2632 EINU AND YURCH TRYPANOCIDAL ACTION CCCLX1.-Trypnocidal ,4ction and Chemical Con-stitution. Part 11. Srylamides of 4-Aminophenyl-arsinic Acid. By HAROLD KING and WILLIAM OWEN MURCH. IN Part I (J. 1924 125 2595) about 25 arylamides of 4-amino-phenylarsinic acid were described with their toxicities to mice and their curative action on experimental trypanosomiasis in mice. Of this series of substances the one clearly indicated for further development was 3’-aminobenzoyl-4-aminophenylarsinic acid (11; R = H) which had a toxicity of 0.6 and a temporary curative action in a dose of 0-18 milligram per gram of mouse. NO2 NH2 R/-\CO*NHr\As03H2 + R/-\CO-NH/-\ASO,H \-/ \-/ \-/ NO2 NH2 NH, \-/ \-/ \-/ (11.) \-/ NO2 \J (I.) -1 R /-\CO*NH/-\AsO,H + R r\CO*NHr\A~O,H, There were two obvious ways of modifying this structure with a view to following the change in its curative action.The one was to replace the m-aminobenzoyl group by other m-aminobenzoyl radicals substituted in the p-position. The parent p-substituted m-nitrobenzoic acids are readily accessible in quantity provided the p-substituent is op-directive by starting from the p-substi-tuted anilines which by the Sandmeyer reaction for the prepar-ation of nitdes saponification and nitration yield the required p-substituted m-nitrobenzoic acids. Anisic acid is however more readily accessible from anethole MeO*C6H1-CH:CHMe a process having been evolved whereby it may be obtained in large quantities in almost quantitative yield. The other way was to re-nitrate the mononitroarylamide {I) in the hope that the main product would be a dinitroarylamide of the type (111) which would give a series of interesting diamines (IV).P I . 1 (IV. AND CHEMICAL CONS!PITUTION. PABT II. 2633 This hw now been done a aeries of p-eubstitutd m-nitrobenzoic acids having been prepared in which R = Me OMe OEt O.CO&t, or Cl. These readily yield the corresponding acid chloridw which can be introduced into 4-aminophenylarsinic acid in ao--sOyo yield by suitable m&cations of the Schotten-Baumann method, e t h y l t + q l CW for instance being relatively sensitive to hydroxyl ions necessitating the use of sodium acetate as halogen acid fixative. The corresponding aminomink acids were obtained by reduction with ferrous chloride and alkali in 55-95 yo yield the et h yhrboncstonit robenzo y 1 -4-aminup7~n ylarsinic acid (U) being first converted into the hydroxynitro-acid (VI) by alkali.(v.1 EtO2C*Or~C0*NH<~AsO3H2 \-/ + NO2 NO2 HOC_>CO*NH<~>A.03H2 (VI.) The maximum dose tolerated by mice expressed in milligrams per gram of mouse of the six isomeric nitro-benzoylarsinic acids (I) with variation of the group R is shown below : R = .................. H. Me. OMe. OEt. OH. C1. Doeis toleruta ...... 0.6 0-2 0.2 0.2 0-8 0-2 whereas for the corresponding amino-acids (11) the maximum tolerated dose and the minimum curative dose on Trypamma equiperdum in mice are given, R = .................. H. Me. OMe. OEt. OH. C1. Doeis to2era.h ...... 0.6 0.3 0.7 0-6 0.5 0.1 D O S i 8 CU~&~VCZ ......0.18 0.15 0.4 0-6 0-2 0-075 ( T = 3) (T = 10) (T = 10) ( r = 7) r signifying the number of days during which the blood-stream remains free from trypanosomes. Of this group two members have effected permanent cures in mice and of these two the p-meth-osy-derivative was superior to the p-chloro-derivative. A m i n o h ~ d r o x ~ b e n z o y l - 4 - a ? n i ~ ~ n y ~ r s i n ~ acid the reduction product of (VI) was of special interest because it contained the o-aminophenol grouping of salvarsan (VII) and could therefore be reduced to an arseno-derivative (VIII) soluble in alkalis and suitable for experimental testing on trypanosomiasis. This -no-base was ten times as toxic (T = 0.05) as its parent amino-acid and on two-fifths of this dose (C = 0.02) a temporar 2634 KING AND MuacH TRYPANOCIDAL ACTION cure of mice was effected for 7 days.The therapeutic indices, CITY for the parent amino-acid and its arseno-derivative are thus identical. The acetylation of the amino-groups in substituted arsinic acids can readily be effected by dissolving in alkali and shaking with excess of acetic anhydride. When applied to the o-aminophenol grouping in 3'-amino-4'-hydroxybenzoyl-4-aminophenylarsinic acid, a practically quantitative yield waa obtained of the ON-diucetyl derivative which on standing in solution in A'-sodium hydroxide, gave the N-acetyl derivative. ON-Diacetylation under such con-ditions has apparently only once previously been observed by Raiford and Greider ( J . Amer. Chem. Soc. 1924 4.6 430) who obtained some diacetyl derivative on acetylation of o-aminophenol in sodium hydroxide solution it having been overlooked by Lumiere and Barbier (Bull.Soc. chim. 1905 33 783) who recommended aqueous acetylation. It seems however to be a general method for the preparation of ON-diacetyl derivatives provided the pro-duct when once it is formed is never subjected to high concen-trations of hydroxyl ions. Applied to o-aminophenol-p-arsinic acid it readily yields the ON-diacetyl derivative. 3'-Acetylamino-4'-acetoxybenzoyl-4-aminophenylarsinic acid proved to be devoid of trypanocidal action. The avenue opened by the curative action of aminoanisoylamino-phenyhrsinic acid was explored by the preparation of its N-acetyl and N-carbethoxy-derivatives of its formuldehydesulphoxylate and its carbamide.Of these four substances the former two were tested therapeutically but proved to be devoid of trypanocidal action. As this might be ascribed to loss of amphoteric character through acylation of the amino-group by non-basic radicals two, more complex tri-nuclear amides were prepared namely the 3"-a;minobenzoyZ-(R = H) and 3"-amin0-4"-anisOyl-(R = OMe) de-rivatives of 3'-amino-4'-anisoyl-4-aminophenylarsinic acid (IX) in which amphoteric character was preserved. These also were devoid of trypanocidal action. Both are f a r inferior to salvarsan. The nitration of a series of p-substituted benzoic acids in which the p-substituent is Me OMe OEt O*CO,Et or Cl has enabled us to effect both a comparison of the relative ease of nitration of these benzoic acids and the relative ease of replacement of the carbosyl group by the nitro-group during nitration.The firs AND CHEMICAL CONSTzTUTrON. PART II. 2635 three acids are readily nitrated on the water-bath by 70:$ nitric acid; the last two are unaffected by acid of this strengbh but are nitrated in 76 and 90% yield respectively by 94% nitric acid. If other comparable observations from the literature be included, the following series is obtained : HNO yo ......... 