330 EXPERIMENT STATION RECORD. 



given. The total population of the transect of this marsh is given as 52,377, 

 Carex species constituting 63 i)er cent, grasses 24 per cent, Carex stricta 40 per 

 cent, and Calamaf/ro.sds canademis 18 per cent. The growth habits of the plant 

 association are described and discussed. Some space is devoted to the dis- 

 cussion of geographical distribution of the species and of the distribution of 

 marsh meadows in North America. 



The range of the depth of the water table of the area under observation was 

 from a few inches to 3 ft. or more. The results of a series of pot experiments 

 to test the influence of the graund water level and to determine whether there 

 is a correlation between the depth of the water table and the root and shoot 

 development are given. Alsike clover, redtop, timothy, and bluejoint plants 

 were grown in peat soils in which the water tables were maintained (1) at 

 the surface of the soil, (2) 4 in., (3) 8 in., (4) 12 in., (5) 24 in., and (6) 30 

 in. below the surface of the soil. Each si>ecies was grown in each cylinder. 

 The result " shows that under the conditions given the roots of all the species 

 tested can extend below the water level and make a vigorous growth in the 

 water saturated soil. The best general plant development was in cylinders 3 

 and 4 where the roots had a zone of soil 8 and 12 in. thick above the water 

 table. In these, root development was more extensive than was the case in 

 cylinders 5 and 6 where there was much deeper water tables. There is, how- 

 ever, a marked uniformity in the root development in cylinders 2, 3, 4, 5, and 

 6, irrespective of the wide differences in the water level. 



" In cylinder 1 all the plants made a dense mat-like growth of roots in 

 the surface 2 in. of soil below which there was relatively little root gi'owth. 

 While redtop and bluejoint sent roots down into the water to a depth of 17 

 and 20 in. these roots did not branch much. In the other cylinders there was no 

 marked zonal distribution of the roots although there were relatively more 

 roots in the upper layers. . . . 



" The greatest differences between the four species in any one cylinder was 

 seen in No. 1. The bluejoint and redtop showed a marked ability to develop 

 with the water level at the surface but this was unfavorable for their best 

 development. 



" In cylinder 6 none of the roots extended to the water level, which was 

 21 ft. below the surface, and the growth that each species here made was sur- 

 passed in at least one other cylinder. This indicates that the water level suit- 

 able for the maximum development has been passed and that no better growth 

 can be expected with a still lower water table." 



The conclusions were as follows: 



" There is sharp correlation between root and shoot development. There 

 is poor development when the water level is constantly at the surface. 

 There is a marked development of roots in water-soaked soil when the water 

 level stands at 4, 8, and 12 in. A high water table (not above 4 in.) is not 

 prohibitive to the growth of these plants, but rather is favorable for the best 

 growth." 



A bibliography of 25 titles is appended. 



A study of the development of the most important meadow grasses in the 

 first vegetative year, E. Sinz {.Jour. Landic, 62 {191^). No. 3, pp. 197-233, 

 pis. 4)- — This describes and gives some I'esults of work carried on at Gottingen 

 in pot cultures with 11 common grasses. Tabulated data on the plants thus 

 grown show the green and air-dried weights, percentage of dry substance and 

 of nitrogen in both the tops and the roots, and the ratio of tops to roots during 

 the first season's growth. In some cases the amounts of nitrate of soda 

 applied as a fertilizer were varied. 



