E.P. OLIVETO, C. GEROLD, Ie;l.
The llp-hydroxyl group, lorig considered to be unreactive toward the usual esterifying agents,2 was found by us to be acetylated in good yield under verv mild conditions, using acetic anhydride and an acid catalyst.g>4This discovery led to the relatively simple synthesis of Compound F 11,21-diacetate and 11-acetate, but further hydrolysis to Compound F was not achieved. .Attention was next turned to the corresponding 1lp-formates, which were found to be not only hydrolysable to the parent 1lp-hydroxy compound, but also to survive the reactions necessary to elaborate the dihydroxyacetone side-chain and the a$-unsaturated ketone in the Our preliminary studies were made on 11/3,17adihvdrosypregnane-3,20-dione(I). While formic acid alone produced no significant change, the addition of an acid catalvst (either p-toluenesulfonic acid or perchloric acid) gave a new compound, lacking in free hydroxyl groups, and therefore as(1) A preliminary account of this work has appeared in A r c h . Bioc h r m . Biophys., 49, 244 (1954). F. Gallagher, J . Bioi. Chem., 162, 539 (1946); L. F. Fieser (2) 'I'. and M. Fieser, "Natural Products Related t o Phenanthrene," Reinhold Publ. Corp., S e w York, N. Y . , 1919, pp. 408, 657. However, Dr. C. W. Shoppee has pointed o u t t o us in a private communication t h a t ?d.Steiger a n d T. Reichstein [ H e l u . Chim. A c t a , 20, 817 (1937)l converted 3P,lli3-dihydroxyandrostan-17-one t o its diacetate b y warming with acetic anhydride a n d pyridine. See also A. Crawshaw, H . B. Henbest and E. R . H. Jones, J . Chem. Soc., 731 (1954),and A. K e m p , A . Kappas, I. Salamon, F. Herling and T. F. Gallagher, J . Bioi. Chem., 210, 123 (1934). (3) ( a ) E. P. Oliveto, C . Gerold a n d E. B. Hershberg, A r c h . Biochem. Biophys.. 43, 234 (1953); (b) E. P. Oliveto, C. Gerold, L. Weber, H. E. Jorgensen, R . Rausser a n d E. B. Hershherg, T H I 3
75, 5480 (1953). (1)Soon after our original communication, A . Crawshaw, H. B. Henbest and E. R. H. Jones [ref. 21 described t h e acetylation of l l p hydroxy steroids with acetyl chloride-dimethylaniline i n chloroform. ( 3 ) At about t h e same time, A . Lardon a n d T. Reichstein [Heiv. Chim. A c t a , 37, 443 (195411 prepared an Ilp-formate with an acetic anhydride-formic acid mixture, a n d found t h a t i t could n o t only be saponified, b u t also was stable during t h e sequence of reactions elaborating t h e side-chain from an etio acid via t h e diazo ketone synthesis.
JOURKAL,
sumed to be 1lp,l7a-dihydroxypregnane-3,20-dione diformate (11). This compound was stable to attempted acid hydrolysis, but easily reverted to I in good yield on treatment with sodium hydroxide or potassium carbonate a t room temperature overnight. Treatment of 3 a , l lp, 17a-trihydroxypregnan-20one (111) with formic acid and an acid catalyst overnight gave the triformate IV in 43.5y0 yield. I n the absence of the acid catalyst, the 3-monoformate V was obtained in 72y0 yield. Complete hydrolysis of IV to the starting triol I11 was accomplished with aqueous sodium hydroxide overnight a t 30'. Reaction of IV with ethanol and p-toluenesulfonic acid gave the 11,17-diformate VI, which was oxidized easily with N-bromoacetamide in high yield to 11. Partial hydrolysis of IV to 3a,ll/3,17atrihydroxypregnan-20-one 11-formate (VII) was accomplished by either refluxing aqueous sodium bicarbonate, sodium methylate in tetrahydrofuranmethanol or Amberlite IRA-400 resin in methanol. The