150 DK. R. J. TILLTARD ON THE 
present. In order to test this conclusion, we next look for the ciibito-median 
Y-vein, and find that it is present quite close to the base of the wing in the 
very characteristic form shown in text-fig. 3. This definitely determines the 
correct homology of the vein Mj-j-Cu^, and enables ns to pick up at once the 
true median above it, viz. the concave vein which had previously been called 
Gu], in spite of the fact that it is not convex. We thus find that M is a 
three-branched vein, excluding M5, of triadic type, just as it is in so many 
of the Palseodiclyoptera themselves. 
We now have only the limits of Rs to consider. The whole of the many- 
brnnched veiu which had previously been considered to be formed of Us and 
M, combined by basal fusion^ is now seen to belong to Rs alone. It arises 
from R, near the base, and its short basal piece is convex. It then branches 
into a concave upper branch (R2+3) and a convex lower one (R^+s) ; and thus, 
as already indicated above, it makes with R^ a true positive triad. The 
numerous branchings of Rs shown in Protereisma need not be a cause of 
wonder, since they are paralleled in quite a number of Palreodictj'opterous 
types. 
In text-figs. 1 and 3, it should be noted that, excluding the clavus, on 
which all the veins are convex, the whole of the veins reaching the wing- 
margin are pLiced alternately convex and concave. The complete notation 
for the wing oi Protereisma has been already given in the table on p. 145. 
Comparison of the Venations of Fossil and 
Recent Mayflies. 
TJje next step is to make a careful comparison between the venation of fossil 
Mayflies and the more archaic representatives of existing types. For this 
purpose, I have chosen the wings of Ameletus ornatus, family Siphlurida; ; 
not because I believe it to be the most archaic of existing tj-pes, but because 
it is archaic enough for our purpose, and happens to be the species on which 
mcjst of the studies of larval wing-tracheation were carried out. 
Comparing the wings of Protereisma (text-figs, la, 3) with the fore- wing 
of Ameletus (text-figs, lb, 4), the first thing we notice is the change in the 
general shape of the wing. That of Protereisma is of the archaic elongate 
oval form seen in most of the Palseodictyoptera and in the fore-wings of 
Perlaria, in which no definite tornus and termen are developed, but only a 
single continuously curved posterior margin from bnse to apex. In most 
recent Miiyflies, Ameletus included, a definite tornus (tn) is to be recognised ; 
the angle of the tornus divides the original posterior margin into a shorter 
basal part, to which tlie term posterior margin is still applied (called the 
dorsum in Lepidoptera, but the term is inapjilicalile to Mayflies, which do 
not fold their wings over their body in a root-like manner), and a much 
longer distal part, called the termen or distal man/in. The result of this is 
that the wing is now no longer of the primitive elongate oval shape, but 
