208 LETTER XLVII. 



s s s 



p p 



s s s 



s s s 



This is of more importance for you to know than 

 you would at first suspect ; for it indicates that Alisma, 

 although formed with only six stamens, has a tendency 

 to produce twelve, and hence that it may belong to a 

 tribe, the prevailing number of whose stamens is 

 twelve, or even more ; and such is really the fact. 

 Even in Alisma itself, the stamens are in other species 

 nine, twelve, or even more ; and in Arrow-head they 

 are in all cases very numerous. Had the six stamens 

 of Alisma belonged to the two first whorls, you would 

 have had no reason to suppose, that although hexan- 

 drous, it might have immediate polyandrous affinities. 



The ovaries of the Water Plantain are about twenty- 

 four (that is, eight times three), arranged in a some- 

 what triangular manner ; they are quite distinct from 

 each other (^fig. 4.), and consist of a single cell, from 

 one side of the top of which the style arises in the form 

 of a curved horn, the upper end of which is broken up 

 into a stigma {fig- 4. a.). There is one ovule {fig. 

 5. a.) attached to the bottom of the cell, by a curved 

 stalk. 



The fruit (fig. 6.) is a triangular head of dry, one- 

 seeded nuts, furrowed at the back, and marked with 

 the base of the style on one side (fig. 7- «•)• 



From what has now been stated, can you tell whe- 



