ALTHOUGH text-books of practical physics are now fairly numerous, the demonstrators in physics of the University of Toronto appear to have found it impossible to adapt any of the existing ones to the requirements of their students. The same difficulty has been experienced in most English laboratories, partly on account of the diversity which exists between the apparatus of the various institutions, and partly because of the different aims of students working in them. Consequently nearly every laboratory possesses instruction-sheets, either in print or manuscript, which have in some instances been edited and issued in book form; the present volume is a case in point. The experiments described in the book are arranged in two courses, elementary and advanced, in the former of which weighing and measuring, light and heat, are included, while the advanced course consists of sound, advanced heat, electricity, and magnetism. Some easy experiments in electricity and magnetism might have been included in the elementary course, otherwise the selection is a fairly good and useful one. The section on sound contains figures and descriptions of Helmholtz's and König's apparatus for various experiments, but many students' laboratories in this country are, unfortunately, not equipped with large electrically-driven tuning-forks and chronographs. Experimental details are sometimes omitted; for example, in describing the use of the spectrometer as a goniometer, the student is not told how to focus the telescope and adjust the slit so as to obtain a parallel beam of light from the collimator. In the same experiment it is erroneously stated that the edge of the crystal must coincide with the axis of the instrument, whereas the two need only be parallel to each other if the collimator is properly adjusted. Again, in calorimetric measurements, errors arising from radiation between the calorimeter and its surroundings are not very adequately discussed; the calorimeter is supposed in all experiments to alter its temperature at a uniform rate, and matters are so arranged that the mean between its initial and final temperatures is that of the atmosphere. In specific heat determinations a preliminary experiment is necessary in order to satisfy the latter condition, and the assumption of uniform rise of temperature is only approximately true, even when the body under experiment is a good conductor of heat. The book is printed in a bold, clear type, and its illustrations are good; it will be found a useful addition to the laboratory bookshelf.-Philosophical Magazine