84: S. L. Penfield — Connellite from Cornwall. 



measured with the following results : 49° 26', 49° 50' and 49° 

 42'. The reflections were good, but of course faint, owing to 

 the small size of the faces. Maskelyne gives for the same angle 

 47° 10'. The zone 2110, 0111 and 2110 was also adjusted on 

 the goniometer in which the pyramid was found to be at right 

 angles to the prism of the second order and finally one pris- 

 matic zone was measured in which the angles approached very 

 closely to 60°. Using the mean angle of p a. p (1011 /s 0111) 

 = 49° 39', the length of the vertical axis 6 = 1-3392 can be 

 calculated. This is probably obtained from more exact meas- 

 urements than those of Maskelyne. Among the many frag- 

 mentary crystals examined under the microscope, the majority 

 had the simple habit of prism and pyramid of the opposite order, 

 a few only showed the combination of prism and pyramid of 

 the same order, while still others were slender and tapering, in 

 habit like an Alston Moor aragonite crystal, but usually termi- 

 nated at the very end by the ordinary pyramid. No crystals 

 were observed like those described by Maskelyne, showing 

 combinations of the two prisms or a dihexagonal pyramid. 



Hardness about 3. Specific gravity as stated above, 3*364. 

 The crystals are transparent and of a beautiful dark blue color, 

 the fine powder is a pale greenish blue. Crystals show 

 under the polarizing microscope parallel extinction and strong 

 positive double refraction, determined on thin prismatic crys- 

 tals by means of the quartz wedge : they exhibit no percepti- 

 ble pleochroism, which agrees with the statement of Maskelyne. 

 No distinct cleavage was detected. 



The chemical analysis was made with great care on 0*0740 

 gram. This seems a very small amount for the purpose, but I 

 was prepared to take advantage of the experience which had 

 been gained in the analysis of spangolite, and as no rare or un- 

 usual constituents were met with, and as the analysis went on 

 without any mishap I am prepared to place great confidence 

 in the results, which are as follows : 









Ratio. 





Spangolite, 



so 3 





4-9 



•061 



1-00 



10-11 



CI 





7-4 



■208 



3-42 • 



4-11 



CuO 





72-3 



•911 



15-00 



59-51 



H 2 





16-8 



•931 



15-34 



20-41 



Loss at 



100° C. 



■4 

 101-8 





Al : 



■ 3 6-60 





100-74 



O equivalent 



to CI 



1-7 





O eq. to CI 



•72 



100-1 99-82 



The ratio here is not very satisfactory except between SO s 

 and CuO, with H 2 slightly in excess. If, however, we assume 

 that some OH is isomorphous with the CI and calculate enough 