product VII, invariably a resin, was next brominated a t C-21 with bromine in chloroform solution, and the bromine replaced by acetate by means of potassium acetate in refluxing acetone to give the 11-formate 21-acetate VI11 in about 5OyOyield. Oxidation of VIII to 11/3,17a,21-trihydroxypregnane-3,20-dione 11-formate 21-acetate (IX) was accomplished in good yield with N-bromoacetamide in t-butyl alcohol-methylene chloride. The over-all yield of IX from IV was considerably improved if no purification of intermediates was attempted. Thus from 5 g. of IV there was obtained 3.37 g. of crude IX (ca. 67y0)if no attempt was made to crystallize the intermediates V -+ VIII. Hydrolysis of IX with sodium hydroxide in aqueous methanol a t room temperature overnight, followed by acetylation with acetic anhydride and pyridine gave 11/3,17a,2l-trihydroxypregnane-3,20-dione 21-acetate (X). This had been converted
17ff-HYDROXYCORTICOSTERONE21-ACETALS
3565
Vb 11P-FORMATES
CHzOAc
I
C=O
CHzOAc
I
c=o
CHzOH
C-=O
c=o
I
c=o OHCO\,,,\
CH~OAC
,!y;OH
-HBr
H+ -
to 17~-hydroxycorticosterone 21-acetate previously.6 Bromination of IX a t C-4, followed by dehydrobromination via the semicarbazone, proceeded in the normal fashion t o give 1 lp,17a,21-trihydroxvA4-pregnene-3,20-dione11-formate 21 -acetate (XII, Compound F 11-formate 21-acetate'). Acid hvdrolysis gave Compound F 11-formate (XIII), while complete hydrolysis with sodium hydroxide, followed by acetylation gave Compound F 21acetate (XIV). The biological properties of XI1 and XI11 will be reported elsewhere. 16) N I. Wendler R P Graber K k Juries and hl 1 ishler. 1 l u X~V A I 14, 3b30 ( l Y 5 2 )
HI'^
__
~
. J
Experimental' 11~,17a-Dihydroxypregnane-3,20-dione 11,17-Diforrnate (II).-A solution of 1.00 g. of 1Ip,l7~-dihvdroxypregnanc3,20-dione (I) and 0.10 g. of p-toluenesulfonic dcid monohydrate in 10 ml. of formic acid (98-100'( ) was allowed t o stand overnight a t 2 5 ' . then poured into water to precipitate 0.99 g. of crude IT, m.p. 2O3-21O0. One crystallization from acetone-hexane gave 0.80 g., m.p. 240-245". The analytical sample, crystallized tr.r ice more, melted a i 25&264", [a]D 4-47.4' (dioxane). The infrared spectrum showed no hydroxyl or conjugated carbonyl bands. Anal. Calcd. for C Z ~ H ~ ZC,O65.29; ~: H, 7.97. Found: C, 68.47; H, 8.17. (7) All m.p.'s are corrected. All rotdtiuus were takeu in chloroform (except where noted) in a 1-dm. tube at 25' and at a concentration of about 1 Vo. Analyses and optical data were obtained by the Micro analytical aud Physical Chemistry Utpartmcnts of t h r a c laboratories
3566
E. P. OLIVETO,C.GEROLD, R. RAUSSER AND E. B. HERSHBERG
Vol. 77
When the reaction was run in the absence of the p-toluwashed with water, dried over sodium sulfate and evapoenesulfonic acid, only the starting material I was obtained. rated to dryness. The resin thus obtained was chromatoHydrolysis Studies on 11. (a).-A suspension of 1.00 g. of graphed on Florisil. The fraction eluted with methylene I1 in 80 ml. of C.P. methanol was treated with l 0 . 