DEFOLIATION OF OATS AND PEARLMILLET AS RELATED TO HERBAGE YIELD Bjr CARL SOREN HOVELAND A DISSERTATION PRESENTED TO THE GRADUATE COUNOL OF THE UNIVERSITY OF FLORIDA m PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA January, 1959 1^0 A SRI. CULTURAI. tlBRARY UNIVERSITY OF FLORIDA 3 1262 08552 2844 ACKNOWLEDGEMENTS The writer wishes to express his appreciation to Doctors G. B. Killinger and D, E, McCloud for serving as chairmen of the Supervisory Coiranittee and guiding the re- search program; to Doctors W, 0., Ash, G, K, Davis, J. H, Davis, D. B* Duncan, G. R, Noggle, and D, 0, Spinks for serving as members of the Committee; to V, Chew for ad- vice on the experimental design; to Dr. H. C. Harris for assistance in laboratoiy analyses; to Mr. P, E, Loggins for furnishing sheep in the grazing study; and to Dr. F^ H, Hull for financial encouragement. Especial thanks are due Mr. L. Burgess for serving as field assistant and enduring the irritations of collecting seemingly innumer- able forage samples. The author is indebted to his wife for much encouragement and exercising unlimited patience throughout the study. Lj/O-] l^ TABLE OF CONTENTS Page LIST OF TABLES iv LIST OF FIGURES x INTRODUCTION 1 REVIEW OF LITERATURE 3 Perennial Grasses ••••••••••••••• 3 Effect of plant height at time of cutting or grazing 3 Effect of stubble height 6 Root growth as affected by defoliation 11 Annual Grasses ••••••••••••••••• 13 OATS 1^ Materials and Methods • .••• IB Effect of cutting treatments, varieties, and irrigation, winter 1955-56 IS Effect of cutting treatments, varieties, and irrigation, winter 1956-57' 24 Effect of sheep grazing treatments, winter 1956-57 29 Results ...••.« 33 Effect of cutting treatments, varieties, and irrigation, winter 1955-56 33 Effect of cutting treatments, varieties, and irrigation, winter 1956-57 53 Effect of sheep grazing treatments, winter 195o-57 65 PEARLMILLET 70 Materials and Methods 70 Effect of cutting treatments and irriga- tion, 1955 70 - ii - Effect of cutting treatments, varieties, row spacings, and irrigation, 1956 72 Effect of cutting treatments and row spacings, 1957 76 Effect of cutting treatments and nitrogen levels, 195^ 79 Effect of planting date, 1956 and 1957 62 Results •••••••••••••••••••• 86 Effect of cutting treatments and irriga- tion, 1955 86 Effect of cutting treatments, varieties, row spacings, and irrigation, 1956 92 Effect of cutting treatments and row spacings, 1957 109 Effect of cutting treatments and nitrogen levels, 1956 121 Effect of planting date, 1956 and 1957 137 DISCUSSION . . . . , 147 Oats 147 Pearlraillet .................. 151 SUMARY AND CONCLUSIONS 158 LITERATURE CITED 162 BIOGRAPHICAL SKETCH 170 - iii - LIST OF TABLES Table Page 1, Season total yields of irrigated oats as influenced by date when clipping ceased, variety, and clipping treatment, winter 1955-56 34 2, Season total yields of unirrigated oats as influenced by date vrtien clipping ceased, variety, and clipping treatment, winter 1955-56 35 3, Analysis of variance: Forage yields of - oats as influenced by irrigation, variety, date vdien clipping ceased, and clipping treatments, winter 1955- 5o 36 4, Summary of oat forage yields as influenced by irrigation, variety, date when clipping ceased, and clipping treatment, winter 1955-56 37 5, Analysis of variance: Forage yields of three irrigated oat varieties as affected by clipping, winter 1955-56 42 6, Effect of clipping treatments on forage yields of three irrigated oat varieties, winter 1955-56 43 7, Crude protein content of oat forage as affected by clipping management, winter 1955-56 48 S, Pounds per acre grain of irrigated oats as influenced by date vdien clipping ceased, variety, and clipping treatment, winter 1955-56 49 9, Pounds per acre grain of unirrigated oats • as influenced by date when clipping ceased, variety, and clipping treatment, winter 1955-56 50 10. Summary of grain yields of oats as affected - iv - Table Page by date vrfien clipping ceased, variety, and clipping treatment, winter 1955-56 51 11. Analysis of variance; Grain yields of three oat varieties as affected by clipping treatment and date ifrfien clipping ceased, winter 1955-56 52 12 • Pounds per acre oven dry forage of three irrigated oat varieties as influenced by cutting treatment, winter 1956-57 54 13. Pounds per acre oven dry forage of three unirrigated oat varieties as influenced by cutting treatment, winter 1956-57 55 14, Analysis of variance; Effect of cutting treatment on forage yield of three oat varieties, irrigated and unirrigated, winter 1956-57 53 15 • Analysis of variance; Effect of cutting treatments on fo rage - yields of irrigated Arlington, Floriland, and Seminole oats, winter 1956-57 59 16, Root and crown yields of three oat vari*- eties as affected by clipping treatment, winter 1956-57. 