438 NORTHERN CHILE. June, 1835. 



but little furrowed by ravines. No considerable river could 

 ever have poured its waters over the bed of shingle, without 

 having excavated a channel similar to those occurring in the 

 southern valleys. I feel little doubt that it was left in the 

 state we now see it, by the gradually retiring sea. The dry 

 valleys mentioned by travellers in Peru, probably owe their 

 origin to a similar agency, and not to the running streams of 

 any former period. I observed in one place, where a ravine 

 (which amongst any other mountains would have been called 

 a grand valley), joined the Despoblado, that the bed of the 

 latter, though composed merely of sand and gravel, was higher 

 than that of its tributary. A mere rivulet of water, in the 

 course of an hour, would have cut a channel for itself; but it 

 was evident that centuries had passed away, and no such 

 rivulet had drained these great valleys. It was curious to 

 behold this machinery (if such a term may be used) for the 

 drainage, all, with the last trifling exception, perfect, yet 

 without any signs of activity. Every one must have remarked 

 how mud-banks, left by the retiring tide, imitate in miniature 

 a country with hill and dale : and here we find a model, 

 only on a grander scale, formed by the waves of a retiring 

 ocean. Let thousands of years replace minutes of the tidal 

 change, and the difference between soft mud and hard rock 

 wiU barely modify the result. If a shower of rain falls on the 

 mud-bank, when left dry, it deepens the shallow lines of ex- 

 cavation : and so will it be with the rain of successive cen- 

 turies on the bank of rock and soil, which we call a continent. 

 We rode on after it was dark, till we reached a side 

 ravine, with a small well called " Agua amarga." The water 

 deserved its name, for besides being saline it was most 

 offensively putrid and bitter; so that we could not force 

 ourselves to drink either tea or mate. I suppose the dis- 

 tance from the river of Copiapo to this spot was at least 

 twenty-five or thirty English miles ; in the whole space there 

 was not a single drop of water, the country almost deserving 

 the name of desert in the strictest sense. Yet about halfway 

 we passed the old Indian ruins near Punta Gorda, which I 



