61 
Having thus, I hope, made good my position as against my critics, 
let us briefly review the phylogeny of the British butterflies as it pre- 
sents itself in its fragmentary condition, with so many gaps, so many 
unrepresented groups. And first let us admit, at the outset, that the 
evidence for the probable phylogeny of the Lepidoptera is only cir- 
cumstantial, and has merely a certain and shifting grade of probability 
to sustain it. Often it is only a bare possibility, and we see how 
Mr. Meyrick treats such probabilities or possibilities with his dog- 
matic phrases, “an offshoot,” &c. ; “‘a development,” &c. ; when in 
truth he knows nothing positively as to what he affirms, and in mis- 
understanding the neuration, and producing defective studies upon 
it, really neglects what might put him upon the right track. 
Having thus schooled ourselves to regard our discoveries as 
evidence not positive, but circumstantial merely, led us to see what 
the neuration teaches us as to the British butterflies especially, and 
the Diurnals in general. First we must separate the Papilionid 
phylum, and for the reason that it presents a character, the down- 
wardly curved internal vein, which all the other butterflies do not 
share. Next we find that the other, Hesperid phylum, offers, in the 
Pieridz, an example of specialisation in two directions ; the breaking 
up of the media and the diminution of the radial veins. The Pierini 
overreach here the highest group of the four-footed butterflies or 
Nymphalide, since the radial veins are reduced at times to three, 
while at the same time the reduction of the media and its system is 
well progressed. The probable phylogeny of the Pierinze stands thus : 
Mancipium and Freris are nearly related and probable developments 
of each other, JZ. brassice seeming to be the youngest form. We then 
come to Ponta, which appears to be a more remote development on 
the ancestral line of the five-branched Axthocharint, which latter 
would represent an older phase of the /vev7z@. Between the three 
and four branched and the five-branched genera come, as lateral and 
peculiarly specialised offshoots, Colzas (Gonepteryx) and Lurymus 
(Colias). ‘The grounds for the view that xhamnz represents a 
specialised mimetic form may probably strike every one. A long way 
from all these stands szzafzs, a strangely isolated form, perhaps in its 
turn a specialisation of an older type of the Whites, and retaining 
some ancestral features, with the middle branch of the media leaning 
turned towards the cubitus on hind wings, an odd reminder of Papz/io. 
Leaving the Pierids, we come to the Nymphalids, and here we find 
our characters for rank mainly determined by the perfection which 
the breaking up of the media attains, and the amount of the 
absorption of II by III on the secondaries. A character to separate 
the purple emperor from the meadow browns, with which Mr. Tutt 
associates it, is found in the position of veins IIT 4 and III 5, which 
are given off to the outer margin as in Vanessa. ‘The swollen vein in 
the male may indicate a lost line of connection between the Nym 
phalids and Satyrids, or it may be an independent acquirement. 
This beautiful butterfly appears to be more specialised than Vanessa, 
