On a neiu Species of Pimelodus, 337 



only by the most indigent race of Indians ; its aspect and 

 the sUminess of its skin render it very disgusting. 



From the enormous quantity of pimelodes that the vol- 

 canoes of the kingdom of (^uito occasionally discharge, one 

 cannot doubt that country contains great subterranean 

 lakes which conceal these fishes ; for the individuals that 

 exist in the little rivers around are very few in number. 

 A part of those rivers may communicate vi'ith the subterra- 

 nean pits : it is also probable that the first pimelodes which 

 have inhabited these pits have mounted there against the 

 current. I have seen fish in the caverns of Derbyshire, in 

 England ; and near Gailenreuth, in Germany, where the 

 fossil heads of bears and lions are found, there are living 

 trouts in the grottoes, which at present are very distant from 

 any rivulet, and greatly elevated above the level of the neigh- 

 bouring waters. In the province of Quito, the subterraneous 

 roarings that accompany the earthquakes; the masses of 

 rocks that we think we hear crumbling down below the 

 earth we walk on; the immense quantity of water that 

 issues from the earth in the driest places during the volcanic 

 explosions; and numerous other phEenomena, indicate, that 

 all the soil of this elevated plain is undermined. But, if it is 

 easy to conceive that vast subterranean basins may be filled 

 with water which nourishes fishes, it is more difficult to ex- 

 plain how these animals are attracted by volcanoes that 

 ascend to the height of 1300 toises, and discharged eithei 

 by their craters or by their lateral vents. Should we sup- 

 pose that the pimelodes exist in subterranean basins of the 

 same height at which they are seen to issue ? How conceive 

 their origin in a position so extraordinary ; in the flank of a 

 cone so often heated, and perhaps partly produced by vol- 

 canic fire ? Whatever may be the source from which thev 

 issue, the perfect state in which they are found induces u3 

 to believe that those vcilcanoes, the most elevated and the 

 most active in the world, experience, i'rom time to time, con- 

 vulsive movements, during which the disengagement ot ca- 

 loric appears less considerable than we should suppose it. 

 Earthquakes do not always accompany those phcenomena. 

 perhaps, in ihe different concameraiions that may be ad- 

 V'ol. 24. No. 9C. May 180ti. Y mitted 



