( ^81 ) 



The species fonud in the three mountain ranges mentioned above may be 

 classed, according to the degree of difference, into three groups : 



(1) Some species are the same in at least two of the localities ; 



(2) Other species are represented in each locality by a more or less different 

 geograj)liical form ; 



(3) Most species are represented iu each locality by an allied distinct 

 species ; 



(4) Some species are known only from one of the mountain ranges. 



As the exploration of these mountains is still very incomplete, not every 

 type of species is known from all three districts ; but so much is certain, that 

 most of these highland species are represented in each separate range by a form 

 belonging to Class 1, 2, or 3. Two further species, which A. S. Meek obtained 

 from a friend who had taken them on Mount Groome, situated at the boundary 

 between British and German New Guinea, confirm the statement, these two 

 Delias (described below under Nos. 4 and 8) being so different from the 

 corresponding forms from the other mountains that we may safely infer from 

 them that Mount Groome also has its own set of subspecies and species. We know 

 nothing of the alpine Lepidojitera of the high mountains in German New 

 Guinea, nor are the Lepidoptera known from the higher altitudes of the Charles 

 Louis Mountains iu the Dutch territory. These places again may be inhabited 

 by different forms. 



From the fact that the various forms are divisible into the above four classes 

 follows that 



(1) No individual mountain range can be called the original home of these 

 alpine Delias as a whole ; and 



(2) An interchange of species between the mountains has taken place at 

 one time or the other. 



The majority of forms which represent each other (Class 3) being specifically 

 distinct in each of the above-mentioned mountain ranges further renders it 

 evident that 



(3) This specific distinctness has come into existence in geographically 

 separate districts. 



Therefore, in order to explain the ahundnnce of species which occur togetlier, 

 it is only necessary to couple points (2) and (3). Now, if a further interchange of 

 species should take place, say between Mount Goliath and the Owen Stanley 

 range, the number of species in either locality would naturally become much 

 enlarged. 



There can be no doubt, we think, that the species which re[)lace each other 

 in these difierent mountains are really distinct — i.e. are so divergent that they 

 could exist side by side without merging into one single species, which is the 

 only test of specific distinctness. Several species, in fact, do exist in the same 

 place with their nearest ally. In the Owen Stanley Mountains, for instance, we 

 have D. meeki as well as D. niepelti, whereas on Mount Goliath Meek only got 

 a good series of meeki, several of the specimens, however, distinctly inclining 

 towards niepelti. On the other hand, the collection from Mount Goliath contains 

 a geographical race of D. eichhorni and a closely allied second species, whereas 

 only eichhorni was found in the Owen Stanley range. D. kummeri and D. iso- 

 charis, which are so similar to each other that we first believed them to be 

 one species, fly together iu the Owen Stanley Mountains, both being common. 



