431 



( ffiuiia is the tact that ncai'lN' all iirimps arc iiilialiitaiils ol dead 

 \\(»(id and (lcl)i'is of the kind that sometimes drifts ahout the 



I oceans, or if not of that ty])e arc nsuallv sli-onii,' in tliiiht. 



' There are no native ]eaf-eatin<i" licctles or grasshoppers or simi- 

 lar insects. 'Jdic l)eetles are ncarlv all wood-horers or iii'onnd 

 hectics which conimoidv hide awav under hai'k. Xcarlv all the 



[ Hynienoptera are borers or forms which nest in logs, etc. Dip- 



' tera are represented mostly by debris inhabiting forms, while 

 of the J.(>pi(loj)tera we have no native l)ntterflies except one 

 com])aratively recent immigrant, but a consider;d)lc numlx^' of 

 moths some of which are strong in flight and others pn})ate 

 imder bark or in similar situations. 



The ("oiudusion to 1)0 drawn from these facts is that these 

 Islands have been in existence for a very great length of time 

 — long enongli for many species to have originated here from 

 a few ancestors — a conclusion which is also snpported l)y 

 geological evidence which points to the existence of the land 

 mass as far back as the Paleozoic. Another crmcdnsion to be 

 drawn from our data is that the islands have always been iso- 

 lated and never a part of a continental land mass, hence re- 

 ceiving no migrations of animals overland bnt only l)y long 

 and very precarions voyages over the ocean in logs and floating 

 debris, and perhaps by flight and carriage by winds. From the 

 very small nnml)er of ancestral types represented by the en- 

 demic species it wonld a|)i)ear that only very rarely did insects 

 and shells succeed in establishing themselves in these Islands. 



The presence here of some very delicate insects is more 

 diflicnlt to explain. They do not inhabit logs nor debris al- 

 thongh some are gall makers, and their span of life is very 

 short, especially short in the absence of living foliage to fur- 

 nish them food. Certain homopterons insects, the leaf hopjKM-s 

 and jumping plant lice, are good examples of this type. Tt 

 is not possible to explain their entrance here by way of a land 

 bridge now disappeared, for if there had been snch a bridge 

 beyond donbt more than the meagre few wonld have become 

 established here. There remains, then, only the agency of 



