Crystallization of Parisite. 539 



elsewhere together with a much greater number of new ones. 

 No explanation is attempted of the prevalence of complex 

 symbols among these forms. The choice of unit is clearly 

 justified by the constant recurrence upon the crystals of the 

 forms having the simplest symbols in the pyramid series ; the 

 infrequent occurrence of correspondingly simple symbols 

 among the rhombohedral forms is curious and appears to be 

 characteristic for at least this occurrence of parisite. The last 

 two columns of Table III, in which the number of faces of each 

 form is given for all crystals, gives better than words the relative 

 abundance of the forms. It may be well to state in explana- 

 tion of that table that where faces of any form are less on a 

 particular crystal than the full theoretical number, the lack 

 may be from two causes : in case of the pyramid series and 

 prism it is generally from the distortion of the crystal reduc- 

 ing the size of some faces so that they are not measurable ; in 

 case of rhombohedrons it is more often due to faces of two or 

 more forms near each other in inclination occurring together 

 as though they composed a single form. 



Table III contains the data upon which the new forms are 

 based, as. well as that which served for the calculation of the 

 axial ratio. For the latter there were used the readings from 

 254 faces of 27 forms on 32 crystals, the value of the element 

 derived from each form being given a weight proportional to the 

 frequency of the readings for that form in making up the 

 final average. The value adopted in this table is stated for 

 comparison with those of other authors, recalculated to this 

 new position and unit as follows : 



Palache c=l'9368 ^> = 1-2912 



DesCloizeaux 1-899 1-266 



Vrba.. 1-9425. 1-295 



Cesaro 1965 1-310 



It will be noted that the new value is in very close agree- 

 ment with that of Vrba. 



Concerning the individual forms there is little to say in 

 addition to the statistical information contained in the various 

 tables. The single scalenohedron observed had two distinct 

 but narrow faces in zone with the forms H and i so that its 

 symbol is well established. The scalenohedron x of Des Cloiz- 

 eaux was not observed and is probably to be regarded as a 

 doubtful form. 



Optical properties of Parisite. — As seen under the micro- 

 scope, very small crystals or fragments are very pale yellow to 

 almost colorless and show a barely perceptible dichroism. For 

 crystals -J mm thick the a> ray is bright yellow, often with a 

 brownish tone ; the e ray is golden yellow. The absorption is 



