Fojtloii, Figure, and Stniaure of the Weljl, Mountains. 221 



along the (hore. it arrives at Caernarvon. From this place it proceeds along the Menai 

 forming the eaftern bank, as far as Bangor Ferry. Hence to Orme's-head it is cut off by the 

 northern extremity of the Snowdon chain, which terminates in the bay of Conway, by the 

 chfFs of Penmaen-mawr and Penmaen-bach. The lime recommencing in the lofty promon- 

 tory of Orme's-head, continues the boundary of the coaft as far as the^mouth of the Dee • it 

 then takes a wefterly direftion, curving to the fouth, as it paffesby Holywell and the upper 

 end of the vale of Clwyd to the Eglwyfo-, rocks in Llangollen vale ; then, paffing due fouth. 

 It appears on the oppofite Hde of the vale ; is broken near Ofweftry by the Ferwyn moun- 

 tams, appears again at Llanymynech, and is at length (lopped in its courfe by a line of 

 pnm.t.ve mountains ftretching northwards out of Radnorfl.ire. The flates occupy the 

 whole intermediate fpace between the ridge of lime and the primitive mountains. 



The primitive, fecondary, and derivative mountains may in general be diftin<.uifhed by 

 peculiarities in their form, as well as by their relative pofition ; the primitive rocks are craggy 

 fteep, and tending more or lefs to a peak, or flender-pointed fummit ; the loftieft mountains 

 are generally about the middle of the chain, which both commences and terminates in abrupt 

 precipices: thefe, together with the infulated peaks that are continually interrupting the 

 outline of the chain, form a very ftriking diftinftive charader. The plates to Mr. Pennant's 

 Sno«'doma will convey a clearer idea of this than the molt laboured defcription. Indeed it 

 is but juftice to obferve, w^ith refpeft to thefe engravings, that they are perfedly accurate 

 reprefentations of the fcenes which they profefs to defcribe. 



The flates are diftlngulQiable from the primitive mount,iins by their inferior height by 

 the evennefs and almoR fquarenefs of the individual hills, and by the eafy (lowing though 

 varied outline of the chains, fuch as that of the Ferwyn mountains already defcribe.l 

 Chap. III. ' 



The lime and fand-ftone hills are confiderably lower even than the (lates ; ridng in ge- 

 neral very gradually at one extremity, and terminating abruptly at the other. The banks 

 of fand rock are however broader and rounder than the lime : where the lime is the hardeft 

 its form is the moft perfeft ; but as it becomes flaty, foft, and mixed with clay, it approaches 

 nearly to the form of the flate-hills, as is remarkably the cafe in the fouthern part of Wen- 

 lock-edge. The fand-(tone, too, where it contains but little iron and clay, bcin-r almoit 

 wholly compofed of fand and lime, refembles moft the limeftone liills. This may be ob- 

 fcrved by comparing the difference of form between the red fand rocks of Nefcllff and the 

 white freeftone of Grinlhlll. ' 



III. I have already mentioned the beds of rounded pebbles that are to be found on the 

 higheft parts of the (late mountains. Their prefent f.luation could never have been that ia 

 which they were formed, for they conf.ft almod univerf.lly of porphyry, quart!^, ferpemlne 

 and other ftony fubftances which lie in large maffes, compofing the primitive mountains • 

 their rounded (hape too, like that of the pebbles on the fea (hore, feems to intimate that 

 they have been carried by the force of water to the places wlilch they now occupy, ^n- 

 otlier circumftance which appears to point out the quarter whence they orlglirdly pro 

 ceedcd IS, that, in proportion to their vicinity to the primitive mountains, is their fize • a 

 crcumflance that might naturally be cxpeaed, f.nce the further they were carried ihe 

 more would they be rounded and comminuted. Still, however, there is a difficulty at ' 

 tciid.ng this Jiypoihca*-, namely, by what means could thefe rounded pebbles have been 



forced 



