THE FLOWERING OF THE G.RASSES. 63 



microscopical in size, no bigger than the dot of an i, and 

 so thin and filmy that they look very much like a midge's 

 wing. So far as I can tell they are of no use at all to 

 the plant as it now stands : they remain there as mere 

 functionless rudiments, apparently on purpose to let us 

 see the essential kinship between the grasses and the 

 lilies. For these are two out of the three original petals, 

 dwarfed almost beyond recognition, but still fairly to be 

 identified by means of intermediate links. As to the 

 third petal, which ought to be within, on the same side 

 as the two calyx-pieces which are united into one, that 

 has disappeared altogether., crushed wholly out of exist- 

 ence between the grain and the calyx. The fact is, the 

 one-sided arrangement of the little flowers on the spike, 

 necessary in order to let their stamens hang out freely 

 to the wind, has distorted all the inner half of the 

 blossoms much as the habit of lying on one side has 

 distorted and blanched the lower half of the sole or the 

 flounder. But we have numerous intermediate forms 

 still existing which lead us from the true lilies, with 

 their colored petals, through the wood-rushes, whose 

 petals are thin and brownish, to certain sedges in which 

 they have become mere rudiments, and to the grasses in 

 which only two of them can be distinguished at all. 

 However, one group of very large and tall grasses, the 

 bamboo tribe, still keeps all three of its petals ; it is 

 the smallness of our English kinds which has made the 

 third and innermost disappear. The stamens are still 

 all right ; they keep up their original number of three ; 

 while in the fruit two of the cells have become abortive, 

 for a reason which we will presently consider, and only 

 one remains to produce a little corn -like grain. Our 

 spike of meadow-brome contains several dozen such 

 very tiny and degenerate lily blossoms. 



