204 Geographical Collections. 



ous in the defence of their country, and tenacious of the manners of their ances- 

 tors ; and while their torrents and eternal snows are oftentimes the source of all 

 abundance, they themselves form uncultivated tracts where all below is fertility, 

 and they sometimes constitute the political, and often the natural boundaries be- 

 tween nations of men. Valleys, intimately connected with the history of mountain 

 chains, and the vast extent of hydrographical basins, have also been the seat of 

 arts and sciences ; but no facts have yet occurred in proof, that high tracts have 

 ever received either inhabitants or their laws from lowland districts. 



The chain of the Pyrenees, into wliose physical characters we are at present 

 going to inquire, apparently extends itself from the ocean to the Mediterranean, 

 in a direction which seldom deviates from the shortest line. The isolation may 

 be considered as perfect on the north : the extensive and fertile valley commen- 

 cing beyond Thoulouse, and terminating only in the Mediterranean, marked by 

 the course of the canal of Languedoc, and comprising in its line the towns of 

 Villefranche, Castelnaudary, Carcassonne, and Narbonne, effects a complete bar- 

 rier to the too general views of those who suppose a junction of the Pyrenees, 

 whether with the Alps or Cevennes, through the medium of the Montague Noire. 

 To the south, we have not had sufficient opportunity of observation to give any 

 determinate opinion. Mr. Charpentier states that it is continued as far as Cape 

 Ortegal in Gallicia. The limits may be considered as marked on the north 

 by the superposition of the tertiary formations ; to the south they have not been de- 

 termined. Its length, as extending only from ocean to sea, does not exceed four 

 degrees of longitude, its breadth varying throughout with tlie extent of the trans- 

 verse chains. The quantity of surface which it may be supposed to occupy, has 

 been estimated at 1198 square leagues. In its central part a solution of imme- 

 diate continuity takes place, together with a divergence from a straight line, its 

 western acclivity receding 1600 toises to the south, but in the same direction as 

 the eastern acclivity. The general direction of the chain, with respect to the me- 

 ridian, is constantly from E.S.E. to W.S.W. and that of the strata is most gene- 

 rally the sanie. On an investigation of their structure, the Pyrenees appear to 

 consist of a series of bands of alpine limestone, old red sandstone, and transition 

 rocks, reposing alternately on mica slate, or granite, — or a mass of intermediary 

 rocks, locked here and there in stratified crystalline beds. The gneiss and mica 

 slate, generally feldspathic, are on the one hand so intimately connected with the 

 transition series, that Am6 Bou^ did not think that their separation was possible, 

 while their intimate relation (according to the same author) with the stratified 

 crystalline deposits, and the accidents of the latter, led him to suspect that tliese 

 were most probably of a date posterior to that of gneiss. Mr. Charpentier, con- 

 sidering the crystalline primitive rocks to form the base of the chain, supposes a 

 gradation as marked by the succession of primitive, transition, and secondary 

 rocks ; and to account for their degradation, and frequent absence on tlie Spanish 

 side, gives an ancient hypothetical section, by which the culminating point of the 

 crystalline mass having been carried away with the other formations to the south, 

 leaves the transition and secondary rocks predominating on the chain. 



The crystalline rocks attain in the east an elevation never equal to beyond 1500 

 toises, while the transition series succeeding immediately in the continuation of 

 the crest, rise to an elevation exceeding that sum, and continue without interrup- 

 tion to the Port de Glare, where the former again forms the crest of the chain. 

 The elevation of the latter is here at its maximum ; but the culminating point of 

 the Pyrenees, was ascertained by the geodesical operations of M. Reboul and Vidal, 

 to be not the Mont Perdu, but the easterly peak of the Malladetta, known under 

 the name of Peak of Anethou. This latter does not form part of the crest of the 

 principal chain, but is situated at the origin of a valley. 



Coursing to the W.N.W. the crest leaves the culminating point of Mont Per- 

 du to the south, and the Pic du Midi d'Ossau, Neouvielle, and Vignemale, to 

 the north, presenting at the appearance of the overlying alpine limestone, acci- 

 dents which give rise to the most striking and beautiful scenery. Beyond the 

 Pic du Midi d'Ossau, the transition rocks are succeeded by the old red sandstone, 



