( 107 ) 



Tci illuftrate this, we muft underftand, that if a grain of fand were 

 divided into parts (fuppofing all to be of a fpherical fliape) each of wiiich 

 was 'aT4?sth part of the diameter of that grain, each of fucli parts would be 

 capable of entering the cavities of the fmall veffels he is treating of: though, 

 at the fame time, the rules of arithmetic teach us, that the folid contents 

 of that grain would be to the folid contents of one of thofe component 

 parts, as 18,392,744,000 to 1. Again, he tell us, that a globule of blood 

 (of which a million are no more than equal to a grain of fand) muft be 

 divided into 64 parts before it could be received into fome of the blood-vef- 

 fels of the Brain ; and that a grain of fand would be equal to 64 millions 

 of fuch parts. But by the fame rule we fhall find, that the diameter of 

 each fuch part is no more than ^^o^h part of the diameter of a grain of fand. 

 After all, it muft be confefted, that thefe reduced fubdivifions are not to be 

 comprehended in idea, though not fo inconceivably minute as the other 

 proportions feem to intimate. — See a note on this fubjeft, taken from the 

 Spectator, in the IntroduQion to the Firft Volume. 





O 2 



