May, 1963
THESYNTHESIS AXD PHARMACOLOGICAL ACTIONOF TREMETOYE
ness and dilute hydrochloric acid added. The crude solid product was collected by filtration, washed with water and recrystallized from methanol-chloroform (charcoal) giving white crystals (2.0 g.), m.p. 248-250’. Further recrystallization from 507, ethanol gave the compound, n1.p 255’; Xmar 229 m p ( e 23,500); 236 nip ( e 23,000); 257 nip ( e 5,400): 263 mp ( e 4,600); 318 mp ( e 7,200). Anal. Calcd for CeHjCl,x30,S?: C1, 22 16; S j 13.13. Found: C1, 21.96; ZJ, 12.99. Catalytic reduction of the compound in alkaline solution under conditions similar to those employed for the reduction of %amino5-chloro-m-benzenedisulfonaniide J ielded 2-amino-ni-benzenedisulfonamide as shown by comparison with the sample previously obtained. 4,5-Dichloro-6-nitro-m-benzenedisulfonamide (X).-4-Amino5,6-dichloro-m-benzenedisulfonamide ( 1.O g.) n ns dissolved in sulfuric acid ( 5 nil.) and the solution added to a niiyture of 207, fuming sulfuric acid (20 ml ) and 3 0 5 hydrogen peroxide solution (10 ml.), cooled in an ice-hath. After 2 hr. a t ice-bath temperature and 4 hr. a t room temperature, the reaction mixture was diluted Kith water (200 ml.) and then allowed to stand overnight. The crude product was collected bv filtration, washed with water and air dried; yield, 0 5? g. ni p 237-2q2’. Recr~stallization was effected from dilnte ethanol yielding the pure product, m.p. 24s-2-19’; X, 214 r n M ( E 41,400);plateau 2S6-200 nip ( e 1,000); 295 mp ( E 2,200): X,,,a, 6 50 p (s). A n a l . Calcd. for CBH,ClZS30&:Cl,20.25; K,12 00. Found: C1, 20.56; S, 11.DG.
315
4,5-Dichloro-6-methoxy.m-benzenedisulfonamide (=).-A mixture of 4,5-dichloro-6-nitro-m-benzenedisulfonamide (1.0 g.), sodium methoxide (2.0 g.) and methanol (20 ml.) was refluxed for 16 hr. The cooled reaction mixture was diluted with water, acidified with dilute hydrochloric acid, and the crude product collected by filtration, washed with water and air dried. Recrystallization u a s effected from water yielding product of m.p. 240-242’; A, 222 mp ( e 17,500); sh. 245 nip ( e 8,600); 269 nip ( e 4,4001; 299 nip ( e 3,200). Anal. Calcd. for CiH8C1,S202S,: C1, 21.15. Found: C1, 21.08. Monosulfonamide of 2,3-Dichlorophenol.-2,3-Dichloroanisole was chlorosulfonated and the resulting product treated with ammonia according to the procedure of Lustig and Katscher.’ The product obtained had m.p. 2d2-253”; A, 241 nip ( e 8,600); 272 mp ( e 3,200); sh. 279 nip ( e 2,800); 291 nip ( E 2,500). Anal. Calcd. for CeHoC12X03S: S, 5.79; C1, 29.29. Found: S,Tj.90; C1, 29.08. The same product was obtained when 2,3-dichlorophenol Tvas used instead of 2,3-dichloroanisole.
Acknowledgments.-The authors are indebted to N r . R. Wayne for the infrared spectral data, Mr. E. Connor for the microanalytical results, and Mr. R. Boraczek for the preparation of certain of the compounds described.
