NEW HOLLAND IN EUROPE. 
69 
Banksia Solonis, Ung. (Fig. 22), — B. foliis lanceolatis v. ovato-Iaiiceolatis 
utrinqiie atteuuatis longe-petiolatis seniipedalibus grosse dentatis coriaceis, 
nervo pnmario valido, neryis secundariid augulo subrecto e primario egredieiiti- 
bus simplicibxi? crebris. — In formatione Eocenica ad Kymme Euboese, 
These fine, large, and well-preserved leaves arc amongst tlie most common 
fossils of Euboea, where I collected them. Their shape and texture (the leaves 
must have been of a robust nature) and their very distinct venation show they 
belong to Banlcsia^ close to B, serrata (Fig. 23), But the differences be- 
tween the two, apart from their size, are as great as theii* I'esemblance. Passing 
over the costa, which, though drawn much too faint in the fossil, is still very 
different from that of the living B. serrata^ I would point out the petiole, which 
in B. Soloms attains a length such as is never found in any Banlcsia and otlier 
Proteaceous leaves of the present creation. This character assigns to the fossil 
a distinct place, which I cannot better indicate than by an allusion to the great 
Greek philosopher. 
Dryandi'oides lignilum^ Ett. (Fig. 24). — D, foliis coriaceis lanceolatis utrin- 
qno attenuates pcfciolatis denticidatis integerrimisqne, nervo medio valido, ner- 
vis secundariis tenuibus camptodromis. — In formatione Eocenica ad Kymme 
EuboesD. 
This foftsilj which was formerly named by me Quereus It^nilum, is al^o widely 
diffused in Central Europe, and is one of the most common at Kumi. Fig. 24 
sliows two of its leaves. Formerly these leaves I held to be related to those of 
Myrica Bennsylvanica^ Lam., but at present I think that those of Lomatia Ion- 
f/ifolia, R. Br., which are either quite entire or toothed, offer a better analogy. 
The venation does not quite agree with any of them, as far aa it can be recog- 
nized. The leaves of this as well as those of the preceding plant are amongst 
the most common fossils at Kumi, and by their abundance give a character to 
the flora of that locality. I may add that they are also amongst the most common 
at Parscldug, showing that the species to which they belong formed the princi- 
pal part of the woods both in Greece and Styria. But as the flora of Par- 
schlug is more recent than that of Kumi, we have a proof of the longevity of this 
species, wliich passed through all the pliases of the Tertiary period. In plat^ 15 
of my * Iconographia ' I hare figured the leaves of this as well as those of a 
closely allied species, and pronounced both to be Oaks, Apart from that mis- 
take, I now tliink that the difference between the two is too insignificant to allow 
them to be regarded as distiaet species. The plants from Kumi are decidedly 
intermediate forms. 
Epacris Sesostris^ 
TJng. (Fig. 26). — E. 
foliis minutis cordate- 
acuminatis integera- 
mis breviter petiolatis, 
petiolo crasso, nervo 
medio soloeonspicuo. 
In schisto margaeeo ad 
Eadobojum Croat iae. 
Fig. 29. 
Fig. 26. 
Fig. 28 
Fig, 27, 
I 
