162 Scientific Proceedings, Royal Dublin Society. 



for automatically shunting the current from the electro-magnets 

 e, e to the electro-magnets e', e' on contact being made with the 

 mercury, or on complete descent of the wire. In any case, as 

 described, the number of wires making up the cable going to the 

 observatory would not, for n instruments, exceed 2« + 1. 



Any number of instruments, however, may, I venture to think, 

 be worked with perfect reliability and certainty by the use of three 

 wires only. Plate lY. will show how this is accomplished, the only 

 extra piece of apparatus needed being a disc,^ carrying insulated 

 contact pieces arranged round its circumference. To these the 

 wires, &c., of the different instruments are attached. Of the three 

 needful wires, then, i. serves to put one after the other of the con- 

 tacts into circuit with the home station through wire ii. By this 

 second wire the readings are taken and the readjustment of the 

 instruments effected. The arrangement by which the current 

 through I. puts the wire ii. in communication successively with 

 the several contacts will be understood from the figure and from 

 the foregoing description of a similar mechanical contrivance in 

 the case of the thermometer. Wire iii. for indication of the 

 contacts, is taken from all the instruments to the galvanometer in 

 the home station. 



One of the wires in the observatory is shown on the diagram as 

 a "test wire." When the movable hand arrives at the contact 

 piece attached to this wire, the current on i. returns through ii, 

 to the home station. If, therefore, at any time a check on the 

 position of the hand of the distributor in reference to the indica- 

 tions of the commutator at the home station is desired, it is only 

 necessary to work the commutator, ii. being put in connexion with 

 the galvanometer. 



The operations involved in taking a reading are obvious. The 

 commutator is first, worked till the required instrument is in 

 circuit — that is, the hand of the commutator is put over the first 

 contact devoted to that instrument. The rheotome on wire ii. is 

 now brought into action. In starting it, a deflection of the galva- 

 nometer on III, shows that the contacts are being truly reckoned 

 from the zero point of the distant instrument. The hand of the 

 rheotome is now steadily turned till a second deflection of the 



~ Plate IV. is diagrammatic only, 



