§o Dr. Bed does’s Obfervations on the 
been traced in our times. It may be proper to premife, that 
under the term bafaltes I comprehend that vaft natural family 
of rocks which is frequently cracked into regular colonnades, 
and may be followed in an unbroken feries from this perfect 
form through endlefs modifications to the mod fhapelefs mafs 
of trapp or whinftone. Though frequently of an iron-grey 
colour and uniform texture, this fpecies of done varies greatly 
in both thefe charadters, even in the fame rock. In particular, 
it paftes, by the mod infenfible gradations, both to the por- 
phyries, with which it coincides in appearance, in compofi- 
tion, and doubtlefs alfo in origin, and to the hornjlein of the 
Germans ; a term including petrofilex and feveral forts of clofe 
grained whinftone, of which I have found in England varieties 
with a conchoidal fracture, femi-tranfparent at the edges, and in 
other refpedts fad approaching * to a filiceous nature. Lillediall 
Hill near Shifnal, in Shropftfire, to mention a fingle inftance, 
affords fuch filiceous, befides Jemi-granilic, porphyritic, and 
common whinftone, containing agate. 
But bafaltes, of which a right knowledge is condudting us 
faft to a juft theory of the earth, is not lefs connedted with 
granite ; infomuch that we may trace thefe rocks gradually ap- 
proaching and changing into one another. I have myfelf had an 
opportunity of examining many con netting links in this gra- 
dual fucceflion ; and this opinion, which has fince been con- 
firmed by other confiderations, was firft forced upon me by 
fpecimens in great variety from volcanic and bafaltic countries. 
But as it is a point by far too important to be admitted on the 
mere authority of any mineralogift, I lhall endeavour to fup- 
port it by the teftimony of obfervers, who cannot be fufpetted 
* Dolomieu, Ifles Ponces, p. in. 185. &c. deferibes juft fuch lavas. 
2 
I 
Of 
