Aug. 7,i9i6 Vegetative Succession Under Irrigation 753 



Astragalus Bodinii. The species of Carex have a high water requirement 

 and belong to the cycle resulting from the continuous use from the first of 

 controlled flooding. The best development of the vetch was coincident 

 to the Agropyron phase; otherwise its growth was limited to isolated, 

 often more gravelly, elevated places of relative dryness. The exact 

 conditions that produce these subphases, as they may be called, are not 

 understood and represent only one of the many points which have yet 

 to be fully worked out. 



The Deschampsia phase is by far the most stable yet considered. 

 Once established, controlled flooding seemed to satisfy its water require- 

 ment so exactly that its character fluctuated but little from year to year. 

 The areas longest under water, however, gave evidence of two possible 

 further changes; either a gradual increase in the abundance of rush 

 (largely Juncus balticus) or a gradual increase in the number of kinds of 

 plants. The former case meant the ushering in of the Juncus balticus, or 

 wire-grass, phase the presence of which typifies swamp conditions and is 

 the forerunner of the bog. Besides wire-grass, the hay harvest from 

 this phase at Rock Creek contained some at least of the following plants : 



SWAMP SPECIES TENDING TO INVADE WETTER PARTS OF MEADOW 



Calamagrostis hyperbored Lange. 

 Glyceria borealis (Nash) A. Nels. 

 Eleocharis palustris (L.) R. and S. 

 Scirpus microcarpus Presl. 



Habenaria viridiflora Cham. 

 Rumex occidentalis Wats. 

 Ranunculus reptans L. 



Of these plants only two are ordinarily of sufficient importance to 

 deserve comment. Eleocharis palustris, naturally an inhabitant of pond 

 margins, finds in the wire-grass phase an environment to which it is well 

 suited and it soon becomes an important factor. Its value as forage is 

 fully as great as that of the rushes. 1 Scripus microcarpus, like the pre- 

 ceding, is indicative of marshy conditions. It is scattering in its distri- 

 bution, but its presence is to be noted with satisfaction, as it is one of 

 the few members of this phase that possess a considerable forage value. 1 



The other possibility, the gradual increase in the number of different 

 species, is yet to be considered. This change is even slower than the 

 other. For one thing, there is now a firmly established turf in which 

 any plant finds it difficult to secure a foothold. Finally, however, now 

 here, now there, some local variation in conditions or some factor which 

 escapes our notice permits the invasion of many species, such as those 

 listed under the heading "Meadow plants appearing at some later sea- 

 son" (p. 758). Only three of these are noteworthy: Cicuta occidentalis 

 is a poisonous parsnip which for its best development needs the conditions 

 that produce the wire-grass phase. Pcdicxdaris crcnulata is no doubt the 

 worst weed of the natural meadows, but strangely enough does not seem 

 to bother the made meadows until late. If it appears, the Carex Gayana 



1 Knight, H. C, Hepner, F. E-, and Nelson, Aven. Op. ciL. 



