Lea — Platinochlorides and Platinum Subchloride. 397 



of it because the time of oscillation of n is greater. There 

 will therefore be a magnetic pull in m throughout this motion. 

 Conversely when both pendulums having reached their extreme 

 elongation begin to return to the right, the bob C will run away 

 from the bob n, and hence there will be no magnetic pull in 

 m. Since the bob n is very light as compared with C, the 

 result of these unilateral impulses applied at the proper time 

 at b is to keep up the motion of the pendulum A. 



In practice however the bobs n and G are never inelastic ; 

 neither is it at all desirable to keep up the contact for so long 

 a time. In other words the record made at D would be irreg- 

 ular from interferences and the batteries would soon run down. 

 Therefore the bob n is hung so that it would meet C when the 

 latter is virtually at the end of its elongation, and a stop e pre- 

 vents n from swinging further towards the right. Furthermore 

 the spring g causes n to be hurled off ballistically and the 

 contact is now quite momentary. Indeed n is made flat and 

 in falling against e there is no rebound. I have found it advan- 

 tageous to surround e with an end of rubber hose, and also to 

 surround the ends of the armature b in the same manner to 

 prevent them from sticking to the electromagnet. Again the 

 pendulum B is made to lean slightly against the stop <?, a slide 

 at/' being provided for this purpose. 



The records show that the time of contact does not reach 

 0*1 sec, and this is the result which I wish to communicate. 

 It is desirable to use electromagnets of high resistance to 

 prevent exhausting the battery, though I have run the pendu- 

 lum for days continuously with such means as I possessed. 

 The apparatus can easily be made symmetrical by placing a 

 second electromagnet similar to m, on the other side of b. 

 This would furnish another set of records differing by about 

 half a second from the first set. 



Washington, D. C. 



Art. LVI. — (1) On some new Methods of obtaining Plati- 

 nochlorides. (2) Probable Existence of a Platinum Sub- 

 chloride ; by M. Carey Lea. 



The methods now in use for obtaining potassium platino- 

 chloride are : 1. Heating platinic chloride to 250-300° C. and 

 treating with potassium chloride. 2. Passing sulphurous acid 

 through a boiling solution of platinic chloride and treating 

 with potassium chloride. To these older methods Thomsen 

 has added : 3, treatment of potassium platinichloride with 

 cuprous chloride. 



