FLORA OF MADAGASCAR. 259 
will probably in the future be found in one or both of the others, 
the proportion of the plants peculiar to the respective Regions 
will not, I am convinced, be seriously disturbed, or the floras be 
shown to be even approximately identical. 
In regard to the Orders, there are several which appear to be 
absolutely confined, and more which are nearly confined, to one 
or other of the three Regions, but these will be noticed further 
on. 
The table on the next page shows the Orders most largely repre- 
sented, and their percentage of the total flora, in the respective 
Regions. In this table the following facts are prominent :—In 
the Eastern Region the two most abundantly represented Orders 
are Filices and Composite ; but the former are more than double 
the latter in the number of species, forming respectively 13:1 
and 6 per cent. of the flora of this Region. It will be noticed 
that Filices do not appear in the second or third column at 
all, the reason beinz that I have not sufficient data for deter- 
mining their relative positions. Possibly they might occupy 
the third or fourth place. In the Western Region the Legu- 
minose stand at the head of the list, and these are followed by 
Euphorbiacex ; but the difference between the two is very great 
the proportion being about 5 to 2. The table shows that 
18:8 per cent. of the flora of the Western Region consists of 
Leguminose. The Composite appear to be poorly repre- 
sented, forming only 3:2 per cent. of the flora. In the Central 
Region, on the other hand, the Composite are at the head of the 
list, with a percentage of 13. Rubiacew, again, which one 
might expect to be largely represented in the Western Region, 
only form 3:2 per cent. of the flora. The Eastern, Central, 
and Western Regions therefore might, if we take the most 
largely represented Orders into account, be fairly called the 
Fern Region, the Composite Region, and the Leguminous Region 
respectively. 
Turning to the table showing the distribution of the species, 
we see that 190 are common to the Eastern and Central Regions, 
and 74 to the Western and Central. But the majority of these 
may be reckoned as intruders which do not far exceed the 
boundaries of one or other of the two Regions to which they 
more properly belong. 
LINN. JOURN.— BOTANY, VOL. XXV. U 
