ARRANGEMENT OF THE PARTS OF THE FLOWER. 



457 



As, for the most part, relations of position are alone discussed in 



tlie following remarks, so the figures or symbols referring to 

 number, union, or disunion of parts are in general omitted as 

 unnecessary for the present purpose. Unless otherwise stated, 

 the formulae here given have been always constructed from per- 

 sonal observation- 

 Superposition exists in a large number of diverse families. It 

 may suffice to cite a few in which the phenomenon is most fre- 

 quently met with or in which it is most apparent, e. g. : 



Berberidaceae. 



Menispermaceae. 



Cistaceae. 



Malvaceae, 



Sterculiaceae. 



Tiliaceae. 



Hypericaceae. 



Ternstroemiaceae. 



Guttiferae. 



Sauvagesiaceae 



Sabiaceae. 



Simarubaceae. 



Vitaceae. 



Rhamnaceae. 



Celastraceae. 



Olacaceae. 



Loranthaceae ? 



Sapotaceae. 



Primulaceae. 



Myrsinaceae. 



Plumbaginaceae 



Lauraceae. 



Iridaceae. 



Burmanniaceae. 



Eestiaceae. 



II. False Superposition. 



Before considering those cases where true superposition exists, 

 mention must be made of instances of the apparent or false super- 

 position which arises from the decussation or crossing of very 

 closely crowded whorls. Some of the cultivated varieties of 



Q 



Hume's 



Blush," afford good examples of this. In these Camellias it often 

 happens that flowers of two different forms are produced. In 

 the commoner case the sepals and petals are disposed in an imbri- 

 cate or spiral manner, the stamens and carpels being replaced by 

 petals. The arrangement is on the ^ type ; and the outline of the 

 flower is circular. In the other and rarer form the petals and 

 petaloid stamens are so arranged as to form a six-angled, six- 

 rayed, or star-shaped flower. The petals are so evenly placed 

 one over the other, that at first sight it appears as if the petals 

 did not alternate with one another, but were superposed to each 

 other. "When, however, the flower is carefully pulled to pieces, it 

 may be seen that the law of alternation is not really interfered 

 with, and that the superposition is apparent rather than real. If 



2l2 



