Notes on Pollen. By Worthington G. Smith. 21 
commonly produce pollen-grains of the most diverse sizes. Fig. 59 
(Crinum) is typical for shape, but many grains in Narcissus are 
linear, while others are triangular, and both these occur in the 
same anthers with typical grains. 
These latter remarks hold good with the species and varieties of 
Tris as found in our gardens. Nothing can be made of the pollens 
of the mere garden forms, but that something may be learned from 
the study of the pollens of true species is shown by Fig. 61, which 
represents the pollen of J. zberica, and Fig. 62 the very differently 
formed pollen of I. Kempferi. These two forms, in good species 
of Iris, are both striking and constant. Gladiolus and Freesia 
have pollen similar in shape with Fig. 62. Fig. 63 belongs to 
Crocus aureus. 
A great deal has been written as to possible hybrids between 
our wild Geraniums and our garden varieties, especially with a view 
to get a blue strain of colour into the garden plants. As far as we 
know, all these attempts have proved abortive ; and from a study of 
the pollens in the Geranium family we are inclined to think that 
no such hydrids will ever be obtained. Fig. 64, Plate CLXX., is the 
pollen-grain belonging to our little wild Geraniwm sanguineum ; 
G. pheum is the same in size, whilst G. pratense is much larger. 
On the other hand, Fig. 65 represents the pollen of Pelargoniwm 
zonale. In Mr. Turner’s fine collection of Geraniums and Pelar- 
goniums the pollens are very similar with the latter. In the 
genus Erodium the pollen-grains are globular. In Oxalis, Fig. 66, 
O. acetosella, and Tropxolum, Fig. 67, 7. majus, the pollen-grains 
agree well with the Geranium family, and in this respect they add 
additional weight to the conclusions arrived at by Messrs. Hooker 
and Bentham, who place these two genera in the Geraniacee. 
The pollen-grains of the Umbelliferze are, as a rule, very cha- 
racteristic. That of Heraclewm sphondylium is shown at Fig. 68, 
Cinanthe crocata at Fig. 69, Siwm angustifolium at Fig. 70. 
The curious genus Hydrocotyle has pollen similar with Fig..71 in 
H. bonariensis, and with Fig. 72 in H. nitidula. The pollen of 
the common Ivy, Hedera helix, is shown at Fig. 73. Dr. Berthold 
Seemann proposed removing the genus Hydrocotyle from the 
Umbelliferze to the Araliaceze, but the characters of the pollen as 
seen In many genera and species of the two families hardly appear 
to us to support Dr. Seemann’s views. 
Fig. 74 is the pollen of the Bird’s-foot Trefoil, Lotus cornicu- 
latus. This may be considered representative of the Leguminose. 
Fig. 75 belongs to Cytisus laburnum, whilst Fig. 76 is Hrythrina 
eristagallia curious departure from the usual type. 
Of the Labiate family, Fig. 77 belongs to Nepeta violacea, 
Fig. 78 to Salvia patens, and Fig. 79 to Scutellaria Mocciniana. 
