COXSTITUTION OF LXPACHOL AND ITS DERIVATIVES. 1355 LXXXVI1.-Tize Constitution of Lapaclzol and its Deri- vcctiws. Part IIJ. The Xtmcture of the Amylene C'hCLi?Z. By SAMUEL C. HOOKER. 0 regarding the constitution of lapachol. It is true that Patern6 assigned to the -C,Hg group the structure -CB:CHCH<CH3, CH3 and that I have adopted this in my former papers, but I have been carefnl to point out that this formula was employed provisionally only (Trans., 1892, 612 ; 1893, 430). In assigning the above constitution to the -C5Hg group, Patern6 was mainly influenced by t h e two following reasons.1356 HOOKER : THE CONSTITUTION OF LAPACHOL I. He had identified isobutylene among the reduction products resulting from the distillation of lapachol over zinc dust. 11. He believed that he had obtained P-iso-amylnaphthalene by submitting Iapachol to the action of hydriodic acid and phosphorus.The substance obtained mas not, however, P-iso-amylnnpht halene, as was subsequently proved by the synthesis of this compound by Roux (BuZZetin, 1884, [2], 41, 380). P-Iso-amylnaphthalene diff ers essentially from Paternh’s hydrocarbon. Thus, since the publication of Roux’s paper, the only two experi- ments bearing on the structure of the amylene chain lead to contrary conclusions. On the one hand, the formation of isobutylene points to the probability that the -C5Hg group contains isopropyl ; and, on tlio other, the dissimilarity between Paternb’s hydrocarbon, CIOHII.C5H1,, and the /3-iso-amylnaphthalene prepared by ROUX, renders the pre- sence of isopropyl improbable.The results of the experiments now to be communicated to the Society clearly prove that the amylene chain of lapachol must be written -CHz*CH:C<c,2, and not -CH:CH*CH<Eg:, as has been previously assumed. It has been shown by Hooker and Carnell (Trans., 1891, 65, 84), that isovaleraldehyde and P-hydroxy-2-naphthaquinone, when heated in alcoholic solution, interact according to the following equation, I. 2CloHs03 + CaHg*CHO = C,Hg*CH(C,oH,03), + HZO; but if the same substances, dissolved in acetic acid, are heated in the presence of a sufficiently large quantity of hydrochloric acid, the following entirely different reaction occurs (compare Proc., 1893, 9, 259). C,Hg*CHO = CloH, I n the first case, the isovaleraldehyde unites with 2 mols.of the hjclroxynaphthaquinone, an action which appears to be a general one, and which was first studied by Zincke and Thelen with benzaldehydo ; in the second, the same quantity of the aldehyde unites with 1 mol. of the hydroxynaphthaquinone only,* and the resulting compound, C,,H,403, is isomeric with lapachol. That the action has occurred as above indicated, is cleai4y demon- strated by the following facts. f Further details of this reaction, so f a r as other aldehydes are concerned, will be communicated later.AND ITS DERIVATIVES. 1357 1. The cornpound exbibits all the properties of a liyclroxyquiiione. Having well developed acid properties, it dissolves readily iu alkalis, and forms intensely coloured, crystalline salts. 2. It yields an acetyl derivative, which is no longer capable of forming salts.3. When oxidised with nitric acid, i t yields phthalic acid. 4. It combines readily with bromine, forming an iinstable additive compound, in which the bromine, as is apparent from the reactions of the compound, is unquestionably situated in the side chain. These properties demonstrate 1. That the quiuone group has taken no part in the reaction. 2. That the hydroxyl group also remains undisturbed. 3. That the side chain must be sitaated in the benzene ring con- taining the quinone and hydroxyl groups, and consequently that it nccupies the /I-position, which is the only one available. 4. That there is a double bond present in the side chain. If, therefore, the origin of isolapachol from /3-hjdroxy-z-naphtha- quinone were alone considered, i t would be necessary to assign t o it the first of the following formulze, 0 013 but as isolapachol differs from the derivatives of &hydroxy-x-naphtba- quinone, in being of it brilliant brick-red colour instead of yellow, we must accept the second formula as the more probable one.(Corn- pare pp. 1363, 1364.) The first of the above formula is that which, up to the present time, has been used as probably representing the constitution of lapachol, and thus the question arises, does the isomerism existing between lapachol and isolapachol merely depend on a difference in the qninone group? This can be readily answered in the negative, for i f t'hese compounds were related as shown in the formulae (1) and (a), it is evident that on reduction they both must yield the same hydro- Inpachol, but as this is not the OH case, the amylene chain in lapachol cannot be1358 HOOKER : THE CONSTITUTION OF LAPACHOL identical with that in isolapachol ; hence the above formula (1) can no longer be iiccepted as representing the structure of lapachol.” I shall show in the €allowing pages that lapachol aiid isolapachol can be converted without loss of carbon atoms into the same com- pounds.The changes involved are simple, and could hardly give rise to any alteration in the skeleton structure of the side chain, which must, therefore, in both cases be written -C*C.C<gI, the car- bon atom to the ext’reme left being attached to the naphthalene nucleus. This follows as a necessary consequence from the synthesis of isolapachol from isovaleraldehyde. Now, from a skeleton of this structure, it is only possible to derive three amylene chains, namely, (1) (2) ( 8 ) and of these the first, being that present in isolapachol, cannot represent the structure of the chain in lapachol itself, whilst the third will not explain the reactions of lapachol, more particularly those connected with the passage of its derivatives into those of iso- lapachol. - There remains consequently the second formula only, and this alone enables all the numerous reactions already studied to be satisfactorily interpreted. The reduction of lapachol, if carried sufficiently far, might, there- fore, be expected t o lead to P-iso-amylnaphthalene, the hydrocarbon which Patern6 believed to be formed by the action of hydriodic acid and phosphorus.As already stated, however, /j-iso-amylnaphtha- leiie has been synthesised by ROUX, and it differs essentially from Paternb’s hydrocarbon. Paternh’s conclusions regarding the composition of his reduction product cannot., therefore, be accepted, for if the compound obtained were really CloH7*C5H11, i.e., an amylnaphthalene, it is evident that the amylene chain in lapachol could not have for its structure the CH3 formula, -CH,*CH:C <CH9’ which the results of my experiments necessitate. I have therefore repeated in more detail Paternb’s work, and have found that the reduction product described by him is not a hjdrocarbon, but a mixture of two isomeric compounds of the formula, C15H160. In order to understand the formation of these compounds, it is only necessary to remember that, by the action of mineral acids, The possibility of stereoisomerism has not been orerlookecl ; the assumption of its existence is however unnecessary. The relation of lslpacliol and isolapachol to each other, as will be preseiitly seen, is made perfectly clear by their reactions.AYD ITS DERIVATIVES. 