PHYSICS. 479 



tact, though iuisuhited from each other, as a continuous flat hiycr about 

 16 inches long, soldered end to end in an alternating series. About 1^ 

 inch in length at each end is bent downward so as to dip into two 

 liquids of different temperatures contained in two long, narrow troughs. 

 The hot one is melted paraffin , kept at 120°; the cold one is a non- vola- 

 tile petroleum. The resistance of 295 pairs of such wires was 95.G ohms 

 at 10° 0.5 and when the difference of temperature in the two baths was 

 100° the electro- motive force was 0.7729 volt, and when it was loO©, 

 1.005 volts. Each element gave, therefore, 0.0000262 volt lor 1°. With 

 this apparatus measurements have been made to the 35-5-00 P^rt of a 

 volt. The author finds it very useful in practice. {Nature, March. 

 1884, XXIX, 513.) 



Von Waltenhofen has observed that the ISToe thermo-battery is capa- 

 ble of taking a charge like an accumulator, When the current of a 

 battery is sent for a few^ moments through one of these generators, it is 

 capable of yielding a counter-current, due, of course, to the well-known 

 Peltier effect. A different effect, however, was observed according to 

 the direction of the charging current. When this current was in one 

 direction, the discharge currents were proportional to these and were 

 always in. the opposite direction. But when the charging current was 

 reversed, the discharge current at first increased to a maximum, then 

 decreased to zero, then began in the reverse direction, or that of the 

 charging current. The author ascribes this anomalous result to the 

 unequal heating of the junctions. {Nature, January, 1884, xxix, 227.) 



Two important papers have appeared having for their object the dis- 

 cussion of the theory of the dynamo-electric machine. ' The first of these 

 is by Clausius ( TT/a?. Ann., xx, 353; Pliil. Mag., January, February, 

 1884, V, XVII, 46, 119; J. Phys., July, 1884, II, in, 313), and the other, 

 on machines with alternating currents, by Lucas ((7. B., March, 1884, 

 xcviii, 670), who used the Meritens machine in his experiments. 



Fitzgerald has sought to produce a non-sparking dynamo by apply- 

 ing the principles of Maxwell's modification of Thomson's electric doub- 

 ler to a dynamo, in which the current passes through two or more coils 

 in parallel circuit. Under these circumstances it is possible to arrange 

 the magnetic field and the brushes so that when the terminals of any coil 

 come into contact with their brushes the terminals shall be at the same 

 difference of potential as the brushes, so that when they break contact 

 there shall be no current circulating in the coil, and therefore no sparks 

 produced. The energy of self-induction, usually wasted on local cur- 

 rents and sparks will now be spent in the i)roduction of useful current: 

 {Nature, July, 1884, xxx, 331.) 



3. Electrical Units and Measurements. 



The International Electrical Conference held in Paris in April con- 

 cluded the business before it and adjourned finally. Each section 

 adoi)ted resolutions as follows : First section, electric units : (1) The 



