200 
James Small . 
was in organic connection, while Coulter (15) and Merrell (49) 
report no such fusion. Observations by the writer prove that in 
some cases, at least, no organic connection exists, but the possibility 
of fusion in some species is not excluded and further research is in 
progress on this point. 
Pollination. 
Camerarius (10) was the first to conduct experiments in 
artificial pollination and to study pollination in the Composite. 
Miller (50) was the first to observe pollination by insects, but our 
knowledge of pollination in the Compositae is founded on the later 
work of Kolreuter (38 and see Chap. Ill) and Sprengel (64). The 
latter noted the advantages of the capitulum and other characters 
in the Composite, and dealt with Tussilago, Silybum, Centaurea, 
etc. The next advance was about 40 years later when Brongniart (8) 
started a controversy on the function of the collecting hairs of the 
Campanulaceae which led to a re-examination of the Compositae 
for comparison by Hassall (31), Wilson (72), Todd (65) and 
Trelease (68). About this period also the study of pollination pro¬ 
gressed rapidly under the stimulus of Muller’s work (51), and the 
modern aspect of the subject together with the literature may be 
found in Knuth’s compilation (36) based on Muller’s contributions. 
The biology of Helianthus annuns and other Compositae was the 
subject of several papers (12, 22, 27, 53), and Darwin (16) notes 
the sterile, pollen-presenting style which occurs in the male 
florets. 
A valuable contribution was made by Juel (34) who dis¬ 
tinguishes four chief types of polleu-presentation: (1) pollen 
presented on the tip of anther-tube, (2) pollen presented on the 
outer surface of the style and stigma, (3) pollen presented on the 
inner surface of the corolla lobes, (4) pollen presented directly from 
the anthers (anemophilous). Types 3 and 4 are comparatively 
rare, while type 1 is divided into two subsidiary types on the 
length of the male stage of anthesis, and type 2 has four forms, the 
Gaillardia type, the Eupatorium type, the Cichoriese type and the 
Arctotis type, which differ in slight details. 
The distribution of the sexes in the capitulum formed the 
basis of Limit’s divisions of the Syngenesia, and this part of the 
subject has been studied exhaustively by Uexkull-Gyllenband 
(69) whose conclusions may be summarised as in Pig. 4 from her 
“ Generalschema fur die phylogenetische Bliitentwicklung.” In 
