Some of the finest figures that are obtained by the effect of distant vibrations on the membrane are repre sented in Plate CCCCLXXXIII. Fig 24-36. When the membrane is ill stretched, it often happens that the lines traced by the sand are very numerous, and that they form kinds of chains, regularly arranged, and apparently the result of concentric lines cut by a great number of diametral lines. See Fig. 37.
From these experiments it follows, that, when the plate and the membrane are parallel, the motion i3 communicated by the air exactly as it would have been if the two bodies had been separated by a common rod perpendicular to their faces; for the number of vibra tions is the same in both cases: since, for each sound produced, the membrane affects a particular mode of division, and the direction of its motion is also the same, since it is perpendicular in the plate and in the membrane. If the vibrating circular plate is held with one of its diameters in a vertical line, the grains of sand have then a tangential motion, and the system of lines in repose have in general the character of paral lelism. By gradually inclining the plate, the figures on the membrane change.
When figures composed of concentric circular lines are obtained, there is often formed between two of these a circular line, composed of the finer particles of the sand. M. Savart is of opinion that this line be longs to a kind of vibration higher than that which is produced, but which co-exists along with the principal vibration. It sometimes happens, also, that the cen tre of the membrane presents an immoveable point, which probably belongs likewise to a higher mode of vibration, so that the membranes appear to produce with facility several kinds of motion at once.
The preceding experiments may be varied in a great number of ways, by making use of membranes whose dimensions, nature. tension, and contour, are differ ent; but they all present analogous results. Thcfigures produced by a rectangular* membrane are shown in Plate CCCCLXXXIII. Fig. 38-45, and those pro duced by a triangular one in Fig. 1. When the diameter of the membranes is less than from half an inch to an inch, it is not easy to observe regular nodal lines, unless when the sound is extremely acute.
The figures which have now been described vary with the tension of the membrane. In those made of
paper, which changes its hygrometric state, and con sequently its tension, continually, Al. Savart observed that the figures changed at every instant. When the same figure is represented several times, it was neces sary only to breathe upon the paper to create a new one, which in a short time disappeared, and returned to its former state through a great number of interme diate figures. Hence M. Savart proposes this as a sure method of detecting small hygrometrical variations in the air. In order to protect the paper membranes from the humidity of the air, they should be covered with a thin coat of varnish made of gum lac.
The membranous vibrations and figures which have now been described may also be produced by the sound of the pipe of an organ, even at the distance of some feet. If we play with a slow motion an air on the flute, at about half it foot from the membrane, the sand will form lines, the figure of which varies unceas ingly with the sound produced. But, what appears more astonishing, the voice produces an analogous ef fect, which is extremely well mat ked, even under the influence of a sound which is neither strong nor sus tained. By whatever method, in short, the air is agi tated, it is capable of communicating to thin mem branes the motion which it has received, and that without any alteration.
These experiments succeed also equally well when the membranes are wetted, or when they have imbibed an oily substance. In this last case, in place of sand, we must cover the membrane with a thin stratum of oil, which is agitated in ripples, that increase in num ber with the acuteness of the sound.
M. Savart next applies these principles to a method Of appreciating very steall quantities of sound. lie stretches a piece of thin vegetable paper or goldbeat er's skin across the mouth of a glass about four inches in diameter. Ile then covers this with sand, and as certains the intensity of different sounds by the dis tance at which they cease to agitate the membrane; and he remarks that the membrane will often be moved by an augmentation of sound which the ear itself is incapable of appreciating. He proposes also to use it for ascertaining the augmentations of sound which arise from the coincidence of vibrations produced by numbers of Nibrations not very distant from each other.