0 m l . of 1 chloride m-as taken up in ether, treated with Darco, filtered, 3 aqueous sodium hydroxide and stirred for 19 hours a t 23". evaporated to dryness and thoroughly dried under vacuum. After neutralization with 1 ml. of acetic acid, the mixture This material, still a resin, had [CY]D f38.0" and was prewas concentrated under reduced pressure until crystalliza- sumed to he 3 a , l lp,l7n-trihydroxypregnan-20-one 1l-fortion was substantially complete, then 200 ml. of water was mate ( \ - I I ) . added and the precipitate was collected by filtration to give A n d . Calctl. for C22H3106:C, 69.81; H , 9.05. Fountl: 0.82 g., m.p. 200-210". One crystallization from acetoneC, 69.69: H , R.::7. hexane gave 0.64 g . , m.p. 214-322'. A inisture m.p. and This experiment ~ r a srepeated several times and the final the infrared spectrum indicated the product t o be identical product mas nearly always a resin which failed t o crystalwith I . (b).-A mixture of 1.00 g. of 11, 30 ml. of C.P. methanol, lize on standing, on treatment with solvents or after chromatograph>-. However, one run did yield some crystalline 1.00 g. of potassium carbonate and 6 ml. of water \vas stirred material oii standing with ether: 0.25 g. with m.p. 160for 20 hours at 25". After neutralization with 1.5 ml. of 166' was obtained from 1.0 g. of IV. The infrared specacetic acid, the solution was concentrated under reduced pressure. % \ h'en crystallization began, the suspension was trum of this material showed a stronger hydroxyl absorption and a weaker formate bond than that of I T : however, diluted with water and 0.82 g. of solid, m.p. 190-196", was satisfactory carbon and hydrogen analyses could not he obcollected by filtration. Recrystallization from acetonetained. The infrared spectra of the resins obtained from hexane yielded 0.55 g. of I, m.p. 211-214', as shown by this and the following hydrolyses matched each other exmixture m.p. and infrared spectrum. actly,and were very similar t o that of the crystalline material ; 301,110,17~~-Trihydroxypregnan-20-one Triformate (IV).A solution of 10.0 g. of 3a,llp,l7a-trihydroxypregnan-20- no important discrepancies existed. (c).-A solution of 0.5 g. of 11' in 25 ml. of tetrahydroEuran one (111) in 100 ml. of 98-100Yc formic acid was treated with and 2.5 ml. of C.P. methanol was treated 172-itha solution 4 ml. of 70-727, perchloric acid (or 1.0 g. of p-toluenesulfonic acid), then allowed t o stand a t 5" for 48 hours. The of 0.7.5 g. of sodium methylate in 10 ml. of C.P. methanol mixture was then poured into water and the precipitate under an atmosphere of argon. After 5 minutes 0.72 ml. One crys- of water was added; 3 minutes later the excess alkali was collected by filtration: 11.26 g., m . p . 98-140'. neutralized by the addition of acetic acid. Concentratioti tallization from methanol gave 5.38 g. of IV,m.p. 167-172'. under reduced pressure and addition of water gave 0.37 g. The analytical sample, crystallized once more from methanol, melted at 174-177', had [CY]D+57.0°, and showed no of a gummy solid. Crystallization from ether-hexane gave 0.14 g., m.p. 158-162". Its infrared spectrum matched free hydroxyl band in its infrared spectrum. exactly that of the crystals obtained in (b), but again no Anal. Calcd. for CZ4HS407: C, 66.34; H,7.89. Fouutl: satisfactory analytical results could he obtained. AtC, 66.60; H, 8.06. tempts t o repeat the reaction invariably gave resins. Sev3a,llp, 17a-Trihydroxypregnan-20-one 3-Monoformate eral variations were tried without success, among them the (V).-A solution of 1.0 g. of I11 in 10 ml. of 98-10OCbL for- use of methanol alone as the solvent, alloning the reaction mic acid was allowed t o stand overnight a t 2.j0, then poured to take place at ice-bath temperature, and using only into water and the precipitate collected by filtration. One slightly more than the theoretical amount of sodium methylcrystallization from acetone-hexane gave 0.80 g. of \-, ate. m . p . 