The Synthesis and Pharmacological Action of Tremetone I > O ~ ~ G11. L ABOTVEZT,~ S JOSEPH I. DEGRAW,JR.,VISOD R. SHAH,ASD WILLIAMA. BOSSER~ Department of Chemistry, Stanford University, Stanford, California Received November 2, 1962 Tremetone (I),the principal levorotatory ketonic constituent of “tremetol,” the crude toxin of Eupatorium The synthetic sequence involved : coumarilic acid-(SaHg)+ hydrocoumarilic acid -(EtOH, H +)+ ethyl hydrocoumarilate -(hIehlgBr)+ 2-(2,3-dihydro-2-benzofuryl)-2-propanol -( Ac.0, SnCl,)+ 2-( 2,3-dihydro-5-acetyl-2-henzofuryl)-2-propyl acetate -(pyrolysis)+ I. Optical resolution was accomplished at the hydrocoumarilic acid stage, and the synthetic enantiomer ( - )-I proved identical in all respects with natural levorotatory tremetone. Preliminary toxicity data for tremetone and tremetol are discussed, as applied to goldfish, several species of insects, mice, rabbits, sheep and chickens. Crude tremetol proved toxic t o chickens, whereas tremetone was harmless, suggesting that the ketone may not be the active tovin in tremetol responsible for “trembles” in cattle and “milk sickness” in humans.
urticaefolium, has been synthesized.
White snakeroot (Ezipatoriunz wticaefoliztm) is a toxic weed which grows extensively in damp areas of the central United States. Its ingestion by cattle causes the veterinary disease tremble^,"^ while human consumption of dairy products from infected cattle engenders the fatal malady known as “milk sickness, a n incurable illness which periodically ravished early pioneer communities of the central west. I n 1929 ,J. F. Couch isolated from white snakeroot a straw-colored oil, “tremetol,” which produced the symptoms of trembles in sheep and which was considered to be the homogeneous active toxin of this weed.3 S o subsequent confirmation of Couch’s ob(1) S a t i o n a l Science Foundation Faculty Fellow, 19G1-1962. ( 2 ) T h e authors are indebted t o the National Institutes of Health for a research g r a n t (RG-8232) which supported this a n d its preceding inwstigations. (3) J. F. Couch, J . A g r . Res., 35, ,547 (1927): see this article for an extensive review of t h e earlier literature. (4) L. R. Tehon, C. C. Norrill, and R. Graham, “Illinois Plants Poisonous to Livestock,” Circular 599, University of Illinois, College of Agriculture, Extensions Service in Agriculture and Home Economics, 1948, p. 43 ff. ( 5 ) J. F. Couch, J. A m . M e d . Assoc., 91, 234 (1928). (6) J. F. Couch, J . Am. Chem. Soc., 51, 3617 (1929).
servatioizs has been reported, and indeed this challenging problem has lain dormant until only recently, when we became interested in its pursuit.’ Application of column chromatography to the crude tremetol described by Couch afforded six individual components : a sesquiterpene hydrocarbon (CI6R,,), two steroids (C2iH460 and C3oH4,O) and three ketones, namely, tremetone (C13H1402), dehydrotremetone (C13Hi,Oz) and hydroxytremetone (G1H1401) .* These ketones proved toxic to goldfish (ride infra) and produced a red color test with sulfuric acid, a test which Couch found characteristic of trembles-producing fractions.3 -4ccordingly, one or more of these ketones was suspected of being the active toxin of white snakeroot, and their structural investigations were therefore undertaken. Degradatire reactioizs and catalytic hydrogenation revealed that tremetone, the principal ketone con(7) W. A. Bonner, J. I. DeGraw. Jr., D. h4. Boaen, and V. R . Shah, Tetrahedron Letters, 417 (1961); W. A. Bonner, Phytochemistry Symposium, Golden Jubilee Celebration, University of Hong Kong, Sept., 1961. (8) W. A. Bonner and J. I. DeGraw, Jr., Tetrahedron, 18. 1295 (1962).
