Asaronic acid is conveniently prepared from 4-O-methyl-β-resorcylic acid and isovanillic acid by oxidatron with alkaline persulphate and subsequent methylation. Useful partial methyl ethers and mixed methyl-ethyl ethers are also made by adopting this method. Similarly asarylic aldehyde is obtained from 4-O-methyl-β-resorcylic aldehyde and isovanillin.

(1) Chrysin, (2) tectochrysin and (3) 5-O-methyl chrysin are subjected to the two stage oxidation process. All of them give good yields of the corresponding 8-aldehydes. In (2) the 8-position (para) is predominantly more reactive than the 6-position (ortho). In (1) and (3) the favourable influence of a hydroxyl and a methoxyl in the 5-position is exerted. The Dakin’s reaction gives poor yields if the product is a quinol and good yields if it is a catechol. Experiments with (1) and (2) lead to the conclusion that the two stage method is available for para oxidations though the yields are poor.

For the synthesis of 5∶6∶8-trihydroxy flavone (6-hydroxy primetin), the two stage ortho-oxidation process is applied to primetin monomethyl ether. A good yield of the 6-aldehyde is obtained, which undergoes oxidation to 5∶6-dihydroxy-8-methoxy flavone. Subsequent demethylation yields 6-hydroxy primetin. Its properties and reactions and derivatives are described.

3-Hydroxy primetin has been synthesised starting with 3∶5-dihydroxy flavone and using partial methylation, para nuclear oxidation and final demethylation. During the last stage no isomeric change is noticed. This is in conformity with the behaviour of similar cases of flavonol derivatives and is different from the behaviour of analogous flavone derivatives. Orthooxidation of 3-hydroxy primetin leading to the synthesis of the lowest member of the gardenin series is found to proceed satisfactorily.

A number of phenyl and halogen substituted coumarins with substituents in different positions have been examined as fish poisons. The presence of a phenyl group in the 3 or 4 position or a halogen atom in the 3-position seems to be most favourable. Higher substitution does not appreciably enhance the toxic properties. In consonance with previous results, a hydroxyl group in the 5-position seems to produce greater toxicity than one in the 7-position.

When boiled with hydriodic acid, 2:3-dimethyl-5:7:8-trimethoxy chromone undergoes besides demethylation, isomeric change also, yielding 2:3-dimethyl-5:6:7-trihydroxy chromone, thus showing that a free hydrogen atom in the 3-position is not essential for this isomeric change to take place and that the nature of the substituent is also important. For purposes of comparison, 2:3-dimethyl-5:7:8-trihydroxy chromone has been obtained by demethylation of the above trimethyl ether with aluminium chloride and 2:3-dimethyl-5:6:7-trimethoxy chromone by independent synthesis using nuclear oxidation as a stage.

A review of ring isomeric change of flavones and chromones is made as also of the reactivity of C-hydroxy phloroglucinol and C-hydroxy phloracetophenone derivatives in methylation, benzoylation and ring closure. The results are consistent showing the greater reactivity of a hydroxyl group of the quinol unit as compared with that of the catechol unit. A tentative explanation is given.

A synthetic study has been made of the chromones ofEugenia caryophyllata. All the intermediates in the synthesis of eugenitin have now been isolated and fully characterised. A convenient route for the preparation of the starting material for this synthesis,viz., 3-methyl phloracetophenone has now been found. Two new syntheses of eugenin are also recorded. The observation of Schmid and Bolleter that eugenitin (6-methyl) undergoes change into isoeugenitol (8-methyl) when demethylated with boiling hydriodic acid and that isoeugenitin methyl ether (8-methyl) does not undergo isomeric change under the same conditions, is confirmed. This is the reverse of what has been observed in the case of C-hydroxy phloracetophenone derivatives. Further it is extraordinary because C-methyl phloroglucinol and C-methyl phloracetophenone behave like the C-hydroxy (methoxy) compounds in methylations and chromone ring closure.

An attempt has been made to develop a comprehensive scheme of biogenesis which includes all the five components occurring inEugenia caryophyllata. The basic compound is taken to be nor-eugenone and all the stages are based on analogy with laboratory experiments.

Okano and Beppu isolated a new isoflavone from soya bean and from its degradation reactions, they assigned to it the constitution of 8-methyl genistein. A synthesis of this compound starting from C-methyl phloroglucinol α-monomethyl ether was also reported by Shriner and Hull.

An easier method of synthesis of a compound of this structure has now been described, in which the main stage is the nuclear methylation of 2∶4∶6-trihydroxy-4′-methoxy phenyl benzyl ketone. The constitution of the product of this reaction is established by an independent synthesis from C-methyl phloroglucinol. Though this synthesis of 8-methyl genistein involves demethylation of its trimethyl ether with hydriodic acid, the absence of any rearrangement during this step has been established particularly by remethylation. The products of the present synthesis are found to be very different from the natural compound and its derivatives described by Okano and Beppu and therefore it is considered that the natural compound cannot have the constitution assigned to it.

