MARYLAND WEATHER SERVICE 195 



well marked plant formations are uplands, lowlands (meadow), fresh 

 marsh and swamp, and salt marsh. In determining this classifica- 

 tion there is apparently (1) a physical factor — difference in level, 

 bringing about a difference in water content, and (2) a chemical 

 factor — the proportion of salt in the substratum. 



Whether the minor differences in plant covering shown by areas 

 at about the same level are to be attributed to differences in the 

 chemical nature of the soils is the subject of a long standing dispute 

 which has waged with more or less acrimony since the time of Tlrar- 

 man (1849). The present writer has come to the conclusion that 

 the district under consideration is not well adapted to throw light 

 on this vexed question. One reason for this is that the soils of the 

 district have been so often worked over by the action of rivers, waves, 

 etc., that they no longer show marked chemical differences. The 

 period which has elapsed since the soil particles were parts of a solid 

 rock mass is so long that the more easily soluble constituents have 

 been washed out, leaving the soils similar, except in the size of 

 the particles. Only in a region where the soil is of a residual na- 

 ture, i. e., overlies the rock from which it has been derived by 

 weathering, can we expect to find peculiarities in chemical composi- 

 tion. Thus the layer of soil overlying limestone is more or less 

 impregnated with calcium salts. An exceedingly favorable locality 

 for the examination of the effect of the chemical factor has recently 

 been studied by Fernald,* who has compared the vegetation of 

 areas overlying potassic rocks (gneiss), calcareous rocks (limestone) 

 and magnesic rocks (serpentine) as found on Mt. Albert, Gaspe, 

 Quebec. This writer finds that while some species are indifferent, 

 a number are characteristic of one or another of the three sorts of 

 soil, and to a certain degree irrespective of drainage. In our dis- 

 trict the only soil which is of a residual nature is the Collington 

 Sandy Loam, which has been derived by weathering from the glau- 

 conite or greensand upon which it rests. Even in this case the soil 

 lias largely lost the potash which is a characteristic element of glau- 

 conite. f It may be said, however, that the Collington Sandy Loam 



•Fernald, M. L. The soil preferences of certain alpine and subalpine plants. 

 Rhodora, vol. ix, 1907. pp. 149-193. 



tSee Soil Survey Report for 1901, p. 186. 



