X 6 s. Nakamura and K. Honda : 



camera was set up in such a way that its objective directed 

 downward was just above the model. Exposures were made, the 

 durations of which were nearly equal to half the periods of 

 oscillations, and the motions of the poAvder were photographed.- 

 Plates VII, IX &c show pictures obtained in this Avay. The curved 

 lines drawn normal to the lines traced by the powder at the places 

 where the horizontal motions are greatest, are the nodal lines. 

 The positions of the nodes thus determined are not necessarily 

 accurate, and first method is always to be preferred for the exact 

 determination of nodes; but this method has the advantage that 

 every detail of the oscillation is very clearly brought to light. For 

 example, at the loops where the horizontal motions are least, the 

 images of the powder ought to come out as points; but Avhen the 

 motion was made too violent, we found that the powder at loops 

 described small circles, showing that then the theory of small 

 oscillations could no longer be applied. 



4. Seiclies and meteorology. 



Generally speaking, the limnogram is a smooth curve on a 

 calm day and is wavy when the lake is disturbed by a shower 

 or a wind. But it often happens that the curve is made very 

 irregular by small indentations superposed on it. In such cases 

 the weather is generally dead calm and there is no breeze to agitate 

 the water into ripples, so that the surface of the lake is as smooth 

 as a mirror. The double amplitude of such small indentation is 

 generally two or three millimeters and the period varies from tAventy 

 seconds to one minute. Such motions naturally escape our eye ; 

 for even the longest Avave, Avhich Ave can folloAV with our eye, has 

 a period of only a few seconds, and its amplitude is much greater 

 than that of the motion Avhich causes the indentations in the 



