■676 



Mr. Elis Hjalniar on 



Ou— CI (\<5'3 A.U.) covered with a thin foil of alumi- 

 nium, thickness 7 /x ; in the following researches on S — Na 

 \5'3<\<12 A.U.) with a very thin goldbeaters'' skin, which 

 was coloured red with erythrosin (cone, alcoholic solution of 

 tetraiodfluorescin) in order to keep out the luminiscence 

 light from the anode. This foil, however, may be recom- 

 mended, even for some shorter wave-lengths. 



The form in which the elements were investigated is to 

 be seen from Table I. For Al and Mg respectively two and 

 three different materials were used. No influence of the 

 chemical compound on the wave-lengths or the intensity of 

 the lines could be observed in these cases. 



Table J. 



Element. Material. Way of application. 



Ni ... \ Pure metallic sheet. Soldered on anode. 



Co ...J 



Va 1 



m- > Acid, powder. Eubbed on Ag-anode. 



*-;;; }soiph.te „ 



oi::::::} K01 ' ■• 



Ba ... BaS0 4 ,, „ Cu-anode. 



Ag ... Metallic sheet. Soldered on ,, 



Mo ... Metallic powder. Rubbed on ,, 



S Cu S+Na 2 S, powder. 



P P 2 5 , powder. „ „ 



Si Pure Si-powder. ,, 



,, f A1C1 3 , powder. 



A1 J Pure met. sheet. Soldered on „ 



f MgCl 2 , powder. Rubbed on 

 Mg ...\ MgO, 



[Pure .net. band. Soldered on „ 



Na ... Na. 2 S0 4 , powder. Rubbed on „ 



The distance between the lines is measured on a com- 

 parator. As the lines are curved to each other, the 

 distance is measured as max. or min. distance with an 

 accuracy of about 0*007 mm. The distance between the 

 turning axis and plate is found to be 126*14 mm. The in- 

 evitable experimental errors will cause a total deviation in X 

 less than 0*01 per cent, of X. The value of the lattice con- 

 stant of calcite has been determined in this laboratory* with 

 a very great precision to be 



log 2d-= 3*7823347- 11, 



* M. Siegbalm, lac. cit. 



