On a "Speed Indicator " for Ships' Propellors. 57 



Iodine so far ssems to deviate most from the above relation; 

 but perhaps, owing to the nature of the body in question, the 

 determination of the latent heat by Favre and Silbermann*, 

 as given in this table, need not be considered final. It would 

 appear that experiments have only once been undertaken 

 with a view to determine the latent heat of iodine. In fact 

 even the boiling-point 200° is only approximate. 



Y. On a "Speed Indicator" for Ships' Propellors. By Sir 

 Archibald Campbell of Blythswood and W. T. GooLDENf. 



[Plate IV.J 



IN February of last year, when visiting one of H.M. 

 Dockyards, it was pointed out to us that a very urgent 

 need was felt of some means of indicating with certainty the 

 speed of naval engines, especially in ships carrying twin- 

 screws, where the engines are not only distinct, but separated 

 from each other by iron bulk-heads which prevent any com- 

 munication between the two engine-rooms. The authorities 

 intimated also that it was desired that several indications 

 should be made simultaneously in various parts of the ship, 

 and that the apparatus should be as far as possible automatic, 

 and require no trouble or attention. 



There are at present one or two instruments in existence 

 for this purpose, but their indications are found to be untrust- 

 worthy when at sea ; and we were warned that no instrument 

 in which the force of gravitation played a part would be. 

 likely to receive any attention at the Admiralty. This 

 consideration excludes the use of ball-governors or of any 

 device depending on the weight of a fluid column ; and led 

 us to consider whether the desired conditions could not be 

 fulfilled by a purely mechanical arrangement founded on the 

 ordinary equation involving angular velocity, 



It was at first difficult to see how this principle could be 

 applied to a direct measure of w, as no simple means are 

 known whereby the variation in the velocity v of a point 

 moving in a straight or curved path can be continuously 

 measured. But it was soon apparent that the variable angular 

 velocity co of a body A can be compared with the constant 



* Annates de CJumie et de Physique (3) xxxvii. p. 469. 



t Communicated by the Physical Society. Read May 24, 1884. 