1359 lapnchol is readily converted into either a- or P-lapachone, or a mix- ture of these substances ; and as the lapachones coiitain a closed side chain, in which the hydroxylic oxygen forms the connecting link between the amylene group and the naphthalene nucleus, i t is evident that some difficulty might be anticipated in the removal of the whole of the oxygen ; hence it is not surprising that the reduction products obtained should have the formula, ClOH3[,< I C&O 0 .No analysis is given by Paternb of his supposed hydrocarbon, bnt. its composition was deduced by him from the following analyses of a picrate prepared from it. Calculated for Paternb found. C l ~ € I j * C ~ H ~ ~ + C,H2(NO2)3*OH. C.. .... 57.41, 57.29, 57.27 59.01 H ..... 4.61, 4.76, 5.15 4-91 N ..... Not determined. 9-83 Patern6 explained the deviation of his analytical results from those required by the formula C15H18,C6Ha(NOz)3*OH, by the supposition that the picrate analysed contained small quantities of free picric acid. The analytical figures given below are the averages of all those obtained ; that is, they include the results of Paternb’s analyses, as well as those of my own ; and the comparison shows that they agree with the requirements of the theory suggested above.This explanation is however, no longer tenable. Calculated for Mean of the CljHlG0,C,jH2( N02),*OH. analytical results.* C ............. 57.14 57-14 H ............. 4.30 4.57 N ............. 9.52 9.61 I have further confirmed the formula C15H160, by the analysis of the reduction product itself (compare p. 1367), and have found t,hat of the two substances of which it is a mixture, the one- that predominating-can be obtained by the reduction of P-lapachone, whilst the other, which appears t o be present in a small quantity only, can be readily prepared by the reduction of a-lapachone. There is, therefore, no doubt possible regarding the composition of Paternb’s reduction prodnct, and the discovery that it contains oxygen has remored the difficulties which, as a hydrocarbon, its existence sug- gested.The conversion of the derivatives of lapachol into those of iso- lapachol, to which reference has been already made, can be best shown graphically as follows. * For details compare p. 1366.Tnutomcric formuloe (3) probably corresponds to the st,a,blc form. cj K P B 1- f- Q .. c3 pi n Q 3 Q !a 0 0 w d Changes spontaneously 2. in alcoholic solution 5 *-p a A 83 QQ A w w Effi bQAND ITS DERIVhTlVES. 1361 The revision of the formula of lapachol renders necessary some modification in the formula: of the remaining substances of this group ; these changes may now be briefly discussed. The conversion of lapachol into the isomeric lapachones has been shown to depend on the absorption of a molecule of water, which is again subsequently eliminated, but in a different direction (Trans., 1892,61,613) ; and there is every reason to believe that in the forma- tion of the intermediate additive compound the general rule has been followed, and the hydroxyl or negative group has attached itself to the carbon atom poorest in hydrogen. Consequently the following formuls must be ascribed to hydroxyhydrolapachol and lapachone respectively.C H, C ,,Hk CH2* C € € , d O H ‘CH3 H y droxjhy drolapachol. C,,H, {::-I CH2*CH,*C<;Z Lapachone. {:H Similarly, chlorhydrolapachol becomes 0 and bromo-P-lnpachon e 0--1 Thus, the side ring of the lapachones consists of six members, instead of five, as has been heretofore assumed. That a similarity exists between the lapacbones, on the one hand, and the anhydrides of the syiithetical compounds oE the general formula X 0 CH 0 I on the other, has been established by Hooker and Carnell (Trans., 1894, 65, 76).In the case of the latter compounds, the anhydride formation can only give rise to a ring consistiug of six members, hence the lapachones are more closely related to this group than was a t first suppmed.1362 HOOKER : THE COBSTITUTIOS OF LAPACHOL The only remaining compound of the Iapachol group t o which it seems necessary to make special refererce in the introductory portion of this paper is Paternh’s so-called isolupuchoute. In a former comniu- nication to the Society (Trans., 1892,61, 624), i t was shown that this compound contains 2 atoms of hydrogen less than lapachone : it has, in cousequence, the formula Cl5Hl1OB, being isomeric with dehydro- lapachone (see next paper) and the isopropylfuran-nnphthaquinones (pp.1370, 1376). The structural formnla provisionally suggested at that time, revised in accordance with the requirements of the new formula for lapachol, is, however, no longer tenable. I shall hope to discuss in the near future the results which have led to this conclusion. In the meantime the compound may be conveniently referred t o as pseudo-dehydrolapachone. EXPERIMENTAL PART. Xynthesis of Iso-/3-lapachol (compare Proc., 1893, 9, 259) .-This compound, to which reference has already been made on p. 1356, was prepared as follows :-lo grams of hydroxynaphthaquinone were heated on a st’eam bath with 175 C.C.of acetic acid. As soon as the substance had completely dissolved, 35 C.C. of isovaleraldehyde, immediately followed by 50 C.C. of concent rated hydrochloric acid 9 sp. gr. 1-20, were added; the flask was a t once returned to the steam bath, and the heating continued for exactly 20 minutes. The solution was then poured into a relatively large volume of water. The dark oil which rose to the surface commenced to crystallise almost immediately. After standing over night, the crystalline crust was lifted off, allowed to drain, and finally repeatedly pressed between porous paper, so as to remove the oil as thoroughly as possible. The substance was then crystallised from a small quantity of alcohol.* Two preparations were made, the one from Kahlbaum’s “valer- aldehyde,” the other from isovaleraldehyde synthesised from iso- butylic iodide.The purified products were found to be identical, equally satisfactory results being apparently obtained with Kahl- baum’s valeraldehyde, although optically active, as with the aldehyde synthetically prepared. The yield is very fair, from 7 t o 9 grams of the purified substance being obtained from 10 grams of hydroxy- naphthaquinone. * The alcoholic mother liquor, after concentration to a small bulk, so as to first yield a second crop of crystals, wan poured into about 400 C.C. of an aqueous 1 per cent. solut,ion of sodium hydroxide. The iso-8-lapachol which remained in the mother liquor was thus extracted by the alkaline solution, from which, after filtra- tion from the resin, it mas precipitated by hydrochloric acid as an orange oily substance, which gi*adunlly became crystalline.AKD ITS DERIVATIVES.1363 Iso-p-lapachol was prepared for analysis by recry stallisation from It separated i n brilliant, brick-red needles, which melted at The alcohol. 120°, showing signs of fusion at a slightly lower temperature. following figures were obtained on analysis. T. 0.2035 gave 0.5528 CO, and 0.1071 H20. C = 74.08; H = 5-84, 11. 