Origin and Development of the Composites. 207 
columns. If, however, the genus shows three types of styles it is 
again counted £ in each column, and as it is only large genera 
which show three forms it is not altogether wrong to represent 
such a genus by a total of 1^ in the table. 
Type I is the characteristic style of the ordinary ray florets in 
all the tribes in which these occur and is noted in the table only in 
the Cichorieae where it occurs without exception in every genus. 
It may be observed that the totals for the tribes do not quite agree 
with those given previously (61, p. 466), but further study has 
revealed differences, distinctions and intermediate forms unrecog¬ 
nised in the previous analysis. 
The main points to be noted in Table VI are the following:— 
Type IV is the only one which occurs in all except the special¬ 
ised tribes; 
Among the tribes in which type IV occurs the Astereas, 
Heliantheae and Helenieae are mainly types VII and VIII, so that 
they can scarcely be regarded as primitive in this respect; similarly 
the Inuleae have many genera of type XII ; the Arctotideae and 
Mutisieae have a considerable proportion of types IX and XII. 
This leaves the Anthemideae, an obviously specialised group, 
and the Senecioneae, which, while showing a large number of 
genera with type IV, has quite a number of other types, thus 
showing the characteristics of a primitive plexus, i.e., a large per¬ 
centage of primitive forms with an extended variability especially 
in the centre of the plexus (in this case the Senecioninae), and a 
specialisation in the outer groups of the plexus (the Liabinae and 
Tussilagininae); 
The occurrence of type III in the Senecioneae and its exclusive 
predominance in the Liabinae and Vernonieae; 
The monomodal curve of complexity, 6, 34, 44, 12, in the 
Astereae and the passage of type Villa of the Homochrominae into 
type II of the Eupatorieae; 
The close relationship between types IV and XII and their 
occurrence in about equal numbers in the Inuleae ; 
The similar close relationship between types IV and VII and 
their occurrence in about equal numbers in the Helenieae; 
The almost exclusive predominance of type IV in the Nas- 
sauviinae. 
These and other less obvious points will be discussed later in 
conjunction with others elucidated by the analysis of the stamens. 
