232 ANNUAL RECORD OF SCIENCE AND INDUSTRY. 



was a fluted figure. Ou closer inspection it appeared that 

 the forms were produced by the superposition of two retinal 

 images of the drop in two extreme phases of vibration, In 

 other words, the drop was really vibrating like a bell sound- 

 ing its first harmonic, and had therefore six ventral seg- 

 ments. To prove this, the vibrating drop was illuminated 

 in a dark room, with sparks from a Iloltz machine, and two 

 curvilinear pentagons, alternate with each other, appeared. 

 When the drop had somewhat decreased in size, four-sided 

 curved figures, alternating with each other, were seen. On 

 opening the shutter, there appeared in the capsule an almost 

 perfectly steady beaded octagon, formed by the union of the 

 two crosses. 



2. Conductivity and Radiation. 



Lodge has described a simple form of apparatus for de- 

 termining the conductivity for heat of rare substances, such 

 as crystals, which cannot be obtained in slabs or rods. It 

 consists of two small tin cans, with a copper arm about 

 eight inches long projecting horizontally from each, the ex- 

 ternal ends being clean and flat. These arms are placed in a 

 straight line, with the crystal between them, and held to- 

 gether by a slight horizontal pressure. Holes are drilled in 

 the copper rods for thermometers, and the curves of tempera- 

 ture being given by these, that for the crystal enclosed be- 

 tween the bars can be calculated very readily. 



Less has measured the conductivity for heat of seventeen 

 varieties of stone and several kinds of wood, the method em- 

 ployed being in general that of Hopkins, though with some 

 modifications of his own. He finds that density and com- 

 pactness favor conductivity, other things being equal. Crys- 

 talline rocks conduct better than sedimentarv, and fine- 

 grained better than coarse-grained stone. lie tabulates his 

 results by placing Pyrenees marble as 1000. Then Saxon 

 granite follows as 804 ; Carrara marble 7G9, etc., to ordinary 

 clay, 275. He corroborates TyndalPs statement that the 

 conductivity in wood is different, parallel and perpendicular 

 to the fibre, but finds the difference much less than Tyndall 

 gives. Since the ratios of the galvanometer deflections are 

 greater in the better than in the poorer conducting woods, 

 it would appear that the deflections are proportional not to 



