In order to be exempt from tax, a cordial must be a cordial "for the making of a non-alcoholic beverage". The respondent's products can only be brought within these words by virtue of a fact which was not given in evidence but was stated by counsel before us at the very end of the argument and not contradicted. That fact seems to be (though it was not stated precisely in these terms) that each of the respondent's products is intended and adapted to be drunk not in its neat state but only after being mixed with water or some other diluent such as soda water. On this basis the relevant "making" of the "non-alcoholic beverage" is the act of the ultimate consumer. This seems to me to be a strange view in itself. It may very well be that the respondent's products, though not normally drunk without dilution, are quite capable of being drunk without dilution, as are whisky and brandy. At least there is no evidence that they are not. Are we to say that the question whether a liquid is "for the making of beverages" is to be answered by reference to the probability of dilution before consumption or to the intention or belief of the manufacturer that there will be dilution before consumption? But, if the view stated above is strange in itself, it becomes even stranger when we look at the context in which the word "cordial" is found. The Schedule exempts "essences, concentrates and cordials for the making of non-alcoholic beverages". The witness Phillips defined an "essence" as an "alcoholic solution of an oil". That, no doubt, is its specific meaning in pharmacy, but that meaning is clearly excluded here because only things which are non-alcoholic are included in item 36 (3). The Oxford Dictionary gives as the pharmaceutical meaning the meaning given by Phillips, but it gives the general meaning of the word as "an extract obtained by distillation or otherwise from a plant or from a medicinal odoriferous or alimentary substance and containing its characteristic properties in a concentrated form". The same dictionary gives the relevant meaning of "concentrate" as "the product of concentration", which is near enough to that given by Phillips, "a substance, the volume of which has been reduced either by evaporation or by sublimation". In connection with the preparation of food and drinks the general idea conveyed by both words is, I think, that of a flavouring substance in concentrated form, and that is what I am of opinion that the words mean here. Essences and concentrates (which may or may not be liquid) are, one would suppose, substances which would normally be used as ingredients in non-alcoholic beverages and could, and perhaps normally would, be used by the manufacturer in the making of a non-alcoholic beverage. It is difficult to believe that it was intended to place on the same footing a product which can only be said to be "for the making of non-alcoholic beverages" in the sense that it could, and probably would, be diluted by the consumer before he drank it.