The Geographical Distribution of the Frankeniaceae etc. 409 
microphylla Cav., Ic. VI from the Puna Region? and Anthobryum tetragonum 
and aretioides Phil. from the Puna Region. 
As to the Puna Region: 14. I believe the F. microphylla Cav. to be 
one of the Caryophyllaceae, because of its free central placentation, which 
is a character not found in Frankeniaceae. From its habit and floral struc- 
ture probably Pycnophyllum. Unfortunately I have not been able to see a 
specimen of this plant. 2. Anthobryum Phil. tirst described as a genus of 
Primulaceae is placed with Frankeniaceae by Reicue and Jonow (Flora 
de Chile I, 1896). If really one of this family it stands so near F. triandra 
in habit, morphology and distribution that we may, for the purposes of this 
article discuss them together. This leaves then, F. triandra Remy as the 
type for the Puna Region. Previous to this time, F. triandra has been 
supposed to be confined to the Punas of central Bolivia (Carangas, D’Or- 
bigny; Biacha near La Paz, Weppett Chlor. And.) but the discovery that 
Pycnophyllum sulcatum Griseb. Plant. Lorentz. p. 28, is F. triandra!) Remy. 
brings to light a much wider distribution, as well as much new material 
from which the characters of the species may be more exactly determined. 
The specimens called P. sulcatum were collected by Lorentz and HIERONYMUS 
near Congrejo at the northermost boundary of Argentine, and by Lorentz 
between Laguna-Blancaand Nacimiento Prov. Catamarca. 
The two species of Anthobryum Phil. were collected in the Puna of 
Tarapaca, A. tetragonum Phil., Polar Grande, A. aretioides Phil., Cerro de 
Copacoya. This indicates a very general distribution throughout the Puna 
Region. 
In its manner of growth F. triandra is one of the polster forming 
plants, common in the high Andean regions. WepnveLL describes it as 
forming »plaques arrondies et rayonnantes«. Lorentz speaks of it as 
occurring in »große Ballen auf salzigen Boden in den Hochthälern 
zwischen Nacimientos und den Laguna-blanca«. The surface of the polster 
or mat is a solid mass of vertical branches 3—5 cm long pressed closely 
together and appearing like mats of moss. The larger, main branches 
extend almost horizontally, and are entirely buried. In the exposed portions 
the internodes are very much reduced and the leaves very closely 
imbricated in four ranks. The flowers are necessarily developed singly 
and terminal. 
The floral structure is normal, i. e., like sect. Eufrankenia, except in 
the number of stamens. There are three in most flowers, but I have found 
four fully developed anthers in an unopened flower, five and even six 
4) I am very much indebted to the kindness of M. Poisson, curator of Herbarium of 
the Paris Museum, who through the friendly correspondence of Professor ScHUMANN 
has sent a small portion of the original plant collected by D'onsiGv, for my exami- 
nation. From comparison with this specimen Pycnophyllum sulcatum Griseb. is found 
to be F. triandra Remy. 
