Aug. 7, i 9 i6 Vegetative Succession Under Irrigation 755 



latent possibilities of this native grain. The advisability of the preser- 

 vation of this phase from year to year is a little doubtful. It might be 

 successful, but it would probably never give the yield to the acre that 

 the Deschampsia phase furnishes and is less nutritious, unless the later 

 phase contains a large percentage of Juncus balticus. Besides, it is not 

 improbable that after a year or two Hordeum jubatum would be able to 

 maintain itself in such abundance that it would ruin the wheat-grass 

 crop (PI. CII, fig. 2). Then, too, controlled irrigation is more expensive 

 than controlled flooding, so that the question of practicability enters. 

 Altogether, the evidence seems to indicate that, whenever possible, the 

 Agropyron phase should be encouraged for a year, or possibly two, in 

 order to eliminate the Hordeum phase (the constant result of controlled 

 flooding) and then should be allowed to pass into the Juncus-Carex or 

 Juncus-Carex-Deschampsia phase, according to the natural tendency. 



In starting the transformation all precautions should be taken to 

 eliminate the worthless Hordeum phase, a stage which, once established, 

 means from one to several seasons lost, with no advantage to the farmer. 

 In case wheat-grass was not present on the upland, controlled irrigation 

 tended to the development of a mixed association. 



Considering the additional expense, it is thought probable that in 

 most cases the Hordeum phase had better be tolerated, as the transition 

 may be more rapid to the Deschampsia phase. 



The rush-sedge phase (the next, the reader will recall, in the cvcle of 

 normal succession) should not be encouraged. It is doubtful whether 

 there is any escaping its presence, although a uniform and constant 

 water supply is probably the chief factor that tends to modify it into 

 the Juncus-Carex-Deschampsia phase. Care must be taken, lest with 

 too much water it revert to Juncus balticus. Whether evolved from 

 the Agropyron or the Hordeum phase, none of its chief components 

 possess the nutritive value of Deschampsia caespitosa. Carex lanuginosa 

 ranks the highest, but it is likely to form a subphase of its own. 



The two principal species of the Deschampsia phase represent extremes 

 in forage value. D. caespitosa is one of the most nutritious meadow 

 plants ' and Juncus balticus is one of the least nutritious. Hence, the 

 importance of maintaining D. caespitosa as the dominant plant of this 

 phase. This accomplished, the farmer has the best natural meadow the 

 region affords, which will average 1% tons to the acre and will remain 

 for years free from meadow weeds. 



If too much water is used, this phase becomes replaced by Juncus 

 balticus. The only species in this phase that in any way makes up for 

 the loss of Deschampsia caespitosa is Scirpus microcarpus. If the farmer 

 allows the Juncus phase to enter, he should consider that his meadow 

 has reverted. The next step is the bog, which would mean the destruc- 



1 Knight, H. G., Hepner, F. E., and Nelson, Avon. ( >j>. cit. 



