Medusa from the River Niger. 31 



somewhat larger size than the tliree forms ah'eady recorded 

 as existing in fresh water *. As a matter of fact the Largest 

 specimens of Limnocnida tanganyikai {sic), Bohm f, measure 

 22 millim., while the diameter of Limnocodium Sowerhyi, 

 Allman and Ray Lankester, amounts to 8 millim., and that 

 of Halmomises lacustris, von Kennel, only to 2'5 millim. 



The fixed polypes of a creeping Hydrozoon of marine origin 

 progressively adapted to fresh water may, in the course of 

 centuries, ascend a river like the Niger, and even surmount 

 its rapids. Without speaking of the changes which may 

 have supervened in the configuration of the country, and of 

 the accidental transport of debris of all kinds to which the 

 animals with which we are dealing readily attach themselves, 

 it is necessary to pay particular attention to the peculiar 

 conditions affecting the river. 



This is not the place to enter into lengtliy considerations 

 of geographical questions. I may be permitted, however, to 

 quote a few passages from the great work by Elisee E,eclus J, 

 which will show, without requiring any comment, how a 

 freshwater fauna is able to disseminate itself with great 

 rapidity through a very extensive region, both up and down 

 a river-system. 



" United to each other to the south of Segou Sikoro two 

 large rivers flow parallel to the Niger ; then, after breaking 

 up into pools {viari(jots)j which convert the Jenneh country 

 into a labyrinth of islands, the common current rejoins the 

 principal stream above Lake Debo. At this spot the waters, 

 which are broad and very deep, are almost stagnant : a dam 

 a few metres in height would transform the confluent into an 

 inland sea ; at the time of the falling of the waters the 

 lacustrine reservoir spreads out on all sides, even flowing up 

 stream towards the upper river. 



" Below the region where the liquid volume of the two 

 Nigers in some places unites into lakes, and in others ramifies 

 in tortuous arms around low islands, the stream, retarded by 



* I purposely omit to recall to mind liere all the instances of Craspe- 

 dote or Acraspedote Medusse {Crambessa, C'allirho'e, Laodice, &c.) which 

 have been recorded as non-marine. I have spoken elsewhere of the 

 majority of these (J. de Guerne, "Meduses d'eau douce et d'eau saumatre, 

 d'apres quelques travuux recents," Bull, scient. du dep. du Nord, 1880). 

 Neither are the Coelenterates which live in what is almost fresh water in 

 numerous localities of the Baltic and of the Gulf of Finland enumerated 

 here. 



t This is how, according to the rules of nomenclature, the name of the 

 Medusa of Lake Tanganyika ought to be written. 



t Elisee Reclus, ' Geographic universelle, XII. Afrique occidentale,' 

 p. 52. 



