542 Mr. P. E. Shaw on 



the air-temperature was much more steady before than after 

 the " remedy " was applied. 



The best plan, therefore, seems to be to give the chamber 

 containing the contacts time to settle to its own state of 

 temperature-equilibrium alter the covers are on. 



What part, if any, dust really plays, the author has not 

 ascertained, for in all the later work it has been left out of 

 account. 



4. Contacts. 



Platinum was first used for both contacts ; but as there 

 seemed to be a slight permanent yielding of the surfaces 

 even under the very small pressures used, iridio-platinum 

 (20°/ o iridium) was substituted for it and used generally 

 afterwards. The iridio-platinum plate (4 mm. square) used was 

 soldered to the surface of the iron diaphragm of the receiver, 

 care being taken to use a small soldering-iron so as to avoid 

 great heating and consequent buckling of the diaphragm. 



It was found requisite to clean the contacts thoroughly 

 and often, so as to keep them in good working order. When 

 the two surfaces are quite right, the sound heard at " make " 

 and " break " is sharp, and has a clear ring ; but when not in 

 good order a dull or " wheezy " sound is heard, which may in 

 some cases continue over a distance of as much as 10~ 5 cm.; 

 evidently in such a case the contact is only partial over this 

 distance. 



On consideration, it will be seen that this disorder may be 

 due to dust or oxide or other impurity on the surfaces ; but 

 conceivably it might be attributed to a gradual grazing contact 

 made between two projecting parts of the surfaces, so that 

 contact is not complete till the contacts aro pressing one 

 another appreciably. 



Whatever the cause, this state of the contact must be 

 removed. In some cases it suffices (1) to press together and 

 release the surfaces several times in succession; (2) to make 

 and break current several times when the surfaces are in 

 contact; (3) to rub the surfaces (in situ) with some clean tape. 

 But it often happens that all such ready means fail : then 

 the contacts must be dismounted, polished, and cleaned as 

 follows : — 



Place each contact in turn in the lathe, (1) rub with the 

 the finest emery-paper, (2) polish with rough paste, (3) polish 

 with fine paste, (4) rub with caustic potash and wash, (5) rub 

 with hydrochloric acid and wash and dry, (6) rub well with 

 clean tape. A clean surface having very small roughnesses 

 should be the result of this process. 



