Origin and Development of the Composite. 7 1 
barbs, but Coreopsis aristosa, Mich, is described (23) as having no 
pappus at all or aristae with downwardly or upwardly directed 
bristles. These forms abound in the same locality and the probability 
of hybridisation is negatived by the absence of Bidens from the 
districts where these forms occur. 
Dispersal Mechanisms. 
Hildebrand (30) considers the pappus at some lehgth in its 
relation to dispersal by wind and by animals. In his general 
memoir (32) also he gives figures and details of a number of hooked 
and pappose fruits in the Composite having previously developed 
the subject of dispersal by hooks in a special contribution (31). 
Hooked fruits in the Compositae are also considered by Huth (41). 
Yapp (71) and others have described various Composite fruits in 
which the pappus is glandular and sticky. 
The presence of a pulvinus at the base of the pappus either in 
the form of a ring or isolated at the base of each seta has been 
noted by Schenk (59), Taliew (65), Haberlandt (22), Hirsch (33), 
Steinbrinck (64) and Yapp (71). 
Taliew distinguishesthree types of mechanism for the spreading 
of the pappus in dry conditions, I—the Lactuca type with an annular 
pulvinus, II—the Tussilago type with no pulvinus but unequal 
thickening of the pappus hairs, III—the Cirsium type with a 
pulvinus to each seta. The first type was recorded by Schenk and 
Taliew in all the Cichorieae and most of the Tubuliflorae examined. 
The second type in which each seta curves on drying occurs 
chiefly in the Tussilagininae. The third type is characteristic of the 
Cynareae, and experiments by the writer show that in the case of 
Centaurea imperialis the mechanism is very sensitive to the degree 
of the humidity of the atmosphere. With a Relative Humidity of 
•78 the pappus remains closed, or, if opened by drying in a warm 
test-tube, closes rapidly when brought into the air. With the 
R. H. -77 it remains open for some time after previous drying and 
with the R. H. -75 it opens spontaneously and remains spread at an 
angle of 45" for an indefinite period. 
There are various other contributions to this part of the subject 
but they are concerned more directly with fruit dispersal and will 
be dealt with in a later chapter. 
Pericarp Structures. 
Capus (10), Hanausek (24-25), Portheim (56), Lavialle (45) and 
Loose (46) have recorded the detailed structure of the pericarp in 
