OF DIMOBPHIC AND TKIMORPHIC PLANTS. 401 



of seeds. Thirteen flowers were artificially and legitimately ferti- 

 lized by pollen from a legitimate long-styled plant, and yielded ten 

 capsules with an average of 95*6 seeds ; so that the application of 

 pollen from a legitimate plant added, as in the two previous cases, 

 to the fertility, but did not bring it up to the proper standard. 



-fi'^jp. XX. This long-styled plant, of the same parentage with 

 the last two mid-styled plants^ and freely fertilized in the same 

 manner, yielded an average from ten capsules of 69*6 seeds, vriih a 

 maximum of 83 and a minimum of 52 ; hence the plant produced 

 75 per cent, of the full number of seeds. 



Class IV. Illegitimate j^^^^ts from the short-styled form fer- 

 tilized hy pollen from the longer stamens of the long-styled form. 



In the three previous classes, plants raised from the three forms 

 fertilized by pollen from either the longer or shorter stamens of 

 the same form, but not of the same individual plant, have been 

 described. Six other illegitimate unions are possible, namely 

 between the three forms and the stamens in the other two 

 forms which do not correspond in height with the pistil. But 



^ 



I succeeded in raising plants from only three of these six unions. 

 From one of them, forming the present Class, twelve plants 

 were raised ; these consisted of eight short-styled, and four long- 

 styled plants, with not one mid-styled. These twelve plants 

 never attained quite their full and proper height, but by no 

 means deserved to be called dwarfs. The anthers in some of the 

 flowers were contabescent. One plant was remarkable from all 

 the longer stamens in every flower and from many of the shorter 

 stamens having their anthers in this condition. The pollen of 

 four other plants, in which none of the anthers were contabescent, 

 was examined \ in one a moderate number of grains were minute 

 and shrivelled, but in the other three they appeared perfectly 

 sound. With respect to the power of producing seed, five plants 

 {Exps. XXI. to XXV.) were observed : one yielded scarcely more 

 than half the normal number ; a second was slightly infertile ; but 

 the three others actually produced a larger average number, with 

 a higher maximum, than the standard. In my concluding re- 

 marks on Lythrum I shall recur to this fact, which at first appears 



inexplicable. 



Exp, XXI. This short-styled plant, freely and legitimately 

 fertilized during 1865 by illegitimate plants, descended from self- 

 fertilized long-, mid- and short-styled parents, yielded an average 

 Irom ten capsules of 43 see3s, with a maximum of 63 and a 

 niinimum of 26; hcnpe this plant, which was the one with all its 



