( Ixxiii_) 
‘‘yemainder for those of Hymenoptera. The great number 
‘of densely hairy flower-frequenting Coleoptera in South 
‘« Africa must also play a large part in plant fertilization.” 
‘© A residence of nearly twenty-five years at Capetown 
‘enables me to state with some certainty that the species 
‘‘ inhabiting the neighbourhood do not number more than 
‘< forty-seven.” 
‘© This remarkable poverty of butterflies is rendered all the 
‘more striking from the circumstance that twenty-nine of 
‘« the species are small Lycenide (twenty-two) and Hesperiide 
‘« (seven), and the bulk of the remainder consists of sombre 
‘‘ Satyrine (ten) of medium size.” 
“It is only when we progress eastward along the belt 
‘«‘ between the first mountain range and the sea-coast that 
‘the Rhopalocerous fauna finds conditions more favourable 
‘for its development. Thus, at Knysna, where extensive 
‘ forests of large trees clothe a large area, I collected, during 
‘¢ nine months’ residence, sixty-two species.”’ 
« Across the Kei river, Colonel Bowker collected one hun- 
‘‘ dred and seventeen species, and it is only when we reach 
‘© D’Urban, on the coast of Natal, that the augmentation in 
‘their ranks is remarkable. At this spot the Rhopalocera 
‘‘ become a constant and beautiful feature of the scenery.”’ 
Mr. Trimen goes on to remark that he took in one day, 
near Natal, no less than fifty-four species, which, however, is 
less than I have taken in one day, with Mr. Salvin, in the 
Italian valleys of the Alps. 
It is not more easy to account for this poverty at the Cape 
than it is in South and West Australia, where, notwith- 
standing a rich, varied, and peculiar vegetation, we have 
extreme poverty in the Rhopalocera. The desert country 
which separates the Cape Colony from tropical Africa on the 
north, is, no doubt, a good natural boundary against the 
extension southwards of tropical forms, for which the climate 
of littoral South Africa would appear to be far better 
adapted than the coast of Korea and South Japan are for 
the tropical Indian forms of butterflies which we find there ; 
or than the coast of New England for some forms of Neo- 
tropical affinity which extend so far north; but though the 