61 17, Analysis of variance; Effect of cutting treatments on root and crovm yields of three irrigated oat varieties, winter 1956-57 62 Id, Air temperatures in degrees Fahrenheit -be- tween rows of irrigated Floriland oats, winter 1956-57 63 19, Analysis of variance; Pounds per acre oven dry forage harvested from quadrats in sheep grazing trial on Arlington oats, winter 1956-57 67 20, Pounds per acre oven dry forage harvested from quadrats in sheep grazing trial on Arlington oats, winter 1956-57 68 21, Fertilization schedule for Starr millet - V - Table Page clipping study, 195^ SO 22. Pounds per acre of nitrogen applied to Starr millet date of planting test, 1956 84 23. Pounds per acre actual nitrogen applied to Starr millet date of planting experiment, 1957 65 24. Total season forage yields of common pearl- millet as affected by cutting treatment and irrigation, 1955 ^ 25. Analysis of variances Effect of cutting treatment and irrigation on the forage pro- duction of common pearlraillet, 1955 ^ 26. Effect of cutting treatment on the crude protein content of irrigated common pearl- millet, 1955 91 27. Effect of cutting treatments on- root yield of irrigated common pearlmillet, 1955 93 2d. Analysis of variance: Effect of clipping treatments on root production of irrigated common pearlmillet, 1955 94 29. Summary of irrigated pearlmillet forage yields as affected by row spacing, variety, and cutting treatment, 1956 96 30. Analysis of variance: Forage yields of irrigated pearlmillet as affected by row spacing, variety, and cutting treatment, 1956 97 31. Forage yields of irrigated pearlmillet as affected by row spacing, variety, and cut- ting treatment, 1956 9^ 32. Number of clippings obtained under various row spacings and cutting treatments of two pearlmillet varieties, 1956 100 33. Analysis of variance: Effect of cutting treatment on the season yield of Starr mil- let grown in 19- and 3 S- inch rows, 1956 101 - vi - Table Page 34, Effect of cutting treatments on the season yield of Starr millet grovm in 19- and 3S-inch rows, 1956 102 35, Crude protein content of irrigated pearl- millet forage as affected by variety, row spacing, and cutting treatment, 195o 106 36 » Effect of variety and row spacing on pounds per acre of oven dry forage from unirrigated pearlmillet cut vdien 30 inches tall to leave a four- inch stubble, 1956 107 37. Analysis of variance: Effect of variety and row spacing on forage yield of unirri- gated pearlmillet, 1956 108 3d, Pounds per acre of oven dry Starr millet forage as influenced by row spacing and cutting treatment, 1957 HO 39, Analysis of variance: Effect of row spacings and cutting treatments on forage production of Starr millet, 1957 HI 40, Analysis of variance: Effect of cutting treatments on forage production of Starr millet at seven-, 19-, and 3d-inch row spacings, 1957 H2 41, Number of clippings obtained from Starr millet as influenced by cutting treatments and row spacings, 1957 H6 42, Percent crude protein in Starr millet as influenced by cutting treatment and row spacing, 1957 H8 43, Pounds per acre oven dry Starr millet forage as influenced by nitrogen fertili- zation and cutting treatment, 195 ^ 122 44, Number of clippings obtained from Starr millet as influenced by cutting treatment and row spacing, 195 ^ -^23 45, Analysis of variance: Forage yield of - vii - Table Page Starr millet as affected by nitrogen levels and cutting treatment, 195o 124 46» Analysis of variance: Effect of cutting treatments on forage production of Starr millet receiving 60 pounds per acre nitro- gen biweekly, 19 5 ^ 125 47» Percent crude protein of Starr millet forage as influenced by cutting treatment and nitrogen fertilization, 195^ 130 4^. Pounds per acre nitrogen contained in Starr millet under three clipping treat- ments and three nitrogen fertiliaer levels, 1953 133 49- Analysis of variance: Pounds per acre nitrogen contained in Starr millet under three