Partial demethylation of 8-methyl genistein trimethyl ether with hydriodic acid gives rise to 8-methyl prunetin which on partial methylation yields 8-methyl-5-hydroxy-7: 4′-dimethoxy isoflavone. This is different from the C-monomethyl-O-dimethyl genistein obtained by Perkin and Horsfall by nuclear methylation of genistein. Therefore the product of Perkin and Horsfall may have the methyl group in the 6-position as suggested by earlier workers.

In view of the discrepancies between the products of the synthesis of 8-methyl genistein described in Part XXII and those of Shriner and Hull, the method of these authors has been re-examined. 2∶4-Dihydroxy-3-methyl-6∶4′-dimethoxy phenyl benzyl ketone and 5∶4′dimethoxy-7-hydroxy-8-methyl isoflavone prepared now are found to have melting points much higher than those reported by Shriner and Hull. However, on methylation, they yield their respective monomethyl ethers which are identical with those described in Part XXII. There is thus complete agreement between the two syntheses.

The natural compound of Okano and Beppu cannot be 8-methyl genistein. However the degradation reactions could be explained on the

One of the substances occurring in soya beans has been considered by Okano and Beppu to be 5∶7∶2′-trihydroxy isoflavone (isogenistein), based on degradation studies. A compound of this structure has now been made starting from 2∶4∶6-trihydroxy-2′-methoxy phenyl benzyl ketone. Partial methylation, treatment with ethyl formate and sodium followed by demethylation yield the required isoflavone. Aluminium chloride in benzene solution is found to be a much better reagent than hydriodic acid for the above demethylation. The products of the present synthesis are found to be different from the natural compound and its derivatives. Therefore the natural compound cannot be an isoflavone of this constitution. The main degradation product, considered to be 2∶4∶6∶2′-tetrahydroxy phenyl benzyl ketone by Okano and Beppu differs considerably from a synthetic product of this constitution. Hence the natural compound cannot be a simple isoflavone or isoflavanone and should have a different structure.

One of the crystalline compounds occurring in soya beans has been considered by Okano and Beppu to be 8-methyl isogenistein and this structure was based on the alkali degradation of the compound and its trimethyl ether. A compound of this structure is now synthesised starting from 2∶4∶6-trihydroxy-2′-methoxy phenyl benzyl ketone, along the lines of synthesis of 8-methyl genistein described in Part XXII. Here also the products are found to be different from those reported by Okano and Beppu and the natural compound should therefore have a more complex structure.

The evolution of the chromano-chromanone unit present in rotenone and related compounds is of much interest. From a consideration of a number of structures occurring in nature, it is suggested that 2-hydroxy-methyl-2′-hydroxy isoflavanone or its equivalent is an intermediate and it undergoes ring closure by dehydration.

Based on this hypothesis, a synthesis of 7-hydroxy-chromeno-(3′∶4′; 2∶3)-chromone has now been achieved. 2-Methyl-7∶2′-diacetoxy isoflavone is converted into a 2-bromomethyl compound by treatment with N-bromo succinimide. On hydrolysis and treatment with potassium carbonate in acetone, it yields 7-hydroxy-chromeno-(3′∶4′∶2∶3)-chromone.

Though attempts to prepare the required diacetoxy isoflavone directly from 2∶4∶2′-trihydroxy phenyl benzyl ketone have not been successful, it could be obtained starting from 2∶4-dihydroxy-2′-methoxy phenyl benzyl ketone and treatment of the intermediate 2-methyl-7-acetoxy-2′-methoxy isoflavone with aluminium chloride in benzene solution, followed by acetylation.

Methylation of 2-methyl-7∶2′-dihydroxy isoflavone with excess of methyl iodide gives rise to the 7-methyl ether, indicating the existence of a considerable difference in the reactivities of the 7 and 2′-hydroxyl groups. The negative ferric reaction of these 2′-hydroxy compounds shows absence of chelation.

It is established that the condensation of 2-hydroxy-4:6:2′-trimethoxy phenyl benzyl ketone and its 3-methyl derivative with ethyl formate as well as with methyl formate produces the corresponding 2-hydroxy isoflavanones. These undergo dehydration to the trimethyl ether of isogenistein and 8-methyl isogenistein respectively. Demethylation of the above ethers can be conveniently effected by aluminium chloride in benzene solution. The 2-hydroxy isoflavanones also yield the same products by undergoing dehydration along with demethylation.