FOBDS OF THE GUATGUAZA. 41 



forming masses like towers, or old ruins, at the summit of 

 the highest mountains.* 



The heat became stifling as we approached the coast. A 

 reddish vapour veiled the horizon. It was near sunset, arid 

 the breeze was not yet stirring. We rested in the lonely 

 farms known under the names of the Hato de Cambury and 

 ' the House of the Canarian' (Casa del Isleno). The river of 

 hot water, along the banks of which we passed, became deeper. 

 A crocodile, more than nine feet long, lay dead on the 

 strand. We wished to examine its teeth, and the inside of 

 its mouth ; but having been exposed to the sun for several 

 weeks, it exhaled a smell so fetid that we were obliged to 

 relinquish our design and remount our horses. When we 

 arrived at the level of the sea, the road turned eastward, and 

 crossed a barren shore a league and a half broad, resembling 

 that of Cumana. We there found some scattered cactuses, 

 a sesuvium, a few plants of Coccoloba uvifera, and along the 

 coast some avicennias and mangroves. We forded the Guay- 

 guaza and the E-io Estevan, which, by their frequent over- 

 flowing, form great pools of stagnant water. Small rocks of 

 meandrites, madrepores, and other corals, either ramified or 

 with a rounded surface, rise in this vast plain ; and seem to 

 attest the recent retreat of the sea. JBut these masses, 

 which are the habitations of polypi, are only fragments im- 

 bedded in a breccia with a calcareous cement. I say a 

 breccia, because we must not confound the fresh and white 

 corallites of this very recent littoral formation, with the 

 corallites blended in the mass of transition-rocks, grau- 

 wacke, and black limestone. We were astonished to find 

 in this uninhabited spot a large Parkinsonia aculeata loaded 

 with flowers. Our ootanical works indicate this tree as 

 peculiar to the New World ; but during five years we saw it 

 only twice in a wild state, once in the plains of the Rio 

 Guayguaza, and once in the llanos of Cumana, thirty leagues 



* At Ochsenkopf, at Rudolphstein, at Epprechtstein, at Luxburg, and 

 at Schneeberg. The dip of the strata of these granites of Fichtelberg is 

 generally only from 6" to 10, rarely (at Scheeberg) 18. According to 

 the dips I observed in the neighbouring strata of gneiss and mica-slate, 1 

 should think that the granite of Fichtelberg is very ancient, and serves as 

 a basis for other formations ; but the strata of grUnstein, and the disse- 

 minated tin-ore which it contains, may lead us to doubt its great an* 

 tiquity, from the analogy of the granites of Saxony containing tin. 