clipping treatments and three nitro- gen fertilizer levels, 195^ 134 50* Number of live Starr millet shoots per one foot of row as affected by nitrogen and cutting treatments, 195S 135 51, Analysis of variance: Effect of nitrogen levels and cutting treatments on numbers of live Starr millet shoots, 195d 136 52. Grams of oven dry Starr millet roots per four-inch diameter core zero to six inches deep as influenced by cutting treatments and nitrogen levels, 1956 13S 53 • Analysis of variance: Effect of nitrogen levels and cutting treatments on grams of oven dry Starr millet roots per four- inch diameter core, zero to six inches deep, 1956 139 54, Pounds per acre of oven dry Starr millet forage as affected by planting date, 1956 140 55, Pounds cer acre of oven dry Starr millet forage as affected by date of planting, 1957 1^1 - viii - Table Page 56. Analysis of variance: Pounds per acre of oven dry Starr millet fbrage as affected by planting date, 1956 142 57. Analysis of variance: Pounds per acre of oven dry Starr millet forage as affected by planting date, 1957 143 58. Percent crude protein of Starr millet planted at five different dates, 1957 146 - ix - LIST OF FIGURES Figure Page 1, Mercury tenaiometers located in oats 22 2* Hand cutting of oats using a two-inch red board 22 3« Hand cutting of oats using a nine-inch blue board 23 4* The irrigated portion of the 64^ plots in the 1955-56 oat clipping study 23 5. A portion of the irrigated 1956-57 oat clipping experiment 2? 6« Minimum thermometer with shields mounted on pole at five- foot level 27 7. Minimum thermometers mounted at heists of three and six inches in oat plots 2S g. Sheep grazing the holding area of oat management study 32 9» Paddocks of oat management study 32 10. Seasonal distribution of forage for three irrigated oat varieties cut when six inches tall to leave a two -inch stubble, winter 1955-56 3S 11. Seasonal distribution of irrigated Floriland oats as affected by height of plants when cut, winter 1955-56 41 12. Effect of irrigation on the seasonal forage distribution of Arlington oats when cut at 12 inches to leave a five- inch stubble, winter 1955-56 46 13. The effect of irrigation on the seasonal forage production of Seminole oats when cut at six inches to leave a two-inch stubble, winter 1955-56 47 - X - Figure Page 14, Seasonal distribution of irrigated Flori- land oat forage as influenced by three cut- ting treatments, winter 1956-57 57 15 • Ihiinjured Arlington oats where plants were not closely graaed prior to freezing tempera- ture on December 2S, 1956 69 16, Total loss of closely grazed Arlington oat stand by freezing 69 17. Over-all view of pearlmillet clipping study 75 Id, Pearlmillet plants clipped when 30 inches tall to leave a four- inch stubble 75 19. Pearlmillet plants clipped vrfien 30 inches tall to leave a 10- inch stubble 76 20. Pearlmillet plants clipped vdien 30 inches tall to leave an 16- inch stubble 76 21. Pearlmillet clipping operation 77 22. Seasonal production of irrigated common pearlmillet, 1955 90 23. Seasonal distribution of Starr millet grown in 19- inch rows and subjected to three cut- ting treatments, 1956 104 24. Seasonal distribution of common and Starr pearlmillet varieties when cut at 12 inches to leave a four- inch stubble, 19- inch row spacing, 1956 105 25. Regrowth of Starr millet after clipping 30- inch plants, 1957 113 26. Thirty- inch plants of Starr millet clipped back to four- and 10- inch stubble heights, 1957 114 27. Regrowth of Starr millet as affected by stubble heights, six dairs after cutting, 1957 11^ - xi - 2S» Seasonal distribution of forage as affec- ted by cutting treatments, Starr millet in 19-inch rows, 1957 117 29, Seasonal production of Starr millet fer- tilized with 15 pounds per acre nitrogen biweekly, 195 S 127 30, Seasonal production of Starr millet fer- tilized with 30 pounds per acre nitrogen biweekly, 195^ 12^ 31, Seasonal production of Starr millet fer- tilized with 60 pounds per acre nitrogen biweekly, 195^ 129 32r Effect of planting date on seasonal dis- tribution of Starr millet forage, 1957 144 - Xll INTRODUCTION Forage plants differ from most other crop plants in that they are commonly subjected to frequent drastic de- foliation of leaves and stems. Too often in the past, the consequences of defoliation of pasture plants have been ig- nored as animals practiced uncontrolled