456 



SCIENCE. 



[N. S. Vol. IX. No. 221. 



in all its force and peculiarities. Over and over 

 again this little experiment was performed ■with- 

 out variation in its results, until, finally, satis- 

 fied, I moved my head off' my arm and 

 stretched my arm out of its cramped position, 

 and felt no more this hete noir of earlier days, 

 now again returned, bringing with it emphatic 

 and unmistakable explanation of its cause. 



G. V. D. 



ASTRONOMICAL NOTES. 



A NEW SATELLITE OF SATURN. 



A NEW satellite of the planet Saturn has been 

 discovered by Professor William H. Pickering 

 at the Harvard College Observatory. This 

 satellite is three and a half times as distant 

 from Saturn as lapetus, the outermost satellite 

 hitherto known. The period is about seventeen 

 months, and the magnitude fifteen and a half. 

 The satellite appears upon four plates taken at 

 the Arequipa Station with the Bruce Photo- 

 graphic Telescope. The last discovery among 

 the satellites of Saturn was made half a century 

 ago, in September, 1848, by Professor George 

 P. Bond, at that time Director of the Harvard 

 College Observatory. 



Edward C. Pickering. 



Harvard College Observatory, 

 Cambridge, Mass., March 17, 1899. 



NOTES ON PHYSICS. 



THE NERNST LAMP. 



The electric lamp recently invented by 

 Nernst, as has been stated in this Journal, con- 

 sists of a small rod of magnesia which is heated 

 to brilliant incandescence by an electric current 

 which is pushed through it by an electromotive 

 force of several hundreds of volts. The rod 

 must be heated nearly to a red heat by a blow- 

 pipe or other independent means before it passes 

 sufficient current to operate. 



A number of these lamps have been made in 

 the Physical Laboratory at Bethlehem, Pa. It 

 has been found that a rod of pure magnesia can 

 scarcely be started even with 1,000 volts and a 

 good blow pipe. The surrounding air becomes 

 electrically too weak to withstand the high elec- 

 tromotive force at a temperature lower than 



that required to make the rod a suflBciently good 

 conductor. This is true even when the rod has 

 been heated to softness beforehand in a tem- 

 porary mounting. 



The conductivity of the rod may be com- 

 pletely controlled by mixing with the magnesia 

 varying amounts of silica and of fusible silicates. 

 A satisfactory lamp is made as follows : Pure 

 calcined magnesia (heavy) is thoroughly mixed 

 with two or three per cent, of powdered silica, 

 one or two per cent, of magnesium sulphate, and 

 one per cent, or less of sodium or potassium 

 silicate (water glass). The mixture is dried 

 until it is just moist enough to pack under pres- 

 sure. A small piece of brass tubing is lined 

 with a roll of several thicknesses of stifle writing 

 paper, and the mixture is tamped into this tube. 

 The tube is then baked until the paper is 

 burned, when the rod of magnesia may be re- 

 moved. This rod is then laid upon a bed of 

 magnesia (powdered lime would, perhaps, 

 answer) and by means of carbon terminals an 

 alternating current is passed through the rod, 

 heating it fii'st to i-edness by a blow pipe. With 

 some care a very hard and compact rod of mag- 

 nesia is thus formed which is then ground to a 

 thin rod with large grooved ends. Platinum 

 wire is wound on these grooved ends and, if de- 

 sired, cement made of water glass and powdered 

 magnesia may be used to cover the platinum. 

 The two platinum wire terminals may then be 

 bound to the sides of a small glass tube as a 

 support. A lamp made in this way may be 

 started easily, although its resistance rises slowly 

 with continued use, owing, perhaps, to the vol- 

 atilization of the potassium or sodium silicate. 

 Calcium silicate would, perhaps, be more satis- 

 factory in this respect. 



A very striking experiment may be perfoi-med 

 with a piece of glass tubing several inches long 

 wound with copper terminals at its ends. The 

 tube begins to pass considerable current at a 

 low red heat, with a few hundreds of volts, and 

 is quickly melted by the current. A thin-walled 

 tube half an inch or more in diameter is best, 

 and it should be heated along one side only so 

 that the cool portion of the tube may for a short 

 time serve as a support for the hot conductive 

 portion. 



W. S. F. 



