T HE recent impetus which has been given to research on pasture improvement in the United States by various educational andresearch agencies has stimulated an interest in improved methods for quantitative and qualitative analyses of pasture swards. A number of methods for evaluating pastures have been described in American and foreign literature, the applications of which are confined to an analysis of particular types of grassland and to the gathering of specific information. Hanson (5) 3 has compared different methods for determining botanical composition of the western North Dakota prairie. His conclusions apply, however, only to the type of vegetation characteristic of the region. Several of the methods now in use in the British Isles and New Zealand are adaptable to the types of sward which are found in the Great Lakes and New England regions. These methods differ in principle which makes it difficult to select, without considerable study, the most useful one for the particular pasture conditions. Some are meritorious in measuring productivity, composition, or quality of the sward with no one method apparently possessing all the qualities that may be desired. The need for an accurate and rapid quantitative method which will measure, without personal bias or error, the relative frequency and productivity of the pasture components in the types of sward found in the humid region of the eastern and middle western states has resulted in the pasture methodology herein reported. It is not possible to collect sufficient data in one season to warrant final evaluation of all the methods. The apparent demand for information as to the efficiency and practicability of the different methods seems to justify a preliminary report of the research now in progress. PROCEDUREThe study of methods is being done on two of a series of fields in a comprehensive pasture project on the University of Wisconsin Experiment Station farm at Madison, Wise. (1). One of the fields is treated annually with commercial nitrogen and the other is left untreated as a check. The two fields, each consisting of 4Y. acres, were sown in 1934 to the following mixture: 14 pounds per acre Kentucky bluegrass (Poa pratensis), 10 pounds timothy (Phleum pratense), 4 pounds redtop (Agrostis alba), 2 pounds white clover (Trifolium repens), 3 pounds alsike clover (Trifolium hybridium), and 3 pounds red clover (Trifolium pratense). A series of 30 randomized counts for each method were taken on each field during May and 'Cooperative investigations of the Wisconsin Agricultural Experiment Station and the Division

T HE relationships between soil and the roots of plants greatly affect plant growth. The roots of plants are important also in conservation and improvement of the soil. The soil environment is, in fact, more readily altered by man than the above ground environment in which the aerial portions of plants develop. Nevertheless, studies of the below ground parts of plants are much less extensive than those dealing with the aerial portions. Investigations by Weaver (ro, r r), 3 his associates and students (r2, 13, I4, 15), conducted mostly in pedocal soils, and by Booth (3), Farris (4), and Sprague (8) in podzolic soils show that the roots of the same species differ when grown in different soils. Accordingly, this study was undertaken to determine the effect of various soil conditions on the rate and extent of root growth of some cultivated grasses and legumes. Materials and MethodsThe study was conducted under both greenhouse and field conditions during 1947 and I948. Smooth bromegrass of Canadian origin (Bromus inermis L.), timothy (Phleum pratense L.), Grimm alfalfa (Medicago sativa L.), and medium red clover (Trifolium pratense L.) were grown in pure stands. The alfalfa and medium red clover seeds were inoculated with suitable bacterial cultures before planting.At intervals, the roots of the plants grown in the field or the greenhouse were separated from the soil by washing with a hose on a screen having a 2 mm mesh after which they were dried at I40°F to constant weight. GREENHOUSE STUDIESThe greenhouse in which the studies were conducted was maintained at a temperature of 65°F. Fluorescent light was supplied to all plants for 7 hours daily in addition to normal daylight. The soil was passed through a .)-4'-inch mesh screen prior to planting. Water sufficient for vigorous growth was applied whenever necessary.

K ENTUCKY bluegrass, Poa pratensis L., is probably the most important source of pasturage in nearly all the north, humid region of the United States, in eastern Canada, and in northern Europe. During the past 8 to ro years increased efforts have been made by investigators in these areas to develop superior varieties of bluegrass for use as pasturage. Thus far no improved strains are grown commercially to any· considerable extent in North America. Nilsson~Leissner and Nilsson (2) 3 have reported that five varieties of Kentucky bluegrass were released in Sweden prior to 1940. Observations indicate that improved European varieties have no particular superiority in Wisconsin.Myers and Garber (3) reported results of studies of replicated clonal plots from 8 r selected plants of Kentucky bluegrass. The plots were overseeded with white clover. Following two years of clipping, simulating pasture treatment, yields were determined in the third season. Significant differences among strains for yield and competitive ability with white clover were noted.Hayes and Thomas (r) found that different clones of Kentucky bluegrass obtained from r so pastures and waste places in Minnesota showed wide differences in yielding ability when grown in clonal plots. There was good agreement in yield between 1941 and 1942 from clonal plots established in the fall of 1939. In another trial, Hayes and Thomas (r) grew 56 selected lots of Kentucky bluegrass from seed in replicated plots "8 by 8 feet in size. Their results show that five strains were superior in total production to the commercial check at the s% level of significance. The seedings were made in the fall of 1943 and results are based on yields obtained in 1944. While several other workers have isolated apparently superior types for pasture or turf purposes, reports of experimental pasture tests of such isolates are unavailable.The large number of selections which need to be tested in a breeding program involving the improvement of bluegrass and the small amount of seed which is usually available almost necessitates making initial selections of superior appearing plants from spaced plantings

O UR knowledge relative to the influence of fertilization and height of cutting on the yield and composition of Kentucky bluegrass is not as complete as would be desired. Because of this fact an experiment to cover these phases in part was begun in 1928 on the East Hill University farm at Madison, Wis., on an area which had for at least 30 years previously been in bluegrass sod. Results for a 7-year period, 1928 to 1934 are presented. The studies were made on a Miami silt loam type of soil sloping gradually to the north and were confined to (a) yields as influenced by various fertilizer, cutting, and irrigation treatments; (b) the effect of various fertilizer treatments on the chemical composition of the grass from the standpoint of its nutritive value; and (c) the seasonal recovery of nitrogen. 3 The IO plats referred to above were selected from a group of approximately 8o on which various fertilizer and cutting studies were conducted by Professor Mortimer. The results and conclusions presented for these ro areas are typical and representative of those which would have obtained for all plats if they had been included in this report. Moreover, the results for each plat may be considered replicated from year to year during the 7-year period these trials were in progress. The general trend from year to year on the same plat and the significant differences as between various plats appears to be ample justification for this assumption.

No abstract