accompanying the Electric Spark. 



295 



the sample it was found, however, to be an old one, and to be 

 partially converted into carbonate. It was thus a mixture 

 of oxide and carbonate. This led to the investigation 

 of other mixtures. It was thus proved that the number 

 given by a mixture of two substances was intermediate be- 

 tween those of its constituents. The following were examined, 

 no particular care being taken to mix in definite proportions, 

 but the substances simply ground up together in a mortar : — 



Ar - , Nos. due to each 

 .Mixture. i-. , 

 constituent. 



No. obtained. 



j Calcic oxide 



56 I 

 40 j 

 80 I 

 40 | 

 60 S 

 68 j 

 60 i 

 80 j 



Exp. 1. 



2. 



3. 



48 

 60 

 64 

 68 



50 

 64 

 64 

 64 



48 

 56 

 66 

 68 





Magnesia alba 



[ Calcic carbonate 



f Cobaltic oxide 



1 Chromic oxide 



baltic oxide 



1 Magnesia alba 



9. Effect on Liquids. — In order to test if the effect could 

 be produced on the surfaces of liquids the following experi- 

 ments were made : — In place of a glass plate with the powder 

 scattered over it, there was held under the spark a shallow 



1 (an ordinary porcelain evaporating dish answers very 

 well) containing the liquid. So long as the surface was 

 looked directly down upon, no effect could be observed ; but 

 when regarded obliquely, minute ripples were seen to be pro- 

 duced upon the surface. No attempt was made to measure 

 these, but they were judged to be of about the same size as 

 those of the powders, that is, between the limits of 88 and 40 

 to the inch. The following were the liquids tested in this 

 way : — water, mercury, alcohol, ether, glycerin, and benzol. 



10. Experiments with Wimshnrst Machine. — It has already 

 been stated that the curves obtained by the aid of the machine 

 are much inferior in regularity to those given by the induction- 

 coil ; in fact, without care they cannot be produced at all. 

 But there are some other peculiarities which deserve notice. 

 Thus, while with every coil used and with varying battery- 

 power it was found that magnesia alba gave 80 lines to the 

 inch, when the machine was used the number varied in differ- 

 ent parts of the figure from 56 to 64. There was no particular 

 mode of variation, that is, no crowding together at a special 

 place, &c. This irregularity in the numbers is not, however, 