14-300,6. 70%. 92-940/,. R = ............... OH,1NMe22. Me,OMe,OEt. F,SBr,’CI,O.CO,Et. (l) Griess Ber. 1887,20,408. (2) Reverdin Ber. 1907,40,2442. (3) Rouche, (4) Hubner Philipp and Ohly Anden, 1867 la 248. This series is substantially the one to be expected from a con-sideration of the relative directive powers of a member of any one of these groups in competition with a member to the right of it, for nitric acid the two substituents being situated in the para.or ortho-positions with respesf to each other. Thus there is ample evidence in the literature that when either OH or NMe is set against Me OMe OEt F C1 Br in the para-position the NO, group enters mainly ortho to OH or NMe,; when the two groups are in the ortho-position to each other the evidence although not so complete supports the same conclusion. Again when Me0 or EtO is set against C1 or Br in the para- or ortho-position the NO,-group is preferentially directed ortho or para respectively to the Me0 or EtO group in both cases but in the relative directive powers of Me and halogens there is an element of doubt.* In 1912 Holleman and Wibaiit (Proc.K . A M . Wetensch. Amsterdam 15 594) from the nitration of o- m- and p-chlorotoluenes drew the conclusion that chlorine induces a velocity of substitution 1-5 times as great as that caused by the methyl radical. Later however Holleman (Re. trav. chirn. 1915 34 283) found that in the nitration of p-bromotoluene the methyl group had undoubtedly the superior orienting power. The latter result is more in accord with our own observations. We have been unable to find in the literature anything bearing on the relative orienting influence of the O=CO,Et group on an entering nitro-group. The relative directive powers for ortho-substitution of a nitro-group may therefore be written in the order OH,NMe > Me,OMe,OEt > F,Cl,Br,O=CO,Et.In every case during nitration partial replacement of the carb-oxyl group by the nitro-group takes place with formation of p-sub-stituted nitrobenzenes and is evidently a general phenomenon. This is contrary to the view of Rouche (Zoc. cit.) who claimed that there was no formation of chloro- or bromo-nitrobenzenes by nitration of the chloro- or bromo-benzoic acids. In addition, * This is confirmed by the recent work of Francis Hill and Johnston (J. Amw. Chem. Soc. 1925,47,2231). Bull. A d . roy. Be&. 1921 534 2636 gzNG AND MlJlWH TRYPBNOCIDAL AC!CION further nitration to 2 4-dinitro-derivatives takes place to some extent depending on the time of heating. The relative yields of non-acidic fractions obtained on pouring the nitration mixtures into water are shown below.R = ............ C1. Me. F.1 OMe. OEt. O-CO,Et Yield yo ............ 1.1 2 11 11.5 14 15-5 HNO yo ............ 94 70 92 70 70 94 Time of heating ... 10 30 120 30 10 10 mins. Rouche loc. cit. Owing to the variation in strength of nitric acid used and the time of heating the only deductions allowable are that 'O-substi-tuents and possibly fluorine greatly facilitate replacement of C0,H by NO, and that O*CO,Et has a different relative orienting power to para-substitution from that which it has to ortho-substitution. It would be of interest to know whether the accumulation of groups in 3 4 5-trimethyl- trichloro- or trifluoro-benzoic acids would enhance the ease of replacement of CO,H by NO just as accumd-ation of the Me0 results in increased displacement of the formyl (Salway J.1909,95,1155) or carboxyl (Harding J. 1911,99 1585) grouping. The sulpho-group also is replaced to the extent of 5'7; by the nitro-group during the nitration of p-toluenesulphonic acid, as wilI be shown in a future communication. For the saponification of p-toluonitrile Herb (,4nnalenY 1890, 258 10) recommended the use of 750/ sulphuric acid. When applied to p-ethoxybenzonitrile EtO*C,H,.CN this had an un-expected result p henol-p - sulp honic acid OH C 6H4*S 03H being formed in good yield and no trace of the required carboxylic acid. A weaker sulphuric acid (60%) however yielded the required acid and amide in satisfactory yield. The dinitration of this series of mononitroarylamides (I) proceeds smoothly but as might be anticipated only yields exclusively one product (111) when R is H Me or C1.When R is OMe or OEt, a mixture of two dinitro-acids is obtained which is from a practical point of view not separable into its components but the com-position of which can readily be determined by an examination of the products of hydrolysis. The isomeric dinitro-acid has the structure (X) and the two arsinic acids obtained on hydrolysis of this and its isomeride (111) are almost quantitatively separable by 0.5N-hydrochloric acid. By this mea.n, it has been shown that whereas in the cases where R = H Me or C1 the secon AND CHEMICAL CON-ON. PART II. 2637 NO,-group enters a Merent nucleus from the one in which the NO,-group is already present when R = OMe or OEt the NO,-group is distributed between the two nuclei in the ratios 3 7 and 3 5 respectively in favour of the nucleus without a NO,-group.The diQmines (IV) were obtained on reduction with ferrous chloride and alkali. The toxicities and curative action of the dinitro- and diamino-arsinic acids where determined are given below : R = ..................... H. Me. C1. H. Me. C1. Dosh tole& ......... 0.1 0-025 0.2 >3-5 1.0 1-25 D O S b ctlfYZ&tk% ......... 0.06 - - 2 0-75 0.5 Dinitro-wide. Diamino-acids. ( r = 2 ) ( r = 29) Comparison of this table with the previous one for the mono-amino-acids shows that the introduction of a second amino-group lowers the toxicity many-fold and at the same time the sub-stances acquire permanent curative properties.It is of interest that 3 3'-dinitrobenzoyl-4-aminophenylarsinic acid causes a tem-porary disappearance of trcvpanosomes from the blood-stream for 2 days. The isomeric 3' 5'-dinitrobenzoyl-4-aminophenyla~c acid was prepared but attempts to isolate the arsinic acid con-taming the m-phenylenediamine group from it by reduction were unsuccessful. We desire to acknowledge our indebtedness to Miss F. M. h h a m and Miss J. Marchal of this department for the painstaking care with which they have determined the toxicities and trypanocidal action of the compounds described in this paper. EXPERIMENTAL. 3 3'-DinitrobenzoyE-4-ami~p~ny~r~nic Acid (III ; R = €I).