knowledge of the so-called physical isomerides is at present very limited.(7.) In five (viz. Nos. 1 2 3 10 and 23) the solubilities are excep-tional not only as regards the melting points but also as regards the symmetry. (The more symmetrical of two isomerides is usually less soluble than the one in which the atomic arrangement is less sjmme-trical.) (8.) I n five (viz. Nos. 4 17 12 22 28) the melting points are not only exceptional as regards the solubilities but also as regards the symmetry. (The more symmetrical of two isomerides usually has the higher melting point.) (9.) I n two (Nos. 16 and 17) one o r other of the isomerides contains water of crystallisation and it is not stated whether the melting point was determined with the anhydrous compound. If made with the hydrated compound the water would be driven off on heating, and collect iii the upper part of the melting point tube and might run down and lower the melting point of the substance if proper care were not taken.(10.) Eight (viz. NOS. 5 6 7 9 12 13 14 15) are not included in any of the above classes and SO far as one can learn from the state-ments published i n reference to these compounds there is nothing to lead one to suppose but that they are real exceptions to the rule. We may say therefore that out of the 1921 cases in which the rule can be applied there are only from 9 to 12 exceptions (= 8 per cent,.) against which so far as one can see a t present no objection can be raised. It is possible however that most i f not all of these excep-tions may on further investigation t u r n out to be due to errors either in the determination of the melting points or of the solubilities o r possibly to clerical errors 79s CARNELLEY AKD THOMSON SOLUBILITY OF The relation between fusibility and solubility may also be illustrated by a great number of cases other than those of isomeric compounds.Among these the following are noteworthy :-(a>) Allotyopic Modi$cations of the same Element.-(1.) Phosphorus. Common or yellow phosphorus melts at $Po and is readily soluble in carbon disulphide chloride of sulphur phosphorus trichloride and phosphorus pentasulphide and slightly soluble in ether turpentine, and the cssential oils ; whereas red phosphorus which does not melt below 255" is insoluble in all these solvents.(2.) SuZpphur. Common or octahedral sulphur melts at 114.5" (Brodie) and is readily soluble in carbon disulphide; whereas plastic sulphur does not melt at 120" (? Brodie) and is insoluble in carbon disulphide. ( 3 . ) Selenium. Amorphous selenium melts at a little above 100" (Berzelius) and is soluble in carbon disulphide and in benzene ; whereas crystalline selenium melts at 217" (Hittorf) and is quite insoluble in carbon disulphide or in benzene. (b.) Payafin 'Wax.-The solubility of paraffin wax in benzene is less the higher its melting point as may be seen from the following table (Wutts' Dict. 7 893) :-Quantity of pai~~ffin wax deposited at 18' by 100 C.C. of refined benzene. M. p. 35.0" 133 grams. 49.6 6 ,> 52.8 4.7 9 , 65.5 1.4 3, 80.0 0.1 >> (c.) Xiztures of Sodium and Potassium Nitrates.-Table VII (p.799) which embodies the results of the determinations already given in detail will show the relations which subsist between the fusibility and solubility of mixtures of sodium and potassium nitrates :-These results are also shown graphically in Diagrams I and 11. Diagram I gives the curves of fusibility and two curves of solubility ; in one of the latter the solubility is referred t o the percentage com-position of the mixed nitrates before solution and in the other to the composition of the mixed nitrates after solution and evaporation to dryuess. Diagram I1 gives the curves showing the actual weights oE sodium and potassium nitrates respectively dissolved Prom various mixtures of the two salts and compares these with t,he curves of total solubiliiy of the mixed salts.An inspection of the above table and accompanying diagrams (a) As regards the fusibility and total solubility of the mixed shows :-nitrates (see Diagram I) : ISOMERIC ORGANIC COMPOUXDS. TABLE VII. Per cent. of NaNO in original mixture before solution. -100 90 80 70 60 50 45 -7* 40 30 20 10 0 lfelting ?oint (cor.) )f mixture 6efol.e solut,ion. -~ 316" 298 283 268 24.2 231 231 231 242 284 306 339 Total weight )f mixed salts] dissolved by } = 100 parts of I water a t 20" J 86 -8 109 -6 136-5 136 '3 137% 106 '1 88 '0 43.5 54 *1 40 -9 33 -6 ;1-1 Veigh t of hsolred. 