effect of being out of the direct line of their natural mode of expression. In Chopin, for instance, the characteristic qualities of modern music, in the treatment of ideas in short and malleable forms specially adapted to their expression, are found abundantly, and in these his genuine qualities are most clearly displayed. His sonatas are less successful, and less familiar to musicians; because, though quite master enough to deal with structure clearly and definitely, it was almost impossible for him to force the ideas within the limits which should make that structure relevant and convincing. They are children of a fervid and impetuous genius, and the classical dress and manners do not sit easily upon them. Moreover the luxuriant fancy, the richness and high colour of expression, the sensuous qualities of the harmony, all tend to emphasise detail in a new and peculiar manner, and to make the sonata-principle of the old order appear irrelevant. The most successful are the Sonatas in B♭ minor for pianoforte, op. 35, and that for pianoforte and cello in G minor, op. 65. In both these cases the first movements, which are generally a sure test of a capacity for sonata-writing, are clearly disposed, and free from superfluous wandering and from tautology. There are certain idiosyncrasies in the treatment of the form, as for instance in the recapitulation, which in both cases is almost limited to the materials of the second section, the opening features of the movement being only hinted at in conclusion. The subjects themselves are fairly appropriate to the style of movement, and are kept well in hand, so that on the whole, in these two cases, the impression conveyed is consistent with the sonata-character. In scherzos Chopin was thoroughly at home, and moreover they represent a province in which far more abandonment is admissible. In both sonatas they are successful, but that in the Pianoforte Sonata is especially fascinating and characteristic, and though the modulations are sometimes rather reckless the main divisions are well proportioned, and consequently the general effect of the outlines is sufficiently clear. The slow movements of both are very well known; that of the Pianoforte Sonata being the Funeral March, and the other being a kind of romance in Chopin's own free manner, which is familiar to players on the cello. The last movement of the Pianoforte Sonata is a short but characteristic outbreak of whirling notes, in general character not unlike some of his Preludes, and equally free and original in point of form, but in that respect not without precedent among the last movements of early masters. In the mind of the composer it possibly had a poetical connection with the Funeral March. The other last movement is a free kind of rondo, and therefore more consonant with the ordinary principles of form, and is appropriate, without being so interesting as the other movements. The total effect of these sonatas is naturally of an entirely different order from that of the earlier types, and not so convincing in oneness as the works of great masters of this kind of form; they are nevertheless plausible as wholes, and in details most effective; the balance and appropriate treatment of the two instruments in the op. 65 being especially noteworthy. The other sonatas for pianoforte, in C minor and B minor, are more unequal. The first appears to be an early work, and contains some remarkable experiments, one of which at least has value, others probably not. As examples may be mentioned the use of 5-4 time throughout the slow movement, and the experiment of beginning the recapitulation of the first movement in B♭ minor, when the principal key is C minor. In this sonata he seems not to move with sufficient ease, and in the B minor, op. 58, with something too much to have the general aspect of a successful work of the kind. The technical devices in the latter as in the others are extremely elaborate and effective, without being offensively obtrusive, and the ideas are often clear and fascinating; but as a complete and convincing work it is hardly successful.
Sonatas which followed implicitly the old lines without doing more than formulate subjects according to supposed laws do not require any notice. The mere artificial reproduction of forms that have been consciously realised from observation of great works of the past without importing anything original into the treatment, is often the most hopeless kind of plagiarism, and far more deliberate than the accidents of coincidence in ideas which are obvious to superficial observers.
As examples of independent thought working in a comparatively untried field, Mendelssohn's six sonatas for the organ have some importance. They have very little connection with the Pianoforte Sonata, or the history of its development; for Mendelssohn seems to have divined that the binary and similar instrumental forms of large scope were unsuitable to the genius of the instrument, and returned to structural principles of a date before those forms had become prominent or definite. Their chief connection with the modern sonata type lies in the distribution of the keys in which the respective movements stand, and the broad contrasts in time and character which subsist between one division or movement and another. Different members of the group represent different methods of dealing with the problem. In the large movements fugal and contrapuntal principles predominate, sometimes alternating with passages of a decidedly harmonic character. In movements which are not absolute fugues the broad outlines of form are commonly similar to those already described as exemplified in Bach's Sonatas, and in the first and last movements of his 'Concerto dans le style Italien.' This form in its broadest significance amounts to a correspondence of well-defined sections at the beginning and end, with a long passage of 'free fantasia,' sometimes fugally developed, in the middle. The clearest example in these sonatas is the first movement of the 3rd Sonata, in A major, in which the corresponding divisions at either end are long, and strongly