-Benzoyl-p-aminophenylarsinic acid (19.3 g.) dissolved in 45 C.C.of sulphuric acid was nitrated at 0" by addition of a mixture of 8-5 C.C. of nitric acid (d 1-4) and 11-4 C.C. of sulphuric acid. The crude product obtained by pouring on to ice was collected allowed to air-dry and digested on the water-bath with 100 C.C. of glacial acetic acid. The product was now crystalline and the filtrate could be used for subsequent batches. The yield was 9276 of the theor-etical. This acid crystallised from 170 parts of boiling glacial acetic acid in fine silky needles forming a monohydrate (Found : Loss at lOO" 5.1 5-0. 1H,O requires loss 4.2%. Found As, 18.1. C,,H,,O,N&s requires As 18.2%). With alkali (1 mol.), it forms sparingly soluble sodium and potassium salts. Hydrolysis of 3 3'-Dinitrobenwyl-4-amirwphenylarsinic Acid.-(a) With acid.When the dinitro-acid (8-2 g.) was boiled for 2 hour 2638 KING AND MDRCH !l'BYPANOCIDAL ACTION with 100 C.C. of 16% hydrochloric acid it was recovered mainly unchanged (6-4 g.). The other products isolated were a small quantity of o-nitroaniline m-nitrobenzoic acid and 3-nitro-4-amino-phenylarsinic acid. (b) With aZkaZi. The dinitro-acid (6.3 9.) was boiled for 2 hours with 100 C.C. of N-sodium hydroxide. Practically quantitative yields were obtained of m-nitrobenzoic acid and 3-nitro-4-amino-phenylarsinic acid. 3 3'-Diaminobenzoyl-4-aminophenylarsinic Acid (IV ; R = H).-The dinitro-acid (16.4 g.) was dissolved a t - 5" in 292 C.C. of 2N-sodium hydroxide and treated with 101 g. (10% excess) of ferrous chloride dissolved in 124 C.C. of water.To the mixture 292 C.C. of 2N-sodium hydroxide were added the temperature throughout being below 0" The ferric hydroxide was filtered off and extracted three times by thorough mixing with 600 C.C. of 0.2Ar-sodium hydroxide each time. The combined filtrates were made neutral to Congo-paper and after keeping for 24 hours a t 0" the separated diamino-acid was collected. The filtrate was made alkaline with concentrated ammonia treated with 50 C.C. of magnesium chloride solution (1 l) and heated for 30 minutes in the boiling-water bath. The precipitated magnesium salt was collected dissolved whilst still damp in N-hydrochloric acid and the acidity to Congo-paper removed by addition of saturated sodium acetate solution. The diarnino-acid crystallised readily (total yield was 7.8 g.or 56%). The diamino-acid so prepared crystallises in clusters of leaflets. It diazotises and couples with alkaline p-naphthol with production of a deep red colour (Found As 20.9. C,,H,,O,N,As requires As 21.3%). 3 5-Dinitrobenxoic AcZ-The following process is an improve-ment on that of Shukov (Ber. 1895 28 1800). Ten g. of fused benzoic acid were dissolved in 100 g. of sulphuric acid and treated with 18.2 C.C. of fuming nitric acid. The solution was heated on the water-bath for 10 hours and poured into a litre of ice and water. The product (10 g.) m. p. 202" was washed with a little hot water and reprecipitated from dilute ammonia. It then melted a t 203-204". 3' 5'-Dinitrobenzoyl-4-amin~frenylarsinic Acid (X ; R = H) .-Sodium p-aminophenj-larsinate pentahydrate (25-26 g .) dissolved in 320 C.C.of 5% sodium hydroxide and cooled to - 3" was treated with 40 g. (2 mols.) of 3 5-dinitrobenzoyl chloride dissolved in toluene. The mixture was stirred for 45 minutes after at1 the acid chloride had been added. After acidification the mixture of acids was collected dried and extracted with ether. The insoluble arsinic acid was reprecipitated from ammoniacal solution (yield There was no o-nitroaniline found AND CHEMICAL CONSTITUTION. PABT II. 2639 84%). This diaitro-minic mid c r y ~ t a b a from 80 part^ of b o a 90% formic acid in fine needles. In glctcirtl acetic acid it is much less soluble (Found As 18.3. C,H,,-,O~& requires AS 18.2%). The maximum tolerated dose for mice is 0-5 mg.per g. of mouse. Attempti3 fo obtain the diamino-minic acid by reduction in alkaline solution with ferrous chloride were unsuccessful. 2-Ndro-p-tohic Acid.-This acid wm prepared by the following process which is an improvement on that of Fittig and Ramsay (AnnuZen 1873 168 251). p-Toluic acid (20 g.) was suspended in 200 C.C. of nitric acid (d 1-4) and heated on the water-bath for 30 minutes. On pouring- into water the yield of nib-acid was 23.5 g. m. p. 186-187". By ether extraction of the acid in alkaline solution 0.4 g. and of its aqueous mother-liquor 0.6 g. of crude p-nitrotoluene were obtained m. p. 51". 2-Nitro-p-toluoyl chhride prepared by the action of phosphorus pentachloride boils at 167-168" (corr.)/l6 121111. and melts at 20-21" (corr.). This acid chloride has recently been described as an oil (Johnson and Soderman J .Anaer. Chern. Soc. 1925 47, 1392). 3'-Nitro-4'-toluqyl-4-aminopheny2arsinic Acid ( I ; R = Me).-This acid was prepared in the same way as p'-nitrobenzoyl-p-amino-phenylminic acid (J. 1924 125 2602). The acid chloride how-ever reacts so slowly that stirring has to be continued for 2 or 3 hours. The mixed acids obtained on acidification were dried and extracted with ether in a Soxhlet apparatus (yield 55%). This arsinic acid is sparingly soluble in boiling acetic or formic acid and crystallises in small needles (Found As 19.6. C&,O,N& requires AS 19.7%). 3'-Amino-4'-toEuoyl-4~m~~~~~~~n~ Acid (I1 ; R = Me).-3'-Nitrotoluoyl-4-aminophenylarsinic acid (1 1-4 g.) was dissolved in 110 C.C.of chilled 2N-sodium hydroxide and 40 g. of ferrous chloride (20% excess) in 50 C.C. of water were run in the tem-perature being maintained between 0" and - 5". Finally 130 C.C. of 2N-sodium hydroxide were added. The ferric hydroxide was filtered off and extracted with two successive portions each of 240 C.C. of 0-4N-sodium hydrozcide. The combined filtrates were neutralised to Congo-paper and the crude acids collected. These were warmed with mccessive portions of N-hydrochloric acid a t 50" until no diazotisable material waa left. On addition of satur-ated sodium acetate solution to the successive f?ltmt~~ the amino-arsinic acid w&8 precipitated in fan-shaped clusters of small white needles. A further small quantity can be isolated from the k t precipitation mother-liquors by addition of ammonia and magnesium chloride and heating the solution.The magnesium salt separate 2640 KING AND IKURCH TBYPANOCIDAL AC'MON readily is dissolved in excess of N-hydrochloric acid and the free acid precipitated by sodium acetate. The yield was 5-7 g. or 54% of the theoretical. This amino-arsinic acid is very sparingly soluble in cold 3N-mineral acids but dissolves readily on warming. The hydro-chloride crystallises in small needles the nitrate in minute needles, and the sulp7uzte in square plates. It diazotises and couples with alkaline p-naphthol with production of a blood-red colour (Found : As 21.2. C,,H,,O,N&s requires As 21.4%). 