TEANSACTIONS OF SECTION A. 565 



5n circuit. As tliis resistance diminishes the temperature of the strip rises, so that 

 the operator may confine Lis attention to the observation of the bodies on the 

 platinum hob, while they are being exposed to a temperature gradually increasing 

 up to that of the melting-point of platinum. Quartz is melted by this apparatus, 

 and orthoclase caused to boil. Many substances are decomposed displaying 

 characteristic phenomena. It reveals that the order of melting-points set forth 

 in von Kobell's scale is incorrect. 



Another form of the apparatus, not yet completed, will permit, it is hoped, of 

 ■nn approximate estimation of the temperature of the platinum hob at any moment. 



8. Eledro-calorimetry. Bij Professor William Stroud, B.A., B.Sc, 

 and W. W. Haldane Gee, B.Sc. 



During the year experiments have been continued both at Owens College, Man- 

 chester, and Yorkshire College, Leeds, with the null method described at the last 

 British Association meeting. Important improvements have been effected in the 

 details of the method and apparatus used. The calorimeters are now made in the 

 form of barrels closed by screw plugs which effectually prevent leakage and loss by 

 «vaporation. For agitation of the contents of the calorimeters paddles and plungers 

 worked by electro-motor or clockwork have been entirely discarded in favour of 

 the simple method of rocking or shaking by hand the calorimeters themselves. For 

 the heating coils platinoid wire has been found the best, each coil being formed of 

 a number of strands of fine wire well insulated by silk and paraffin. Experiments 

 made with the bolometer have shown that, compared with the thermo-pile, it is 

 greatly inferior to the latter for the purpose of ascertaining the equality of tempera- 

 ture of the calorimeters. A number of determinations have been made which con- 

 firm the accuracy of the method. 



9. On Figures produced by Electric Action on PJwtographic Dry Plates.^ 



By J. Brown. 



A rapid bromide photographic dry plate is laid film side up on a sheet of metal 

 in connection with one terminal of a good induction coil. A wire from the other 

 terminal is brought down on the centre of the film. A single discharge from the 

 coil is given, and the plate developed in the usual way. 



If tbe wire touching the film be the negative pole, a beautiful figure with 

 branches carrying palm-like fronds is produced ; if positive a number of irregular 

 branches of a quite different character accompanied by a quantity of light radiating 

 irregular tracery extending beyond the darker branches. 



If the discharge take place with the two poles of the coil touching the uncoated 

 side of the plate, other figures are produced by the electricity induced on the film, 

 the positive and negative differing both from each other, and from those given as 

 •described above by the discharge on the film. 



If a coating of tinfoil be placed on each side of the plate, after the manner of a 

 Franklin's pane, that on the film having letters cut out like a stencil plate, the result 

 after electrification of the foils and development is a dark irregular marking round 

 the edge of the stencil plate foil on the film, including the edges of the cut-out 

 parts, apparently produced by a discharge from the edge, together with irregular 

 Wotchings, apparently corresponding to wrinkles in the foil. If a piece of gutta- 

 percha tissue be placed between the foil and the film, the effect is much the same. 

 If four thicknesses intervene there is only irregular blotching without any appear- 

 ance of the outline of the letters cut in the foil. 



These last effects, together with some peculiarities of the figures produced by 

 the discharge on the film itself, appear to show that the figures are produced, 

 partly at least, by a direct action of the electricity on the film without the inter- 

 vention of light or purely photo-chemical causes as usually understood. Further 



' Phil. Mag. Dec. 1888. 



