55 



And, independently of the interest of the subject in ab- 

 stract botanical science, I have often found this cell-charac- 

 ter useful practically. Thus, in little accidental embarrass- 

 ments or disarrangements of the garden and seed-frame, 

 I have been able very easily, and with surprising certainty 

 and novelty, to pick out, simply by this diagnostic, all the 

 seedlings of exotic Onagracese, now so generally cultivated 

 and admired, from seedlings of other orders ; and seedlings 

 of Mesembryaceae, which had been sown in pots, and got 

 into perj)lexing confusion with other pots of seedling 

 Crassulacese— both orders of succulent flowering plants — 

 were as surely and quickly distinguished in like manner. 

 This kind of practical diagnosis also proved equally good 

 and convenient in plants at every period of their existence, 

 and even shapeless fragments of their dead and rotting 

 stems were thus plainly distinguishable. For example, a 

 reserve bed, in which had been planted, and intended for 

 removal when required, a Willow-herb, various Evening 

 Primroses, Phloxes, Campions, and Rockets, had got into 

 troublesome confusion, when nothing was easier than to pick 

 out alltheOnagraceEB solely by this raphidian character in the 

 roots, subterranean buds, and dead and rotten leaves or stems, 

 and at a time before growth had revived in the winter and 

 early spring. Only there was an abundance of raphidian 

 cells, beautifully marked in the tissue of the root and under- 

 grown stems of anotheiplantinthatbed, and certainly neither 

 a Willow-herb, an Evening Primrose, nor any other 'Onagrad.' 

 Here then was a puzzle ; some perplexing exception ? Not 

 at all. For after it had been put into a pot, and grown 

 there for further observation, this questionable plant turned 

 out to be a fine specimen of Woodruff, a species belonging 

 to that very order, Galiacca;, which I had already shown to 

 be characterised by these raphidian cells. Thus plants 

 may be known by their inmost structure ; and for that rea- 

 son by characters most eminently natural, but which use of 

 them is yet either unknown to, or neglected by, all syste- 

 matic writcn ; since in their works you will look in vain 



