THEORIES OF CRYSTALLOGRAPHY. 87 



fei by Him whom the ereateft of pasan philofophers calls the ^='^s of the 

 Eternal Geometrician? This recalls reflections to my mmdcryHais, and 

 which I cannot fupprels. Converfing one day with the Abb6 theory thence 

 Haiiy, he was taking a curfory view of all the modern difco-^^ " ^* 

 veries ; when he could not help remarking, that there was 

 not one of them but what furniftied victorious arms to the 

 caufe of religion. My anfwer was, that in future the nam« 

 of God would be as diftindlly written on a cryftal as it had 

 hitherto been in the heavens. The obfervation of this moft 

 religious and ingenious man reminds me of the faying of lord 

 Bacon : '* A little philofophy eftranges us from religion, but 

 a great deal reclaims us again." Even d'Alembert could not 

 help faying, ** An atheift in the Cartefian fyftem is a philo- 

 fopher miftaken in the principles; but an atheift in the New- 

 tonian fyftem is fomething worfe, an inconfequent philofo* 

 pher." 



But to return to the mathematical part of our author's the- 

 ory : the branch of mathematics, and the manner in which he 

 treats it, are almoft new. The theory of polyedrons had been 

 nearly negleded by geometers, both on account of the diffi- 

 culty to reprefent a polye<Iron on a plane, and becaufe they 

 did not feel the utility of the purfuit. Neverthelefs, ftrange 

 to fay, all the regular figures that are to be found in one of 

 the three kingdoms of nature are polyedrons. In this point of 

 view, the branch of mathematics illuftrated by the Abb6 be^ 

 comes very interefting ; and it is not a little fo, to fee with 

 what ingenuity he extricates himfelf from the difficulties 

 he meets with in his refearches. He forms all the polyedrons, 

 however complicated, of little equal rhomboids or parallelo- 

 pipedons, and by that means he reduces the theories of every 

 poffible polyedron to that of the rhomboid, which is extremely 

 fimplified by two very limple remarks : 1ft, That in all equi» . , 

 lateral rhomboids, whatever may be the fpecies, their projec-^ , . 

 tlon on a plane perpendicular to their axes will always be a 

 regular hexagon : 2dly, That the axes will always be trifeded . 

 by perpendiculars drawn from all the lateral folid angles. 

 His theory has alfo led him to difcover in a variety of cryftals 

 geometrical properties, which muft be highly gratifying to . 

 geometers. But the great advantage to be derived from it is, 

 that it enables us with the feweft poffible data to calculate 

 the cryftalline forms juft as aftronomers do the motions of the • 



heavens. 