3 3'-Dinitro-4'-toluoyl-4-ami~~nylarsinic Acid (I11 ; R = Me). -The mononitro-acid (7-5 g.) dissolved in sulphuric acid (20 c.c.) was n i b t e d a t 0" with a mixture of 1.9 C.C.of sulphuric acid and 1-4 C.C. of nitric acid (d 1-4). The yellow solid obtained on pouring on to ice was collected and when digested on the water-bath with glacial acetic acid became crystalline (yield 8 g . ) . This dinitro-acid is sparingly soluble in boiling acetic or formic acid and crystallises in needles (Found As 17.7. C,,H,O,N,As requires As 17.6%). Hydrolysis of 3 3'-Dinitro-4'-toluoyl-4-aminopheny~~~in~c Acid. -Two g. of the acid were boiled for 2 hours with 30 C.C. of N-sodium hydroxide. The acids obtained (2.05 g.) on neutralisat.ion to Congo-paper were extracted in a Soxhlet apparatus with ether. The ether-soluble portion (0.8 g . ) consisted of pure 2-nitro-p-foluic acid and the ether-insoluble portion of 3-nitro-4-aminophenylarsinic acid (1.25 g.) the yields being practically quantitative.3 3'-Diamino-4'-toluoyl-4-am~~~nylarsinic Acid (IV ; R = Me).-The dinitro-acid (8.5 g.) was reduced in the way described for the acid without the methyl group (above) save that it was found advantageous to increase the excess of ferrous chloride used from 10% to 20% and with it the amount of alkali to secure faint alkalinity at the end of the reduction. The alkaline filtrates from the ferric hydroxide extraction on neutralisation to Congo-paper gave no precipitate of diamino-acid. The acid was however, precipitated as the magnesium salt in ammoniacal solution by heating on the water-bath. The magnesium salt was dissolved in 70 C.C.of N-hydrochloric acid and the diamino-acid liberated by addition of saturated sodium acetate solution (yield 4-2 g. or 57%). This diamino-arsinic acid separates when liberated as described above in sphmo-crystals (Found As 20.4. C,,H,,O,N,As requires As 20.5%). Diazotised with sodium nitrite in hydro-chloric acid solution it turns yellow and couples intensely with alkaline @-naphthol with production of a brownish-red colour. Preprution of Anisic Acid.-Pure crystalline anethole (11.1 g.) was stirred vigorously with 50 C.C. of water a t ruom temperature, and 34.4 g. of potassium pemanganate in 1076 C.C. of water wer AND CHEMICIIL COHSTlTUTION. PART II. 2641 added at a consta4t rate within 76 minutea. No attempt was made to regulate the temperature which at its xmximwn ~ & 8 below 35".The very slight excess of permaqanate was reduced by warming with alcohol. After ftlhtion and extmction of the manganese oxides with dilute alkali the combined filtrates gave, on acidification 9-0 g. of almost pure anisic mid. An aliquot portion of the filtrate extracted with ether indicated the presence of a further 1-5 g. of equally pure anigic acid. The total yield is about 92% of the theoretical. Oxidation at mom temperature with 32 g. of permanganate (4 atoms of oxygen) and with addition of 12 g. of potassium hydr-oxide gave 4-6 g. of aniaic mid and 2.3 g. of mkldehyde (semi-carbazone m. p. 216-217" con.; Walbaum J . pr. Ck. 1903, 68,235 gives m. p. 203-204"). Oxidation at 0" gave 6-6 g. of a mixture of anisic acid md d y l -ketocarboxylic acid and 0-85 g.of anisaldehyde. Oxidation of pure anethole by Ladenburg and Fib's method as applied to oil of anise using potas8ium &chromate and mlphuric acid gave a 47% yield of anisic acid. 3-Ndro-4-ani8oyl ChlOt.ide.4-Nitroanigic acid was prepared by Auwers's method (Ber. 1897 30 1477) by heating 20 g. of anisic acid with 200 C.C. of nitric acid (d 1.4) on the water-bad& for 30 minutes. From 75 g. of anisic acid there were obtained 3-7 g. of a non-acidic fraction which on distillation gave 2-65 g. m. p. M 5 " b. p. (external bath temperature) 170"/20 mm. These constants agree with those of 4-nitroanhole. The non-volatile reaidue (1.0 g.) on two crystaktions from alcohol gave 2 Pdi-nitroanisole (0.45 g.) m.p. 87". In another experimenf which was strictly comparable with the nitration of p-foluic acid there were isolated from the nitration of 20 g. of anieic acid 20.2 g. of nitmanhic acid m. p. 190"; by ether extraction of this nitro-acid in alkahe solution 2.3 g. of nitroanhles m. p- 84"; and by ether extraction of the original aqueous mother-liquor 1.45 g. of nitro-anisola m. p. 45". The nitroanisic acid was converted info 3-nit~o-4-ankyZ W i d e by phosphorus pentachloride. This boils a t 210"/15 mm. is sparingly soluble in low-boiling petroleum but readily soluble in warm ether from which it crystallises in broad needles m. p. 52.5-53.5" (corn.). 3 ' - N i t r o - 4 ' - a n i s o y l - 4 - a m ~ ~ ~ y ~ 7 ~ n ~ Acid (I ; R = OMe).-This nitro-acid w&8 prepared in the same way as the corresponding toluoyl derivative.The yield waa 48.37& It is soluble in boiling formic acid and c r y s t a h therefrom in needlea but is sparingly soluble in boiling acetic acid from which it separates in woolly needles (Found As 18.9. C,,H,O,N& requires As 1809%). VOL. CXXVTI. 4 2642 KING M D MURCE "BYPANOCIDAL ACTION 3 ' - A r n ~ n o - 4 ' - u n ~ ~ l ~ - a ~ ~ ~ ~ n y ~ ~ ~ n ~ acid (II ; R = OMe) waa p r e p d exactly as described for the corresponding toluoyl derivative. The yield waa 95%. When libemted from concen-trated solutions of its salts with acids by mans of saturated sodium acetate it separates in a gelatinous state but from dilute solutions in needles. The gelatinous form passes into the crystalline on contact with the needles.It is not soluble in acids weaker than 3N in the cold but dissolves readily on warming. The most characteristic salt is the hydrochlot.de which crystallises well in wedge-shaped plates. The 8uZphude separatea in sphaero-crysbls, the nitrate in microscopic woolly needles. It diazotises and couples with alkaline Ij-naphthol with production of a bright red colour (Found Bs 20-2. Cl,H1505N& requires As 20*5y0). 