114 THE EFFECTS OF FOKESTS ON MOISTXTRE, 



the former period the neighbouring mountains were well wooded, but 

 at the time of his visit the mountahis had been almost entirely 

 stripped of their wood. He further adds that other cases, similar to 

 those alrei.dy detailed, might be cited, and he proceeds to show, by 

 several examples, that the waters of other lakes in the same regions, 

 where the valleys had always been bare of wood, or where the forests 

 had not been disturbed, had undergone no change of level." 



The following statement is also made by Boussiiigault : — *' In the 

 island of Ascension there was an excellent spring, situated at the foot 

 of a mountain originally covered with wood. This spring became 

 scanty, and at last dried up, after the trees which covered the 

 mountain had been felled. The loss of the spring was ascribed, and 

 rightly so, to the cutting down of the timber. The mountain was 

 therefore planted anew, and a few years afterwards the spring re- 

 appeared by degrees, and by and by flowed with its former abundance. 



" The metalliferous mountain of Marmato is situated in the province 

 of Popayan, in the midst of immense forests. The stream along which 

 the mining works are established is formed by the junction of several 

 small rivulets, which take their rise in a country thickly wooded. 



" In 1826, when I visited the mines for the first time, Marmato 

 consisted of a few miserable cabins, inhabited by negro slaves. In 

 1830, when I quitted the country, Marmato was covered with work- 

 shops, had a foundry of gold, machinery for grinding and amalga- 

 mating the ores, and a free population of nearly 3000 inhabitants. 

 In the course of these four years an immense quantity of timber had 

 been cut down, for the construction of machinery and of houses, as 

 well as for fuel and the manufacture of charcoal. But the clearing 

 had scarcely been two years effected before it was perceived that the 

 quantity of water for the working of the machinery had notably 

 diminished. The volume of water was measui-ed by t!ie work done 

 by the machinery; and actual gauging at different times showed the 

 progressive diminution of the water. Now, in the island of Ascension, 

 and at Marmato, it is improbable that any merely local and limited 

 clearing away of the forest should have had such an influence on the 

 constitution of the atmosphere as to cause a variation in the mean 

 annual quantity of rain. Besides, as soon as the diminution of the 

 stream at Marmato was ascertained, a rain-gauge was set up, and in 

 the second year of observation showed a larger quantity than in the 

 first year, though the clearing had been continued, and though there 

 was no appreciable increase in the size of the running stream. 



" Two years' observations are insufficient to show any definitive 



