398 W. L. Bray. 
surrounds the embryo. 3. They retain their vitality at least for a period 
of years !). 
Anatomical. 
The Frankeniaceae offer some of the most extreme types of xerophytic 
structure, although generally growing near salt lagoons where more or 
less moisture is present in the soil. The presence of the salt is of itself 
sufficient to cause the plant to assume a xerophytic habit, as is well known, 
but aside from this, and notwithstanding the presence of moisture in the 
soil, the plants are exposed to all the extremes of temperature, intensity 
of sunlight and dryness of air characteristic of the most arid regions. The 
so-called extreme types have arisen as the result of the different methods 
of adaptation to these extreme conditions. This is expressed in the whole 
plant structure, but more particularly in the leaves, in whose form and 
function, oceurence on the stem, in the epidermal structure, arrangement 
of palissade cells and mechanical tissue, lie characters useful in distinguish- 
ing certain species. But any attempt to make a detailed classification of 
species based on the leaf anatomy, such for instance as proposed by 
Vesque?), results in bringing together species the most widely separated 
genetically. 
In the sections Eufrankenia and Toichogonia cosmopolita, i. e., those 
groups referred to in the following pages as the modern developement 
of the family, the leaves throughout, are of the same general type and occur 
in the same manner on the stem. In these and all other members of the 
family, excepting possibly F. triandra, the flower is subtended and par- 
tially enelosed by the last four leaves, which grow together to form a cup, 
thus leaving no interval between the two upper pairs of leaves. 
The presence of the so-called salt glands is one of the most noteworthy 
features in the special adaptations of the Frankeniaceae. These glands are 
found without exception in the epidermis of the assimilative structures of 
every species. Doubtless they furnish one of the most distinguishing 
characters of the family?). 
4) I have planted seeds which had been in the capsules of herbarium specimens 
for ten years, of which more than 50 % germinated. 
2) Contributions a l'histologie systematique de la feuille des Caryophyllinées. 
Ann. Sc. Nat. Ser. VI Tom. 45 p. 119. 
3) On the function of these glands see »Pflanzenfamilien« Ill 6 S. 290; 
VoLkENs, Flora der ägyptisch-arabischen Wüste; ManLorm, Berichte d. 
deutsch. botan. Gesellschaft 1887 S. 349; SraurL, Bot. Zeitung 1894 S, 439. 
Later I hope to offer the results of my study upon the mechanism of these glands. I can 
not forbear to suggest here, that the presence of these glands in precisely the same 
manner in all the genera of Tamaricaceae, excepting the North American Fouqueriae 
indicates even a closer relation of this family with the Frankeniaceae than has been 
previously suggested, — a fact which is given probably greater value when one considers 
