TRANSACTIONS OF THE SECTIONS. 133 
certainty of the plants selected for inoculation being themselves previously destitute 
of hyphe or apothecia. Of course, small portions from various places in a tuft of 
any given alga could be previously well examined, which, though if, indeed, found 
to represent the alga “ pure and simple,” would not render it absolutely conclusive 
that some other portion of the tuft might not already have been invaded by the 
“parasite.” However, having selected some plants for experiment they should be 
well inoculated with spores, and portions removed from time to time for examina- 
tion and experiment. If found satisfactory it would be interesting to try spores 
from the same and from different species, in order to see the result, and whether the 
seeming fixity of the forms and the apparently extreme choiceness of the parasites 
be true or not, or ultimately whether the theory itself be true or not. Whether, 
for the time being, the truth of the new theory be previously assumed, or its un- 
tenability be presupposed, would matter very little, if only the suitable opportunity 
and ready field of operation were at command of the observer. It woul seem as 
if in this way only can either presupposition be justified or negatived. 
On the Form of Pollen-grains in relation to the Fertilization of Flowers. 
By Aurrep W. Bennett, F.L.S. 
Although a common form of pollen-grain not unfrequently runs through a 
whole group of plants, yet more often the form is found to be adapted to the par- 
ticular requirements of the species in respect of its mode of fertilization, and varies 
even within a small circle of affinity. In those plants which are fertilized by the 
agency of insects, there are three general modes in which the form of the grain is 
adapted for the purpose. We have, first (and this is by far the most common 
form), an elliptical grain with three or more longitudinal furrows, as in Ranunculus 
Ficaria, Aucuba japonica, and Bryonia dioica; secondly, spherical or elliptical, 
and covered with spines, as in many Composite, Malvacez, and Cucurbitacee ; 
and thirdly, where they are attached together by threads or a viscid excretion, as 
in the Fuchsia, Evening Primrose, and Richardia ethiopica. In those plants, on 
the contrary, which are fertilized by the agency of the wind, as most grasses, the 
Hazel, and Populus balsamifera, the pollen is almost perfectly spherical when dry, 
unfurnished with any furrows, and very light and powdery. The genus Viola 
furnishes two very markedly different forms of pollen-grain: in one, the section to 
which V. canina and odorata belong, they have the ordinary elliptical 3-furrowed 
form, and every point of the structure of the style and stigma is favourable to fer- 
tilization by bees; in the other, the section to which V. tricolor belongs, the grains 
are very much larger, and either hexagonal or pentagonal, and the style and stigma 
are adapted for fertilization by very minute insects, such as Thrips. In all species 
of the order Cruciferze at present examined, the pollen has the most common form. 
Pringlea antiscorbutica, the “ Kerguelen’s Land Cabbage,” has been shown by 
Dr. Hooker to be in all probability wind-fertilized, from the following considera- 
tions :—the absence of petals, the absence of honey-glands, the exserted style, the 
stigma being covered with long Lg aly and the apparent entire absence of winged 
insects in Kerguelen’s Land. The form of the pollen supports the same view, being 
very minute and perfectly spherical, extremely different, therefore, from every other 
known plant of the order. In the cowslip and primrose there is a uniform dif- 
ference in size between the pollen-grains from flowers belonging to the two dimor- 
phic forms, that of the short-styled being always considerably larger than that of 
the long-styled form. 
On the Embryogeny of certain Species of Tropeolum, 
By Professor Drcxson. 
On an Abnormality in Chrysanthemum leucanthemum. 
By Professor Dickson. 
