Origin and Development of the Composite p. 231 
Conditions being favourable these forms continued to exist, and, 
requiring something to replace the lost pappus, developed a 
substitute by rendering the corolla of the ray florets persistent. 
When the disc florets were fertile the awns were present or, if the 
loss mutation extended to all the flowers in the capitulum, tuber- 
culations of the pericarp in the achenes of the disc were sometimes 
developed for dispersal purposes to replace the lost stuctures. 
These cases and the original development of the achenial hairs 
into a pappus are examples of the general rule that lost structures 
are not re-developed but are replaced by something similar 
developed from parts which are still present. 
The other end of the Central Sea in the Mediterranean region 
was also active geologically during the middle Miocene. The 
continued evolution of the semi-desert region, with dryness and 
strong insolation as predominant characteristics, produced the 
spiny group (Scolyminae) of the Cichorieae. The changes in climate 
resulting from the alternating sinkings and uprisings of considerable 
stretches of the land surface were doubtless responsible for a 
considerable proportion of the diversity of forms produced in the 
Cichorieae at this time. 
One of the most interesting events of the middle Miocene was 
the origin of Helichrysum from the Gnaplutlium line in South Africa. 
Senecio and Gnaphnlium had by that time become established in 
this region, and finding the climate in these similar latitude srather 
like that of their Andine home, as far as aridity, insolation, etc. 
were concerned, these two genera multiplied abundantly. Gnapha- 
lium , more accustomed to living near the snow-line, yielded earlier 
than Senecio to the difference in altitude, and Helichrysum was 
produced. This genus was so suited to the conditions on account 
of its epharmonic origin that its numbers soon exceeded in a very 
marked degree those of its parent. 
Probably on account of its lower physiological differentiation 
Senecio was slower in yielding, but in the succeeding period, the 
upper Miocene, it gave Ursinia by the changes which usually 
resulted from the attainment of a more favourable environment, 
where food material was abundant and wind-dispersal rather uncer¬ 
tain. The ray florets became neuter and the reduction in the 
number of fruits per capitulum was carried further by the sterilisa¬ 
tion of the inner disc florets. At the same time fusion of the pappus 
setae was accompanied by a development of the fimbrillae of the 
receptacle to form flat setae. 
