II LYTHRARIE^ 193 



It was also shown that perfect fertility can only be 

 obtained by fertilising each form with pollen from stamens 

 of the corresponding length. Thus the long-styled form 

 is naturally fertilised by pollen from the long stamens of 

 the two other forms ; but it can be so, though imperfectly, 

 by its own two sets of stamens, and by the shorter 

 stamens of the two other forms. It can, therefore, be 

 fertilised, to use Darwin's expression, " legitimately " in 

 two ways, and " illegitimately " in four ways. The 

 same is the case with the other two forms, so that 

 eighteen modes of union are possible, of which six are 

 natural or "legitimate," twelve are "illegitimate" and 

 more or less sterile. This case is, therefore, indeed 

 most complex. The long-styled and short-styled forms 

 are almost sterile if " illegitimately " fertilised ; the mid- 

 styled produces about half as much seed as if " legiti- 

 mately " fertilised. Darwin suggests -^ that the tri- 

 morphous condition of this plant may be advantageous, 

 because if it were dimorphous only there would- be but 

 an equal chance in favour of any two plants being of 

 different forms, and therefore capable of self-fertilisation ; 

 whereas, being trimorphous, the chances are two to one. 

 In the Cowslip and Primrose, where large numbers of 

 plants grow together, this, he thinks, would not be so 

 material. However this may be, the stigma and the 

 two groups of stamens appear to correspond with the 

 three divisions of the body (viz. the head, thorax, and 

 abdomen) of the bee, Cilissa melanura, by which it is 

 almost exclusively fertilised. 



The genus Lythrum is remarkable for the great 

 differences existing between different species. For 

 instance, L. OrcBfferi, like L. Salicaria, is trimorphous ; 

 while L. tliymifolia is dimorphous, and L. hyssopifolia 

 is homomorphous. 



The fruit is a capsule which opens along the centre 

 of each valve. The seeds are numerous, small, plano- 

 convex, yellowish, or testaceous, and nearly smooth. 

 They float in water. Cilissa melanura, as already 



^ Journ. Linn. Soe. viii. (1864). 



