30 HARPER: PLANT POPULATION OF MICHIGAN 
can be filled only by means of a floating mat of vegetation growing 
out from the edges.* And plants not connected with the soil, 
whether epiphytic or floating, cannot grow very fast on account of 
the dearth of mineral plant food. The peculiarities of bog vegeta- 
tion have been commonly attributed to the acidity of the water, but 
the acids in bog water are not a fundamental part of the environ- 
ment, but are derived from the vegetation, so that offering such 
an explanation is only reasoning in a circle. 
_ The size of the basin in which the pond or bog is located also 
influences the vegetation through the amount of seasonal fluc- 
tuation of water (as was observed in Florida a few years agof). 
For considerable fluctuation of water in or above the ground hastens 
the decay of dead vegetation and the liberation of the food in it, 
and peat is formed best in places where the water-level is nearly 
constant.§ 
The occurrence of typical slow-growing (sometimes erroneously 
called xerophytic) bog vegetation in flat, slowly drained areas as 
well as in deep stagnant basins is probably to be explained largely 
by the fact that seeping water, having just emerged from the 
ground, is considerably cooler in summer than shallow standing 
water.|| The dense growth of conifers in such places still further 
protects the water from the heat of the sun, after they are once 
established. Another important factor is that in a perpetually 
saturated soil the lack of aération restricts the availability of the 
mineral plant food in the soil, particularly the potassium com- 
pounds. Of course it may be said that the soil of a marsh is also 
perpetually saturated, which is true enough; but in a marsh, 
whether stagnant or estuarine, the water has been exposed to 
evaporation much longer than that in a seepage or spring-fed 
swamp, and thus the soluble salts in it are more concentrated. 
* Probably the best description of this process is that by C. os Davia: in Rep. 
Mich. Geol. Surv. 1906: 125- 907. 
{ For a summary of bog theories, in which however the question of mineral nu- 
trients is hardly considered at all, see G. B. Rigg, Plant World 19: 310-325. ‘Oct. 
[Nov.], 1916 
t See Tocris TI; 225-234. 1911; Ann. Rep. Fla. Geol. Surv. 6: 202, 203. figs. 
46-48. 1914. 
§ See Ann. Rep. Fla. Geol. Surv. 3: 211. IOI, 
|| The temperature of springs, other than thermal springs i is — very close 
to the average annual temperature of the locality where they 
