TECHNOLOGY. 589 



may be diffused by suitably arranged reflectors and globes, 

 would be utterly unsuited for the wants of the private dwell- 

 ing. For this purpose there is demanded not a single light 

 of a thousand candles, but rather fifty lights of twenty can- 

 dles; and not only this, but the luminosity of the fifty lights 

 must be as controllable as our jjaslisfht is, so that one or all of 

 them may be used as may be required ; and each of the fifty 

 must be so manageable as to be competent to yield such a 

 proportion of its light up to its maximum of twenty candles 

 as may happen to be needed just as a gaslight may be turned 

 on or oft". And all this must be accomplished without loss 

 of current, and without' additional expense. The distribution 

 of the current, and its indefinite ramification, from central 

 stations must be controlled with little comparative loss, as 

 gas is now conveyed in the mains and distributing pipes; 

 and the preventives against interruptions of communication 

 must be practically perfect in operation; and the mechanism 

 of the lamps, or burners, must be simple, and not liable to 

 become disordered by reason of ordinary usage. These are 

 the salient requirements of a practical household light, which 

 any substitute for gaslight must possess; and though they 

 are by no means impossible to realize in the electric light, it 

 will be obvious that inventors have still much to do before 

 the problem is successfully solved. In recording the prog- 

 ress made during the past year in this field of research, the 

 name of Jablochkoff*, with his electric candle, is perhaps most 

 prominent. In America, several notable contributions to the 

 subject have been made. The most novel, and perhaps the 

 most suggestive, is the plan proposed by Messrs. Houston and 

 Thomson, of Philadelphia, to employ the "extra spark," a 

 phenomenon which appears whenever an electrical current, 

 which flows through a conductor of considerable length, is 

 suddenly broken. As this "extra spark" will appear, al- 

 though the current is not sufficient to sustain an arc of any 

 appreciable length at the point of separation, the system 

 which these investigators propose, permits the use of feebler 

 currents for producing an electric light than that ordinarily 

 required. In this plan one or both the electrodes (of carbon 

 or other suitable material) are caused to vibrate to and fro 

 from each other, at such a distance apart that in their mo- 

 tion towards each other they touch, and then recede to a 