VI1 N a Hg
7
VI11 MeMgBr 192%
99'7
COOH
VI
IX
0 V
XI
r0.4c
W
($1 and
(-) 1
THESYSTHESIS
May, 1963
A S D PHARNACOLOGICAL
ACTIOSO F TREMETONE
317
mouse, only a slight lethargy was noted when either tremetone or crude tremetol was administered. In toxic intraperitoneal doses of tremetone (500 mg.,’kg. and above) , jerks and clonic convulsions were observed prior to death. Neither oral nor intraperitoneal adXIV xv ministrations (550 mg./kg. and 150 mg./kg., respectively) of crude tremetol to rabbits produced tremor symptoms, although the latter administration elicited muscular weakness and increased respiration. I n the hope of establishing in a single experiment whether tremetone mas the responsible toxin of Eupat o r i u m urticaefolium toward cattle, a single feeding of XVI pure tremetone (130 nig. kg., administered with alfalfa ium is reported to contain a toxic principle, m.p. 6 7 O , feed) was given to 22.6 kg. sheep. 50 untoward which, like that of white snakeroot, is capable of passymptoms were observed during 7 days of subsequent sage through milk into suckling ~ a t t 1 e . l ~Although close observation. I n view of the results of howin melting point to our hydroxytremetone (m.p. 70ever, such a single negative result could not be con71°),* this toxin is said to resemble coumarin and not sidered definitive, and further assays were deemed to give reactions characteristic of crude tremetol. necessary. Unfortunately, our limited supply of pure More recently Zalkow and co-workersIRhave established tremetone precluded further tests with animals as large the presence of both dehydrotremetone (11)and “toxol” as sheep [ (- )-2-isopropenyl-2,3-dihydro-3-hydroxy-5-acetylben- I n 1945, after preliminary experiments with a variety zofuran (XY)] in the crude tremetol of the rayless of hosts, B u t l e F achieved promising results using goldenrod ( A p l o p a p p u s heterophylltes) of Kew Mexico. chickens in toxicity assays of tremetol and other fracOriginally discovered in 1870, euparin has no reported tions obtained after extraction of the rayless goldenrod physiological action, *9 while toxol has been shown to be ( A p l o p a p p u s heterophyllus) of the Southwest. With bacteriostatic toward Bacillus cereus, Staphylococcus the hope of obtaining critical data, therefore, me have albus and Corynebacterium hoagii.I8 repeated Butler’s assays on our samples of tremetoiie Comparison of the structure of the well known fish and crude tremetol, with the following tentatively depoison rotenone (XVI) with that of tremetone (I) finitive results. Whether administered orally or by indicates an almost exact similarity in the substituted injection (in sesame oil) into the breast muscle, pure benzofuran nuclei of these toxins. I t was on the basis tremetone elicited no symptoms of distress in white of this similarity that we originally employed goldfish leghorn cockerels during a 7-day period. Crude tremas test organisms for our tremetol fraction^.^ 8 Semietol a t the same dose level, on the other hand, killed quantitative fish toxicity tests, eniployiiig goldfish in our test chickens in 3-7 days by either mode of adminthe procedure of Gersdorff ,2n suggested that on a molar istration, and produced extreme lethargy aiid breathing basis tremetone (I), dehydrotremetone (11) and hydifficulties within 24-36 hours prior to death. drosytremetoiie (111) had toxicities toward goldfish If therefore appears tenatively valid to conclude that, approximately half that of rotenone. while it is a potent fish poison and shows the sulfuric Derris preparations have also enjoyed widespread acid color test which Couch associated with tremblesuse as insecticides against a wide variety of pests, and producing a c t i ~ i t y , tremetone ~ (I) is probably not the the toxicities