3 ' - A c e t y l a m i n o - 4 ' - a n ~ ~ ~ - 4 - a m ~ ~ ~ n y ~ r ~ n ~ c acid is most con-veniently prepared by shaking the amino-acid dissolved in N-sodium hydroxide (4 mols.) with excess of acetic anhydride. The yield is quntihtive (Found AB 18-8. C,,H,,O,N& requires As, 18-4Y0). This acid is almost insoluble in boihg acetic mid but extremely soluble in cold 90% formic acid.From more dilute formic acid it separates anisotropic in sphaero-crystals. The maximum dose tolemted by mice is 1.5 mg. per g. of mowe. Attempts to prepare the propionyl derivative by the same method were fruitless. anisoy14aminuphenylursinic Ac;&.-!l!he amino-acid (5.5 g.) wag dissolved in 15 C.C. of N-sodium hydroxide (1 mol.) and 2.4 g. of pure sodium formaldehydesulphoqlate were added. The solution WBS heated for 15 minutes in boiling water cooled and poured into a large volume of spirit. A gum separated which was obtained in a solid powdery condition by grinding under absolute alcohol (yield 2-6 g.) (Found Loss at 95" 0.5; As 15.8; S 7-0. C,,H,,O,N~AsNa requires As 15.4; S 6.6%).This sodium salt so prepared has no free amino-groups. It is however unstable towa;rds N-hydrochloric acid at 50" and evolves sulphur dioxide. 3'-Car~~ama'no-4'-anisoyl-4-ami~~ny~rorphenylarsinic acid was pre-pared by adding ethyl chloroformate (1.1 c.c.) in two portions to the amino-acid (3.7 g . ) dissolved in 15 C.C. of N-sodium hydroxide. The product was acidified and the precipitated solid extracted with N-hydrochloric acid at 50" to remove diazotisable material (yield 3.7 g . ) (Found As 17.5. C1,H1,O,N&s requires As 17.1%). The acid is practically insoluble in boiling acetic acid but readily so in boiling 90% formic acid and crystallises therefrom in micro-scopic leaflets. The maximum dose tolerated by mice is 0.75 mg. per g. of mouse. A d h of S d k m F~ldehy&esulphxi$i& ort ~'-ATw~o-~' AND m C A L CONSTITUTION.PART II. 2643 The s-Carbam& of 3 f - A m i n o d ' - c m n ~ ~ ~ m ~ ~ ~ y Acid.-The amino-acid (3.7 g.) dkolved in 100 C.C. of ha& saturated sodium wetate solution with the aid of 5 C.C. of BN-Sodinm hydroxide waa shaken with several molecules excess of carbony1 chloride in toluene (35 C.C. of 1205% solution). Tbe product obtained on acidification waa extracted with warm N-hydrochloric acid and precipitated finally from dilute ammonia by acid (yield 33%). This ciwbamide is precipifa;ted in the gelatin-ous state from its salts. It is insoluble in boiling glacial acetic acid but from boiling 90% formic acid in which it is very spar-ingly' soluble it crysta.llises in microscopic needles (Found As, 3" -Nitro-4" - a n ~ l - 3 ' - u m i n o - 4 ' - a n i s o y l - 4 - a m i ~ ~ n ~ ~ r a ~ ~ ~ Acid (corresponding with IX; R = OMe).-Aminoanisoylaminophenyl-arsinic acid (7-3 g.) dissolved in 50 C.C.of 10% sodium hydroxide at - 5" was treated with 8.3 g. of nitroanisoyl chloride in 15 C.C. of toluene and stirred vigorously for hours. Toluene WE)^ removed and the aqueous solution acidified. The precipitated solid was exfracted with N-hydrochloric acid to remove amino-arsinic acid. The dried solid was extracted with ether to remove nitroa,nisic acid and reprecipitated from dilute ammonia. This complex nitro-arsinic acid is precipitated as a voluminous gelatinous solid from solutions of its salts. It is very sparingly soluble in boiling glacial acetic acid but separates well in clusters of needles.It is somewhat more readily soluble in boiling 90% formic acid (Found As 13.6. C=%O&As requires Bs 1307%). !"he maximum dose tolerated by mice is 0-075 mg. per g. of mouse. 3"- Amino - 4"- anboyl-3'- amino-4'- aniaoyl - 4 - aminupknylur&nie Acid (IX; R = OMe).-The nitro-acid (5-2 g.) was reduced in the usual way with ferrous chloride and alkali. The combined alkaline extracts of the ferric hydroxide were made neutral to Congo-paper. The precipitated solid was collected and whilst s t i l l damp made into a thin cream with water and added to 2000 C.C. of N-nitaic acid (free from nitrous acid) at 50". The solution was rapidly filfered treated with charcoal and refiltered. On addition of saturated sodium acetate solution the hino-acid waa pre-cipitated in an amorphous condition in 75% yield.This acid is soluble in hot N-hydrochloric acid and deposits an indefinitely c m e but anisotropic solid. on cooling. It is almost insoluble in boiling N-sulphuric acid the gulphate crystallising in micro-scopic needles. In N-nitric acid it is soluble on heating and the nitrate crystallism in balls of needles (Found As 14-3. C,$O,N& requires As 14.6%). The maximum dose tolerated by mice is 0.02 mg. per g. of mouse. 19-6. C~80,1N,BS re~uirea As 194%). 4T 3 " - N i t r o b e n z o y E 3 ' - ~ ~ - 4 ' - a n ~ ~ ~ a m ~ ~ ~ n y ~ r ~ n ~ acid (corresponding with IX; R = H) was prepared in the same way as the corresponding 3"-nitroanisOyl compound (above).From 7.3 g. of 3'-amino-4'-anisoylaminophenylarsinic acid and 3-nitz-o-benzoyl chloride (2 mols.) there were obtained 5-8 g. of the required acid as a gelatinous precipitate. It is sparingly soluble in boiling acetic or 90% formic acid separating from the former in short, poinM needles and from the latter in squm tablets (Found: As 14-7. C,,H,,O,N& requires As 14.6%). The maximum dose tolerated by mice is 0-1 mg. per g. of mouse. 3"-Aminobenzoyl-3'-amino-4'-an~oyl-4-clminopinylur&nie acid (IX; R = H) was prepared in the same way as the corresponding 3"-aminoanisoylarsinic acid except that 1500 C.C. of N-nitric acid at 50" were su%cient to dissolve the amino-acid. The yield was 3-1 g. from 5.0 g. of nitro-acid. This complex amino-acid separates in needles when liberated from dilute acid solutions by addition of sodium acetate (Found As 15-4.C,,HM0,N3As requires As, 15.4%). With N-hydrochloric acid it forms a very sparingly soluble hydrodoride crystallising in microscopic rods. In 2N-sulphuric acid it is readily soluble the &ph& crystallking on keeping in h e woolly needles ; and in warm N-nitric acid it dissolves and gives an indefinitely crystalline but anisotropic nitrate on cooling. The maximum dose tolerated by mice is 1.0 mg. per g. of mouse. Nitration of 3'- Nitro 4'-anisoyl-4-amimphn ylursinic A~id.