496 
PACIFIC SCIENCE, VoL VII, October, 1953 
Fig. 2. Newly exposed postglacial ash profile on Tanaga Island, August, 1950. Ash, humus, and clay strata 
are clearly defined. 
Unalaga, and Shemya, and parts of Kanaga, 
Tanaga, etc. These undoubtedly served as 
refugia for a number of hardy plants which 
were able to withstand the climatic rigors and 
the proximity of nearby ice caps on Adak, 
the northern parts of Kanaga and Tanaga, 
Amchitka, and Semisopochnoi. 
POSTGLACIAL PLANT SUCCESSION 
The ecological concept of succession is 
hard to apply to communities which continu- 
ally fluctuate in an unstable way so that a 
so-called climax grouping today may be 
shown later to have been only temporary. 
Under such conditions, succession is difficult 
indeed to define and more difficult to under- 
stand, especially if brief field studies attempt 
to name characteristic or dominant species 
as criteria of certain stages. In the Aleutians 
what one thinks of as a successional stage 
or climax community is often found to con- 
sist of characteristic species which occur 
widely in other habitats and make up a large 
part of the local flora. Moreover, the com- 
munity may be replaced by another during a 
relatively short period of time only to re- 
appear again in slightly different form. Hulten 
has called attention to this by remarking 
(1937^) that the Aleutian plant associations 
are very similar to those of Kamchatka, but 
in the Aleutians the associations are much 
less stable. There is considerable shift back 
and forth in species within communities, 
which results in the dropping out of various 
members. There are also abrupt shifts in en- 
tire plant communities which result in a rad- 
ically new vegetative character of a compara- 
tively large area. 
In postglacial times, as Aleutian plant life 
reinvaded the newly ice-free areas from va- 
rious refugia, there undoubtedly occurred ma- 
jor disruptive phenomena which seriously in- 
