226 
James Small. 
By the end of the lower Oligocene all but the more restricted 
groups of the Composite had been initiated, so that the considerable 
variety in the Cypselites of that period, even in the Mediterranean 
region, is what might be expected. 
During the middle Oligocene a considerable amount of differ- 
entation took place, chiefly along the lines already laid down. 
Thus Inula was developed in the Mediterranean region by a series of 
orthogenetic changes all of which had already been initiated in 
Senecio. These changes included the spreading of the stigmatic 
papillae all over the inner surfaces of the style branches, the elon¬ 
gation of the anther tails, the greater aggregation of the cauline leaves 
into the calyculus, and the increased development of bilabiate 
florets. This last change was probably due to the moist climate, 
where food material was abundant both for the greater vegetative 
growth shown by Inula and for the extra floral material required to 
produce numerous rows of bilabiate corollas. In the Mediterra¬ 
nean also, the same moist environment led to a development of the 
beaked achene with which Lactuca had been experimenting at an 
earlier date and the Scorzonera group was originated. 
Events were meanwhile progressing in the Mexican region, 
where considerable geological changes were occurring. Similar 
conditions to those of the Mediterranean prevailed for a time and 
Evigevon was developed from Aster by an increased outlay in 
bilabiate corollas and a decrease in the number of protective in- 
volucral leaves, but, as the western end of the Great Central Sea 
continued drying up, the climate, at least locally, became much 
drier and a proportion of the Erigerons began to find conditions less 
favourable. The bilabiate florets were therefore reduced to 
filiform female florets and at the same time the protective action 
of more aggregated cauline leaves was required. Erigeron in these 
local dry regions thus became Conyza. 
Before this time the moist conditions of the wooded parts of 
the Mexican region had induced Epilanthes to reduce the protective 
calyculus to a few leaves, and in the absence of opportunities for 
wind-dispersal the pappus and other achenial hairs ceased to be 
developed. Other changes had led to the production of a number 
of genera now classed as Vei besininae, but these particular develop¬ 
ments produced Siegesbeckia. Continued existence under such 
favourable conditions gave the innate economy an opportunity to 
act by sterilising a number of the disc florets, thus reducing the 
number of fruits produced by each capitulum. This economy, 
