, 
THE GENUS NOTOTRICHE. 207 
. species the hairs on the venter of the carpel and on the beak are similar, but in others 
the venter may be covered with small stellate hairs while the beak is beset by silky 
hairs, often very long and spirally twisted. 
All kinds of stellate hairs are met with on the fruits, and the nature of these, together 
with the length of the beak, are of great value for the determination of the species. 
'The seeds in all eases are irregularly reniform and often show a longitudinal dorsal 
furrow (Pl. 30. figs. 19, 23, 32). 
GEOGRAPHICAL DISTRIBUTION. 
The range of the genus appears to extend from the high mountains of Antisana, 
Pichincha, &c., on the borders of Ecuador and Colombia, as far south as the cordillera 
of Santiago in Chile, with an eastern extension through the south of Bolivia into the 
provinees of Salta and La Rioja (Sierra Famatina) in Argentina. 
The genus appears to reach its maximum development in the south of Peru and in 
Bolivia, but it is possible that a fuller knowledge of the flora of Northern Peru may 
necessitate a modification of this statement. 
One curious and unexplained fact in connection with the geographical distribution 
of Nototriche is the sharply defined northern boundary in the mountains of Ecuador. 
No specimens have been collected north of Pichincha, although the cordillera in 
Colombia would appear to afford conditions as favourable for the growth of these 
| plants as elsewhere in the range. It is possible that this marked break in the flora 
i in the north of Ecuador may be apparent rather than real and may be due to the 
fact that the country lying on the boundary between Colombia and Ecuador has 
not yet been properly explored: for in the region of the * knot" of the cordillera, 
about 1? north of the Equator, where the triple ridge of the Colombian cordillera passes 
into the Cordillera occidental and oriental of Ecuador, there exists an area but little 
known to botanists and without doubt of great interest in connection with the relation- 
ship of the Colombian flora to that of the Andes of Peru and Bolivia. 
From the tabular statement of the geographical distribution of the various species, 
Which is placed at the end of this section, it will be seen that certain species appear 
to be confined to very restricted areas. In some cases no doubt they are very local 
in their occurrence, while in others our ignorance of the Andine flora in certain 
regions makes them appear so, and this applies more particularly to the Andes of 
Northern Peru and of Chili and Argentina. 
As examples of species with a very restricted range the following may be cited : 
| N. turritella, which appears to be confined to the slopes of the volcano El Misti 
(Dep. Arequipa), N. argyllioides from Vincocaya (Dep. Arequipa) and N. sajamensis 
Which has been found only on the volcanoes of Sajama and Tacora. The species from 
. ihe Atacama desert, from Ecuador, and several others are also very local. 
On the other hand, several species, as can be seen from the Table, are known to 
have a considerably extended range: among these may be mentioned JN. flabellata, 
N. pedicularifolia, N. Mandoniana, N. Orbignyana, N. anthemidifolia, and some others. 
