86 



The Waterlilies. 



prolongation of the floral axis and are immersed in the excavated 

 receptacle with about half of the ventral suture exposed above and the 

 other half fused with the central axis. The back of each carpel is more or 

 less prolonged above the receptacle (except in N. gigantea) into a stylar 

 process. We shall speak of this as a carpellary style (Fig. 39, c). The 

 exposed portions of the ventral sutures are connivent or united into a 

 broad, round disc, cup, or funnel, the whole surface of which is stigmatic. 

 At the center of this the floral axis projects as a process (Fig. 39, a) which 

 is ovoid or nearly spherical in Eu-castalia, narrowly ovate in N. tetragona, 

 linear and rounded in Xanthantha, linear and acute in Hydrocallis, broadly 

 conical and obtuse in Lotos and Apocarpiae. It is of a dull white color. 



Fig. 39. 



Fig. 40. 



Fig. 38. Diagram of ovary (Brachyceras) in longitudinal section : a, axile process ; c, carpellary style ; 



n, air-oanals ; r, receptacle : a, stigma tic papllhe ; t, torus, with bases ot floral leaves. 

 Fig. 40. Diagram of ovary in transverse section : a, sy ncarpous ; b, apocarpous. 



The stigmatic surface (Fig. 39, s), is usually yellow and densely papillose. 

 The papillae completely cover the upper surface of the small rounded dorsal 

 projections of the carpels of N. gigantea, and in other species they extend 

 for a short distance up on the inner face of the carpellary styles. The 

 styles themselves are, as indicated, wanting in N. gigantea. In Brachy- 

 ceras they are short, stiff, orange-colored bodies, grading from mere points 

 in N. elegans to erect processes in N . flavo-virens or stout, incurved bodies 

 0.6 cm. long in N. zanzibariensis. In Xanthantha and Chamaenymphaea 

 they are more or less ligulate, and stand out horizontally in the basal part. 

 In Eu-castalia and Lotos they are 0.6 to 1.3 cm. long, ligulate and flexible, 

 and in Hydrocallis of about the same length, slender and terete below, with 

 a large clavate end. 



In the structure of the ovary Caspary (1865) was first to recognize two 

 fundamentally different types. In most of the old group Cyanea of De 

 Candolle and Planchon he found the walls between the cells of the ovary 



