23 
the discoidal cell is a branch of the median, and not of the subcostal, 
nervure. Now it is evident that in our Rhopalocera the nervure 
that comes from the centre of the front of the discoidal cell is 
homologous in all species, and that if the upper branch of the median 
be called 6 in Papilio , it should be called 6 in all other species. 
Now if you turn to the next diagram, that illustrating the 
neuration of Aporia crataegi, you will notice a very decided difference. 
If the nervures be counted upwards from the bottom, we shall find 
that the nervure numbered 6 in Papilio, i.e., the upper branch of the 
median nervure is now the 5th from the bottom. This 5th nervure in 
Aporia, then, is homologous with 6 in Papilio, and to keep the homology 
correct we must still number it 6, whilst a little careful work shows 
3 to be missing. In this respect, therefore, Aporia is more specialised 
than Papilio. But Aporia differs somewhat from Papilio in the subcostal 
nervure (7-11), since the branch 9 is very much reduced in size, another 
distinct tendency to specialisation. 
Comparing these with the next figure, that of Pieris, we see 
that the absence of 3 noticed in Aporia, is still prominent, whilst 
9 is still more reduced, pointing to another slight advance in 
specialisation. 
Now if we compare with the next figure, that of Euchloe, we 
shall observe that the branch of the median numbered 3 is still 
missing, but a complication of the subcostal occurs, 9 not only being- 
present, but also a supplementary nervure which is not present in 
Papilio, Aporia or Pieris. Now our authorities on neuration would 
explain this by saying that 3 had migrated to 4, 4 to 5, 5 to 6, 6 to 
7, and so on, and that as Euchloe has 12 nervures, they are 
homologous with the 12 in Papilio. I maintain that this is not so. 
I assert that it is impossible for a branch of the median to migrate so 
as to become a branch of the subcostal; 6 is to me the terminal of 
the median ; and I would explain the presence of branches 9 and 9a 
to the bifurcation of 9, due to the necessities of wing-support, since, 
in this species, the discoidal cell is much restricted, and the distance 
of the apex from the cell is comparatively great, but I believe this 
increase in the branches of the subcostal to be a comparatively recent 
development due to local considerations, and in no way to vitiate 
the general law, that the more highly specialised lepidoptera have the 
fewer number of wing nervures. 
That this explanation is probably true may be seen from the next 
diagram, that illustrating the neuration of Leucophasia, where, owing 
to the contraction of the discoidal cell, the branches of the subcostal 
nervure, numbered 10 and 11, are also removed from the top of the 
cell, and appear as branches of the main subcostal stalk, 8. 
I trust that I have made my meaning clear. The absence of 3, 
throughout the whole of the Pieridae, I take to be of the utmost 
importance as placing the specialisation of Pieridae above Papilio, 
whilst the complication of 9 in the tribe Antliocaridi, I look upon as a 
special recent development due to local needs and circumstances. I 
have attempted to explain the meaning of an apparent paradox, i.e., 
how the reduction in the number of nervures, represents, as a rule, a 