86 -8 96 *4 98 -0 90 '0 66 *O 53-3 45 -6 20 -8 9 -4 0 --Veight of dissolved.0 13 -2 38 *5 47 *6 40 *1 34 - 7 35 -5 33 *3 31 - 5 33 *6 --Per cent. of VaNO i n the nixture af iev 3olution and evaporation to dryness. --100 88 71.8 65 '4 62.2 60 - 6 56 * 2 35 -5 22 9 0 --Melting ?oint (cor.) of the mixed nitrates after solu-tion. -316" 294 267 256 247 242 238 231 278 339 --(1.) That the fiisibility gradually increases from pure NaN03 (m. p. 316") until the mixture contains about 50 per cent. NaNO,, and melts at 231" ; the melting point then remains constant until the sodium nitrate falls to 40 per cent. after which the fusibility gradually diminishes until pure KNOs (m.p. 339") is reached. A mixture having the molecular composition (NaNO + KNOJ contains 45.7 per cent. NaNO+ According t o Schaffgotsch (Pogg. Am,. 102, 293) the salts when mixed in molecular proportion melt at a lower temperature than when present in any other. (2.) That starting with pnrc sodium nitrate the total solubility quickly increases until the proportion of the NaN0 in the mixture be-fore solution falls to 80 per cent. after this i t remains constant until the NnNO = 60 per cent. beyond which the solubility rapidly diminishes until pure potassium nitrate is reached. A mixture containing about 45.7 per cent. NaN03 (= NaNO + KNO,) dissolves to about the same extent at 20" as pure sodium nitrate whereas all mixtures con-taining a smaller proportion of sodium nitrate are less soluble than the pure salt.The above refers to the proportion in which the salts are mixed before solution ; if however the solubilities be referred to the proportion between the two salts in the residue obtained by eva-poration of the saturated solution we find that though the general form of the curve is very siniilar to that in the first case yet the solu-bility is constant over a much smaller range of variation in the com-position of the mixed salts. Thus the maximum of solubility i 800 CARXELLEY AND THONSON SOLUBILITY OF obtained when the sodium nitrate falls to 72 per cent. and continues constant at this until the sodium nitrate falls to 65 per cent. after which it rapidly diminishes to pure potassium nitrate. It will thus be seen that the depression of the curve based on the composition of the mixture ufter solution and evaporation lies throughout its whole course within that of the ciirve based on the composition of the mixture befoye solution.This shows that before tho maximum of solubility is reached the proportion of sodium nitrate in the dissolved portion is less than in the mixture before solution whilst beyond that point it is greater. In other words starting with pure sodium nitrate, there is a smaller proportion of sodium nitrate in solution than in the original mixture until the point of maximum solubility is reached, after which the proportion of sodium nitrate in solution is greater than in the original mixture before solution. ( 3 . ) That the general form of the four curves (two of fnsibility and the other two of solubility) in Diagram I is very similar showing an intimate connection between them.The two curves of solubility, however differ from those of fusibility in one important particular, in that starting from the pure sodium nitrate end the maxima of the two solubility curres occur earlier than those of the curves of fusibility. Thus whereas the middle of the maximum of fusibility occurs when the composition of the mixture corresponds with the formula (NaN03 + KNO,) that of the two solubility curves corre-sponds approximately with the composition (3NaN03 + KNO,). That is to say the addition of potassium nitrate to the pure sodium salt affects the solubility more quickly than it does the fusibility whereas the addition of sodium nitrate to pure potassium nitrate increases the fusibility more rapidly than it does the solubility.(p.) As regards the solubilities of the sodium and potassium nitrates respectively in the presence of one another (see Diagram 11) :-(4.) That the solubility of pure potassium nitrate a t 20" (Curves B and B,) is scarcely if a t all affected by the addition thereto of the sodium salt unless the amount of the latter present in the mixture reaches about 45 per cent. before solution or 60 per cent. in the dis-solved portion. Thus a solution of pure potassium nitrate saturated a t 20° contained 33% parts by weight of the pure salt in 100 parts of water while a saturated solution obtained by treating a mixture con-sisting of 20 pcr cent. NaN03 and 80 per cent.KN03 contained 33.3 parts of potassium nitrate and one obtained by treating a mix-ture consisting of 45.7 per cent. NaNO, and 54.3 per cent. KNO, con-tained 34.7 parts of potassium nitrate. The rcsidues obtained by evaporating t,he several saturated solutions to dryness weighed 33.6, 54.1 and 88.0 grams respectively and contained 0 38.5 and 60.6 per cent. of sodium nitrate. When the mixture contains more tha hozcr~i ('hp?i/ ,So,> Sep?1%88. CAPXLLLEY & THOMSGN I DlAORAM OF THE SOLUBILITY AND FUSIBILITY OF MIXTURES OF SODIUM AND POTASSIUM NITRATES. 