grazing* In recent years rotational and strip grazing have been increasingly used to augment productivity per acre. The problems facing the grazier using more intensive pasture system have been outlined by Myers (62) in Australia. Two questions must be decided in rota- tional grazing or close folding! (1) the hei^t of the pasture at the beginning of grazing which depends on the frequency of grazing suid (2) the hei^t of the pasture at the finish of grazing (i.e., the severity of grazing). The above mentioned questions have been answered to a large degree for cool season pei*ennial grasses and to a more limited extent for warm season perennial grasses. Annual forage grasses have received little attention inso- far as plant management is concerned. Consequently, oats (Avena sativa L. ) and pearlmillet (Pennisetum glaucum (L.)) have been selected for this study, the former being a cool season plant and the latter a warm season species. Both have been cited in Florida work (56, 55) as being the most productive dairy pasture species during their respective - 1 - - 2 - growing seasons. The questions raised at the inception of this in- quiry concerned the consequences of defoliation on total yields and seasonal distribution of forage and to a lesser extent the effects on crude protein content and root pro- duction. This series of experiments attempted to evaluate the effects of plant height when defoliated and the stubble height on forage suid root production of oats and pearlmil- let. The effects of varieties, row spacings, irrigation, and nitrogen levels on forage yields were studied to a more limited extent. REVIEW OF LITERATURE Perennial Grasses Effect of plant height at tlms of cutting or grazing The literature is replete with references to vari- ous frequency of clipping experiments with cool season per- ennial grasses. One of the earliest studies was made by Crozier (21) with timothy and reported in 1S97. He found that frequent mowing reduced total yield to one-fourth of the yield under hay conditions. Although the percentage of crude protein was highest in forage from frequently cut plots, Crozier did not feel that it made up for the loss in yields. Cotton (IS) in 1910 ascribed the deterioration of pastures in New York and New England to overgrazing, caus- ing less food storage in perennial plants. Several years later Carrier and Oakley (16) and Hutcheson and Wolfe (42) reported that heavy grazing of Kentucky bluegrass resulted in larger gains per acre than light grazing. Somewhat dif- ferent conclusions were made by Ellett and Carrier (2?) in 1915 from their work with Kentucky bluegrass sod. They found that the total yield of forage varied inversely with the number of times the plants were cut but that the de- crease in protein content at the mature stage more than counterbalanced the increase in dry matter. Early work by - 3 - - 4 - Waters (S?) pointed out the importance of maturity of top- grovrth on subsequent yields and permanence of stands of timothy. Stapledon and Beddows (74) in England reported in 1926 that repeated cutting of orchardgrass reduced the num- ber of new shoots. Another English worker, Fagan {2S), compared orchardgrass and Italian ryegrass cut monthly and every fourteen days, finding that the crude protein content was highest from plants cut most frequently but the total protein quantity secured during April and May was greater from plants cut at monthly intervals. Work by Hudson (41) in New Zealand also points out the decrease in yields from frequent clipping of ryegrass and clover. Graber (30) in 1931 at the University of Wisconsin found that the amount of root and top growth of bluegrass, red top, fescue, and timothy varied inversely with the fre- quency of defoliation. Similar results were reported by Robertson (68) in 1933* In contrast, Ahlgren (1) obtained increased productivity from cutting bluegrass when four to five inches tall instead of at heading time, Mott (61) in Indiana found that bluegrass-white clover clipped when the herbage reached four to six inches produced three times more forage than when clipped weekly. Dibbem (26), work- ing with bromegrass, substantiated previous investigations in that dry weights of tops and roots were in inverse - 5 - proportion to frequency of clipping, Tincker (Si) noted that a plant produces more additional dry weight in a month by