i8o 



HARDWICKKS SCI ENCE-GOSSJP. 



four-angled, but this is scarcely correct ; there arc, 

 however, two distinct ridges on opposite sides of the 

 stem that change places between each pair of leaves, 

 just like the rows of hairs in Veronica chamadrys. 

 These ridges do not appear to be the vascular bundles 

 for the leaves above, because it may be seen on 

 reference to the drawing that they do not enter the 

 leafstalks but stop just between the stipules on the 

 opposite side of the stem, the next pair starting out of 



at their edges, and form a sort of three-cornered box 

 that holds some ten or a dozen stamens springing 

 from a central disk. The filaments are long, slender, 

 and erect, the anthers are pendulous from a sub- 

 globose connective, and the dehiscence is extrorse. 

 In a unisexual flower such as this the pollen should 

 be examined in order to decide whether it be of the 

 light smooth kind adapted specially for wind-carriage, 

 or whether it have any of those roughnesses that 



Fig. 159. — Early Leaf ot 



ditto. 



Fig. 15S.— Very young Male Plant 

 of Mercury. 



Fig. 163. — Stamen and 

 Pollen (greatly mag.). 



Fig. 161.— Staminate plant 

 (Nat. size). 



Fig. 164. — Sta;ntn. 



Fig. ifio. — Stem of 

 Mercuri.-»lis. 



Fig. 162.— Flower greatly enlarged. 



Fig. 165.— Early State 

 of Stigmas. 



v^>-i(f5^^-^' 



Fig. 167.— Ovary. 



Fig. 16S.— Section of Ovary. 



Fig. 166.— Pistillate Flower 

 (Nat. size). 



the axils of the leaves above to stop between the next 

 stipules, and so on. The flowers of Mercurialis are 

 di(Xcious and small. The tall axillary spikes of the 

 staminate flowers arc certainly conspicuous, but the 

 pistillate are far less numerous and hide themselves 

 amongst the leaves so that they are not very easily 

 seen. The staminate flower is a very simple little 

 thing, being content with one floral envelope com- 

 posed of three sepals, that in the bud just fit together 



characterize the pollen of entomophilous flowers and 

 that facilitate its removal by insects from one flower 

 to another. The pollen of Mercurialis is perfectly 

 smooth and most easily scattered, but then the pollen 

 of the Liliacese is also smooth and oval. The lily- 

 tribe is now greatly visited by insects, but Hermann 

 Miiller considers that the colour of the perianth must 

 have been originally greenish and unattractive, as the 

 dog's mercury is now. The long filaments are of 



