"It does not apply to cases where the conjunction unites portions of the _predicate_, instead of the _subject_, of a proposition. If I assert that a gentleman of my acquaintance drinks brandy _and_ water, he might not relish the imputation of imbibing separate potations of the neat spirit and the pure element. Stradling _versus_ Stiles is a case in point: 'Out of the kind love and respect I bear to my much honoured and good friend, Mr. Matthew Stradling, Gent., I do bequeath unto the said Matthew Stradling, Gent., _all my black and white horses_.' The testator had six black horses, six white horses, and six pied horses. The whole point at issue turns upon the question whether the copulative _and_ joins sentences or words. If the former, the plaintiff is entitled to the black horses, and also to the {280} white, but not to the pied. If the latter he has a right to the pied horses but must forego his claim to the rest. And if the latter interpretation be adopted, must we say that _and_ is a preposition, not a conjunction, or must we modify the definitions of these two parts of speech?" The following definitions are finally proposed in place of the ordinary ones: "A preposition is a part of speech annexed to a noun or verb in a proposition, and serving to connect it with a noun or pronoun by which it is limited, as the subject or predicate of that proposition." "A conjunction is a part of speech serving to unite two propositions as parts of the same complex assertion, or two words as similar parts of the subject or predicate of one proposition. By _similar parts_ it is meant that the words so united stand in similar relations to the term to which they belong. For example, 1. As attributes, both qualifying a subject, 'Sic bonus et sapiens dignis ait esse paratus.' 2. As prepositions, both introducing limiting nouns 'without money and without price.' 3. As substantives, both forming parts of a collective subject, 'two and three are five.' Whereas with the preposition, the words united are not similar, but opposed, the _limiting_ and the _limited_ notion." While differing from some of PROFESSOR BOOLE'S views on the relation of logic to mathematics, I fully agree with him that the true functions of the several parts of speech must be determined by an analysis of the laws of thought. Both grammar and logic might be considerably improved by an accurate development on psychological principles. H. L. MANSEL, B.D. St. John's College, Oxford. Has not your correspondent G. BOOLE fallen into an inaccuracy whilst contending about the accuracy of another's logic? He seems to employ the proposition, _all trees are endogens or exogens_, as an example of an accurate proposition.