-Thi~ mononitro-arsinic acid was further mononitrated as described for the previous dinitro-arsinic acids. The product consisted of a mixture of 3' 5'-dini~o-4'-anisoyl4-aminophenylarsinic acid and 3 3'-dinitro-4'-anisoyl-4-aminophenylarsinic acid in the propor-tion of 3 7.These acids could not be separ-ated by fractional c-tion from 75% acetic acid nor by fractional crystallisation of the amhonium salb (Found As 17-0. C1,H,,0~& requires As 17.0%). The composition of the mixture waa r d y deter-mined by hydrolysis as follows. Three g. of the dinitro-acids were boiled for an hour with 45 0.0. of N-sodium hydroxide and then neutralised to Congo-paper. The mixture of ether-soluble arsenic-free acids was not examined in detail but 3 5-dinitro-4-hydroxybenzoic acid was identified. The ether-insoluble acids consisted of a mixture of a 4-aminophenylarsinic and 3-nifro-4-aminophenylarsjnic acid. These could be separated h o s t quantitatively by making use of the observation that 4-amh.10-phenylarsinic acid is readily soluble in 0-5N-hydrochlo1.i~ acid, whereas the nitro-acid is not appreciably soluble.From the h a 1 aqueous mother-liquors of the hydrolysis precipitation as mag-nesium salt in ammoniacal solution gave a further crop of 4-amino AND CHEMICdL WXSTITUaON. PABT II. 2645 phenylarsinic acid. In this way 1.25 g. of 3-nitro-4-aminophenyl-aminic and 0-4 g. of 4-aminophenylaminic were obtained or 96% of the theoretical yield. The reduction of the mixed dinitro-acids by ferrous chloride w88 Hydroly& of 4-Eth~xy&mitrile.-(a) By 75% sulphurk a d . Twelve g. of the nitrile were boiled for 1 hour with 72 g. of sulphuric acid and 24 g. of water. Ether extraction of the mixture diluted with water gave 0.36 g.of a phenolic fraction partly crystalline and giving a blue d o u r with ferric chloride in alcoholic solution and a violet colour in aqueous solution. The sulphuric acid liquom were worked up as barium d f s when after removal of barium sulphate a very soluble barium salt was isolated (yield 9 g.). On c r y m t i o n from water it separated in long glistening needles of barium phenol-4-sulphonate. It waa compared with a sample prepared bythe action of sulphuric acid on phenol (Found on air-dried salt loss at 95" 9.3; on salt dried at 95" Ba 27.8. Cl,HloO~Ba,3~0 requires loss of 2&40 804%. C&EC,,O&&l3a,~~O requires Ba, 27.9%). If the original solution be treated with ether inafead of water the free phenol-4-sulphonic acid crysfallises but is very hygroscopic (compare Ahin BUU.Soc. dim. 1887 47 879). Twelve g. of 4-ethoxybenzonitde were boiled for 30 minutes with 56-5 g. of sulphuric acid and 37-5 C.C. of water. The solid obtained on pouring into water waa separated by means of sodium carbonate solution into 3-1 g. of 4-ethoxyben.z-amide md 6-1 g. of 4-ethoxybenzoic acid. The amide was com-pletely hydrolysed to the acid by boiling with 60% sulphuric acid for 2 hours. 3-Nitro-4-ethozybenzoic Acid.4-Ethoxybenzoic acid (21 g.) was heated with 210 C.C. of nitric acid (d 1-42) on the mter-bath until solution had just been effected. The product waa poured into water the solid collected dissolved for the most part in ammonia, and ezctracted with ether which removed a low-melting crysfalline solid A weighing 3 g.The ammoniacal solution on acidification gave an 80% yield of 3-nitro4eth~xybenmic a&. This acid crystaUises well from spirit in recfangular plates or rods m. p. 2o0-201" (Found C 51-0; H 4.4. C,&O,N requires C 51.2; H 4.3%). The ether-soluble material A (4.5 g. from two batches) on fractional distillation under reduced pressure gave 1.6 g. m. p. 5s-5Qo b. p. 188" (external bath temperature)/22 mm. The residue 2-65 g. on recrystadisation from alcohol gave pure 2 4 dinitrophenefole m. p. 83" in excellent yield whilst the low-llnsatisf~tory. (b) By 60% sulphur& mid melting solid on c-htion from alcohol melted at 62" and proved to be 4-nitrophenetole. 3-Nitro-4-ethoxybenzql chloride is very sparingly soluble in boiling petrol but is more soluble in dry ether from which it crystallises in slender prisms m.p. 8 1 4 2 ' (con.). It boils at 215L216" (corr.)/20 111111. 3 ' - N i t r o 4 ' - ~ b e n z o y l - 4 - a m i ~ ~ n y ~ r ~ n ~ acid (I ; R = OEt) was prepared in the same way aa the corresponding 4'-ChlOrO-derivative using method 5. The yield was 49%. It is very spar-ingly soluble in boiling acetic acid and sparingly in 90% formic acid. It crystallises in fhe needles (Found As 18.5. C15H1507N&s requires As 18.3%). 3'-Amino-4'-et~~nzoyl-4-ami~~nylarsinic Acid (11 ; R = OEt).-The mononitro-acid (8-2 g.) was reduced as described for the corresponding 4'-toluoyl compound. On neutralisation of the alkaline extracts of the ferric hydroxide to Congo-paper the acid was precipitated in an amorphous condition.When dissolved in 200 C.C. of N-hydrochloric acid at 45" and precipitated by addition of saturated sodium acetate it separated crystalline (yield 5.8 g.). The originel mother-liquors gave a further crop of 0.7 g. of acid by precipitation as the magnesium salt from hot ammoniacal solution (total yield 86%). This amino-acid crystallises in micro-scopic woolly needles (Found As 19.3. C1,H,,O5N& requires Aa 19.7%). It is solublg in warm N-hydrochloric acid but very spazingly soluble in hot 3N-acid through formation of the hydro-c7bZur& which crystabes as a sandy powder composed of small tablets. It is soluble in 2N-sulphuric acid but rapidly crystallises as the sulpmate in small pointed prisms. In 3N-nitric acid it is readily soluble and the nitrate c r y s e e s from concentrated solu-tion in fine needles.The diazotised acid couples with alkaline /3-naphthol with production of a bright red colour. Nitration of 3'-Nitro-4'-~ybenzoyl-4-amino~~nylarsinic Acid. -Thi8 acid was re-nitrated as described for the corresponding 4'-anisoyla,rsi11ic acid. A mixture of 3 3'-dinitro-4'-ethoxybenzoyl-4-Rminophenylarsinic and 3' 5'-dinitro-4'-ethoxybenzoyl-4-amino-phenylmsinic acids in the ratio 5 3 was obtained as w&s proved by hydrolysis and quantitative separation of the arsinic acids as described for the anisoyl acids (Found As 16.5. C15Hl,0JST3As requires As 16-5y0). 4 ' - C r m o r o - 3 ' - n ~ r ~ n z o y l - 4 - a m ~ ~ ~ n y ~ r $ i n ~ Acid (I ; R = Cl) .A-Chloro-3-nitrobenzoic acid was prepared in 90% yield by heating 4-chlorobenzoic acid with nitric acid (d 1-5; 4 vols.) on the wafer-bath until solution was effected; 20 g. of chlorobenzoic acid gave 23.6 g. of chloronitrobenzoic acid m. p. 180" and as by-product 0.22 g. of p-chloronihbenzene m. p. 77" by extraction of the acid in dtrrrline solution and 0-5 g. of p-chloronitrobenzene, m. p. 82" by extraction of the original aqueous mother-liquors made alkaline. Boiling nitric acid (d 1-42) has no action on 4-chloro-benzoic acid. The acid cyoride waa prepared in the wual way, b. p. (external bath temperature) 180-19Oo/ZO-22 mm. It wm introduced with diiliculty into 4-am.jnophenylar~h.ic acid by the Schotten-Baumarm method using two mo1ecu.h proportions of acid chloride.No. Solvent. Temperature. Sodium Hydroxide. Yield yo. A vaxiety of conditions to improve the yield were tried. 