AltSENlATED C0PI*ER4 237 



The fpeclfic gravity of the fourth fpecies is 42S0, and is 

 confequently perfedly analogous to that ot tlie arfeniated cop- 

 per of the third fpecies, but at the fame time greatly infejiof 

 to that of the firfl, which is ^881. 



Its harrlnefs, much below that of the third fpecies uhich we 

 have jiifl feen that it refembles in weight, is alfolefs than that 

 of the fli ft fpecies, by which it is cut. 



Its forms, vVhich are greatly multiplied, vvhile there exifts 

 but one iti the firft fpecies, differ effcntialiy from thofe of the 

 firft. 



All thefe forms appear to me to be derived from the right Reafons ht 

 tetrahedral prifm with equilateral triangles for bafes; and al I f°™^^S a fif * 

 thofe which I have endeavoured to recognize and have given 

 in my memoir, feemed to me to be very readily derived from 

 thatj as I have alfo faid : thefe cryftals are always extremely 

 fmall, and I was unable to meafure them; a few groups of 

 them exhibited this prifm in fuch a manner that I could ob- 

 ferve it perfe(!lly complete. The fcarceft cryfials, after thofe 

 belonging to the Varieties whicii I have reprefented at figures* 

 15, 16 and 17, and which are ufualiy fo grouped as to pene- 

 trate each other, and thus to become very ditficult to be known, 

 are the very acute complete rhomboid and its incomplete va- 

 rieties, fuch as are reprefented at figures 22, 23 and 21- of m/ 

 memoir. I even hefitated, the divifion (clivage) not having 

 fhown any thing to guide me, vvhether I (hould not take this 

 rhonibf)id for the primitive form. In this inftance, the only 

 manner which feemed to me natural and fimple of connefting 

 this cryftal Wilh the obtufe oflahedron of ihe third fpecies, is 

 to fuppofe this octahedron become rhoraboidal by an increafe 

 having taken place by the fuperpofition of laminae, or by the 

 colledion of rows of moleculae, growing fmaller, on only one 

 of the faces of each pyramid, and in an oppofite direction oa 

 each of them, as occurs in the fpinel, and a number ot fub-* 

 fiances having a right odahedron for a primitive cr)ftal, and 

 which I have alfo fcen in the diamond. Bui \u this cafe, eithef 

 the increafe muft be made Gn the moft inchned faces, and then 

 calculation fliows that the planes of the rhomboid Ihould have 

 37« 39'' and 122° 21'' for the meafure of the angles ol lis plane; 

 or the fame increafe muft be made on the leaft inclined faces, and 

 then the meafures of the angles of the plane of the rhomboid 



* Phil. Tranf. 

 Vol. VIII.— August, 1804., S lliould 



