The Botany of Iceland - 01.12.1942, Qupperneq 17

THE TARAXACUM-FLORA OF ICELAND
243
always be formed? Does it not mean that the work will be impossible,
and that the number of species will grow indefinitely ? No, investiga-
tions of the genus Taraxacum in the Scandinavian countries and espe-
cially in Finland (see G. Marklund 1940) have proved that the
number of species is in fact limited; and it must be supposed that in
nature the forming of new species does not take place to such an
extent as has been proved by Thorv. Sörensen for garden-
cultures, or that the large number of new species are not able to survive
in the struggle for existence in places where they are not tended, as in
an experimental garden.
The fact, however, that Taraxacum-species can form new, constant
species closely resembling the mother-species in their whole character,
calls for the greatest care in the identification of new-found species.
Complete conformity with individuals from different areas must be
required for their correct reference to the same species. If the conformity
is not complete, there is every reason to believe that the plants belong
to different species. My experience during my cultivation of the
numerous Icelandic species, has indeed proved this. I am thinking
especially of the spontaneous species; for the species favoured by cul-
ture the case is different. As mentioned already, the Danish Vulgaria-
species introduced into Iceland were so deformed, so untypical, probably
on account of the moist climate, that they were hardly determinable
in the field. Seeds of these plants, however, when cultivated in Den-
mark, produced quite normal specimens, easily determinable.
The experience I have gained from my studies of the Icelandic
Taraxacum-species may be summed up as follows:
1. In the identification of spontaneous species perfect conformity
must be required, before the determination can be considered certain;
only groups of isoreacting individuals, i.e. groups of individuals which
react uniformly under the same life conditions, but differently from
all other individuals, can be referred to the same species.
2. In the identification of introduced species, especially species,
introduced into localities outside the natural area of distribution, a
determination may be right, even if perfect conformity is not present.
But only by cultivating the species concemed is it possible to obtain a
perfectly safe determination.
3. Cultivation of the species to be identified under the same cir-
cumstances, is in all cases the best way to obtain a safe result; this test,
the isoreaction test, is the only decisive method.
4. And finally, investigators cannot strongly enough be wamed