32 ?? 18 1 Benzene 20-25' 10% 2 None 20-50 3 Benzene "10 40 5 Toluene -5-4 7 7 40 4 Benzene 20-30 & 22 Unchanged nitrochlorobenzoic acid wits readily recovered by ether extraction and unused 4-aminophenylarsinic acid by neutralising the aqueous mother-liquors to Congo-paper and evaporating until sodium chloride began to separate. 4 ' - C h l o r o - 3 ' - n ~ t r ~ ~ ~ - 4 - a ~ i ~ ~ n y ~ r ~ ~ ~ a d is very spar-ingly soluble in boiling acetic acid more readily in boiling 90% formic acid from which it crystatllises well in needles (Found: Cl 8-6. C,H,,O,N,ClAs requires Cl 843%). 4 ' - C ~ o - 3 ' - a m ~ ~ n z o y l - 4 - a ~ ~ ~ k n y l a r s 4 n i c Acid (II ; R = C1).-The nitro-acid (6.65 g.) was reduced with ferrous chloride (7 mob.) at - 5" as described for the toluoyl compound.The combined filtrates from the ferric hydroxide extractions on being made neutral to Congo-paper deposited the amino-acid in a crystalhe condition mixed with a small quantity of amorphous impurity. The amount contained in the mother-liquors and pre-cipitable as magnesium salt was negligible. The crude acid waa heated at 80" with 800 C.G. of 3N-hydrochloric acid; the amino-acid then dissolved. On addition of saturated sodium acetate to the rapidly filtered solution the amino-acid separated in leafleta (yield 80%) (Found Cl 9.4. C,,H,0,N2ClAs requires Cl 906%). It is very sparingly soluble in 3N-hydrochloric acid at 100" but from stronger acid the kydr&ide c m h in oval leaileta.In boding 2N-sulphuric acid it is insoluble but from a much stronger acid the sulphate crystallises in square tablets. It is soluble in hot 3N-nitric acid and this solvent freed from nitrous acid would probably be preferable to hydrochloric acid for its extraction in the above preparation. The diazotised acid couples with alkaline p-naphthol with production of a bright red colour. 4'-#lmo-3 3 ' - d i n i t r ~ e n z o y l - 4 - ~ y ~ r ~ n ~ A d (Iu ; R = a).-The mononitro-acid (8 g.) dissolved in 24 C.C. of sulphuri 2648 AND XUkCH TBYPAXOCIDAL ACXFION acid waa nitrated at - 5" with 2 g. of sulphuric acid and 2 g. of nitric acid (d 142); the mixture when at room temperature was poured on to ice.The product was collected dried and digested on the boiling-water bath with 40 C.C. of glacial acetic acid for 8 hour; it then became crystalline (yield 8.5 9.) (Found (3 8.1. c13~08N3c1As requires Cl 8.0%). The acid is very v g l y soluble in boilmg glacial acetic acid somewhat more soluble in boiling 90% formic acid from which it crystdises well in needles. Hydrolysis of 4'-chloro-3 3l-dinitrobenzoyl-4-arni~~nylarsinic Acid.-Two g. of the dinifro-acid were boiled for 30 minutes with 30 C.C. of N-sodium hydroxide. When cold the solution was neutrahed to Congo-paper and the precipitated acids were collected, dried and extracted with dry ether in a Soxhlet apparatus. The ether-soluble acid weighed 1.0 g. m. p. 175". A mixture of the substance with 4-chloro-3-nitrobenzoic acid which itself melted a t 182" melted at 177".The ether-insoluble material weighed 1-1 g. and was unchanged in weight after extraction with O-FiN-hydro-chloric acid to remove any possible 4-aminophenylarsinic acid. This proved to be 3-nitro-4-aminophenylarsinic acid. The original aqueous solution on ether extraction gave 0.05 g. of crystalline acid, m. p. 172". Mixed with chloronitrobenzoic acid this melted at 157" ; with 3-nitro-4-hydroxybenzoic acid m. p. 185" however, the melting point was raised to 175". 4'-chloro-3 3'-diaminobenzoyl-4-amino;plZenylarsinic Acid (IV ; R = Cl).-The dinitro-acid (6.5 g.) was reduced exactly as described for the corresponding dinitrotoluoylarsinic acid. The combined alkaline extracts of the ferric hydroxide when neutralised to Congo-paper gave an amorphous brown precipitate but on keeping a few hours the diamino-acid (3.0 g.) crystallised.The filtrate was made alkaline with ammonia and heated with magnesium chloride; a magnesium salt then separated which gave an additional 1-25 g. of acid. The combined crude acids were dissolved in 40 C.C. of N-hydrochloric acid with addition of 1 C.C. of concentrated acid and precipitated by addition of saturated sodium acetate until neutrality to Congo-paper was reached (yield 76%). The acid 80 prepared crystallises in rosettes of pointed plates (Found Cl 9.2. Cl,H,,0,N3ClAs requires C1 9.2 %). Preparation and Nitration of 4-Ethylcarbonatobenzoic Acid.-4-Hydroxybenzoic acid (34.5 g.) dissolved in 500 C.C.of N-sodium hydroxide (2 mols.) was shaken with 26.0 C.C. of ethyl chloroformate (1-1 mols.) added in four portions. Ether extraction removed a viscous pleasant-smelling oil (2.6 g . ) presumably (compare analogous case of salicylic acid D.R.-P. 117,267). Et0,C-O*C,H4=C0,C0,Et Th AND CHEMICAL CONSTITUTION. PABT II. 2649 aqueous SoIution on acidification gave 49.3 g. of 4ethyih*-benxoic acid (94% yield). This acid is best r e a p h b d * from 85 volumet3 of boiling water and separatea in long glistening needles, m. p. 154-156". It is very soluble in the usual organic solvents (Found C 57-0; H 4.9. CJIl0O5 requires C 5'7-1 ; H 4.8%). Ndrution. The acid (48.5 g.) was dissolved in 182 C.C. of fuming nitric acid (nitric acid d1-42 has no action) and hated for 10 minutes on the boiling-water bath.The product wm poured info water the solid collected dissolved in sodium bicarbonate solution, and extracted with ether which removed '7.5 g. of non-acidic d i d , A. The aqueous solution on acidification gave 44-6 g. '('76% yield) of 3 - n i t r o + t - e t h y h ~ n & d. This is best r-from 15 volumes of boiling benzene and separates in clear kr*-shaped plates m. p. 168-169" (Found C 47.3; H 3.7. C,&&+O,N requires C 47.1; H 3.6%). The non-acidic solid A was distilled in a vacuum when 6-35 g. passed over at 18 111111. and an external bath temperature of 192". This fraction on crystallisation from alcohol separated r d y in fine needlea m. p. 67-68O (corr.) in agreement with the propertie8 described by Ransom (Ber.1898 31 1064) for ethyl 4-nitrophenyl-cabonate. On hydrolysis with alcoholic soda it yielded 4-nitro-phenol. The non-volatile residue 1.6 g. was highly coloured and did not crystallise on keeping but was proved to be ethyl 2 4dinitro-phnykizrhnate by hydrolysis with alcoholic sodium hydroxide, which gave 2 4-dinitrophenol. 