462 Scientifie Proceedings, Royal Dublin Society. 



It has a resistance of under an ohm per cm., and is fused by a current of 

 about 11 amperes. 



It is occasionally more convenient to use platinum.' A spool of ribbon 

 2 metres in length and 2 millimetres in breadtli (No. 0-440, Johnson and 

 Matthey) can be obtained for 35s., and the short length of such ribbon 

 required in a sublimation chamber of the foregoing dimensions is in- 

 expensive, and, in the absence of substances corroding platinum, may be 

 used repeatedly. If a ribbon be used, it must be twisted into a horizontal 

 position after clamping between the forceps. 



A convenient resistance is necessary to ensure suitable temperature control 

 of the strip. If desired, an approximation to the temperature of the strip could 

 be obtained by observation of tlie heated carbon through different thicknesses 

 of neutral tinted glass. When an opaque cover plate is used, the deposition 

 of the sublimate may be watched in a mirror placed underneath the rod. 



Many of the effects obtained by tliis procedure are also produced in 

 ordinary blowpipe practice, but it is possible by this method to obtain many 

 effects which are not possible with the blowpipe, and, owing to tlie ease and 

 rapidity with which these tests may be executed, the electrical metliod is 

 more trustworthy, and at the same time more comprehensive than the older 

 blowpipe process. 



A small heap of the powdered substance is placed upon the flattened 

 surface of the carbon support, with the upper cover-plate in position. A slow 

 stream of dry gas from an automatic generator is admitted if it is desired 

 to examine the substance in an atmosphere other than air. Although not 

 always necessary, a continuous current of gas through the sublimation 

 chamber is sometimes advantageous for the purpose of transporting heavy 

 sublimates, and for preventing the formation by diffusion of explosive 

 mixtures. An explosion, however, in a sublimation chamber of the above 

 dimensions is not serious unless the upper cover-plate is fastened down. 



The temperature of the strip is now raised at any desired rate, tlie substance 

 being observed meanwhile througli tinted glass. If a mixture or alloy is 

 being examined, it may be necessary to raise the temperature slowly, and to 

 stop at any certain point to effect a partial separation of the constituents, 

 leaving a certain residue for further examination in a second experiment. 

 The deposit on the cover-plate may be subjected to furtlier wet or dry tests. 

 With high temperatures, and when it is desired to subsequently heat the 

 plate, transparent or opaque silica is the best substance ; and where colours in 

 transmitted light are to be examined, glass is preferrable. For showing mere 

 traces of colour, white biscuit ware (which confers great richness upon the 

 coloured deposits) should be used. 



' Joly, Proo. Roy. Irish Acad., 3id ser., vol. ii, 1891. 



