ARRANGEMENT OF THE PARTS OF THE FLOWER. 471 



ing the position that stamens ordinarily fill, and which they do 

 really occupy in the male flower. 



8. Torsion of the axis. — This between two successive whorls, or 

 torsion of the constituent elements of the whorl, may bring about 

 an appearance of superposition. Such cases are not uncommon in 

 the case of leaves ; but I do not recall any instance in the case. of 

 the normal flower, except in certain cases of obvious monstrosity, 

 as where the parts of the flower are displaced by a twisting of the 

 axis, conjoined or not with a tendency of the parts to pass from 

 a verticillate to a spiral arrangement*. 



V. Objects of the Diversities in mode of Arrangement. 



The alternate position seems to be in most instances a provi- 

 sion for ensuring space to closely packed organs, so as to prevent 

 undue amount of pressure on the growing tissues. In cases where 

 the development of the stamens lags behind that of the petals, the 

 question of space is not of so much moment as under other cir- 

 cumstances. * It may often be an advantage to the plant to have 

 the stamen concealed in the cavity of the petal, as in Blumen- 

 bachia among Loasacese, Seaphopctalurn in Byttneriace®, and in 

 very many other instances, where such an arrangement may be 

 beneficial either for protection's sake or in various manners con- 

 nected with fertilization. 



In other cases the existing arrangement may be a relic of some 

 ancestral condition which has now lost its significance ; or, on the 

 other hand, could we but read it aright, it might offer a forecast 

 of some future development. 



VI. Illustrations of the Relative Position of the Parts of Perianth 

 and Androecium in various genera o/'Tiliacese and 01acace*e. 



1. Tiliacecd.—hx illustration of the manner in which the relative 

 position of the parts of the flower may vary within the limits 

 of the same Natural Order, 1 may refer to some of the genera of 

 Tiliaceae and Olacacea* which have come under my own ob 

 tion. Other instances of a similar character relating to Malvales 

 are specified in a paper published in the Journal of this Society \. 



In Tiliaceae in general the stamens are indefinite in number, but 



* Some instances of this, especially amongst Orchids, are given in 4 Vegetable 



Teratology/ p. 91 &c. 

 + "On some Points in the Morphology of the Malvales," Joan. Linn. 

 *>1. x. p. 18. See also Mast, in Olivers 4 Flora of Tropical Africa/ i. p. 177 



Soc, 



vaceoe, Stereuliacea?. Tiliaceae 



MKN. JOURN. BOTANY, TOL. XT. 2M 



