184 NATURAL HISTORY BULLETIN 



More frequent visits to the several localities selected for the 

 table, and an examination of larger areas in each locality, would 

 no doubt have filled many of the vacant places, for there are 

 variations in the abundance of many of the species from year to 

 year some apparently disappearing for a time, and different 

 (even closely contiguous) parts of the same prairie show varia- 

 tions which may result in part from differences in topography 

 or which may be due to the accident of distribution. 



But with the exception of the species appearing as invaders, a 

 more complete discussion of which follows, the flora of any of the 

 types of prairie here discussed will be practically contained with- 

 in the limits of the foregoing table. The exact lists are not al- 

 ways the same, but such variation as exists is not due to differ- 

 ences in the types of prairies, but may occur within any one of 

 the areas. The variation is well illustrated in the table. 



The invasion of additional species which do not properly be- 

 long to the flora of the prairie takes place chiefly from four dis- 

 tinct sources: 



1. From, wooded areas. This may usually be observed only in 

 a narrow belt along the border between forest and prairie and 

 is of little significance, as forest plants seldom extend beyond the 

 scrubby border and that only during moist years. It would 

 probabl.y be better to say that here the prairie invades the for- 

 est, as none of the inesophytes of the forest extend to the open 

 prairie Avhile along the drier borders of groves there is a distinct 

 scattered invasion of prairie plants. 



An illustration of the mixing of species in these border belts 

 is given in the list of plants from the border between evapora- 

 tion stations (1) and (2) at Missouri Valley. 



2. From prairie hogs. The plants which invade the prairie 

 from this source occur on dry prairies only occasionally, and 

 then in places which at least during some part of the season, or 

 during some seasons, are comparatively moist. Thus Habenaria 

 hie pilar i glottis seemed to have disappeared in large part from 

 some sections of the state during the dry years of the middle 

 nineties, occurring only here and there in what seemed to be 

 dry places, yet the restoration of the prairie bogs which oc- 

 curred in the wetter years which followed caused a reappear- 