adding to its old growth than when it is clipped and produces new leaves. Wagner ($4) subjected mixtures of orchard-ladino, brome-ladino, orchard-alfalfa, and brorae -alfalfa to various cutting intervals and stubble heights of six and twelve inches. He concluded that frequency of cutting had the greatest over-all effect on total yield and distribution of production. The adverse effect of frequent defoliation on for- age production of native range grasses is well known. Studies with native grasses of mostly cool season species in the western United States by Sarvis (72), Sampson and Malrasten (71), Biswell and Weaver {&) , and Gemert (29) all point out the decline in forage production under frequent clipping or grazing. The time of the year when frequent defoliation oc- curs may be important. Blaisdell and Pechanec (9) foiind that the greatest reduction in yield of bluebunch wheat- grass the following year occurred from clippings made in late May and early June, Stoddart (77) reported that Agro- pyron spicatum in Utah was most severely damaged by clip- ping in midsummer due to drouth conditions normally exist- ing at that time. This same grass was also found to be - 6 - most vulnerable to clipping damage at the middle of its vegetative stage and that cutting intensified maximum vul- nerability and prolonged its duration (56), Cutting or grazing management may alter the season- al distribution of forage. English workers (75) found that when cutting does not begin until late May, the May to No- vember grovrth curve is much more level than when cutting begins earlier in the year. Bird (7) emphasized that stage of plant growth at the time of the first cutting determines the period of aftergrowth. Frequency of clipping studies with warm season grasses were reported by Leukel, Camp, and Coleman (52) at the Florida Experiment Station in 1934« Bahia, carpet, and centipede grasses cut frequently maintained more vegetative growth and made a better sod. In 1935 Leukel and Bamette (51) reported that although Bahiagrass cut when mature pro- duced the most dry matter, the largest quantity of nitrogen was produced by frequently cut plants. Studies with napier- grass in Trinidad (65) and Hawaii (69) also indicate de- creased dry matter production with frequent cutting. The Hawaii workers found that the greatest amount of palatable forage was produced by cutting every eight weeks, Lovvom (53, 54) in North Carolina studied the ef- fects of clipping on Dallis, carpet, Bermuda, and Kentucky bluegrass, concluding that although Dallis was most - 7 - seriously injured by frequent cutting, all the species re- quired more careful management under high fertility. He also found that frequent cutting was more deleterious at high temperature, Prine and Burton (6?) studied the effect of nitrogen rate and clipping frequency on Coastal Bermuda- grass in Georgia, They foimd that increasing the clipping interval from one to eight weeks increased hay yield, stem and leaf length, plant height, seed head frequency, and intemode length and number. This had little effect on protein yield and percentage nitrogen recovery and de- creased the protein percentage and leaf percentage of this grass. To recapitulate then, forage grass yields are gen- erally in inverse proportion to the frequency of defolia- tion; the response to clipping being influenced by the species, time of year when clipping is practiced, and soil fertility. Protein yields, on the other hand, may be as high or higher imder more frequent defoliation, Unfortxmately, most clipping studies have been based on a chronological schedule rather than on the height of the plant, Kohler (47) points out in his studies on the chemistry of grasses that cutting treatments in variety and fertilizer tests should be on the basis of physiological age rather than on a time interval. The problem immediately arises as to what is meant - g - by plant height. Heady (37) defines the height of a plant as "the perpendicular distance from the soil at its base to the highest point reached with all parts in their natu- ral position." In clipping studies with plant varieties having different growth habits, it is difficult to use this definition and remove an equivalent amount of forage from each of the plants. In the case of decumbent and upright oat varieties, too often clipping experiments have disre- garded the amount