184-188". The analytical sample, crystallized twice (d).-A solution of 1.0 g. of IV in 50 ml. of C.P. inethmore, melted a t 1S7-190" and had ID +43.9'. aiio1 was stirred for 2 hours a t 25' with 4.0 g. of Amberlite Anal. Calcd. for C2tH310j:C, 69.81; H,9.05. Found: IRA-400 resin (basic form). The resin was removed by C, 69.91; H, 8.72. filtration and the filtrate evaporated t o dryness, leaving 0.88 g. of a resin. Its infrared spectrum was identical with The triformate I\. could be obtained from V by treatment tliose of the resins obtained in ( b ) and ( c ) . with formic acid and an acid catalyst. Complete Hydrolysis. (a).-A mixture of 0.50 6 . of I\., Hydrolysis Studies on IV. Partial Hydrolysis. (a).--A mixture of 5.0 g. of IV and 0.5 g. of p-toluenesulfonic :ici,l 40 nil. of C.P. inethanol and 3 ml. of 1 r\i aqueous sodium liydroside \va\ stirred o:~ernight a t 30". After neutralizamonohydrate in 100 nil. of absolute ethanol was stirred a t tion ivitli acetic acid, the solution was concentrated under 37-38" for 2 hours, then stored a t 5' for 16 hours, m t e r reduced pressure, 200 nil. of water was added, and 0.21 g. added and the clear solution extracted with methyletie of solid rcinoved hy filtration, m.p. 170-195'. Recryschloride. The organic extracts were washed with dil. 50tallizatioii from acetoile-hexane yielded 0.11 g., m . p . 205dium bicarbonate solution and water, dried over sodium 2 0 ! ) O . Its infrared 5pectrum matched t h a t of 111. sulfate and evaporated to dryiiess. The resinous resitiui., ~ C U1 , lfi,17a,21-Tetrahydroxypregnan-20-0ne11-Formate 4.6 g., crystallized easily upon trituration with ether; 21-Acetate (VI11 1 . ----Theresinous product VI1 from the m .p. 168-182". Rccry~tallization from acetone-hexane liydrolysis of 1 . ( I g. of 11- as described in ( a ) was takeu ul) gave 2.31 g. of ?m,l lB,1~a-trihydroxypregiian-2O-o11~ 1 1 .17-diformate ( V I ) , m . u . 187-190", and a second crop o f in 20 ml. of C.P. chloroform and brominated by the dropwise addition of :I solution of 0.41 g. of bromine in 10 ml. 0.74 g., m.p. 17.i-lX03. Two additional crystallizalio!ii of the major portion raised the m . p . to 194-1Scio, [ C ~ ~ I I of C.P. cliloroforin. Two grams of potassium acetate was added, the iol\-ent removed under reduced pressure, the +47.5O. residue treated iT-ith .-,(I nil. of acetone and 2 g. of potassium Anal. Calcd. for C2>1-1>,Oti: C , fi7.9.7; €1,8.43. 1:outid: acetate, a i i d t h e resulting suspension refluxed for 3 hours C, 67.78; H, 8.59. Ivith stirring. The acetone was removed under reduced This compound ( V I ) tended t o solvate a t times, but tliv pressurc. i n t e r Lidded and the product collected by filtrasolvate could be broken by heating a t 90" under reduced tion: 1 . I 4 g . , i i 1 . p . 105-140" dec. .4r1 ether sludge g a w pressure. The structure was established by oxidation with 0..57 g.. i i i . l ' . 12t.i~-lBO"dec. Recrystallization from aque9-bromoacetamide to yield 11. A solution of 0..i g . of \.I ous methanol ga\-e 0.40 g. of VIII, m.p. 160-16So. Thc in 5 ml. of methylene chloride and 3 ml. of t-butyl alcohol aiialytical sample was crystallized twice more and had m.p. \vas chilled in an ice-bath, treated with 0 . 5 g. of S-broiiio18f5-190", [ a ] D+86.0". acetamide and allowed t o stand 5 hours in the dark a t 5". A n t i / . Calcd. for Cn4HseOi: C, 66.0:3; €1, 8.31. Found: The solution then was washed with dilute sodium sulfite solution, the washes extracted twice with methylene chlo- C,66.27; H, 8.35. 