The Botany of Iceland - 01.12.1945, Qupperneq 99

THE VEGETATION OF CENTRAL ICELAND 441
Thus Carex rufina is very conspicuous in analysis XIII, 11, but C.
microglochin in XIII, 12, and Festuca rubra and Juncus balticus in
XIV, 8. Thc following moss species are recorded from this association:
Philonotis fontana, Calliergon turgescens, Meesea trichoides, Gymno-
cybe palustris, Leptobryum piriforme, and Mnium cuspidatum. These
mosses show most clearly that we are hert; concerned with a moist soil;
this is, indeed, saturated by spring-water, which irrigates the whole
area in which the association was found. It was only met with in Hvan-
nalindir, where it extends over a fairly large area in the upper part of
the oasis. As pointed out, the soil is damp, but the layer of soil is very
thin. As far as I can make out, the association is here at the limit of
the shrub heath. Anderson & Falkin their paper record a “wil-
low community” from Hvannalindir indicated by C on the accompany-
ing sketch map. According to this, it is the same community which
I here call a Salix association rich in moss. Still there is a difference
in the list of species. Anderson & Falk list Salix phylicifolia and
S. lanata as dominant species, while S. glauca is not found in their
list, which is quite contrary to my observations, according to which
Salix glauca is the dominant species everywhere, while S. lanata is of
more scattered occurrence, and S. phylicifolia is not found. The last-
mentioned species occurs, indeed, in Hvannalindir, but only on the
banks of brooks, where it may form fairly big shrubs here and there;
probably, however, it is the hybrid S. phylicifolia X S. glauca more
frequently than the main species which is found there. I can only
explain this disagreement on the assumption that a confusion of species
has taken place in the paper of these authors. Otherwise Anderson
& F a 1 k’s zone C comprises both this association and associations 2
and 10.
The three associations of the shrub heath dealt with so far are so
closely related to the jaðar formation that I was in doubt as to which
of these formations they should be referred. I have placed them under
the shrub heath because they have both a higher A and Ch percentage
than the jaðar vegetation in general, and because they cover larger
continuous areas than the jaðar formation; and finally, what decided
the case was that in these associations Salix glauca dominates every-
where physiognomically, which it never does in the typical jaðar forma-
tion. In Amardalur associations 1 and 2 are interspersed with jaðar
associations, but it was distinctly visible that the shrub heath occurred
in the highest-lying and driest localities surrounded by the jaðar belts,
so it was quite evident that we were concerned with two formations.
The Botany of Icland. Vol. III. Part IV.
30