XI 



been regarded as new, many of them types of distinct genera. 

 But it would have been a work rather of local interest if it had 

 been confined to the objc^cts in the British Museum ; besides, to 

 determine species satisfactorily, a general study of all the allied 

 species is necessary. Therefore its scope was extended to de- 

 scribing also those species which at present are not represented in 

 the British Museum. In this form (that of a handbook con- 

 taining descriptions of, or references to, all the species known) it 

 promised to be particularly useful to the student of ichthyology, the 

 traveller, and collector. The last general works were that of Lace- 

 pede and Schneider's edition of Bloch, published at the beginning of 

 this century, and containing between 1400 and 1500 species, of 

 which about 1100 are still recognized. The great work by Cuvier 

 and Valenciennes remained incomplete. 



"The species and genera have been critically examined; and I 

 have come to the conclusion, after the study of long series of exam- 

 ples, or after autoptical comparison of typical specimens, that it 

 was necessary to eliminate from the system a great number of 

 species, as well as genera, established on insufficient grounds. 

 Zoological science is never advanced by general works compiled 

 mechanically and without critical discernment. In the descrip- 

 tions, I have been satisfied with giving the most important cha- 

 racters, without entering into a complete account of the organi- 

 zation, as this necessarily would have been, for the most part, 

 merely a reproduction of the labours of others ; these, however, 

 are conscientiously referred to. But whenever I thought an ob- 

 servation made by me new and original I have added it. I have 

 paid particular attention to the formation of more natural families, 

 in which endeavour I have laid greater stress upon the structure of 

 the vertical fins and of the skeleton as family characters than my 

 predecessors. Still thinking that the subclasses proposed by Midler 

 are most expressive of the fundamental difiercnces in the organiza- 

 tion of fishes, I found myself compelled, on the other hand, to 

 abandon the order of Pharyngognaths, on establishing which he had 

 bestowed so much labour. 



" I am well aware of the many imperfections of this work ; many 

 have been already corrected by others ; but if it should form the 

 basis for the future development of a collection at i>rc!<ciil unri\allcd 



