152 BOTANICAL GAZETTE | SEPTEMBER 
the Cornacez; to the Ebenales, the Sapotacex, and Styracacee ; 
and to the Gentianales, the Oleacez, and Apocynacee. The sub- 
orders Primulales (with Myrsinacee 6), Polemoniales (with Con- 
volvulacee 2 and Solanacez 1), Rubiales (with Caprifoliacee 4), 
and Asterales (with Composite 1) complete the list of additions, 
and give us a hint as to the method of evolution. 
Miocene dicotyledons included 66 families, and the species 
were more than seven times as many as in the Eocene. Here 
the petalous plants constituted 64 per cent. of the whole, of which 
nearly 16 per cent. were gamopetalous. The great increase in 
the number of species was accompanied by a rapid multiplication 
and modification of previously existing types. Thus we find 
three more families added to Ranales, three to Caryophyllales, 
one.to Geraniales, three to Malvales, three to Myrtales, one to 
Sapindales, one to Umbellales, one to Polemoniales, two to Gentt- 
anales, and one to Rubiales. The Parietales, Guttiferales, Per- 
sonales, and Lamiales appear here for the first time. A closer 
examination of Schimper’s list of Miocene plants indicates that 
in passing from the Eocene to the Miocene, the percentage of 
species of Ranales was not changed, while that of the Caryo- 
phyllales was increased, the Malvales decreased, the Primulales 
unchanged, the Ericales decreased, the Ebenales slightly 
increased, the Rosales unchanged, the Myrtales, Celastrales, and 
Sapindales slightly increased, the Umbellales decreased, the 
Rubiales and Asterales increased. If we examine the dicoty- 
ledonous vegetation of the earth today we may observe that to 
a limited degree these tendencies to increase or decrease are 
maintained to the present. This is shown in detail in the fol- 
lowing tables (see page 153). 
These facts are still more suggestive when presented in dia- 
grammatic form (fig. 2, page 154). 
After making due allowance for the imperfection of the pala- 
ontological record, and our limited knowledge concerning it, it 
is still safe to say that earlier dicotyledons were of considerably 
different types from the later, and that from period to period the 
relative numbers of higher types were increased. 
