J94 Geographische Verbreitung, Reisen. 



very many species is testified to by the length of tlie series wliicli he collected. 

 In this connection, he considers it very doubtful whether close interbreeding 

 of insects on oceanic Islands has much effect on their fertility, and gives some 

 striking examples where this does not appear to be at all the case. As to the 

 supposed paucity of species it need only be said that over 3300 insects have 

 been discovered, over 2700 species of v^^hich are considered to be en- 

 demic: and this in spite of the fact that, in his opinion, at most half the species 

 have been found and the groups are very unequally known. A section is devo- 

 ted to the principal causes of extinction of those species wliich have disap- 

 peared. As stated above imported cattle have destroyed large areas of forest, 

 and by so doing have been largely responsible for the disappearance of some of 

 the species of Passerine birds within the last 20 years. The introduced ant 

 Pheidole megaccphala, which seems to be aggressively dominant in nearly every 

 country which it reaches, has cleared large tracts of forest almost entirely of 

 their endemic fauna. With few exceptions the highly specialised forms composing 

 this remarkable fauna are unable to maintain themselves against imported ene- 

 mies or altered conditions, 



Perkins considers that the endemic fauna is probably derived from 

 a very tew natural immigrants. In viewing it as a whole one can hardly 

 fail to be Struck by the "great series of congeneric species and smaller series of 

 allied genera". That is, there are very many genera each of which comprises 

 an inordinate number of closely allied species: and there are a number of groups 

 of very closely allied genera each comprising many, or (as in the case of the 

 Drepanidid birds) few, species. On the other band there are "extraordinary gaps 

 in the fauna of whole families of wide distribution and containing countless 

 species'', many of wliich would doubtless thrive in the islands, could they but 

 reach them and become estabhshed in them — some of which are actually 

 known to thrive in them when imported by man. Surely these facts point to 

 the origin of the fauna from a few ancient immigrants, which have arrived long 

 ago and probably at long intervals, and become the progenitors of large genera 

 of endemic species or of groups of endemic genera? 



There can of course be much diversity of opinion as to whether these ende- 

 mic genera or groups of genera have a "single or multiple origin", that is, 

 whether they are derived from one or several immigrant species. In some cases 

 one can be almost certain that they are of multiple origin, e. g. in the case of 

 the endernic solitary wasps {Odynerus), of which one group of species is clearly 

 of Oriental affinities and quite different from the other members of the genus. 

 But in many cases the facts appear to point to a "single" origin. The fauna is 

 compared at some length with the flora, and Perkins thinks the pheno- 

 mena very similar in the two cases. 



The comparison of "introduced", "immigrant" and "endemic" 

 species is very interesting. There are certain insects known to have been in- 

 troduced by man many years ago, probably sprung from one or very few intro- 

 duced specimens, and most unlikely to have been re-introduced. Some of these 

 have passed through numbers of generations in these islands so different in cli- 

 mate and in other respects from their native homes, yet they show no greater 

 tendency to Variation than they do in their native homes, in some cases even 

 less. As Perkins says "one is Struck with the stability of specific characters 



[and] one would infer that much time is as a rule required for the appre- 



ciable modification of a stable species". But certain naturally immigrant species 

 (not endemic, but known from outside the islands) present a different })lienome- 