0.1991 ,, 0.5400 ,, ,, 0.1057 H20. C = 73.97; H = 5.89. ClsHlrOa requires C = 74-38 ; H = 5-78 per cent. Dilute aqueoiis solutions of sodium and potassium hydroxide dis- solve iso-P-lapachol readily, becoming intensely purple. From these solutions the corresponding salts can be easily obtained in a crystal- line condition by the addition of a concentrated solution of the respective hydroxide ; when dry, the salts are dark violet, almost black.In alcohol, more especially when in dilute solution, i t slowly undergoes change, the odour of acetaldehyde becomes noticeable, and a yellow, granular, although crystalline, substance, sparingly soluble in alcohol, separates. This dissolves to some extent in 1 per cent. sodium hydroxide, but is apparently mostly converted into a salt by the alkali mi thout passing into solution. Iso-P-lapachol is dissolved by concent,rated sulphuric acid to a solution, which, after passing through various shades, soon becomes crimson, the odour of sul- phurous acid being distinctly perceptible. The precipitate obtained on the addition of water became resinous on drying, and consisted of more than one substance.Several attempts were made to form an additive compoiini: with hydrogen chloride and hydrogen bromide respectively, as it wa0 thought possible that if such a compound were obtained i t might serve as a stepping-stone to the preparation of lapachol itself: these experiments have proved entirely unsuccessful. The constitution of iso-p-lapachol has been already discussed in the theoretical port~ion of this paper. Reference may, however, be made somewhat more in detail to the fact that I have assigned to i t the structure of a P-naphthaquinone derivative. This has been done became it is extremely probable that the difference in colonr between a- and p-naphthaquinone derivatives of the types occurring in the lapachol group is a sharp and distinct one.Iso-p-lapachol is very soluble in the ordinary organic solvents. In previous papers I have regarded the structure OH VOL. LXIX. 4 P1364 HOOEER : THE CONSTITUTION OF LAPACHOL as an nnst'able one, but the existence of iso-/3-la,pachul, as well as of similar compounds which have since been prepared in my laboratory, renders it probable that although frequently unstable this structure is not necessarily so. In the case of the internal anhydrides of the general formulze Red. Yellow. the relation of colour to structure can be readily established. Thus we have the red anhydrides which, without exception, form azines, and the yellow ones which do not ; and whilst it is not possible with any degree of certainty t o apply the azine test to the hydroxy- naphthaquinone derivatives themselves, owing to the mobility of the hydroxylic hydrogen, it may nevertheless be presumed, with a fair amount of probability, that a similar relation to colour in their case also holds good.Hence the red and yellow hydroxynaphthaquinones must, for the present at least, be regarded as ortho- and para-qninone derivatives respectively. Acetyl Derivative.-An acetyl derivative was readily obtained by boiling for two or three minutes 2 grams of iso-P-lapachol with 4 grams of anhydrous sodium acetate and 13 C.C. of acetic anhydride ; the solution was then poured into a large volume of water ; the oil which separated soon became crystalline. The acetate was purified by crystallisation from alcohol, in which it dissolves very readily, and from which it separates as yellow needles.The portion f o r analysis was again recrystallised, and then melted sharply at '74'. The analytical results were as follows. 02128 gave 0.5568 COz and 0.1071 H,O. Cl,H13(COCH3)0, requires C = 71.83 ; H = 5.63 per cent. The acetate, being yellow, is most likely an a-naphthaquinone derivative. Thus i t is probably derived from isolapachol, the /3-hydroxy-a-naphthaquinone derivative isomeric with iso-/3-lapachol. Concentrated sulphuric acid gives with the acetate a violet colora- tion, which rapidly changes into crimson. The acetyl group is readily removed by caustic alkalis ; the compound was boiled for a short time with a 1 per cent. solution of sodium hjdroxide until it had completely dissolved ; hydrochloric acid precipitated iso-6'- lapachol from the alkaline solution, and this, when once crystallised from alcohol, fused a t about 119.5".Por the analyses of iso-/3-lapachol, and also those of the acetate, I C = 71.36; H = 5.59.AND ITS DERIVATIVES. 1365 am much indebted to MI*. C. C. Burger, who also rendered valuable assistance in the preparation of these compounds. Reduction of LapchoZ.--'rhis was conducted essentially as de- scribed by Paternb (Garzetta, 1882, 12, 329). One payt of lapachol, one part of amorphous phosphorus, and four parts by weight of hydriodic acid, sp. gr. 1.7, were heated together until the action which is at first quite brisk, appeared to be entirely ended. The lower or oily layer was separated, washed slightly with water, and then distilled in a current of steam.The oil which collected in the receiver passed ovei* with very great difficulty, and the distillation mas continued for some da,ys. As a further means of purification, the oil wzs converted into the picrate described by Patern6. This was c~yst~allised from alcohol and then decomposed by a dilute solution of ammonia, the liberated oil being extracted by agitation with ether. The ethereal solution was repeatedly washed with water, dried over calcium chloride and distilled ; after the ether bad been driven off, the t,emperature rose rapidly and no further distillate was obtained until the thermometer registered over 300". The first por- tions of the oil vr-ere discarded: that collected for analysis passed over at about 3 1 0 O .The following figures mere obtained. I. 0.2256 gave 0.7075 COz and 0.1542 H,O. 11. 0.24'38 ,, 0.7596 ,, ,, 0.1671 H,O. Found. Calculated for 7-7 7--A- C ...... 85.52 84.97 84.90 90.90 7 I. 11. C15H160. ClOH7*C5Hll. H . . .... 7-59 7.61 7-54 9.09 It is apparent from these analyses that the oil is not an aniyl- naphthalene. The analytical results point rather to the formula CI5Hl6O, and this was further confirmed by the experiments given below. The action of hydriodic acid in giving rise to a reduction product, C15H160, is a two-fold one. The lapachol first merely under- goes the change which is brought about by all the stronger mineral acids, and which invariably results in the formation of a- or B-lapa- &one, or of a mixture of these substances. The quinone group is then completely reduced, and the product, Cl5HI6O, is formed.Theoretically, therefore, the formation of two isomeric substances is possible, the one derived from a-lapachone the other from p-lapa- chone ; and indeed it was subsequently foiind that the oil nnttlysed was a mixture of these two isomerides. They will in future be referred to as a- and 8-lapachan respectively. The reduction product as prepared above was amber coloured 4 ~ 21366 HOOKER : THE CONSTITUTION OF LAPACEOL when freshly obtained, but, subsequently, it darkened considerably, even though protected from the light. It does n9t appear to distil entirely without decomposition at the ordinary atmospheric pressure, and the slight colour of the freshly prepared substance was probably due to this cause.About nine months after preparation, a few perfectly colourless prismatic crystals had formed, whilst by far the larger quantity of oil still remained in a liquid condition. Efforts were then made to obtain enough of the substance in n crystalline condition f o r analysis, and in view of the theoretical explanation of the reduction process suggested above, a- and p- lapachone were in turn submitted to the action of hydriodic acid and phosphorus. It was thus found that the reduction of a-lapachone gave rise to a substance which crystallised readily, and which was identical with that deposited in crystals from the oil. p-lapachone, on the other hand, gave an oil which could not be obtained in a crystalline condition, and which was recognised by its picrate as being identical with the permanently fluid constituent of the oil obtained by the reduction of lapachol.A portion of the picrate obtained from the lapachol reduction product was several times recrystallised from alcohol and then an aly sed. 0.2564 gave 0,5367 C0,and 01036 H,O. C = 57.08 ; H = 4.48. 