110,17a,2l-Trihydroxypregnane-3,20-dione 11-Forride, the combined extracts washed twice with water, dried solution of 1.43 g. of VI11 in over sodium sulfate and evaporated to yield 0.5 g. of crys- mate 21-Acetate (IX).-A 15 nil. of methylene chloride and 15 ml. of t-butyl alcohol tals, m.p. 260-265". Recrystallization from acetone hexane gave 0.43 g.. ii1.p. 263-26G0, which was show11 t u t)r \vas chilled in an ice-bath and 1.34 g. uf N-bromoacetamide ~ i l d r d . . I t t r y 4 hours in the dark a t .-la, the solution wa5 identical with I1 b>fmixture n1.p. and infrared spectruii. M urkecl u p iri thr maliner described 'ibuve. 'The reiitlue (b).-.4 mixture of 5.13g. of I Y , 200 nil. of C.P. niethdiiul, ' U I I treatment w i t h rthcr: 1.4:3 g . , I I I . ~ . 3.9U g. of sodiuin bicArboivate arid 40 1111. uf wafer wa.; rc ~llizxtiuufroin xretoiit~ licyaiic> ga fliisecl fur 2.5 hours. After iieutralizatioii with acetic acitl, aii:dytic,Lil 5aiiipIv, c tlie wlution \vas tliluted w i t h water ancl est ructed 5ewr:il 2L1.-I0, ! 1 1 1 l l 1 9 ! l . ~ I ' ' . tiriirz j v i t l i iiirtlivlriie cliloi-iclr. 'rile coiiil)iiirtl exlr;ict s weiw
July 5 , 1955
6-O-P-h!fALTOSYL-a-D-GLUCOPYRANOSE HENDECAACETATE
Anal. Calcd. for C24H3407:C, 66.34; H , 7.89. Found: C , 66.00; H , 8.08. I t was found that purification of intermediates was not necessary in the preparation of I X from I V . Thus starting with 5.0 g. of I V and carrying the material through the hydrolysis, bromination, acetoxylation and oxidation steps without isolation, there was obtained 3.37 g. of I X , m.p. 194-200°. One crystallization from acetonehexane yielded 2.58 g., m.p. 209-212". 11P,17c~,2 I-Trihydrorypregnane-3 ,20-dione 2 I-Acetate (X).-A solution of 0.50 g. of I X in 20 ml. of C.P. methanol was combined with 6 ml. of water and 11.5 ml. of C.P. methanol containing 0.115 g. of sodium hydroxide, all under an argon atmosphere. The solution was allowed to stand for 18 hours a t 27", the excess alkali was neutralized with acetic acid, water added and the methanol removed under reduced pressure. Filtration yielded 0.21 g., m.p. 143-170"; this was acetylated with acetic anhydride and pyridine, and ChroDmatographed on Florisil to give 30 mg. of X , m.p. 189-195 . Its infrared spectrum was identical with that of an authentic sample.8 4-Bromo-1 1p,17af2 1 -trihydroxypregnanel ,ZO-dione 1 1 Formate 21-Acetate (XI).-A solution of 3.0 g. of I X in 30 ml. of methylene chloride and 30 ml. of t-butyl alcohol a t 25" was brominated by the dropwise addition of 1.14 g. of bromine in 25 ml. of methylene chloride. The bromine color discharged after standing 4 hours. The methylene chloride was removed under reduced pressure and the remaining solution poured into water. Filtration of the crystalline precipitate gave 3.76 g., m.p. 163-169° dec. Recrystallization from aqueous acetone gave 1.75 g. of XI, m.p. 185-187' dec., [(Y]D f86.3" (acetone). Anal. Calcd. for C24H330,Br: Br, 15.57. Found: Br, 15.37. Debromination of the mother liquor gave 1.07 g. of X , m.p. 213-221". 1 lp,1701,2I-Trihydroxy-A4-pregnene-3,ZO-dione1 I-Formate 21-Acetate (XII).-A suspension of 2.63 g. of X I in 53 ml. of t-butyl alcohol and 40 ml. of C.P. chloroform, stirred in an argon atmosphere, was treated with 0.81 g. of semicarbazide and stirring was continued for 2 hours. The reaction was concentrated under reduced pressure to ca.
-
( 8 ) Kindly supplied by Merck and C o . , Inc.
