1910] ANDREWS—TWIN HYBRIDS 199 
The sepals are only about one-half the length of the calyx tube, 
whereas in laeta they are about three-fourths of its length. This 
difference is very evident on comparison. The sepals of velutina are 
also much more hairy than any of the other forms. 
The cells of the outside epidermis of the sepals of velutina are 
about uniform in size with those of laeta, but the walls are much 
stronger. The cells of the inner epidermis are more uniform in size 
than in /aeta. 
The petals are more deeply indented at the top than in /aeta, are 
somewhat smaller, and light yellow. The cells of the inner and 
outer epidermis are more nearly isodiametric than in /aeta, and their 
cell walls are much more zigzag, and thickened at the angles. 
The stamens are the same length as the stigmas. In this respect 
they differ strikingly from Jaeta and O. Lamarckiana, but resemble 
O. biennis. As in laeta, they open some of their anther cells at 
least, on sunny days, but on damp and rainy days they remain closed. 
I have noticed instances, as in Jaeta, where some of the anther cells 
never open, no matter how ideal the conditions. I have found it 
necessary to use considerable force in order to cause even thin cross- 
sections of the anthers to open their cells. This, as in /aeta, is due 
to the fact that the tissues at the point of opening never separate, but 
remain more or less completely grown together. This non-opening 
of the anthers of Oenothera forms, however, is not confined to those 
mentioned above. I have visited another primrose garden at Haar- 
_ lem which is conducted by Mr. A. R. ScHouTEN and Mr. J. Jeswrer. 
In this garden are grown, among other forms, O. scintillans and O. 
sigas, and Mr. Jeswrer kindly informed me that they often find it 
necessary to open the anthers with a needle in order to free the pollen. 
Much of this pollen, as in our twin hybrids, is not sterile, but in some 
cases would never escape naturally from the anthers. Mr. JESWIET 
tells me that of the types they have at Haarlem all of the anthers of 
O. scintillans must be opened with a needle to obtain the pollen, 
and also many of the anthers of the form O. gigas. 
The ovary of velutina exceeds all the other forms as to density 
of pubescence, 
s The capsule resembles that of /aeta, but is curved near the top. It . 
'S strikingly different from Jaeta in regard to its denser pubescence. 
