1913] COOPER—ISLE ROYALE 203 
causes for its occurrence have been suggested by MACMILLAN 
(38), SHaw (52), Davis (19), and ATKINSON (5). None of these 
explanations could be applied with certainty to the phenomenon 
as observed upon Isle Royale, but that offered by ATKINSON seemed 
most plausible, that is, that during the early development of the 
sedge mat the sphagnum was excluded by the shade cast by the 
near-by forest growth. Only after the bog substratum had built 
out beyond the shaded area did the moss become established upon it. 
< at i 33 
Fic. 40.—Tension zone between sedge and sphagnum-shrub societies: Carex 
limosa ay Chunanedatinie keep pace with the growth of the sphagnum by upward 
elongation; Raspberry Island. 
The third point to be considered is the manner of invasion by 
the sphagnum, including its relations to the sedges and other plants 
which it finally replaces, and to the shrubs which accompany or 
follow it. The sphagnum area spreads marginally, and at the 
same time colonies of young plants originate among the sedges in 
advance of the main mass. Several such colonies are shown on the 
map (fig. 36). By the coalescence of these and the solid mass 
behind, the sphagnum zone extends itself at the expense of the 
sedge mat. 
Certain plants of the sedge zone persist for some time after the 
sphagnum has gained control. They do this by a process of upward 