of forage left on the decumbent plants. Cattle grazing, however, tend to remove much more of the basal growth on the decumbent plants. For this reason it would seem highly desirable to lift decumbent forage into an erect position to determine its true "height" or length before cutting. Effect of stubble height The influence of stubble height on top and root growth of perennial grasses has been investigated by a num- ber of workers, Stapledon and Milton (76) reported in 1930 that orchardgrass plants cut to leave a six-inch stubble considerably out yielded those cut to the ground level, Graber and Ream (32) found that closely clipped bluegrass did not respond to additional nitrogen but that leaving a one and one-half- inch stubble greatly increased production both with and without additional nitrogen fertilizer, A number of other studies point out that bluegrass can be - 9 - clipped or grazed closely without ill effect (31, 60, 5# 12), Bromegrass, on the other hand, requires that higher stubble be left if forage production is to remain high (90, 2), Quite obviously, the growth habit of a forage species deteimines the response to close clipping, Lang and Barnes (49) and Newell and Keim (63) noted that native short grasses produced greater yields when cut frequently to ground level than when cut only once at the end of the season, Harrison (35) and Harrison and Hodgson (36) deduced from clipping studies with several cool season grasses that the shorter the stubble, the less top growth and roots were produced, Harrison (35) concluded that the deleterious ef- fect of frequent cutting may be offset partially by cutting the plants at a greater height above the soil. Dexter (25) and Johnson and Dexter (44) found that quackgrass could be severely injured and often killed by close clipping, Aldous (3) found in clipping studies with big and little bluestem that a four- inch stubble height gave 25 percent more forage than a two-inch stubble, Holscher (3^) reported that with bluestem, wheatgrass, and blue grama even a stubble height of four centimeters was not suffi- cient to maintain yields and prevent killing of the plants. Research on tropical grasses by Paterson (66) in Trinidad showed that close clipping sharply reduced yields - 10 - of para, elephant, Guinea, and Guatemala grasses. He re- commended that a cutting height of six to ten inches should be used for these tropical grass species, Torstensson (S2) in Sweden found that meadow fescue and Italian ryegrass grown in the greenhouse were most pro- ductive when cut to leave a five-centimeter stubble every 15 days, Jantti and Heinonen (43) in Finland reported that a mixed sward of fescue-orchard-timothy-red and white clo- ver showed a striking reduction in yield when clipped closely. Defoliation to leave a stubble of one centimeter, while the soil moisture was at or near permanent wilting percent down to a depth of 40 centimeters, arrested growth almost completely but the growth rate of plants with a four- or 12-centiraeter stubble was 70 to 90 percent of the corresponding rate on moist soil. They concluded that de- foliation alone causes heavy reductions in forage yield even in moist soil, and these reductions are more important quantitatively than those resulting from the interaction with soil moisture. Similar results were noted by Canfield (15) with black grama and tobosa grasses where close clip- ping nullified the value of high rainfall years under range conditions, Tobosa grass clipped to leave a four-inch stubble yielded 110 percent more forage than two-inch stub- ble over an 11-year period* Robinson and Sprague (70) found that clipping - 11 - bluegrass-vrhite clover sod to a heigjit of two inches pro- duced a more dense sod of grass with less clover than when shorter stubble heights were used. Brougham (10, 11) found th^ a perennial ryegrass, white, and red clover mixture in New Zealand defoliated to five inches yielded 20 percent more herbage than pasture cut to one inch during a 32-day period. During the first 14 days of regrowth the yield difference was 100 percent. He inferred that excess light changed the growth habits The more intense the defolia- tion, the lower the initial leaf efficiency. In contrast to the preceding papers which point out the advantages of higher stubble. Cooper (17) found that the yield of Nevada bluegrass and sedges from a native