360 Copeland . — The Mechanism of Stomata . 
suggested that by progressive thickening and a feasible 
change in the hinges a stoma might easily pass from the 
Amaryllis to the Helleborus or Achillea type. But this has 
not been found to occur. The stomata of Quercus imbricaria , 
Dent aria diphylla , Lamium ample xicaule and Salvia lyrata , 
and the Azalea already figured, all show their characteristic 
thickening before the guard-cells separate. 
The ultimate ground for the difference between the pre- 
dominant types of stomata in early spring and in summer lies 
of course in the need of protection against drought in summer. 
It does not follow that pronounced xerophytes can never have 
stomata of the Amaryllis or Medeola type. The stomata of 
Equisetum are nearest the Medeola type, but are protected by 
their position. The occasional difference between the stomata 
of the stems and leaves of the same plant may be partly due 
to the different demands on the smaller number present on 
the stems, and should be correlated with the larger and 
longer epidermal cells common on stems, instead of being 
ascribed altogether directly, as by Westermaier, to the 
tensions on the tissues of stems. This difference has already 
been described on Euphorbia dentata. Specialized subsidiary 
cells appear in cross-section only on the stems, or much better 
developed there, on many plants ; for instance, Lobelia pub erula> 
Campanula Americana i Eupatorium coelestinum and Chelone 
glabra . The surface-view of the Chelone stem shows several 
cells in contact with the dorsal wall of each guard-cell, in the 
walls between which are thin places, analogous to hinges in 
that they are to remove resistance to the movement of the 
guard-cells. The leaf of Lobelia shows a similar structure, 
but less clearly. On the stem of Lamium the walls between 
the subsidiary cells strike the dorsal wall nearer the end, 
instead of about midway as on the leaf. On Ipomoea , Convol- 
vulus , Campanula , and Salvia (the species as in the list above) 
the stomata of the stem show a tendency toward the Amaryllis 
type, in the thickening. 
It would be ridiculous to undertake to tell in detail why 
each plant makes use of its particular stomatal mechanism. 
