EXPERIMENTS ON THE VIBRATIONS OF PLATES. 371 



The figures will not always be produced at the first attempt. 

 First see whether the plate is fixed precisely in the middle or at the 

 desired point ; then vary the inclination of the bow and the 

 pressure applied to it, until you succeed. The sand should be spar- 

 ingly thrown upon the plate with a castor ; if there is too much, its 

 weight impedes the vibrations. If, instead of sand, powdered blue 

 (cobalt) glass (prepared by heating a piece of blue glass to redness 

 and throwing it while hot into cold water, after which it can be 

 pounded in a mortar) be used, the figures may be preserved. For 

 this end prepare a gummed sheet of paper, carefully remove the 

 plate from the clamp without disturbing the figure, place it on the 

 table, cover it with the wet gummed side of the paper, lift the paper 

 and let it dry. The sand figure will adhere and remain upon the 

 paper. 



If circular plates are made to vibrate from the centre instead of from 



the edges, the figures do not solely consist of various arrangements 



}f diameters. Attach to the middle of the glass plate a small piece 



of sealing-wax, about l cc in size, and fix a glass tube perpendicular to 



the plate by means of the wax; the tube should be from 30 to 80 cm 



long, and from 5 to 10 mm thick. Holding the tube with the thumb 



ind forefinger of the left hand somewhere below its middle and 



aearer to the glass plate, and rubbing it with the moistened fingers 



}f the right hand, a musical note is produced ; the exact point at 



,vhich the tube must be held in order to produce the note most 



easily should be found by repeated trial. The first three fingers of 



;he right hand, which are from time to time to be dipped in water, are 



ised for producing the note. They are loosely placed round the 



yube, somewhat in the manner in which a pen is held whilst writing 



vith it, and drawn either from the middle of the tube, which is held 



[iiite vertical, upwards to the free end, if the plate is below, or from 



he free end upwards to the middle, if the plate is above. In the 



,atter position we have the advantage that no water can drop upon 



he plate ; in the former the apparatus is used more conveniently. 



v.fter ascertaining the precise point for holding the tube in order 



10 make it sound, the plate is wiped in case it should have been. 



vetted, sand is thrown over it, and the apparatus again made to 



ound. The figures thus obtained vary with the relative propor- 



ions, as to size and thickness, of the two glass parts of the contri- 



ance. A definite combination of tube and plate gives only one 



efinite figure when sounded. Fig. 221 shows two such figures, 



T hich are respectively produced by means of a disc 10 cm in clia- 



icter, and of two tubes, one 55 cm , the other 25 cm long. 



Bells may be considered as circular concave plates. 



B r, 2 



