Even when all details are omitted, the general drift of this work must be sufficiently apparent to the reader. Although conducted according to a fixed method, it does not escape the difficulties inherent in _every_ classification of the emotions. In the first place, the order of filiation is not always very well marked. The author himself recognises that an arrangement in series is not possible, but this difficulty has also presented itself in zoology and botany. We meet with repetitions, _i.e._, forms of sentiment figuring several times over in different categories. This, too, is inevitable. The complex emotions (or some of them, at least) are formed by anastomoses: they are rivers formed by converging streams coming from various sources lying in different directions. One may legitimately refer them to one or other of these origins; but the attribution will be partial and arbitrary. The religious sentiment, for instance, is included in the class of intellectual emotions. But its social character is undeniable (a point we shall return to in the proper place); let us recall, in passing, the worship of ancestors and deified heroes, and the strictly national religions of antiquity, the communities, orders, confraternities, corporations, the missionary work carried on in modern times, and, above all, the contagious character of religious emotion in general. It is false, moreover, to say that this emotion “tends neither to the preservation nor the destruction of the individual.” It might therefore be just as well—or just as ill—placed in the third class. As soon as we pass from simple to complex emotions, it is of more importance to determine their composition than their filiation. Now, this procedure belongs rather to chemical than to zoological classification. III. A third type of classification, peculiar to the intellectualists, consists in classing according to the intellectual states, in so far as these are accompanied by affective elements. This system sprang from the psychology of Herbart, is based on it, and is met with in the works of the principal representatives of his school, Waitz, Drobisch, and especially Nahlowsky in _Das Gefühlsleben_ (pp. 44 _et seq._). This method is peculiar to Germany, and its influence is still perceptible even in Wundt, and more recently in Lehmann’s book (_op. cit._, pp. 338 _et seq._). In England, Shadworth Hodgson approaches this type. Apart from the procedure common to all, these classifications agree still less in detail than those of the first two types. Taken broadly, they have an academic aspect; they are frittered away in divisions, sub-divisions, distinctions, whence there arises more darkness than light. There is, however, a dichotomy peculiar to them corresponding to a reality which is not met with in the two previously-mentioned types, and, on this account, deserves notice.