0.1847 ,, 0.3846 CO, ,, 0.0746 H20. C = 56.79 ; H = 4.48. 0.2013 ,, 0*4211 CO, ,, 0.0756 HzO. C = 57.05; H = 4.17. 0.2007 ,, lost ,, 0.0793 H,O. H = 4.39. 0.2484 20.8 C.C. moist nitrogen at 19.6' and 761 mm. N = 960. 0.1512 ,, 12.3 C.C. ,, ,, 162O r, 771 mm. N = 9.61. C15H,60,C6HZ(N02)3*OH requires C = 57.14 ; H = 4.30 ; N = 9.52 P.C. Subsequent to the discovery that there are in the lapachol reduc- tion product two distinct substances, a portion of the picrate analysed was carefully examined; i t was found to consist entirely of thc fi-lapachan derivative, the smaller quantity of the corresponding a-lapachan product having been entirely eliminated in the alcoholic mother liquors.In the substance which these deposited on el-apora- tion, the presence of a-lapachan picrate could be readily demonstrated. Reduction of a-Lapachone.-Twenty grams each of a-lapachonc and amorphous phosphorus were gently heated with 110 C.C. of hydriodic acid of sp. gr. 1.5. The action was moderated by occa- sional withdrawal from the source of heat, and when it appeared to be ended, the temperature was raised to, and maintained at, the boiling point f o r a few minutes. The resulting oil, after thorough washing with hot water, was dissolved in alcohol and freed from phosphorus by filtration.The a-lapachan present was then converted ,,AND ITS DERIVATIVES. 1367 by the addition of picric acid (15 grams) into its picrate which separated readily, and was purified by recrystallisation several times from alcohol. Rather more than 10 grams of the picrate was thus obtained in a satisfactory condition, although its colour remained persistently darker than that of the pure substance. From the alco- holic solution of the picrate, a-lapachan was obtained by the addition of sodium hydroxide dissolved in a little water, in quantity theoreti- cally sufficient to combine with the picric acid. I t was prepared for analysis by recrystallisation from alcohol, a small quantity of a much less soluble compound being thus removed. The substance first nnalysed was slightly coloured, treatment with animal charcoal having failed to produce a perfectly white product.C = 84.24; H = 7.35. I. 0.2375 gave 0.7336 C02 and 0.1573 H,O. 11. 0.1824 ,, 0.5657 (202 ,, 0.1202 H2O. C = 84-58 ; H = 7-32, 111. 0.1698 ,, 0.5252 CO, ,, O.113OH2O. C = 84.35 ; H = 7.39. C15H,,0 requires C = 84.90 ; H = 7-54 per cent. As the above fjgures were not entirely satisfactory, the preparation was further purified by distillation with steam, followed by recrystal- lisation from alcohol. It then melted sharply at 112.5 to 113*5O, was perfectly white, and gave analytical results as follows. 0.1822 gave 0.5655 CO, and 0.1255 H,O. a-Lapachan crystallises in long and remarkably brilliant needles readily soluble in hot alcohol. I n concentrated sulphuric acid, it dissolves to a yellow solution : the addition of water produces in this a milkiness which slowly gives way to a formation of microscopic crystals.With picric acid, it combines very readily, the acid chang- ing instantly to bright red on coming in contact with an alcoholic solution of a-lapachan. The picrate crystallises well from alcohol in red needles, being only moderately soluble in the cold: dilute s o h - tions are, however, apt to deposit some crystals of a-lapachan simultaneously with the picrate. The picrate melts sharply at 140°, and gave the following ana- lytical results. C = 84.64; H = 7.65. 0.2181 gave 0.4380 CO, and 0.0821 H20. C = 57.27 ; H = 4.18.C,5H,,0,C,H2(N02)3~OH requires C = 57-14 ; H = 4-30 per cent. Reduction of P-Lapachone.-This was conducted essentially as pre- viously described for the corresponding a-compound ; the resulting oil, still enclosing amorphous phosphorus, was distilled with steam. The P-lapachan, which collected as an oil in the receiver, was filtered off from the water, dissolved in alcohol, and converted into its picrate. After recrystallisation, the latter fused a t 143-144*, and was recognised as being identical wit'h the picrate previously1368 HOOKER : THE CONSTITUTION OF LAPACHOL obtained from lapachol, of which analyses are given on page 1366. I n spite of the melting point, being only 3-4" higher than that of the picrate of a-lapachan, and that i n appearance the two are essentially identical, the picrate of P-lapachan can be readily recognised by the intensely blue-green colour which it yields when slightly warmed with coucentrated sulphuric acid, a colour which p-Iapachan itself develops under the same circumstances, but which in the case of a-lapachan and its picrate is entirely wanting, the last two substances giving only a yellow solution.I n consequence of their mode of formation, the following formuh must be amscribed t o a- and B-lapachan respectively. 0-7 a-Lapachan. &Lapachan. In the foregoing experiments I am indebted for valizable assishance t o Mr. H. L. Wood, who undertook the somewhat tedious prepara- tion of the lapachol reduction product, and also made several of the analyses of its picrate. Action of Sdphzt~ic acid on Dihydroxyh yds.olapacho1.If dihydroxyhydrolapachol be dissolved in concentrated sulphuric acid, the solution, at first orange-red, almost instantly passes into a brown, and then more slowly into a dingy purple-red. Action takes place simultaneously in three directions, giving rise to the formation of hydroxy-/3-lapachone, isopropylfuran-a-naphthaquinone, and iso- propylf uran-/3-naphthaquinone. The changes involved are as follows. I. Hydroxy-P-lapachone is formed, Dih ydroxyhydrolapachol. Hydroxy-B-lapachone. 11. Dihydroxyhydrolapachol is converted into hydroxyisolapachol, Not isolat,ed.AND ITS DERIVATIVES. 1369 0 0 0 0 Hydroxyisolapachol. 111. Hydroxyisolapachol then gives off water, and is simultaneously converted into its internal a- and P-anhydrides, 0 - 0 Hjdroxy isolapachol.Isopropy If uran-a-naphthaquinone. 