Ch. VII. SOUTH AMERIGA. 1^3 



Cordelleros and Peru. It would naturally be ex- 

 pected, that the feveral branches of this river, which 

 abounds in fifh, would caufe a great plent}^ of them 

 in the city but it is quite otherwife, and the fmall 

 quantity caught near it is far from being good, and 

 fo full of bones, that none but the inhabitants can eat 

 them without danger. Their badnefs in the neigh- 

 bourhood of the city is probably owing to the brackifh 

 water ; but fome leagues above the city, the river af- 

 fords a great fupply of what is very excellent. In 

 fuch hot climates, however, they cannot be kept with- 

 out fait ; and it is very feldom the fifliermen venture: 

 to carry any to the city, left, after all their labour^j 

 they Ihould be obliged to throv/ them away. 



The coafts and neighbouring ports abound in very 

 delicious fifh, fome of which are carried to the city^ 

 as keeping better than the fpecies in the river ; and 

 thefe, together with feveral of the teftaceous kind, 

 conftitute a confiderable part of the food of the in- 

 habitants of Guayaquil. In the falt-creek are taken 

 very large and fine lobfters, of which they make de- 

 licious ragouts : and from Jambeli creek, on the 

 coaft of Tumbez, are brought great quantities of 

 oyfters, which, in every refpe6t, furpafs thofe of all 

 the coafts from Panama to Peru, v;here there is alfo 

 a great demand for them. 



The fame caufe which drives from that part of the 

 river near the city the fineft fifli, fome to the fait and 

 others to the frefti waters, according to their refpedive 

 natures, renders good water very Icarce^ at Guayaquil, 

 efpecially in fummer , none being to be had at a lef^ 

 diftance than four or five leagues up the river, ac-^ 

 . cording to the height of its waters. Many balzas 

 are therefore employed in fetching water, and felling 

 it to the inhabitants. During the winter, this trade 

 is partly at a ftand, as, by the increafe of the rivers, 

 the water at Guayaquil is rendei*ed fit for ufe. 



M 2 . Instead 



