( 514 ) 



and Papuan Regions ; while again seven series are found in tlie Indo-Malayan 

 Region, three of them iilso in Celebes, all seven being absent from the Papnan 

 Region. 



The lesser Snnda Islands have therefore not a single series of representatives 

 which is purely Australian. From tliis fact we have to conclude that the immigration 

 of the Papilio fauna cannot have taken place from Australia ; and as the physio- 

 graphical and meteorological conditions of the islands resemble in many respects 

 more those of Northern Australia than those of the larger Sunda Islands, the reason 

 why no Australian types immigrated cannot lie in the biological conditions of the 

 islands, and hence must be accounted for by the assumption of another barrier 

 against immigration. This barrier most probably was, at the time wlieu the lesser 

 Sunda Islands became first populated by Papilios, the same which we perceive at 

 the present epoch, namely the wide Timor Sea separating Timor from Australia. 



If we leave out of consideration as being iudiHerent the types which occur in 

 the Indo-i\Ialayan and Papuan Regions, there remain eight series of representatives 

 which the lesser Sunda Islands have in common with the Indo-Malayan Region, 

 and which are absent from Australia and New Guinea, four of them being repre- 

 sented also on Celebes, and one also on the Northern Moluccas ; opposite these eight 

 types stands one single Celebensian type that is absent from the Indo-Malayan 

 Region. The Indo-Malayan element, therefore, is so strongly predominating that 

 we must assume that there never existed any mechanical barrier of consequence 

 preventing migration of the Papilios from the greater to the lesser Sunda Islands 

 (or in the opposite direction). 



Nevertheless there is a marked discrepancy in the fauna of the nearest of tlie 

 larger Sunda Islands, Java, and the lesser Sunda Islands, as fourteen Javan 

 types are not represented on any of the islands east of the Strait of Lombok, while 

 six types occurring on the lesser Sunda Islands (except Lombok) are not known to 

 have a representative on Java. From the fact that out of the eighteen series of 

 representatives of the lesser Sunda Islands twelve occur on Java it is evident that 

 the absence of fourteen Javan tvpes from the lesser Sunda Islands, and the absence 

 of six lesser Sunda Islands tyi)es (five of which are represented in other ])arts of the 

 Indo-Malayan snbregion) from Java, cannot acce])tably be explained by the sugges- 

 tion that there existed at a former period a very much wider strait between the 

 lesser Sunda Islands and Java than there is now. And taking into account the 

 Papilio fauna of Lombok alone,* whicli island has not a single Papilio type that 

 is absent from Java, it is obvious that there is no evidence in favour of the as- 

 sumption that the Strait of Lombok has changed in width since tlie appea;rance of 

 the Pajnlios on the Malayan islands. If we exclude Lombok from the lesser 

 Sunda Islands, and comj)arc the Papilio fauna of the rest of the islands with that 

 of Java, there is an important numerical discrepancy whicli amounts to more than 

 60 per cent, of the total number of the series of representative forms (as opposed to 

 the total number of representative forms) inhabiting Java and the lesser Sunda 

 Islands. As this amount t is very high compared with the numerical discrepancy 



* The island has recently been visited by three explorers (W. Uolierty, A. Everett, and 

 n. Fruhstorfer). 



t As Flores is lepidopterologieally almost a blanlc the numbers m.ay become quite difiercnt in con8(Squence 

 of the exploration of this large island. The al)sencc of many types from small islands which are found on 

 neighbouring large islands is quite natural, considering that the small islands do not afford so great a 

 variety in the conditions of life as would be required by a multitude of diverse inhabitants. 



