CHAP. V. HETEROSTYLED TRIMOKPHIC PLANTS. 201 



styled, but not one short-styled. None of these plants 

 were in the least dwarfed. I examined, during the 

 highly favourable season of 186(5, the pollen of four 

 plants : in one mid-styled plant, some of the anthers of 

 the longest stamens were contabescent, but the pollen- 

 grains in the other anthers were mostly sound, as 

 they were in all the anthers of the shortest stamens ; 

 in two other mid-styled and in one long-styled plant 

 many of the pollen-grains were small and shrivelled ; 

 and in the latter plant as many as a fifth or sixth part 

 appeared to be in this state. I counted the seeds in 

 five plants (Nos. 26 to 30), of which two were mode- 

 rately sterile and three fully fertile. 



Plant 26. This mid-styled plant was freely and legitimately 

 fertilised, during the rather unfavourable year 1864, by numer- 

 ous surrounding legitimate and illegitimate plants. It yielded 

 an average, from ten capsules, of 83 5 seeds, with a maximum 

 of 110 and a minimum of 61, thus attaining 64 per cent, of the 

 normal fertility. During the highly favourable year 1866, it 

 was freely and legitimately fertilised by illegitimate plants 

 belonging to the present Class and to Class V.. and yielded 

 an average, from eight capsules, of 86 seeds, with a maximum 

 of 109 and a minimum of 61, and thus attained 66 per cent. 

 of the normal fertility. This was the plant with some of the 

 anthers of the longest stamens contabescent as above mentioned. 



Plant 27. This mid-styled plant, fertilised during 1864 in the 

 same manner as the last, yielded an average, from ten capsules, 

 of 99 '4 seeds, with a maximum of 122 and a minimum of 53, 

 thus attaining to 76 per cent, of the normal fertility. If the 

 season had been more favourable, its fertility would probably 

 have been somewhat greater, but, judging from the last experi- 

 ment, only in a slight degree. 



Plant 28. This mid-styled plant, when legitimately fertilised 

 during the favourable season of 1866, in the manner described 

 Tinder No. 26, yielded an average, from eight capsules, of 89 

 seeds, with a maximum of 119 and a minimum of 69, thus pro- 

 ducing 68 per cent, of the full number of seeds. In the pollen 

 of both sets of anthers, nearly as many grains were small and 

 shrivelled as sound. 