of rotenone aiid other derris components toxin in tremetol which is responsible for trembles in toward higher animals have been the subject of a cattle and milk sickness in humans. The responsible nuniber of iiivestigations.?l Accordingly, we have also toxin, which must clearly be a more potent minor cominvestigated the insecticidal properties of tremetone, ponent of crude tremetol, is now being actively sought and wish to report the following preliminary observaafter. tions. Against mosquito larvae, the coninion housefly and the Gerniaii roach, tremetoiie had approximately 7, Experimental 3.5 and 2%, respectively, of the insecticidal potency Coumarilic Acid (VI).-The following improved method for the of pyrethrins. Against the boll weevil. cotton aphid, oxidation of 2-acetylbenzofuran to coumarilic acid was developed two-spotted spider mite and southern army worm, after preliminary experimentation with previously reported proneither tremetone nor crude trenietol showed significant cedures.*3~ 24 “Sanichlor” household bleach ( 1500 ml.) was warmed to 60”, then treated over 20 min. with stirring and mild toxicity. cooling with a solution of 2-acetylbenzofuran (64 0 g . ) in rnethToxicity tests on both the rabbit and the mouse indianol (400 ml.) a t such a rate that gentle reflux was maintained. cated that these mere not suitable hosts for the evaluaToward the end of the addition the solution darkened. but could tion of trembles-producing activity. With both oral he restored to its original 4 ellow color by intermittent addition and intraperitoneal doses of 50-300 mg., kg. in the of more “Sanichlor” (100-200 ml ). The small excess of hypoI
04
‘I
(17) C. P a k and B. E. Read, Chinese J . Phusiol., 1’2, 263 (1937); C . A . , 32, 2221 (1938); H. L. Hi and C. Pak, Chinese J. Phusiol., 12, 275 (1937); C. A . , 32,2222 (1938). (18) L. H. Zalkow, S . Burke, G.Cabat, and E. G . Grula, J. .Wed. Pharm. Chem., 6, 1342 (1962). (19) J. U.Lloyd, as reported by H. Trimble, A m . J . Pharm., 62, 73 ( 1 8 9 0 ) ; C. C. Manger, i b i d . , 66, 120 (1894). (20) W. A. Gersdorff, J . A m . Chem. Sac., 62, 3440, 5051 (1930). (21) Cf. A. M. Ambrose, F . DeEds, and J. B. M o S a u g h t , I n d . Eng. Chem., S4, 684 (1942). for bibliography of earlier studies.
chlorite remaining was destroyed by the addition of sufficient sodium bisulfite. The clear, yellow solution was cooled, washed with ether and acidified t o precipitate 62.1 g. (96%) of crude conmarilic acid as nearly colorless granules. The sample was (22) S. 0. Butler, “Fractions of Tremetol and their Toxicities,” 11. S. Thesis, Oklahoma Agricultural and Jlechanical College, Stillwater, Oklahoma, 1945. (23) M. W. Farrar and R. Levine, J . A n . Chem. Soc., 72, 4436 (1950). (24) R. C. Fuson, J. U’.Kneisley, and E. W. Kaiser, “Organic Syntheses,” Coll. Vol. 111,J o h n Wiley a n d Ssna, New York, N. Y., p. 209.
rerrystallized froni 50' i, etli:inol t i , yield siilficientl>,pure product m . p . I!t!.,?-l!J4' (repiirted2i n1.p. l95-IWo), i n ; over-:ill yield. Hydrocoumarilic Acid ~ V I I ~ . - - ~ C l i i i i i i : i , r i l:ic*id i(~ 15.3 g. I W : L ~ suspended in water (150 nil.) :ind neutralized t t i tlte pliwolphthalein end point with 6 .I s' ~ d i i i ~hydroxide. ii I: s n i n l l : i i i i o i i I i t of solid reniained undissolved iintil r t ~ l u c ~ t i c i Ijeg:iii. ii Tile solu1 , i i i n \viis shaken vigor~~iisly in :i stoppered vt.ssel Ivitli 1'( ; sodiLi1ii :iiii:ilg:tiii2b (X30 g.), resulting iii spciiit:ineoiis \r:irniing. *ifter > i t i i n . the mercury \viis sep:ir:itrd, tlie :icjueoris sl!luticln v,xs filtered (l.'ilter-cell :ind t h e filtrate was :ic*idified. cviiiled :Ind estractetl witli etlter. Solvent removal afforded I.i.4 E. 1iytlrr~c:ciiiiiiarilicacid, 11i.p. 115-1 17.3" report \rlii14i wis used directly in suhsequent work. 1,esa of c*ouriiarilic.:icGl pave s a n i p i t ~c i f Ii>dnicoiini:i nw-e lxirely solid :it rooin teniper:iture :ind rwlu tion f r o m tienzene before tiicy were :iccept:il)l~-pitre. Resolution of Hydrocournarilic Acid.