THE WATER-LILY TRIBE. 211 



attention to the rule of evidence in Systematic Botany 

 is required, in order to form a correct judgment. The 

 stem of this plant affords no precise character, either 

 one way or other, as between Exogens and Endogens. 

 Its leaves, moreover, are referable, as much to the 

 type of the one as of the other. Its flowers (Plate 

 XLIX. ^'fg. 1.) consist of about twenty-five, thickish, 

 oblong leaves, of a dazzling white colour, and the five 

 external ones are more or less green at the back, in 

 representation of a calyx ; these leaves grow gradually 

 smaller and smaller towards the centre, till at last 

 their points become callous and yellow ; at length bear a 

 pair of short, anther-lobes, in the room of the yellow cal- 

 losity (fig. 3.) ; these again narrow into straps, having 

 more perfect anthers at the points (fig- 4.), and finally, 

 next the ovary, shorten, diminish, and produce less 

 perfect anthers. What I have called anther-bearing 

 petals, are ob\'iously stamens. Do not suppose that 

 in this respect the Water-Lily ofiers an exception to 

 general rules ; in all cases the stamens are nothing but 

 contracted and altered petals provided with anthers ; 

 only in the Water-Lily the transition is gradual and 

 apparent, in others, it is too abrupt to be perceived. 

 The number of the stamens is about fifty, but it is 

 not fixed, nor indeed easily ascertained. 



The ovary is in a curious state (jig. 2.) ; instead of 

 beinf? either altoo-ether free, or altof^ether united with 

 the calyx, it has the lower floral leaves free from it, 

 and the upper united with it, so that the anther-bearing 

 petals or stamens grow from just below the stigmas. 

 It has ten or eleven cells, the partitions of which are 



r 2 