0-7 Isopropplf uran-8-nap~haquinnne. Whilst it is not possible to isolate hydroxyisolapachol under the above circumstances, because it so readily undergoes furkher change, its formation as an intermediate product can be demonstrated by employing sulphuric acid somewhat diluted, but even in this case small quantities only escape further action. That the change into the isopropylf uran-naphthaquinones takes place through the interrnediat!e stage as above shown is further proved by the reconversion of both the anhydrides into the same hydroxyisolapachol by boiling aqueous solutions of the alkalis. The hydroxyisolapachol thus isolated, when snbmitted to the action of concentrated sulphuric acid is again con- verted into a mixture of both anhydrides.The above interpretation of the dehydration of di hydroxyhydrolapachol resulting in the for- mation of hydroxyisolapachol and the isopropylfuran-naphthaquinones, is the only one suggesting itself which is in perfect accord with the whole of the facts accumulated in the study of the compounds of this group. Its acceptance would seem to be rendered necessary by the possibility of converting iso-P-lapachol into hydroxyisolapachoi and the isopropylfuran-naphthaquinones (pp. 1360, 1379), and also by the existence of the lomatiols and dehydrolapachone (see follomng paper), which hare most probably the following formulze respectively.1370 HOOKER : THE CO-\JSTITUTION OF LAPACHOL Lomatiol and isolomatiol. Deh ydrolapachone.Whilst the above changea are those occurring with concentrated eulphuric acid in the cold, there are yet others which are effected hy the somewhat dilute acid at the boiling temperature. In this case however, the principal product of the action is isopropylfuran- a-naphthaquinone, small quantities of hydroxy-a-lapacho)?e and anhydrodihydroxyhydrolapachol being simultaneously formed. The last two products are liere met with for the first time. The relation of hydroxy-r-lapachone to hydroxy-/3-lapachone is the same as that existing between a- and 6-lapachone, but the methods which are applicable for the conversion of a- into p-lapachone, and vice vers6, have failed to produce corresponding changes with the hydroxylapachones. Anhydrodihydroxyhydrolapachol is formed by the removal of one molecule of water from dihydroxyhydrolnpachol ; it has consequently the formula CISH,,Oa, being isomeric with the hydroxylapachones.Its structure is undoubtedly correctly represented as follows. I \)\/OH 0 This may be inferred from the following facts. I. Anhydrodihydroxyb ydrolnpachol dissolves in alkaline solutions readily, forms intensely coloured stable salts, and has the properties of a hydroxynaphthaquinone generally ; consequently the hydroxyl group attached to the naphthalene nucleus has not been disturbed. 11. It can be dissolved in concentrated sulphuric acid, and pre- cipitated therefrom unchanged. This proves the absence of the original hydroxyl groups attached to the second and third carbon atoms of the side chain, as in the presence of an hydroxyl group at either of these points, an internal anhydride would undoubtedly be formed by exposure to t.he action of the acid.Isoprop y &.wan-a-naphtha pinone. This compound was first prepared by heating dihydroxyhydro- lapachol in acetic acid Bolution with a small quantity of sulphuric acid. This method of preparation has its disadvantages, and as it was subsequently improved on, it is only given in detail here, because,ABD ITS DERIVATIVES. 1371 in addition to isopropylfuran-a-naphthaquinone, two secondary pro- ducts were isolated as the result of the action, which hare not yet been obtained as satisfactorily in any other way. Twelve grams of dihydroxyhydrolapachol mere dissolved in 200 C.C. of acetic acid, to which 5 C.C.of concentrated sulphuric acid, sp. QY. 1.84, had been previously added. The solution was boiled with a reflux condenser for 20 minutes, during which time it changed in colour from orange to greenish-brown ; it was then immediately poured into a large volume of cold water. A dark oily substance was precipitated, which soon commenced t o crystalli~e, and on the following day was readily filtered off. In addition to dark coloured resinous products, the crude substance consisted of a mixtiire of iso- propylfuran-a-naphthaquinone and another compound, fusing point 179*5O, which was subsequently proved to be acetoxy-a-lapachone. These compounds were sepa,rated by repeated crystallisation from alcohol, animal charcoal being a t first freely used. It was found after the resin had been removed by one or two crystallisations that isopropylfiiran-a-naphthaquinone, which crystallises very readily from sufficiently concentrated solutionP, first separated.It was thus obtained in heavy needles, from which the supernatant liquid, still retaining the larger quantity of the acetoxy-a-lapachone, was readily poured off. The solution on furhher standing, deposited both substances, acetoxy-x-lapachone predominating, however, in the mixture. A preliminary separation having been thus effected, no difficulty was encountered in subsequently completely purifying the compounds. The yield of isopropylfuran-a-naphtha.quinone amounted t o about 33 per cent., and that of acetoxy-a-lapachone to about 6 per cent. of the dihydroxyhydrolapachol employed.Isopropylfuran-a-naphthaquinone fuses a t l l O o , and, when abso- lutely pure, crystallises in canary-yellow needles, but as it is difficult to remove the last traces of colouring matter, the compound as usually obtained is apt to be light brown. From moderately impure solutions, the needles deposited are almost black, and frequently so much shortened that they appear as heavy grains. If a moderately concentrated alcoholic solution be rapidly cooled by immersion in cold water, care being taken not to agitate or otherwise disturb it, the solution apparently solidifies as the compound separates, in pale yellow, flattened needles, grouped together in globular masses, If now a few fragments of the crystals of the compound as ordinarily obtained be dropped upon the surface of the crystalline mass, they gradually increase in size and number at the expense of the more bulky variety, replacing i t entirely in the course of a few hours.This change is very striking and characteristic. Analysis gave the following figures.1372 HOOKER : THE CONSTITUTIOX OF LAPACHOL 0.1617 gave 0.4423 CO? and 0.0751 H,O. C1,H,,03 req5ires C = 75-00 ; H = 5-00 per cent. Isopropylfuran-a-napbthaquinone can be distinguished from all the remaining compounds so far obtained from lapachol by the colonr of its solution in concentrated sulphuric acid, which is intenselj crimson. On dilution, the substance is reprecipitated from the acid unchanged. Isopropylfuran-a-naphthaquinone is best prepared as follows. 600 C.C. of dilute sulphuric acid (one volume of acid, sp.gr. 1.84, and two volumes of water) are heated to boiling, and then transferred to a flask containing 8 grams of dihydroxyhydrolapachol. A few fragments of a porous tile, &c., are added to prevent the solution from becoming snperheated and obviate bumping. A reflux con- denser is then adjusted, the heating immediately resumed, and t,he solution kept briskly boiling for 15 minutes. Hydroxy-p-lspachone appears to be formed as the substance first dissolves, a clear orangc- red solution being obtained; as the action is continued, the liquid becomes turbid, an oil separates which grndually darkens, and becomes ultimately greenish-brown. After the boiling has been con- tinued for the required time, the solution is allowed to stand until the dark oily substance has conipletcly crystallised ; this is $hen col- lected,* washed, and digested for about 24 hours with a 1 per cent.solution of sodium hydroxide, in order to remove small quantities oE nnhydrodihydroxyhydrolapachol (compare p. 1378). The crude sub- stance is then purified by crystallisation from alcohol, the addition of mima1 charcoal being desirable. The weight of the purified substance approximates to about half of that of the dihydroxyhydrolapachol euiployed in its preparation. C = 74.59 ; H = 5.16. Acetox y- a- lapachone a d H y drox y- a- lap a choqae. The compound obtained in the preparation of isopropylfuran- a-naphthaquinone (p. 1371), cryst,allising from alcohol in pale yellow woolly needles, and fusing at 179-5", was analysed, with the following results.0.2044 gave 0.5072 CO, and 0,0993 H,O. C = 67-67 ; H = 5.39. 