-( h i r iriitial :ittenipty t o ~:iiipli~y brucine as :t resolving agent for this ark1 after the provedure of FredgaI2 were unsuccessful. In our experience racemic I)ruciiie liydroi~ouniarilatrfniled t,ii crj.st:illizc, :is did also the I)rric.inc. wits of tile resolved s:iniples 1111t:iined tieln~v. 1I-e sul~seiiueritlyfound that aniplietnriiine served corivenientl!- 111 rcsiilve 1iyclr~icouiii:irilic:acid and, sin(*(.it is c.oiiiiiier&illy availd ) l e i n Ivitli enantioineric fornis, Ii:itl tlir. :idtlitionul :idvant:tgci of Iwrtiiitting procurenient of hoth enantirrniers of this acid. Fredg:L 1i:is used miphetaxiiine only on the nioilier liqunrs f r l i r i i his I l r i i iaiiie resolution.'2 1 experiment hydrocourii:irilic acid (4.30 g.) w:is lie :imine liberated froni ( + ):imphetarnine sulfate ie mixture x a s dissolved in w:iter (GO nil.) a t 80". ('cioling deposited 3 . 7 3 g. cif salt, n1.p. 130-IS1 '. Itcc.r>-st:illiz:i' of tlie salt from m t e r (50 nil.) afforded 1.73 g. of snit. m.p. -187". the acid recovered from ~ I i i c l had i m.p. C)2-:)Go. l a ]%I 0.0" ( c , 6.1, et1i:rnol). A second recrystallization pave 0.34 g. of s d t Iinving imp. 186-1Si0, tlie acid (0.22 g.) from wliicli lint1 ii1.1). 100--104° and [ a ] ? %22.6" f c , 7,:3, etlimol), in agreenient with previously reported values. I ? 13y conversion I J f tlie free acid obtained from the mother liquors of this resolution into its ( - j :tniplietaniine s d t , the en:rntiorneric ( - ) Ii!-droc~oiini:irilil. acid could similarly be Suhsecluently it w:is fciuntl that recrystnllization of such salt's from 0547, etli:inol permitted soniewliat more rapid resolution. In larger scale resolut,ions full purific:~tiim of the amphetamine salts \vas not. undertaken, since cciiiiplete resolution w:is nttended by hcavy losses and it w:is furtlier found that the succeeding crystalline acetate XI could lie rendered optir:illy prire n-itli less owr-:ill loss bj- simple recrystallim tion. Ethyl Hydrocoumarilate (VIII).--The enantinmers of ethyl lt~tlri~c~~iuiii;irilate were prepared by applying the esterification prcicwlurt: i i f voii A u w e r ~ directly '~ t o the enantiomeric nmplict: i n r i i i e s:tlts of Iiydrc~couniarilic acid. .4 mixture of the ( - j :iniplictaniine salt ( i f ( - )hydrocoumarilic wid (GS.8 g . ) , absohitc c~tltanol(1.50 nil.) ant1 sulfuric acid ( 2 5 nil.) was lieated under reflux for 1 ..ihr.>then was cooled, diluted with water and processed :is iisual. The crude ester wvas distilled, h.p. 130° ( I 1 iiiin.j, t o givtb 30.4 g. (69' I j of pure etliyl ( + ) liydrt)coiiiiiarilate, n ? S n l..i172. [CY]% +17.9°(c,:3.98; hexane); +12..(io (neat). . I r r c i i . (hlid. for ( ~ i I € l l z O ~C. : 6S.73: II, 6.29. Found: c'. (i11.10: 11, 0.18. X s:inipIe of tlie ( )-amphetamine salt of ( )-hydrocoumarilic :ic.itl \vv:is cmnverted t o ethyl ( - )hydrocouniarilate ( 7 7 % ) in the s:inie ni:inner, h.p, 130" (!) nini.), ~ 1 % 1.5190,] CY]% -11.6"
+
+
2-[2,3-Dihydro-2-benzofuryl)-2-propanol (IX!.-3Iethvlmagt)rriniide wts prep:tred f r o m rnngnesiuni (4.86 g.) in ether ( 130 nil.) and sufficient inethyl bromide. Addition of n solution of r:iceiriic ethj-1 Iiydrocou~i~arilate (14.4 g.) in ether (50 m1.j ovclr 10 niin., anti customary processing afforded 12.3 g. (9257) of distilled produvt, b.p. 107-107.5° ( 2 mni.), the infrared spect r u m IJf which was identical with t h a t of our previously reported s:imple.'i The enantiomeric forms of this compound were prenc4ritri
May, 1968