1W 80 20 F s P I E E ISOMERIC ORGANIC COMPOUNDS. 801 45 per cent. of sodium nitrate the weight of the potassium salt dis-solved increases until the former salt reaches about 60 per cent.after which it rapidly diminishes until the salt consists of pure sodium nitrate. The weight of potassium nitrate dissolved by a given weight of water at 20" from a mixture containing 82 per cent. of sodium nitrate is also the same as that dissolved from pure potassium nitrate by the same weight of water at the same temperature. (5.) That the solubility of pure sodium nitrate at 20" (Curves C, and C,) is at first somewhat increased by admixture with potassium nitrate ; this continues until the latter reaches about 20 per cent., after which the solubility of the sodium salt at first slowly and then rapidly diminishes until the salt consists of pure potassium nitrate. 111. InJEuence of the Nature of the Solvent. If it be true that the more fusible compound is also the more soluble it follows that:-Rule (3).For any series of isomeric compounds the order of solubility i s the same no matter what may be the nature of the solvent. We have been able to apply this rule in 666 cases and find that the whole of these without a single exception accord with the rule. A few examples in illustration of this rule are given in Table IT. It holds good not only with acids but also with their salts so far at least as the solubilities of such salts have beeu tested in solvents other than water (cf. Rule 2). It further holds in many other cases, e.g. the two allotropic modifications of phosphorus of sulphur and of selenium (see above p. 798). With the view of submitting this rule to 5~ very stringent test we have determined the solubility of meta- and para-nitraniline in 13 different solvents with the results given in Table VIII p.802.X This table shows therefore that no matter what the nature of the solvent the meta-nitraniline is always more soluble than the para-compound. The table further shows that :-Rule (4). The yatio of the solubilities of the two isomerides in any given solvent i s very nearly bonstant and i s therefo~e independent o j the nature of the solvent. This ratio varies from the extreme limits of 1.15 in the case of methyl alcohol in which the nitranilines are most soluble to 1.48 in water in a-hich they are least soluble. The ratios for the other solvents however lie very much nearer to the mean value 1.29 the greatest deviation from it in their case being confined to the second decimal place.* For details of these determications see Table I. VOL. LIII. 3 802 SOLUBILITY OF ISOMERIC ORGANIC COMPOUNDS. TABLE VIII. Solvent. Water HzO Methyl alcohol CH,O . Ethyl alcohol C,H,O Propyl alcohol CBH,O Isobutyl alcohol C,HloO . Isoamyl alcohol C,H,,O . Ethyl ether C,HloO . Benzene C6H6 Toluene C7Hs Cumene CSHIz . Chloroform CHCl . Carbon tetrachloride CC1,. . Carbon disulphide CIS2 . ~ Weight of substance dis-eolved by 100 parts by weight of solvent. Meta-. -0.11.4 11-06 7.05 5 -65 2 -64 8 *5l 7 *89 2 -45 1 - 7 1 1 *15 3-01 0 -21 0 -33 --I Para-. -~ 0'077 9 *59 5-84 4 -35 1. '91 6.29 6.10 1 -98 1.31 0.90 2 -31 0.17 0 -26 -Ratio. meta-para-1-48 highest 1-15 lowest, 1 .21 1 *30 1 *38 1-35 1 - 2 9 1 *24 1 *31 1 -28 1 -30 1 *24 1 - 2 7 1-29 - I Mean = I It will be noted that the solvents employed are numerous and very varied in character whilst a t the same time the range in solubilit,y is very great varying in the case of the para-compound from 0.077 in water to 9.59 in methyl alcohol in the latter of which therefore it is 124 times more soluble than in water.This forms a very crucial test a t least so far as regards the order of solubility. Up to the present we have only completed experiments with this single pair of compounds but we are extending the work to other series of isomerides. If this further work confirms the above results, then the solubility of any pair of isomerides A and B will be represented by the following expression which is independent of the nature of the solvent :-= constant (for any given tem- Solubility of A in any solvent Solubility of B in the same solvent persture).So that if the solubility of A in any solvent at a given temperature be known that of B a t the same temperature niay be calculated. The influence of temperature on the above ratio is at present under investigation and we are also engaged with the deter-mination of the solubility of isomerides at temperatures equidistant from their me1 ting points. The bearing of the above results on the general question of the nature of solution will be discussed lat'er in connection with other data obtained on entirely independent lines but all leading to the same point