﻿114 



Cincinnati Society of Natural History. 



creased gradually in number and became more and more sensitive 

 till they reached their present state of perfection. 



It may be, that when the development of the blade and the 

 petiole had proceeded a certain distance, the hairs of the blade 

 caught and retained insects. If this proved beneficial in any way, 

 then the enlargement of the petiole went no further, and the blade 

 hairs commenced to be modified in the new direction. Some of 

 them became the sensitive filaments; others l doubtless, came to 

 form the minute glands which Darwin says cover the upper surface 

 of the leaf, but no other portion.* 



The order Droseracece itself has been placed in quite different 

 positions in the sequence of orders by different authors. Bentham 

 and Hooker and Decaisne situate it between Crassulacece and 

 Hamamelacece, and consider it closely allied to Saxifragcce. De 

 Candolle places it between Violacece and Polygalacece. Dr. Gray had 

 it first between Violacece and Cistacece, but now has it between the 

 latter and Hypericacece, though allied to Violacece. Nuttall consid- 

 ered it closely related to Hypericacece; Wood put it near this last 

 order; Chapman between Cistacece and Pamassiacece (now with 

 Saxifragacece), and Lindley, in his " Vegetable Kingdom," in close 

 alliance with Papaveracece, Sairaceniacece and Violacece. 



Considering, then, that the Dioncea leaf is one form of a devel- 

 opment of which Sarracenia is the other, it is proposed to take 

 Dioncea from Droseracece and place it in a section [Dionceace) of 

 Sarraceniacece. Then some of the other orders should be arranged 

 differently. If a lineal system is to be observed, it seems the clos- 

 est approximation to a natural arrangement to place them as 

 follows : Nymphceacece, Sarraceniacece, Droseracece, Papaveracece, 

 Hypericacece. 



* In this connection the thought arises that the "quadrifid processes" of Aldro- 

 vanda and Utricularia may be the modified internal stellate hairs of some water plant, 

 just as the glands of Dioncea may be the modified internal stellate hairs of the hy- 

 pothetical water-lily. 



