374 ACCOUNT OF A GRAFHORETER. 



figures, the end a is mounted with a brafs fcrew, having a 

 hole through it correfponding with the infide of the tube, the 

 leg b c is filled with mercury, and d is hermetically fealed: 

 d c is divided into atmofpheres. 



XI. 



Account of a Graphometer for meafuring the Angles of Cryflah 

 - In a Letter from Mr. Robert Bancks, No. 411, Strand. 



To Mr. NICHOLSON. 

 SIR, 



Grc 3 t advantage \ N£ED not ; nt t t nd { {he learned readers 



ot dittinguilhing r J 



minerals by their ot your Journal, how great the advantage will be, whenever 

 ngurc. ' the lame may be realized, of difiinguifliing fubjects of the 



mineral kingdom by their external appearance. This has 

 long been done, with confiderable precifion, by operative 

 men who have acquired their fkill from continued practice, 

 but without being able to communicate the knowledge they 

 poffefs by any fimple indications, fuch as might be given in 

 writing, or through the medium of the prefs. Neither need 

 I on this occafion point out how much we are indebted to the 

 Cryftallography. labours of Bergman, Rome de l'lile, and above all, Haiiy, 

 for fcientific inveftigations of the forms of cryftals, which 

 at prefent bid fair to afford us criterions of the mod extenfive 

 ufe. My prefent object is to communicate what I hope will 

 be thought one ftep, however fmall, towards facilitating the 

 CarangeatTs admeafurement of their angles. In your firft vol. at page 132, 

 cr*ftabT Cttr t0t y 0Vi ^ ave S' ven an account of the graphometer of Carangeau, 

 which is now confiderably known and efteemed. I have 

 rendered that inftrument fomewhat cheaper, and eafier in the 

 execution, and more correct in its ufe. For the fake of thofe 

 who may not have that volume at hand, I (hall briefly ftate, 

 that the inftrument confiiis of a femi-circle, like that which 

 I am about to defcribe, and a pair of compares or legs having 

 their centre in the centre of the femi-circle, but capable of 

 having their points drawn back, fo as to admit of their ap- 

 plication to any fmall cryftals. The arc of the femi-circle is 



divided 



