506 
spruces in such formations. Such tamarack swamps with central 
spruces are well developed in Cass county, Minnesota, especially 
in the vicinity of Gull lake, where I have studied them. 
In any given instance it is apparent that several considerations 
should enter into a judgment concerning the probable origin of 
a solid or spruce-centered formation of tamarack. If the forma- 
tion be a small one it may properly be questioned whether 
a stage with central moor had ever intervened. Furthermore, 
the contour of the bottom and depths of water in the original 
pond or lake would always have to be regarded in any generaliza- 
tion. For example, I have convinced myself that in some cases 
successions of muskeag openings with intervening tamarack arise 
from the filling of a lake with bars or reefs upon the bottom, the 
original positions of which is perpetuated by the lines of larches. 
Had the slant in the lake been strongly off-shore and the pond 
been deep in the middle a moor might then have been established 
at first with open water in the center, and only after a long lapse 
of years could the trees have crept out upon the peat. Had, on 
the other hand, the slant been slight and the pool sufficiently 
shallow at the center it might probably have filled without passing 
through any moor stage. Therefore some, at least, of the circular 
tamarack swamps with or without spruce cores could scarcely be 
regarded as necessarily derived from moors with tamarack or 
tamarack and spruce border-rings. This would, on the other 
hand, not hold true of all, and a consideration of the size and age 
of the trees from circumference to center will indicate that many 
of the solid tamarack swamps must have developed by this process 
of closing in a ring of timber upon a constantly diminishing moor. 
Exceedingly pretty examples of the dispersion and oblitera- 
tion of plants by such movements of tension-lines as here de- 
scribed may be observed. After the tamarack formation has be- 
come solid the sphagnum often manages to persist in little clumps 
and mats at the bases of trees where considerable rain water 15 
drained into a small area by the tree trunk with its radiating branch 
system above. Sarracenias, Vaccimums, Empeliums, the dwarf cor 
nel and other moor plants continue thus among the trees. The 4a 
_ Eniophorwns, Salices and many others do not seem to find the shade 
ocd grateful and commonly disappear altogether from the formation. 