0.2058 ,, 0.5108 ,, ,, 0.0995 ,, C = 67.69; H = 5.36. C1,H1605 requires C = 68.00; H = 5.33 per cent. The formation of a compound of the formula CnH&j from dihydroxyhydrolapachol is i n part due t o the aclion of the acetic acid used as a solvent, Thus, * The sulphnric acid filtrate, in addition to isopropg lfuran-n-naphthaquinone and anhydrodihydroxyhydrolapachol, also contains small quantities of hpdroxy-u- lspachone.AKD ITS DERIVATIVES. 1373 I. C,,H,,OS = C,5H1401 + H,O. 11. C1,H,,O, + CH,3*COOH = C,,H,,(O*CO*CH,)O, + H,O. The dihydroxyhydrolapachol is thus first conrerted into an internal anhydride, which then Fields an acetyl derivative; the acetate is yellow, and is derived from the hitherto unknown hydroxy-a- lapachone The correctness of these conclusions was proved as follows.I. By the removal of the acetyl group and the isolation of hydroxy- a- 1 apachone. 11. By the reconversion of hydroxy-a-Iapachone into dihydroxy- hydrolapachol by the action of a boiling aqueous solution of sodium hydroxide. 111. By the reconrersion of the hgdroxy-a-lapachone into the acetate fnsing at 179.5' by the action of acetic anhydride. Acetoxy-a-lapachone dissolves in concentrated sulphnric acid, g i v i n g an orange-red solution from which, when freshly prepared, the addition of water reprecipitates the compound apparently unai - tered. The acid solution, on long standing, however, slowly under- goes a change, and eventually, after a week or two, becomes crimson in coloiir, doubtless owing t o the formation of isopropylfuran-a-naph- thaquinone. The conversion of acetoxj-a-lapaclione into hydroxy-a-lapachone cannot be accomplished by dilute caustic soda (1 per cent.), as it was fomd that in addition to effecting the removal of the acetyl group, the alkali simultaneously opened the side ring, converting the corn- pound into dihydroxyhydrolapachol. From the behaviour of the lapachones previously studied this was to have been expected.Whilst apparently pure, the dihydroxyhydrolapachol obtained did not fuse sharply even after several recrystallisations from alcohol ; arid the hydroxy-P-lepachone into which it was conyerted for further identification, whilst also apparently pure, fused through a com- paratively wide range.It is therefore probable that the dihydroxy- hydrolapachol contained a small quantity of its acetyl derivative. The acetyl group may be removed from acetoxy-a-lapachone by the action of dilute sulphuric acid. The strength of the acid is of importaiice, as if too weak, the hydrolysis does not occur, and if too strong, the hydroxy-a-lapachone first formed immediately undergoes further change, and cannot be isolated. After a number of experi- ments, it was found that good results may be obtained as follows.1374 HOOKER : THE CONSTITUTION OF LAPACHOL For 1 gram of acetoxy-a-lapachone, 150 C.C. of dilute sulp'nuric acid (1 volume of acid, sp. gr. 1.84, and 3 volumes of water) are used. The substance is ground, and then thoroughly moistened with a small portion of the acid, this being most readily done, by adding a drop or two a t a time, to the substance still in the mort8ar..It is then rinsed with a few C.C. of the acid, kept in reserve for the purpose, into the main portion previously heated to boiling in a vessel provided with a reflux condenser. The heating is continued for precisely six minutes from the time the solution recommences to boil, and if the substance has been carefully ground, it mill dissolve almost completely in this time. The solution is now cooled as rapidly as possible, and immediately filtered, being poured back, if necessary, until quite bright. Hydroxy-a-lapachone soon commences to separate in small, bright, yellow crystals, but crystallisation occurs slowly, and is complete after some hours only.2.02 grams of hydroxy-a- lapachone were obtained from 2.84 of the acetat.e. For purification for analysis, the compound was crystallised from alcohol, from which it separates slowly in a ricb, yellow crust con- sisting of numerous small rosettes, fusing at about 187". 0.1948 gram gave 0.4955 CO, and 0.0946 H,O. C = 69-37 ; H = 5.39. CI5Hl4O4 requires C = 69.76 ; H = 5.42 per cent. Hydroxy-a-lapachone was reconverted into its acetyl derivative as follows: 0.18 gram was mixed with 0.36 gram of dried sodium acetate, and boiled for a few minutes in a test tube with 5 C.C. of acetic anhydride. The liquid was then poured into water, and the oil which first separated soon solidified to a pale yellow, crystalline substance, which, after cry stallisation from alcohol, was recognised by its melting point, 179O, by its crystalline form and other proper- ties, to be the compound sought for.By the action of dilute alkalis, hydroxy-a-lapachone, like hydroxy- P-lapachone (Trans., 1892, 649) is converted into dihydroxyhydro- lapachol. 0.18 gram was boiled for a, few minutes with about 13 C.C. of 1 per cent. sodium hydroxide ; the substance dissolved readily to a claret coloured solution. Acetic acid was then added in slight excess. No precipitation occurred immediately, but yellow crystals separated slowly on standing, which were found to be identical in all particulars with t'hose of dihydroxyhydrolapachol. Hydroxy-a-la pachone dissolves in concentrated sulphuric acid in the cold to an orange-red solution. The addition of water discharges most of the colour, leaving the solution jellow, but does not cause the immediate formation of any precipitate ; unchanged hydroxy-a- lapachone separates, however, slowly on standing.Noderate heating with somewhat dilute sulphuric acid convertsAXD ITS DERIVATIVES. 1375 hydroxy-a-lapachone into isopropylf uran-cx-naphthaquinone, and for this reason it is necessary to carefully follow the directions given above in the preparation of hydroxg-z-lapachone from its acetate, otherwise the substance liberated will undergo this further change. 0.19 gram of hydroxy-u-lapachone was boiled with 10 C.C. of dilute sulphuric acid (acid sp. gr. 1-S4, 1 volume, water 2 volumes) for about seyen minutes. The substance dissolved to a clear, Sellow sohition, which soon became turbid.When cold, the partly crgstallised, brown deposit was filtered off and purified by crystallisation from alcohol. It was then recognised by its fusing point, 109*5', by the crimson colour with which it is dissolved in concentrated sulphuric acid, and by crystallising in the two characteristic forms, as isopropyl- furan-a-naphthaqninone. Hy&oz yisol apachol. When isopropylfurm - u - naphthaquinone is boiled with dilute caustic soda, the furfuran r i n g is opened, and hydroxyisolapachol is formed. The change does not take place as smoothly as in the con- version of the various lapachones into the corresponding hydroxyl compounds, and the action of the alkali is not so energetic. The operation was coiiducted as follows: 6 grams of isopropyl- furnn-x-naphthaquinone were boiled under a reflux condenser with 690 C.C.of a 1 per cent. solution of sodium hydroxide for. .nearly three hours. The substance first fused, then gradually passed into solut>ion, leaving a solid residue, consisting of a new compound, which did not appear t o be further attacked by the alkali. As the highly coloured aolution cooled, some unchanged isopropyl-furan- a-naphthaquinone separated? which had probably passed into solu- tion in consequence of the reduction of its qninone groiip, and which in proportion as air gained access to it was oxidised and reprecipitated." To complete the oxidation, air was drawn through the cold alkaline solution f o r two or three hours; the precipitate was then filtered off, and the filtrate acidified with dilute hydro- chloric acid.The hydroxyisolapachol which then separated as a curdy precipitate weighed, when dry, about 3.4 grams. Hydroxyisolapachol is very soluble in alcohol even when dilute. It separates from this solvent in yellow, silky needles, which, when pure, melt at 133*5-134°. 