1889. J BOTANICAL GAZETTE. 257 



In a trimorphic Pontederia discovered by Fritz Mtiller in 

 Brazil, the relative sizes of pollen as determined by Darwin 

 are 16 — 13 — 9, or 100 to 55. Our own species, therefore, by 

 reducing to the same terms, is as ioo to 54, which accords to 

 Pontederia cordata the greatest range for pollen that has yet 

 been found in any flower. 



The variation of pollen among dimorphic flowers does not 

 approach this, as the following short table of the most strik- 

 ing illustrations shows : 



Forsythia suspensa as 100 to 94. 



- Pulmonaria angustifolia " 100 " 91. 



~ Polygonum Fagopvrum " 100 " 82. 



~ Pulmonaria officinalis " 100 u 78. 



-Pkloxsubuiata " 100 u 75. 



-Primula vulgaris l< 100 " 71. 



•Primula veris " 100 •■ 67. 



When the contents of the pollen grains are taken into con- 

 sideration it is seen that those of the largest stamens of Pon- 

 tideria cordata are to those of the smallest as 8 to i. Mr. 

 Leggett 3 was aware of this remarkable difference at the time 

 of his study of the flowers, but he was in doubt as to the 

 smaller being perfect. To throw some light upon this point 

 six small wells in an artist's slab were filled with sugar solu- 

 tion, and into two of them the largest pollen was placed— one 

 lot from flowers with mid styles and the other from a short- 

 styled form. A corresponding set was made for each of the 

 - other sizes of pollen. The largest grains germinated with 

 the greatest promptness and vigor, and the tubes were natu- 

 rally much broader and in every way larger than with the 

 smallest sized grains, which as above stated contained onlv 

 one-eighth as much of substance. There was no observed 

 difference in pollen of the same set of stamens from flowers 

 with different lengths of styles, and all sizes were equally able 

 to germinate when sufficient time was given. The largest pol- 

 len tubes need to penetrate a style nearly an inch in length, 

 while the short styles are exceedingly short. As far as I know 

 it is invariably true with both dimorphic and trimorphic 

 plants that the longest grains are borne by the longest sta- 

 mens, and designed for the longest styles. Therefore, other 

 things remaining constant, the length of style may be an in- 

 dex of the size of the pollen. 

 JVew Brunswick, N. % 



2 Different forms of flowers on same species, p. 186. 

 'Torrey Bulletin, Nov. 1875. 



