162 
.Note. 
cell-walls, situated in the micropylar prolongation of the sac. The embryo has no 
suspensor. In this respect and in its globular form it closely resembles the young 
embryo of Penaea A The endosperm is copious. In many of the South African repre- 
sentatives of the Proteaceae only a small percentage of the flowers set seed. In Protea 
Lepidocarpon only about 30 per cent, of the flowers in the capitulum are fertile, while 
in some of the other species the percentage is lower. Mr. Phillips found that of the 
1260 flowers on five capitula from different plants of Protea grandiflor a, Thunb., 
1067 contained aborted ovaries. A careful search for fertile seeds of this species has 
not yet met with any success, though seedling plants are occasionally established. In 
Proiea mellifera , Thunb., it appears that less than 50 per cent, of the ovaries become 
aborted. 
Irregular protuberances of the nucellus — the origin of which calls for further 
investigation — penetrate the tissue of the integuments. These are presumably 
haustorial in function. The nucellar tissue disappears very soon after the appearance 
of the embryo. 
The tetragonal pollen-grains show some peculiarities. The two divisions of the 
pollen mother-cell are rapid. Here also the reduced number of chromosomes is 
twelve. After the first division of the mother-cell a thick transitory wall appears 
between the nuclei. The microspores are in tetrads by the time the megaspore 
mother-cell is dividing. 
These investigations were carried out in the Botanical Laboratory of the South 
African College, Cape Town, at the suggestion and under the supervision of Professor 
H. H. W. Pearson. The work was done during the early part of 1908, before the 
Winter Session at Edinburgh commenced. 
A. J. BALLANTINE. 
University of Edinburgh. 
1 Stephens, E. L. A Preliminary Note on the Embryo-sac of certain Penaeaceae. Annals of 
Botany, vol. xxii, p. 329. 