0.2311 gave 0.5893 COz and 0.1119 H20. C = 69.54; H = 5.38. CI5H,,Oa requires C = 69.76; H = 5.42 per cent. * The unchanged substance, mixed with the compound above referred to, was again submitted t o the action of sodium hydroxide, and this resulted in the further gain of a small quantity, about 0.65 gram, of hydroxyisolapachol.137 ri HOOKER : THE CONSTITUTION OF LXPACHOL It dissolves readily in dilute alkaline solutions, the colour pro- duced being intermediate in shade between the claret-red of lnpachoI and the orange-red of /3- hydroxy-a-naphthaqninone.T’he cautious addition of hydrochloric acid to a moderat,ely dilute alkaline solution causes the liquid to set to a pale yellow, jelly-like mass. Hydroxyisolnpachol is dissolved by concen tmted sul phuric acid to an orange-red solution, which almost instantly changes t o a brown, arid ultimately becomes a dark, dull red. 0.25 gram was dissolved in zbout 5 C.C. of concentrated sulphuric acid, which was allowed to act for about five minutes. The acid was then poured into water. When the resulting orange-red precipitate had become crystalline, the microscope revealed a mixture of yellow and red needles. These were separated by crystallisation from dilute alcohol, and were then I ccognised by their melting points, crystalline form, and coloui* re- actions with concentrated sulphuric acid as isopropylfuran-a- and iso- propylf uran-/3-naphthaquinone (see below) respectively.Isopropy ljtcran- P-napht haquiuone. If either dihydroxyhydrolapachol or hy droxyisolapachol be d is- solved in concentrated sulphuric acid, isopropylfuran-&naphtha- quinone is formed. I n both cases, howecer, it is only one (compare pp. 1368, 1369) of the products of the action, and hence its prepara- tion from either of the above substances by this method is tedious, and the yield small. If hydroxjisolapachol be reduced to the corresponding hydro- quinone, OH /\/\CH:C (OH) .CH <:z I I jOH 3 \/\/ *‘ OH the anhydride formation still takes place readily.The action, how- ever, under the conditions given below, is almost entirely confined to the hydroxyl group occupying the a-position in the naphthalene nucleus. The resulting compound can then be readily converted by oxidation into isopropylfuran-P-naphthaquinone. The operation is conducted as follows. 2-5 grams of hydroxyisolapachol is dissolved by the aid of heat in 100 C.C. in acetic acid, and 80 C.C. of water. 2.5 grams of zinc dust, followed by 40 C.C. of dilute hydrochloric acid (3 volumes water and 1 voluume acid, sp. gr. 1.20), are then added, and the solution is boiled under a reflux condenser for five minutes. The excess of zinc is next filtered off, and 0.65 gram of chromic acid dissolved in 25 C.C.XSD ITS DERIVATIVES. 1377 of water added.commence to separate shortly afterwards. melted sharply at 94-95?. Red needles of isopropylfuran-P-naphthaquinone For analysis, the substancg was recrystallised from alcohol until i t 0.2096 gave 0.5741 CO, and 0.0956 H,O. C,,H,,O, requires C = 75.0; I3 = 5.0 per cent. If a small quaiitilly of isopropylf uran-p-naph thaquinone be dissolved in a few drops of acetic acid, a deep, orange-red solution is obtained, which becomes intensely crimson on the addition of a few drops of concentrated sulphuric acid, and as the quantity of sulphuric acid is gradually increased becomes purple, and finally dark green. It is dissolved by concentrated sulphuric acid alone to a rich blue-green solution, from which water, if added soon afterwards, reprecipitates the substance essentially unaltered.The prolonged action of con- centrat ed sulphuric acid gradually, however, produces a change, resulting in the formation, amongst other prodnct,r;, of some isopropyl- f iiran - x-naphthaquinone. Pseudodehy drolapschone (Paternb's isolapachone), which is iso- meric with isopropylfuran-P-naphthaqninone, also develops almost exactly the same shade of green wit'h concentrated snlphuric acid ; it may, however, be readily distinguished by the fact that the green passes in a few minutes into a dark purple, whereas no such change occurs with isopropylfuran-,k?-naphthaquinone. The melting point of the two substances also differs widely, as pseudodehydroIapachone melts a t 140-141'. They can, moreover, be readily distinguished by their behaviour with dilute caustic alkalis ; in both cases acid substances are formed, but that obtained from isopropylfuran-p- naphthaquinone alone is stable, whilst that to which pseudodehydro- lapachone gives rise (compare Trans., 1892, 61, 623, 624), when liber- ated from its salts, immediately passes into its internal anhydride.Dehydrolapachone (see following paper), the red anhydride obtained by Rennie (Trans., 1195, 67, 792) from lomatiol, has also the formula C,,H1203, being isomeric with the above compounds, but its melting point, 11O-11lo, as well as the orange-red colour, passing into a brown, developed with concentrated sulphuric acid, serves to dis- tinguish it from them both. Isopropylfuran-P-naphthaquinone (0.50 gram) was boiled with a 1 per cent. solution of sodium hydroxide (50 c.c.) for about 30 minutes.The crystals fused to a red oil, which gradually decreased in quantity and darkened, until finally little besides small quantities of a, blue substance* remained undissolved. The alkaline solution was filtered off from this and acidified with acetic acid. The yellow, C = 74-70; H = 5.07. * This was found by microscopic examination to be distinctly crystalline.13713 HOOKER : THE COSSTITUTION O F LAPACHOL curdy, precipitate, consisting of microscopic tufts of needles, weighed 0.37 gram, and aQter crystallisation from dilute alcohol, was recognised by its melting point and other properties as hydroxyisolapachol. Isopropylfuran-P-naplithaquinone dissolves in concentrated hydro- chloric acid with difficulty, forming a purple solution.Under these circumstances, it is gradually changed, heing ultimately converted into the corresponding a-naph thaquinone compound. When action is allowed to take place in the cold, it is possible to demonstrate the presence of an intermediate product soluble in alkalis, and, doubtless, corresponding to chlorhydrolapachol, which was previously shown (Trans., 1892,61,621) to be formed by the action of hydrochloric acid on p-lapachone in its conversion into a-lapachone. The change can be readily effected i n an hour or so by digesting the substance in a relatively large quantity of hydrochloric acid, sp. gr. 1.20, at a tem- perature of about 75". The