But we imagine that a very serious additional defect arose from the use of paper cases. If these be adopted, the diameter of the whole rocket becomes so rapidly enlarged in proportion to that of the composition, that the resistance of the air increases in a great ratio, and thus the flight becomes materially retarded, and the range consequently diminished. Such increase of re sistance is often opposed to the ascent of the smaller signal rockets, by making the diameter of the head larger than that of the case, a practice very properly exploded in the one pound signal ones made for the service or government. It is not improbable that this cause was superadded to all the rest in the case of these very large ones; as the value of the air's resist ance was not at that time thoroughly estimated. Mr. Robins's experiments on this subject, as applied to the purpose of calculating the fall of a shot, had not long been made, and were not sufficiently before the public to have attracted the attention which they merited.
We cannot therefore agree in all the conclusions of Mr. Ellicott on this subject. In his report, read be fore the Royal Society, he says, that " not only is the making of large rockets very expensive, but much more uncertain than those of a lesser size." "It is also evident from the experiments, (those, namely, above mentioned,) that rockets from two inches and a half to three inches and a half in diameter, are suffi cient to answer all the purposes they are intended for," Ste. Scc. They are unquestionably much more expensive and difficult to make when of a large size; nor, as signal rockets, is it necessary that they should be so large, as it is very seldom that it can be required to make or receive signals from distances of fifty miles. But it is far from true, that the greater ranges may be obtained from the smaller sizes if the rockets are well made ; and though these ranges are not re quired, either for signals or for purposes of amuse ment, they are of great importance in using rockets for military service.
This matter has been proved, and the question set at rest by Sir William Congreve's experiments; he having, by the adoption of iron cases, materially re duced the resistance, while he has also made various improvements in the proportions, in the composition, in the sticks, and in all the minor details, the neglect of which before his time had been the common causes of failure. The amplitude of his ranges, and of course the vertical elevation, (were that wanted,) have been thus materially augmented: but we shall have occasion to notice the nature and construction of the iron rock ets, in that part of our article which relates to military fireworks.
When Mr. Ellicott's report was presented to the Royal Society, the heights given for the flight of these rockets was received with incredulity, and were sup posed to have arisen from the imperfect method of measuring the elevations by means of one quadrant only. There is no doubt, however, that if they were not absolutely correct, the general results are suffi ciently true. They have, indeed, been. amply con firmed by computing the vertical elevations of the Con greve rockets from the lengths of the ranges ; a task not difficult ; as from the continued action of the rocket during the time of flight, the path forms a much more regular and manageable curve than that of a shot through the air, where the action of gravity in causing what would be only such a regular deflection as would produce the parabola, is exceedingly modified by the retardation produced by the air, which commences from the first moment that it quits the gun,and acts unequally till the very moment of its descent.
But that no doubts on this subject might remain, we repeated the experiments on the ascents, with sig nal rockets ; using two quadrants at once, that the errors of observation might be diminished, and reject ing all the flights which, materially deviating from the perpendicular, would have given false results and disturbed the general averages. The mean ascents of pound rockets were thus found to be 480 yards, the greatest being about 520 and 530, and the least at 450 and 460. We shall conclude this part of the subject by adding, that the vertical power of ascent, in the Congrcve rockets, as in others fired at an elevation of 45°, may be computed from the theory of the parabola. The flight of a rocket is not indeed exactly in this curve; but the difference of its path from that line is There are various modes of firing rockets when in tended for purposes of amusement or ornament, but we shall first consider the discharging of single ones, whether used with that view, or as intended for signals. So much in the performance of a rocket, whether in the altitude of its ascent, or the verticality of its direc tiun, depends on the care used in firing, that it is im possible to be too particular on this point, as the very best made ones may fail for want of proper atten tion to many minute points.