496 REPORT — 1886. 



The above results are plotted in the form of curves (Plate X.), the 

 pressures between the surfaces being taken as abscissae, the correspond- 

 ing forces transmitted as ordinates. 



From these curves it is clearly seen that by reducing the area of the 

 surface in contact a less pressure is required to transmit a given force ; 

 thus the curves 1, 2, and 3 rise above each other in regular order (3 being 

 the curve corresponding to the chilled iron spherical surfaces) ; and, 

 moreover, that while the two first are straight lines the last rises more 

 rapidly as the pressure increases, and, in fact, gives at last a coefficient of 

 friction 



/i=0-262. 



Finally it may be remarked that, whereas at lower pressures and with 

 larger surfaces in contact the presence of oil upon the latter causes con- 

 siderable variations in frictional resistance, it was found that at the 

 highest pressures used in the last set of experiments oil poured continuously 

 upon the rolling surfaces had no appreciable effect whatever. 



On ImjproveTnents in Electric Safety Lamps. 

 By J. Wilson Swan, M.A. 



[A communication ordered by the General Committee to be printed in extenso 



among the Keports.] 



I SHOWED an Electric Safety Lamp in this Section last year. I have 

 now improved the lamp in various ways, and I propose to describe and 

 exhibit the improvements. 



The objects I had in view in altering the original design were : — 



1st. To reduce the weight as much as possible consistently with the 

 giving of sufficient light. 



2nd. To simplify the construction, with a view to minimising the cost 

 of manufacture and the cost of keeping in order. 



3rd. To make the lamp better able to resist a blow. 



4th. To seal in the liquid, so that the lamp could be held in any 

 position. 



5th. To add a fire-damp indicator. 



The lamps on the table embody these improvements, and are here for 

 the inspection of members. 



The details of construction and the record of experiments are given 

 in the diagram. 



With regard to the first point — namely, the size and weight of the 

 apparatus — that, of course, depends, in a great degree, on the amount 

 and duration of the light. Guided by the fact that the best of the oil 

 safety lamps scarcely gives the light of one standard candle, and that 

 most of those in common use give only a quarter or half a candle, I have 

 assumed that an average light of one candle, over 12 hours for datal men, 

 and 1:^ candle over 9 hours for hewers, is sufficient, and I have fixed the 

 size of the battery in the new lamps on this assumption. No alteration 

 of the design is necessary when more or less light than this is required, 

 but merely an increase or decrease of the size and weight of the battery 

 contained within the lamp body. 



