283 
upon by civilization. The wet meadow vegetation along these 
streams when viewed at a little distance does not differ much in 
aspect from that of the dry prairies, except that it is taller, many 
of the shrubs being as high as a man's head and the herbs knee- 
high. The species in the two habitats are of course almost 
entirely different, but their numbers happen to be about equal. 
This prairie was originally bordered all around by forests, 
mostly of the oak type, but the border-line has been nearly 
everywhere obliterated by civilization. At some places south of 
Hicksville only a single row of fields at present intervenes between 
the prairie and the oak forest, but in most places the original 
boundary of the prairie could now hardly be determined within 
half a mile. Before the country was settled the oaks were 
presumably encroaching on the prairie from all sides. But in 
the few places where pine forests border the prairie I have never 
been able to determine which way the tension-line is tending 
to move. 
The cause of the treelessness of prairies has probably been 
discussed in geological, semi-popular, and non-botanical literature 
more than any other strictly botanical problem, and perhaps 
even more than it has by botanists, but no explanation has yet 
been found to fit all cases. Some of the partial explanations 
which have been suggested for the well-known prairies of the 
upper Mississippi valley will apply as well to the one under 
consideration, and some will not.* In a paper of such limited 
scope as this it would be out of place to attempt to review all the 
prairie theories, or even to mention all who have speculated on 
the subject; and only the briefest summary can be given here. 
Among the western prairie theories which will not apply on 
Long Island are deficient rainfall, extreme variations of tempera- 
ture, and impervious subsoil. Our prairie is subject to a good 
deal of grazing, frequent fires, strong wind, and excessive evapo- 
ration, like the western ones, but these factors are the result rather 
than the cause of treelessness, so that they could hardly have 
* The interesting papers of Shimek (Proc. Ia. Acad. Sci. 7: 47—59. pl. 4. 1900; 
Iowa Geol. Survey 20: 426—474. 1911; Bull. Lab. Nat. Hist. State Univ. Iowa 6: 
169-240. pl. 1—14. April, 191r) and Gleason (Bull. Torr. Bot. Club 36: 265-271. 
1909) should be examined in this connection. 