isopropylfuran-a-naphthaquinone obtained in this way possessed all the characteristics of that prepared as pre- viously described from dihydroxyhydrolapachol.Isopropylfuran-P-naphthaquinone, in virtue of its orthoquinone group, gives a characteristic azine with orthotolylenediamine, which crystallises in yellow, silky needles, and melts with decomposition a t about 132", darkening? and showing signs of fusion some degrees lower. The azine is coloured dark green by concentrated sulphuric acid, but the solution, when seen in sufficiently thin films, appears pink ; the addition of a small quantity of water precipitates a dull-red salt, which is decomposed on further dilution. The axine undergoes a change when its alcoholic solution is allowed to stand for a few days a t the ordinary temperature. The solution becomes darker, and exhibits increased fluorescence, depositing fluffy, orange-red crystals, which develop a csrmine colour with concentrated sulphuric acid.For much valuable experimental assistance in the preceding study of the dehydration of dihydroxyhydrolapachol, I am greatly indebted to MY. J. G. Walsh, junior, whose painstaking and careful work has greatly contributed to the successful conclusion of this research. A~tl~ydrodihydroxyhydrolapachol (compare p. 1370). This substance is formed in small quantity in the preparation of isopropylfuran-a-naphthaquinone from dihydroxyhydrolapachol. In order to isolate it, the latter, in its crude condition, is digested with a weak solution of sodium hydroxide (compare p. 1372). The alkaline extract is acidified with dilute hydrochloric acid, a n d the resulting precipitate purified by crystallisation several times from alcohol, animal charcoal being a t first used.The crude substance, previous to recrystallisation, amounted to only about 4 per cent. of the di- hSdroxy hydrolapach 01.AND ITS DERIVATIVES. 1379 Anhydrodihydroxyhydrolapachol crystallises in small, yellow tufts of short needles, which fuse at 190.5-191°, and dissolve in alkaline solutions with a rich crimson-red colour ; the substance is reprecipi- tated by acids in a distinctly crystalline condition. Like all the other lapachol derivatives previously studied, i t appears to be perfectly stable in alkaline solution ; even after boiling for about five hours with a 1 per cent. solution of sodium hydroxide, the substance had undergone no change. The folIowing analytical results were obtained.I. 0.1405 gave 0.3579 CO, and 0.0686 H,O. C = 69-47 ; H = 5-42, 11. 0.1638 ,, 0.4176 CO, and 0.0818 H20. C = 69.54; H = 5.54. CI5H,,Oa requires C = 69-76 ; H = 5-46 per cent. Anhydrodihydroxyhydrolapachol dissolves in concentrated sulph- uric acid to an orange-red solution, from which, water, if added soon afterwards, reprecipitates the substance essentially unaltered. If, however, the solution, previous to dilution, be allowed to stand for two or three days, the substance appears to be complctely changed, and a brown precipitate is then obtained, which differs in its pro- perties from anhydrodihydroxyh ydrolapachol, but still remains almost entirely soluble to a red solution in dilute alkalis. Conversion of Iso- p- lapachol into Isoproy y ljuran-p-n aphthaquinone.Seven grams of bromine, dissolved in 30 C.C. of chloroform, were gradually added to 10 grams of iso-P-lapachol, dissolved in 65 C.C. of chloroform. The bromine appeared to be completely absorbed, and the solution of iso-0-lapachol became lighter in colour as it was added. The chloroform was a t once distilled off from a water bath ; the residue, which was resinous, was taken up in 50 C.C. of alcohol. No perceptible quantity of hydrogen bromide passed over with the chloroform, hence an additive product had undoubtedly been formed. The alcoholic solution was allowed to stand eight days at the autumn temperature of the laboratory, during which time it darkened in colour, becoming intensely orange-red ; the alcohol was then partially distilled off.During the operation, hydrogen bromide and other pungent fumes passed over. When reduced to a small bulk, water was added to the solution ; the dark red resin precipitated was washed and then gently warmed with a small quantity of a 1 per cent. soh- tion of sodium hydroxide, to remove the traces of acid still present ; the substance shortly commenced to cry stallise. After standing over- night in contact with the alkali, i t was crystallised from dilute alcohol; 3.4 grams of beautiful red needles were obtained. The alcoholic mother liquor was concentrated, bnt, as no .further crystals separated, the substance in solution was again submitted to VOL. LXIX. 4,1380 CONSTITUTION OF LAPACHOL AND ITS DERIVATIVES. the treatment with 1 per cent. sodium hydroxide, above described. This resulted in a gain of an additional gram of the red needles; thus, in all, the yield of the pure substance amounted to 44 per cent. of the iso-P-lapachol used. For analysis, the substlance wsts again crystallised from alcohol. I. 0.2516 gave 0.6875 COz and 0.1111 H20. C = 74.52 ; H == 4.90. 11. 0.1489 ,, 0.4091 ,, ,, 0.0669 ,, C = 74.93; H = 4.99. C,,H,,Oa requires C = 75.00 ; H = 5.00 per cent. The substance has thus the composition C,5H,,0s, and is otherwise identical with isopropylfuran-/3-naphthaquinone. It melts at 94-95', dissolves in concentrated sulphuric acid with a characteristic blue-green colour, is converted by alkalis into hydroxyisolapachol, and by hydro- chloric acid into isopropylfuran-a-naphthaquinone ; it gives an azine with orthotolylenediamine, melting at about 132O, thus behaving in all particulars similarly to the isopropylfuran-/3-naphthaquinone described above, with which it is, beyond all doubt, identical. Various attempts were made to isolate the intermediate products in a pure condition, but these have not met with success. The crude substance, which is left in a resinous condition on the evapora- tion of the chloroform (see above), has doubtless the formula CloH4 CHBrGHBr*CH7. {:; This was treated with 1 per cent. sodium hydroxide in the cold ; it From the solution, impure dissolved slowly and almost completely. hydroxgisolapachol was precipitated by the addition of acids. The condensation of f3-hydroxy-a-naphthaquinone with isovaler- aldehyde, resulting in the formation of iso-P-lapachol, and the further change of the latter under the influence of bromine into a furfnran derivative of /3-naphthaquinone, which can then be converted into the corresponding derivative of a-naphthaquinone, would seem to justify the anticipation that we have in these reactions general methods, firstly, for the preparation of alkylene derivatives of hydr- oxynaphtha#quinone, in which the alkylene chain occupies the ,%position next to the hydroxyl group, and, secondly, f o r the conver- sion of these compounds into furfuran derivatives of both a- and P-naphthaqninone. This subject is also of interest as being likely t o furnish further material for general dednctions regarding the formation of internal anhydrides, and the conversion of ortho- into para- and of para- into ortho-quinone derivatives.HOOKER : LOMATIOL (HYDROXYISOLAPACHCIL). 138 1 In view of these possibilities, I have felt justified in undertaking a The results of my experiments will further study of these reactions. form the subject of a future paper. Philadelphia, U. 8. A .