2 l6 



NATURE 



[August 19, 1909 



students. One shudders at the thought of a student 

 attempting to commit to memory such a mass of 

 detailed information as is compressed into this volume 

 It has developed into a portable book of reference, and 

 as such is eminently useful and trustworthy for filling 

 up gaps in one's knowledge as occasion requires. 



Whilst deprecating the use of books of this type as 

 class text-books, we do not share the feeling expressed 

 by some that the bulk of new facts accumulated year 

 by year in organic chemistry have little or no value, 

 or that the motives w-hich lead to their production are 

 unworthy. The worst that can be said of multiplying 

 compounds is that, without adding anything to the 

 complexity of the subject, they fill in, as it were, the 

 missing blocks in the picture puzzle, and merely com- 

 plete what was anticipated ; and as to the motives of 

 those who produce them it may be pointed out that 

 much of the research work of the compound-making 

 type is done by young chemists as an exercise in that 

 kind of skilful manipulation which counts for so much 

 in every branch of chemical investigation, and for 

 \\ hich organic chemistry seems so exceptionally well 

 fitted. J. b'. C. 



THREE FISH-FAUNAS. 



(1) Catalogue oj the Fresli-watcr Fishes of Africa in 

 the British Museum (Natural History). Vol. i. By 

 G. A. Boulenger. Pp. xi + 373; illustrated. (Lon- 

 don : Printed by Order of the Trustees ; sold by 

 Longmans, Green and Co., and others, 1909.) Price 

 32.S. bd. 



(2) The Fishes of Illinois. By S. A. Forbes and R. E. 

 Richardson, Nat. Hist. Survey of Illinois. Vol. iii.. 

 Ichthyology. Pp. cxxxi + 387; plates, maps to ac- 

 company above, pp. 103. (Illinois, n.d.) 



(3) Andrew Garrett's Fische der Siidsee. Part viii. 

 By A. C. L. Giinther, Hamburg, Journ. Museum 

 Godeffroy, vol. xvi. Pp. iv + 26i-388; plates, 141- 

 :6o. (Hamburg : L. Friederichsen and Co., 1909.) 

 Price 60 marks. 



(i) 'T^HE zoological survey of the Nile, undertaken 

 J- by the Egyptian Government during the ad- 

 ministration of Lord Cromer, and the explorations of 

 the great Central African lakes, initiated in that 

 country, together with those of the Congo, carried out 

 by the Belgian authorities, have resulted in an enor- 

 mous expansion of the collection of African fresh- 

 water fishes preserved in the Natural History branch 

 of the British Museum. That collection, moreover, as 

 we are informed in the introduction to the first of the 

 three works forming the subject of the present notice, 

 contains a very large proportion of the type-specimens 

 of the many new species which have been described 

 as the result of the aforesaid explorations. It was, 

 therefore, from all points of view highly desirable that 

 a descriptive catalogue of this vast collection should 

 be published, as such a work will serve as a basis 

 for the discussion of the many points relating to the 

 distribution and origin of the African fish-fauna, and 

 likewise as a book of reference for workers in Africa 

 itself, from which it can readily be ascertained whether 

 specimens belong to already described species. 

 NO. 2077, '^'OL. Si] 



Mr. Boulenger, to instance only his volume on those 

 of the Congo, published by the Government of the 

 Congo Free State, has already devoted much study 

 to African fresh-water fishes, and for this reason, 

 coupled with his official position at the Museum, he 

 was obviously the man to undertake the laborious 

 task of writing this catalogue, which, it is considered 

 probable, will run to three volumes. 



The present volume, at all events to others than 

 ichthvological specialists, will probably prove the most 

 interesting of the three, as it includes some of the 

 most distinctive and aberrant types of the African 

 fish-fauna. To many naturalists it will be of special 

 interest to learn that a shark (Carcharias zamhesiensis) 

 inhabits the Zambesi at a distance of 120 miles from 

 its mouth, and also that a saw-fish ascends this and 

 probably other African rivers to a considerable distance. 

 Of the characteristic and peculiar Ethiopian types, 

 two of the most remarkable are the bichirs (Polypterus) 

 and Calamoiehthys, the sole survivors of the fringe- 

 finned ganoids, and the mud-fish (Protopterus). No 

 fewer than ten species of bichir are now recognised, 

 although the allied genus is still represented by a 

 single known member ; and there are two kinds of 

 mud-fish. 



Next in point of interest to these ancient types are 

 the remarkable fishes forming the exclusively Ethio- 

 pian family Mormyridae, many of the members of 

 which display such extraordinary vagaries in the 

 matter of beak-development — a development which has 

 suggested for the group the not inappropriate name 

 of elephant-fishes. Of these strange fishes eleven 

 generic types are now admitted, some of which, such 

 as Mormyrops (with thirteen), include a large number 

 of species. It may be hoped that before long the 

 author will take an opportunity of giving us his views 

 as to the origin of this family, which, if determinable, 

 will add considerably to our knowledge of th origin 

 and relationships of the African fauna generally. 



The other important family treated in this volume 

 is the Characinidae, which has a distribution similar 

 to that of the Lepidosirenidae, being common to Africa 

 and Central and South America. The number of 

 African genera admitted in the volume before us is 

 twenty. The characinids, like the lepidosirenids, have 

 been frequently quoted as affording evidence in favour 

 of a land connection between Africa and America, but 

 before it can be decided whether they are of any value 

 in lending support to that theory, it is essential that 

 their past history should be known. 



Although we have not much to say in the 

 way of criticism, it may be mentioned that 

 Flops sauriis (p. 25) has recently beer, shown 

 bv Mr. C. T. Regan to occur only on the 

 American side of the Atlantic, and that there are now 

 three African species of the genus, viz. senegalensis, 

 machnata, and lacerta, the last of which is alone ad- 

 mitted in Mr. Boulenger's volume as a valid species. 

 Then, again, family rank miglit well be granted to the 

 genus Chanos. On the other hand, it is satisfactory to 

 find that the author has recognised the correctness of Dr. 

 Gill's removal of the Kneriidae from among the 

 Haplomi, and their transference to the neighbourhood 