[CONTRIBUTION FROM
THE
3567
half-volume, water added and the distillation continued t o the precipitation of solid. Filtration yielded 2.65 g., which was taken up in 25 ml. of acetic acid, combined with a solution of 1.22 g. of 94% pyruvic acid and 1.05 g. of sodium acetate in 7.5 ml. of water, and refluxed for 5 minutes. After the addition of 100 ml. of hot water, the mixture was chilled to precipitate 1.91 g. of X I I , m.p. 190-200". The analytical sample, crystallized several times from acetone-hexane, melted a t 199-201", [a]D +176.1", emax 15,100 a t 239 mp (95% EtOH). A T L U ~Calcd. . for CnrHa~O7: C, 66.65; H , 7.46. Found: C, 66.46; H , 7.22. 1 14,170r,Z1 -Trihydroxy4 I-pregnene-3,ZO-dione 1 1-Formate (XIII).-A solution of 0.50 g. of XI1 in 5 ml. of C.P. chloroform and 17.5 ml. of C.P. methanol was chilled and combined with 1.O ml. of concentrated hydrochloric acid in 1.8 ml. of water. The reaction was allowed to stand for 48 hours a t 25', diluted with water, and extracted with methylene chloride. The organic extracts were washed with dilute sodium bicarbonate solution and water, dried and evaporated. The resinous residue, 0.48 g., was taken up in benzene and chromatographed on Florisil. The fraction (0.17 8.) eluted with 1% methanol in methylene chloride crystallized on trituration with ether, m.p. 168-178'. Recrvstallization from acetone-hexane vielded 0.08 8 . of X I I ~ ,m.p. 184-187', [ ~ I D+162.7", ,,e, 15,700 at-238 mp (95% EtOH). Anal. Calcd. for CzzHaoOs: C, 67.67; H , 7.74. Found: C, 67.63; H , 7.92. l7a-Hydroxycorticosterone 21-Acetate (XIV).-Under an argon atmosphere, a mixture of 0.50 g. of XI1 in 30 ml. of C.P. methanol and 0.116 g. of sodium hydroxide in 6 ml. of water was allowed t o react 20 hours a t 25". The excess alkali was neutralized with acetic acid, water was added and the methanol removed under reduced pressure. Only a small amount of solids separated, so the aqueous residue was extracted several times with methylene chloride, the extracts combined, dried and evaporated t o give 0.37 g. This was acetylated with acetic anhydride and pyridine, thencrystallized from acetone togiveO.ll g., m.p. 215-218'. Its infrared spectrum was identical with that of an authentic sample of XIV. BLOOMFIELD, K. J .
DEPARTMENT O F CHEMISTRY O F THEOHIO STATE UNIVERSITY]
6-O-p-Maltosyl-~-D-glucopyranose Hendecaacetate BY A. THOMPSON' AND M. L. WOLFROM RECEIVED JANUARY 13, 1955 The reaction mixture2 obtained by treating 6-O-B-maltosyl-~-~-glucopyranose hendecaacetate successively with titanium tetrachloride, mercuric acetate and acetic anhydride-pyridine, has been investigated chromatographically and found to contain 6-0-~-ma~tosy~-cu-D-g~ucopyranose hendecaacetate, P-maltose octaacetate and 8-D-glucopyranose pentaacetate, in addition to the starting material. A proof of the identity of 6-O-P-maltosyl-cu-~-glucopyranose hendecaacetate is presented.
-
The trisaccharides of D-glucose containing mixed (1 4 ) - a - and ~ (1 -P 6)-01-Dlinkages are of interest in studies on the structure of amylopectin. Of the three possible trisaccharides, containing these linkages, that should be preformed in amylopectin, panose3 (4-0-a-isomaltopyranosyl-~-glucose) has been isolated from a r n y l ~ p e c t i n . ~An impure preparation of a second member of this group, 6-0-01maltosyl-~-glucose hendecaacetate,has been reported by Asp and Lindberg2 as resulting from the action (1) Corn Industries Research Foundation Associate, (2) L.Asp and B. Lindberg, Acta Chem. Scand., 6, 665 (1951). (3) S. C. Pan, A. A. Andreasen and P. Kolachov, Science, lt2, 115 (1950); S. C. Pan, L. W. Nicholson and P. Kolachov, THIS JOURNAL, 73, 2547 (1951). (4) M. L. Wolfrom, A . Thompson and T. T Galkowski, ibid , IS, -1093 (1951). ( 5 ) A Thompsou and A I . I , . Wolfrum, i b i , l , 7 5 , JS4Y ( 1 9 5 1 ) .
of titanium tetrachloride on 6-O-P-maltosyl-P-~glucopyranose hendecaacetatea with subsequent treatment with mercuric acetate and reacetylation with acetic anhydride and pyridine. Such a transformation of a (1 6)-P-~-glucosidicto the (1 + 6)O~-D form has been established by Lindberg' for the conversion of &gentiobiose octaacetate to 8-isomaltose octaacetate and has been verified in this Laboratory. In the hope that we could obtain 6-0-a-maltosylP-D-glucopyranose hendecaacetate in a state of purity, we have made a chromatographic study of the reaction mixture obtained by Asp and Lindberg, a t the stage in which all products appear as (6) S H Nichols, J r , a' 1, Evans and H D hlcDowell, zbtd , 62,
-
1754 (1940) (71 B Lindberg, A ( l a
f h d m
. S r u i i d , 3 , 1355 ( I 9 4 Y l
