tAEENTAL (WRE AMONG FRESH-WATER FISHES. 51 5 



iiiv comiiaratively deep. " It will be a dry summer," and vice versa. This is based 

 on the supposition that in dry summers the evaporation will exceed the rain- 

 fall, and that the fish place their nests in deep water to prevent them from 

 being left high and dry. 



Doctor Abbott justly adds that " the absurdity of this is apparent. 

 The nests are occupied but for a portion of one month, and what the 

 summer may be can in no wise affect them; but of this the vilhige 

 sages never think. The truth is the same spots are used year after 

 year, whether the water be high or low." 



The sunfish is to a considerable extent used as a pan fish and is at 

 least as good as a perch, but on account of its small size (it averages 

 less than G inches) it can not be ranked as an important market fish. 

 It has been, nevertheless, the object of considerable demand in different 

 )3arts of the United States, and the United States F'ish Commission 

 raised and distributed a large number, especially in the early nineties. 

 The demand, however, was doubtless to a large extent from dealers 

 in fancy fishes and for artificial poiuls and household ornamentation- 



THE CICIILIDS. 



The next family that may be (considered is that of the Cichlids. 

 Its chief I'epresentatives are much like most of the Centrachids, but 

 tluM-c is considerable diversity in some respects. The form varies 

 from a high and much compressed body to an elongated one like that 

 of a pike; the scales are generally of moderate size, in some cycloid, 

 in others ctenoid; the lateral line in most is interrupted and double 

 for a short distance on the tail, but in a few merely deflected ; the head 

 is scaly; the nostrils are always single; the palate is never dentiger- 

 ous; there are mostly five branchiostegal rays, and never more than 

 six ; the lower pharyngeal bones are united, at maturitj^ at least, into 

 a single bone. Further, systematists lay much stress on the fact that 

 they have two rows of filaments to the last branchial arch as well as 

 others usual in fishes, but not in the Pomacentrids, to which they 

 are most nearly related, and that there is no subocular shelf. 



The family is almost entirely confined to fresh waters and further 

 to those of tropical Africa and America. A very few have extended 

 beyond these bounds; from Africa several into neighboring Asia 

 including Palestine; from tropical America one species into sub- 

 tropical Texas. 



Until near the close of the nineteenth century it was supposed 

 that America was far richer both in genera and species than Africa, 

 but long after the discovery of Lake Tanganyika explorers of the 

 animal riches of that great lake were rewarded with an unexpectedly 

 rich harvest of new types of fishes as well as mollusks. No less than 

 73 species, representing 24 genera of Cichlids, had been discovered 

 therein up to the middle of the year 1906. 



