102 EXPERIMENTAL EVOLUTION LECT. 



one of the most striking which can be quoted in 

 respect of floral morphological variation. Such varia- 

 tion is always present among different varieties of the 

 same species. For instance, C. E. Bessey l has investi- 

 gated the floral structures of different varieties of 

 apples, and while it is generally thought that no 

 difference obtains, he has detected considerable 

 differences in the form of the stigmas and styles, and 

 found that the pistil varies much in length, breadth, 

 hairiness, &c. And the proof thereof is seen in the 

 engravings which accompany his paper, and refer to 

 five well-known varieties of apples red Canada, 

 Talman Sweet, Rambo, Wagner, &c. 



The very smell of flowers is also subject to variation, 

 as Dalibard 2 showed by direct experiment nearly a 

 century and a half ago. He planted mignonette in 

 different soils, using seeds from the same mignonette 

 plant, possessing its well-known fragrancy. While 

 the seeds sown in rich garden soil became vigorous, 

 and were well perfumed, the seeds sown in sandy soil 

 produced plants which remained weak and small, and 

 had no perfume. It even seems that the latter 

 did not acquire any odour when transferred to rich 



1 Can Varieties of Apples be distinguished by their Flowers ? 

 American Nattwalist, vol. xx. 1886, p. 162. 



2 Observations stir le Reseda a few odorante, in Me moires de 

 Mathematiqucs et de Physique de V Academic des Sciences, 1750, 

 P- 95- 



