For the definitions of the derived dynamical units and the investigation of their dependence on the fundamental units,
the reader may refer to the article on Mechanics.From what has been said it will appear that the whole difficulty in fixing upon a system of dynamical units lies iu the determination of the fundamental units of length, mass, and time in such a manner that their constancy can be relied upon. The unit of mass offers very little difficulty in this respect. Long experience has taught us which are the most permanent of the varieties of matter we have at com mand. We have good reason to believe that a piece of platinum or an alloy of platinum and iridium may be exposed to pure air at ordinary temperatures for an in definite time without any increase or diminution of its mass whatever. Such a piece of metal may therefore with propriety be chosen as a national standard of mass, the absolute constancy of the quantity of matter constituting it being accepted on definition, as we are unable to test it by any standard in which we have more confidence than we have in itself. The British and French national standards of mass are of platinum, but the new standards recently constructed in Paris consist of an alloy of platinum and iridium.
The determination of a unit of length is not so simple as that of the uuit of mass. In this case, as in the preceding, we avail ourselves of the properties of a material standard, but we know that however indestructible the standard itself may be its dimensions depend upon the pressure to which it is exposed, its temperature, and in some cases upon other accidents, such as the magnetic force in the neighbourhood, ifec. Hence the only course open to us is to determine as far as possible all the causes of variation in the length of our standard, and carefully to define its condition with respect to these variables, so that it shall be a standard only under the circumstances thus defined. Having thus defined the condition of the material standard with respect to all the variables upon which we know its length to depend, we must consider the absolute constancy of its length at all times and places to be a matter of definition until we have discovered other causes which affect it. It has been pro posed that the wave length in vacuo of a particular kind of light, as for instance that corresponding to one of the sodium lines, should be taken as the unit of length, and its period as the unit of time. These units are probably more constant than those afforded by any material standards or vibrating springs which we can construct ; but a belief in their absolute constancy implies complete confidence in the constancy of the properties of the interstellar medium, and of the sodium molecule.
The determination of a satisfactory means of measuring time seems to offer greater difficulties than the measure ment of mass or of space, though the difficulties are of the same character as those we have just considered. The great difficulty" consists in defining what is meant by the equality of two intervals of time which do not commence simultaneously. Remembering that it is upon the proper ties of matter alone that we can rely for assistance, we might construct a spring and define as equal lapses of time those intervals during which this spring executes the same number of vibrations, the temperature, <fec., being kept constant. But if we were to construct a number of such springs, though a perfect agreement might obtain between them at first, we should find after a considerable period that the measurements of time derived from different springs did not agree, while our knowledge is in sufficient to enable us to apply to each the corrections necessary to lead us to a consistent result. Now there may be no reason why we should prefer one spring above all the others, and thus it appears that a definition of equal intervals of time based upon the behaviour of any single spring is too arbitrary to be satisfactory. If, however, we found a large number of springs, constructed of different materials and differently affected by temperature and other known causes of variation, continue to give perfectly consistent results, the theory of probability would lead us to place a high value upon the measure of time thus afforded. Now, we have stated that our highest conception of the measurement of time is derived from the dynamical principle expressed in the first law of motion, but when we come to apply this it is impossible to determine a priori whether in the case of two given bodies there is no stress acting between them or between one of them and some third object. Consequently, the only course open to us is to examine the motion of a large number of material systems, making such corrections for the action of stresses which we know to be in operation as our theoretical dynamics will enable us to determine ; and, if after this we find that several independent systems afford the same measurement of time, while those systems which lead to a different result disagree also among themselves, we must accept the measurement of time afforded by the first set as the true measure, and attribute the discrepancies manifested by the other systems to some unknown stresses, which it should be our subsequent business to discover.
Work.—The contemplation of a stress, together with a relative displacement of the portions of matter between which it acts, introduces us to the conception of worlt. If we con sider a stress, together with the distance through which the solicited bodies are capable of moving relative to one another in obedience to the stress, the object of our contem plation is the work which may be done under the given conditions of the system, and this we call energy. The sub ject of which natural philosophy treats is the transformation of energy, which in all its phases takes place in accordance with two great principles, known respectively as the principles of the conservation and the dissipation of energy. Of these two principles the former rests upon a much higher scientific basis than the latter. In order to lose our faith in the principle of the conservation of energy we must give up our belief in the fundamental principles of dynamics expressed in the laws of motion ; but as regards the dissi pation of energy we can say little more than that all the operations of nature with which we are acquainted take place in accordance with this principle. Clerk Maxwell has, however, shown that it is possible to subvert the prin ciple of the dissipation of energy by the simple exercise of a sufficiently high order of intelligence. For the statement and discussion of these two principles see Energy.
It is the work of the natural philosopher to explain the operations of nature in accordance with the principles of dynamics, and we consider that we understand any pheno menon when we have shown it to consist of a motion of matter and determined the character of this motion. Thus it is that dynamics forms the foundation of every branch of natural philosophy, and a thorough appreciation of the prin ciples of conservation and dissipation of energy is the only safe guide in physical investigations.
various explosive preparations containing nitroglycerin. The first practical application of nitroglycerin, discovered by Sobrero in 1847, was made by Alfred Nobel, who in 1863 used gunpowder soaked with it for blasting. In 1864 he found that it could be exploded by the initiative detonation of fulminating materials ; and in 1867, owing to the uncertainty and danger attending its employment, he conceived the idea of mixing it with some solid and absorbent inert substance. The siliceous infusorial earth called in Germany Kieselguhr proved to be well adapted for this pur pose, since it took up as much as three times its weight of
nitroglycerin without becoming more than damp to the