8 
EOTAL HORTICULTUBAL SOCIETY. 
take on the pallor or din^iness of decay. "Where the anthers 
emerge of a purple hue, and change from that to brown, it will 
probably be found that they have discharged their pollen while 
still included in the flower. These observations, made without 
any reference to the question now in hand, require to be renewed 
and tested ; and in them, as in all that precedes, I am open to 
correction. I have not noticed in any grasses the yellow anthers 
with yellow pollen, and the purple anthers with green pollen, 
such as are combined in the same flower in many genera of Ly- 
thracese, TiliacejB (Mollia, Luliea, &c.), Leguminosae {Dimor- 
pliandra, &c.), and of some other orders. 
Of grasses with bisexual flowers, there are two ways in w4iich 
the ovary may be fertilized, viz. either by the pollen of its own 
flower (closed or open), or by that of other flow^ers, after the 
manner of the diclinous species. In the latter case, the pollen 
may be transported by the wind, or in the fur of animals (as I 
have observed the seeds of Selaginellas in South America), or in 
the plumage of birds. The agency of insects has not been traced in 
the fertilization of grasses, but may exist. The little flies I have 
seen on the flowers of grasses seemed bent on depositing their 
eggs in the nascent ovaries, but may also have aided in cross- 
fertilization. In the Amazon valley grasses are often infested 
by ants, which, indeed, leave nothing organic unvisited throughout 
that vast region ; and they also, I think, cannot help occasionally 
transferring grains of pollen from one flower to another. 
The flow^ers of palms and grasses agree in being usually small 
and obscurely coloured, but contrast greatly in the former being 
in many cases exquisitely and strongly scented, whereas, in the 
latter, they are usually quite scentless. The odour of palm- 
flowers often resembles that of mignonette ; but I think a whole 
acre of that " darling " weed would not emit more perfume than 
a single plant of the great Pan-palm of the Eio Negro (ITauritia 
Carard, Wallace). In approaching one of these plants through the 
thick forests, the sense of hearing w^ould perhaps give the flrst 
notice of its proximity, from the merry hum of winged insects which 
its scented flowers had drawn together, to feast on the honey and 
to transport the pollen of the male to the female plants ; for it is 
chiefly dioicous species of palms that have such sweet flowers. 
The absence of odoriferous flowers from the grasses seems to 
show that insect aid is not needed for effecting their fecunda- 
tion, but does not render its accidental concurrence a whit less 
unlikely. 
