INTRODUCTORY REMARKS. XVII 



longed) operation with which we are concerned ; for the fact of a 

 species having been found once in any part of a given country, is 

 sufficient for its name to be entered into that country's fauna. But 

 when it is an archipelago that we have to deal with, instead of an 

 unbroken tract, a Catalogue (if it is to be worth anything, in point 

 of accuracy) must not only record the united productions of the whole, 

 but likewise those of each individual part ; and the labour will con- 

 sequently be increased, in proportion to the number of islands which 

 it is our duty to examine. Nor is the question materially affected 

 whether the latter be extensive or minute, for the real difficulty lies 

 not so much in prosecuting our researches on them when there as 

 in reaching them at all, and that, too, sufficiently often to enable us to 

 gain a knowledge of what is found in them at different seasons of 

 the year. Each island is, literally, a country in itself (whether large 

 or small), and must be investigated separately, the commonest 

 species of each having to be added up with as much care and veracity 

 as if that particular island were the only one we had to ransack ; 

 and when we consider that, in the present instance, some of the 

 islands are well-nigh inaccessible, and that their extremes are re- 

 moved from each other by at least 400 miles of stormy ocean, it will 

 be admitted that I do not exaggerate the difficulty which a thorough 

 exploration of the whole of them must of necessity involve. 



In the Atlantic clusters which have supplied the material for this 

 monograph, the islands (exclusive of mere rocks, which of course 

 cannot be taken into account) are 14 in number ; and some of the 

 uninhabited ones are so dangerous to approach that they are scarcely 

 accessible during the winter months. In the case however of the 

 three Desertas of the Madeiran Group, I think that there is no real 

 need to enumerate the species of each of them separately (although 

 I have done so, nevertheless, and have used the utmost caution in 

 preventing an intermixture) ; for not only are the islands exceedingly 

 small, so that they could not singly be contrasted with the others in 

 the archipelago, but they are likewise so barely separated inter se 

 that they form a little system of their own, and there can be no pos- 

 sible doubt that they were once united. Perhaps, too, the same 

 might be said of the Salvages ; for although they are removed from 

 each other by as much as nine or ten miles, the distance is but slight 

 compared with that which isolates them from the Madeiras and Ca- 

 naries ; whilst, as in the instance of the Desertas, their area is so 

 diminutive that we may well be permitted to treat them also as one 

 at any rate until we have acquired a more perfect knowledge of 



b 