THESYXTHESIS ASD PHARMACOLOGICAL ACTIOSOF TREMETOSE
1.080; R f in silica thin layer chromatography: synthetic, 0.20 (benzene), 0.83 (ethyl acetate); natural, 0.19 (benzene), 0.54 (ethyl acetate); infrared spectra (neat), superimposable. Semicarbazones: synthetic, m.p. 222", [a] 2 8 -56.6' ~ (c, 0.74, chloroform); natural, m.p. 22'L0, mixture m.p. 222", [ a ] -.56,2' ( c , 0.84; chloroform): Rr on silica plates: synthetic, 0 . E : natural, 0.26 (ethyl acetate); identical infrared spectra in c*lilorofiirmsolution. .And. Calcd. for CIIHliS102:C, 64.84; H , 6.61; S , 16.21. Found: (synthetic), C, 64.W; H, 6.91; S , 16.35; (natural), C,65.01; H, 6.61; Pi, 16.49. 2,4-Dinitrophenylhydrazones: synthetic, m.p. 184.2-184.To : natural, m.p. 183.8-184.2", mixture n1.p. 183.8-181.2°. Ri on silica plates: synthetic, 0.96; natural, 0.96 (ethyl acetate); identical infrared spectra in chloroform solution. Anal. Calcd. for CI,H,,N405: C, 59.68; Ii, 4.74; S , 14.65. Found: (synthetic), C, 59.8T; H, 4.99; ?IT, 14.63; (natural), C, 59.50; H, 4.61; X, 14.57. .4 sample of unnatural (+)tremetone was prepared by pyrolysis of the dextrorotatory acetate (+)XI. Vapor phase chromatography of the crude pyrolysis product showed the usual mixture of components. The tremetone fraction was isolated as before, [ a ] " +55.3' ~ (c, 4.5, ethanol). Goldfish Toxicity Tests.-In general these were cond tcted as followszo: The goldfish (1.5-2 g.) was placed in a large beaker of water a t room temperature, the toxin in ethanol solution (20 nig./ml.) was added in such quantity as to give the indicated final c-oncentration, and the time required for death of the goldfish was noted. The criterion for death was cessation of gill motion. Complete paralysis of the fish, other than gill motion, was generally noted in approximately 337; of the time required f i n gill motion to cease. Crude tremetol was administered to goldfish a t concentrations of 50, 100 and 200 nig./l. Ileath times were, respectively, about 10, 5 and 3 min. Several of the pure compounds previously isolatedi.8 from white snakeroot were administered to goldnsh a t a dose level of 30 nig./l. The approximate death time observed for each of these compounds was: Terpene I, non-toxic: dehydrotremeton, 20 niin. ; tremetone, 24 min.; hydrosytremet,one, 21 min. A rough comparison of the fish toxicity of tremetone R-ith that of rotenone was made in this fashion. At 10-6 JI rotenone gave a death time of about 45 min., in agreement with the previous literature.*O At dl tremetone required 110 min., while a t 2 x 10-5 N i t required about 53 min. Mouse and Rabbit Toxicity Tests.-Tremetone was given orally and intraperitoneally t o groups of white mice ( 3 per dose level). The doses were 50, 100, 200, 300, 500, 750 and 1000 mg./kg. Tremor was not observed a t any dose level. The only symptoms observed in doses of 100-300 mg./kg. was a slight decrease in activity. In toxic doses (500 nig./kg. and above, intraperitoneally only) jerks and clonic convulsions were present. These data suggest that this compound is not a tremor-producing agent in the mouse. The substance was also run in a standard battery of anticonvulsant tests in mice, but showed no anticonvulsive activity. Essentially siniilar results were obtained in mouse tests employing crude tremetol. I n addition, crude tremetol was given to rabbits with the following observations: oral administration a t 550 mg./kg. produced no tremors; intraperitoneal administration a t 150 mg./kg. likewise produced no tremors, but did appear to elicit general muscular weakness and increased respiration rate. It is apparent that neither the mouse nor the rabbit is a suitable animal for screening these types of compounds. IVe are indebted to Dr. Eugene L. TVorach and
319
