LEAVE MANAOS. 345 



waters, it extends over a much wider area. In the same 

 way, fishes which gather near the month of a rivulet, at 

 the time of low waters, will be found as high as its origin 

 at the period of high waters ; while fishes which inhabit 

 the larger igarapes on the sides of the Amazons when they 

 are swollen by the rise of the river, may be found in the 

 Amazons itself when the stream is low. There is not a 

 single fish known to ascend from the sea to the higher 

 courses of the Amazons at certain seasons, and to return 

 regularly to the ocean. There is no fish here corresponding 

 to the salmon, for instance, which ascends the streams of 

 Europe and North America to deposit its spawn in the cool 

 head-waters of the larger rivers, and then returns to the sea. 

 The wanderings of the Amazonian fishes are rather a result 

 of the alternate widening and contracting of their range 

 by the rise and fall of the waters, than of a migratory 

 habit ; and may be compared to the movements of those 

 oceanic fishes which, at certain seasons, seek the shoals 

 near the shore, while they spend the rest of the year in 

 deeper waters. 



"Take our shad as an example. It is caught on the coast 

 of Georgia in February, on the Carolina shores a little 

 later ; in March it may be found in Washington and Balti- 

 more, next in Philadelphia and New York ; and it does not 

 make its appearance in the Boston market (except when 

 brought from farther south) before the latter part of April, 

 or the beginning of May. This sequence has led to the 

 belief that the shad migrates from Georgia to New England. 

 An examination of the condition of these fishes, during the 

 months when they are sold in our markets, shows at once 

 that this cannot be the case. They are always full of roe, 

 and, being valued for the table at this period, they are 



15* 