3 ' - N i t r o - 4 ' - ~ h y l c a r b o n a t o b e n z o y l - ~ - u ~ i ~ ~ n y l a r Acid (V). -The above-described 3-nitro-4-ethylcarbonatobenzoic acid (25.5 g.) wae shaken with 21.0 g. of phosphorus pentachloride until reaction took place. The phosphorus oxychloride was removed under reduced pressure by gentle warming. The residual syrupy nitro-e t h y h r W e n q 1 chloride which crystallisea readily in a freezing mixture and remains solid above room temperature after being evaporated to dryness once or twice with dry ether was added in three portions with vigorous shaking to 15.5 g.of sodium 4amho-phenylarsinate pentahydrate dissolved in 300 C.C. of half-saturated sodium acetate solution. The solid (34 g.) precipitated by making the solution definitely acid to Congo-paper waa divided by ether exfntction into 15.6 g. of unchanged nitro-acid and 18-1 g. (80% yield) of the required arsinic acid. !l!his acid is' soluble in boiling acetic wid more readily soluble h boiling 90% formic acid and crysttallises in needles (Found As 16.5. C,,H,,O~& requires As 1605%). 3 ' - N i t r o - 4 ' - A y d ~ ~ ~ n ~ ~ - 4 - u ~ i ~ k n y ~ r ~ n ~ Acid (VI) .-The crude acid (18-1 g.) as obtained above wa8 dissolved in 160 C.C.of 4 T 2650 T B Y P A X ~ A L ACTION AND CHEMICAL CONSTITUTION. PART n. N-sodium hydroxide and the solution just brought to its boiling point. Addition of acid to the cold dution precipitated a t an intermediate stage the s o d k n salt of the nitrohydroxy-acid in r e c w m plates but finally (acid to Congo-paper) the free acid a8 a primrose-yellow solid (yield 97%). This acid is sparingly soluble in boiling acetic acid crystaWg therefrom in minute clusters of needles but readily soluble in boiling 90% formic mid, from which it crysfallises in long silky needles (Found AS 19.4. Cl&Il,O,N& requires AS 1906%). Reduction of 3'-Nitro-4'-hydroxyben~l-4-ami~wph.t?ny~rsinic Acid. -(a) By hypomdphite.The nitro-acid (3.8 g.) in 25 C.C. of N-sodium hydroxide (2.5 mols.) waa treated with 6 g. of sodium h p u l p h i t e added in portions. The end-point waa determined by removing samplea and observing the absence of yellow colour on adding alkali. The separated solid consisting mainly of the odium salt of the required amino-acid was collected and dissolved in 240 C.C. of N-hydrochloric acid at 50". The iiltrate made neutral to Congo-paper by addition of saturated sodium acetate solution, deposited the pare amino-acid (2.1 g.) in small needles. This reagent used as frequently de-scribed in this paper reduced the nitro-acid very smoothly. The amino-acid was obtained quite pure in 936% yield by neutralisation of the al.kaJine filtrate from the ferric hydroxide.3'-Amino-4'-hydro~benzoyE-4-ami~wp~ny~~8~n~ acid iEi readily soluble in warm N-hydrochloric nitric or sulphuric acid the salfs of the former two crystallising in needles that of the latter being amorphous and gela,tinous. The acid diazotises with production of a pale yellow colour and even from very dilute solutions the diazo-oz&de separates in pale yellow needles. This couples with algaline p-naphthol with a cherry-red colour. The acid is soluble in satur-ated sodium hydrogen carbonate solution but the sodium salt ~oon separates in micro-crystals (Found As 21.5. C13H130,N&s requires As 2103%). 3'- Amino-4'- hydro~benzoyl-4-aminoarsenobenxene (VIII) .-Amino-hydroxybenzoylaminophenylars~c acid (3-3 g.) was suspended in 16-5 C.C. of hypophosphorous acid (d 1.137) with addition of 33 c.c of 50% acetic acid and a crystal of potassium iodide.The mixture was stirred at 50-55" for 1 hour; the whole of the originally crystalline suspenrjion had then become orange-yellow and amor-phous. The product wa8 centrifuged off washed several times with air-free water and treated with sodium hydrogen carbonate solution until permanently alkaline. The liberated base was re-centrifuged well washed and dried in a vacuum (yield 2-7 g.). This arseno-derivative is soluble instantly in sodium hydroxide, (b) By ferrowr chloride THE ACTION OF " I T S ACID UPON A'MTT)lW ETC. 2651 but not in sodium W ~ O M ~ . It is insoluble in hydrochloric acid of any strength but vanishes almost instanfly on addition of nitrite with intermediate development of a deep colour.It then couples with rtlkrrline @-naphthol (Found As 24.7 25-1. 3 ' - A c e t y l a m 6 7 a o - 4 ' ~ ~ A - . . n n ; ~ ~ ~ ~ ~ ~ n ~ Acd.-The aminohydroxy-acid (3-5 g.) waa dissolved in 40 C.C. of N-sodium hydroxide or sodium carbonate and 5 C.C. of acetic rmhydride were added in 1 C.C. portions with vigorous shaking. The solution waa acidified to Congo-paper and the precipitated solid well washed with water and dried (yield 3.8 g.). From weakly slkAline solution this acid is precipitated in fine needlea which are almost insoluble in boiling glacial acetic acid but extremely soluble in cold 90% formic acid. From more dilute formic acid it crysfallises well in fine, soft needles (Found As 16.9. C,,HI,O,N& requires As 17.2%). The maximum dose tolerated by mice is 1-75 mg. per g. of mouse. 3 ' - A ~ y l a m i n o - 4 ' - h ~ d r ~ ~ n ~ l - 4 a m 6 ~ ~ e n y k r r s ; n Acid.-When a solution of 1 g. of the preceding acid in 10 C.C. of N-sodium hydroxide (4 mols.) was kept for 20 hours at room temperature, then diluted and acidSed with 3N-hydrmhloric acid the N-dyZ acirl was obtained as a microc-e ahtropic powder (Found : As 192. CI5H,,O6N& requires As 19.0%). This acid is almost insoluble in boiling glacial acetic acid but very readily soluble in ~ a r m 9 0 ~ ~ formic acid from which it sepasates in microscopic needles. 3 - A ~ y ~ m i n o - 4 - a n ~ ~ ~ ~ n ~ Aeid.-3-Amino-4hydroxy-amink acid (2.33 g.; 0.01 mol.) wm dissolved in 25 C.C. of water with the aid of 3.5 g. (4 mols.) of sodium bicarbonate and a,cetylatd by addition in 1 C.C. portions of 5 cx. of acetic anhydride. When the solution was acidified with strong hydrochloric acid the &c acid separated in needles (yield 2-5 g.). It is soluble in boiling glacial acetic acid and readily soluble in cold 90% formic acid (Found As 23.7. C,~,,O,NAEI requires As 2306%). CZ6%04N4& r e q h 24.8%). THE NATIONAL INS- FOR MEDICAZ RESEABCH, 33.Am?SmA.D N.W.3. [Received Augwt 5th 1926.