Dr. Guy Everett of the Abbott Laboratories, North Chicago Illinois, for conducting and interpreting these experiments. Lamb Toxicity Tests.-A 250-g. portion of alfalfa hay was mixed thoroughly with a solution of tremetone ( 3 g.) in chloroform and the solvent was allowed to evaporate. I n a control, the alfalfa was mixed with chloroform only and allowed to dry. Both diets were prepared immediately before feeding to the test' animals, which were two female lambs, 22-25 kg., fasted for 24 hours prior to the test. Each animal consumed the test meal and was observed subsequently during 7 days. S o untoward symptoms were noted in either animal. This test was performed by the Diablo Laborat,ories, Berkeley, California. Chicken Toxicity Tests.-ButlerZz has studied the toxicities of tremetol preparations from rayless goldenrod in a number of animals incaluding the cat, dog, guinea pig, rabbit, rat, duck and chicken. Only the chicken proved sufficiently sensitive to merit detailed study. We have accordingly modified But!er's technique for our own purposes, employing 1-month old white leghorn cockerels (approximately 90 g. each) and using both parenteral and oral administration. Parenteral injections were made directly into the breast muscle using 200 mg. of toxin dissolved in 1 ml. sesame oil, uncontaminated sesame oil ( 1 ml.) serving as a control. Oral administration involved 200-mg. samples of toxin placed in S o . 5 gelatin capsules which were forced down the subject's gullet with a glass st.irring rod, sesame oil in a capsule being used as the control. Samples were administered by each method once every 24 hr. I n either parenteral or oral administrations, no effects were noted using tremetone after 5 or 6 doses. \Then crude tremetol was administered parenterally the chickens died between the 5th and i t h dosage. When trernetol was administered orally death occurred after the 3rd or 4th administration, and the chickens showed a typical crouched position, great lethargy and difficult breathing after the 2nd or 3rd dosage. Insecticide Tests.-A peries of insecticide tests n-as performed on tremetone by the procedures described and with these results. Mosquito larvae (Ciclex quinquejasciahs) in batches of 25 were placed in aqueous tremetone solutions at a series of concentrations, and after 24 hr. the dead and surviving were counted. LCW proved to be 0.7 ppm. The common housefly (Musca domestica) was exposed to a series of concentrations of tremetone dissolved in oil and placed on the interiors of shell vials ( 2 mm.3/vial). Female flies, 4-5 days old, were placed in the vials in groups of 25 and exposed for 1 hr., then removed to holding cages, given sugar and water, and examined for mortalitj- 24 hr. later. LDsr proved to be 20 pg./vial. I n the case of German roaches (Blatdla germmica), tremetone in acetone solutions of varying strength was applied topically to the thorax of males, and the mortality was read afte; 4s hr.: I,l)jr, proved to be 50 pg. Typical data in these tests for pyrethrin insecticides are: mosquito larvae, LCjo, 0.05 ppm.; common housefly, I,I)jo, 11 pg./vial: German roach, LIIjn, 1 pg. These tests, interpretations and data were gcnerously furnished by Prof. IT. 11. Hoskins, Ilepartment of Entomology and Parasitology, University of California, Berkeley. I n addition, both tremetone and crude tremetol were subjected to standard laboratory tests involving toxic behavior toward the boll weevil, cotton aphid, two-spotted spider mite and southern army worm. Pieither sample proved toxic to any of the above insects in these tests. We are indebted to Mr. B. A. Butt, Pesticide Chemicals Research Branch, Entomology Research Division, U. S. Department of Agriculture, A4griculturalResearch Service, Beltsville, SIaryland, for generously conducting and interpreting the latter tests.