mountain meadow in Oregon decreased as the cutting height was increased from two to four or six inches. Protein per- cent of the forage showed only a slight increase with high- er stubble height. The lack of response to higher stubble heights here may probably be attributed to the low quality forage plants and unfavorable wet environment. Root growth as affected by defoliation Forage plant roots are often decreased in size and number by frequent defoliation of the top growth. Dibbem (26) concluded that clipping of smooth bromegrass was more adverse in its effects on root growth than top growth. Sturckie (73) reported that any cutting of Johnsongrass - 12 - reduced roots talk development and the more frequent the cutting, the greater was the reduction. Similar results have been reported for orchardgrass (76), Kentucky blue- grass (46), and wheat grass (33). Parker and Sampson (64) noted that frequent removal of aerial growth of purple needlegrass and soft chess caused poorly developed roots with a reduction in the diameter of the stele and entire root. Baker (6) in England found that cutting of top growth reduced the number of roots and tillers per ryegrass plant but increased the* number of roots per tiller. The root weight of cut plants was found to be lower than on uncut plants a few weeks after defoliation. Crider (2), in a detailed study of 13 cool and warm season grasses including bunch, rhizomatous, and stoloni- ferous types, showed that root growth stopped for a time after each removal of aerial growth. He concluded that the growing top cannot be reduced more than half without ad- versely affecting the functioning of the root system and the plant as a whole. At variance with this work. Laird (4S) concluded that mowing of centipede and Berraudagrasses increased the root growth. The difference of opinion as expressed by the latter paper can probably be attributed to poor control of environmental factors in root investiga- tions. This fact is stressed by Williams and Baker (88) - 13 - who point out that root studies should be conducted on plants grovm in their natural habitat, or on field experi- mental plots, rather than in the artificial environment provided by containers. Annual Grasses Clipping studies with annual forage plants have been limited in number. Many of these experiments were concerned with grain production of oats, rye, and wheat. Field studies in Pennsylvania by Washko, Edge, and Haskins (S6) showed that fall clipping of oats did not affect grain yields but that spring clipping decreased grain yields, Robinson (69) in England found that if grazing of winter wheat is continued into raid or late April the loss of grain will be 20 to 25 percent of the potential crop. Applying 100 to 150 pounds per acre of ammonium sulfate immediately after the last grazing reduced the loss of grain by one- half. Itorris and Gardner (59) in north Georgia foimd that clipping to Ilarch 15 resulted in 75 percent or more reduc- tion in grain yields of oats, wheat, and rye. Research work in north Florida (5^) indicated that clipping oats until February 15 reduced grain yields 25 percent while an additional two weeks of clipping resulted in 50 percent yield reduction. High nitrogen fertilization was effective in maintaining yields of oats and rye clipped to mid - 14 - February, Cutler, Pavez, and I'&.lvey (22) in Indiana found that clipping wheat during early April increased the yield and quality of grain. Clipping reduced plant height by shortening the lower intemodes and thus decreased lodging of the grain, Davies (23) in England reported that vrinter grazing of oats caused a much sharper decrease in grain yields than did spring grazing. His results indicate that the reduced yields of grain are probably due to the height of the plants grazed and the stubble heights used but he attri- butes the difference to season of the year. In this ex- periment, the plants were tvio to three inches tall and grazed to leave a three-quarter-inch stubble on the winter grazed paddocks v;hile in the spring grazed paddocks the plants v^ere six to seven inches tall and grazed to leave a one to one and one-half -innh stubble, Washko (35) found in Tennessee sheep grazing trials on erect and prostrate types of oats, wheat, barley, aurid rye that the habit of grc^th had no effect on forage pro- duction. Grazing was found to be detrimental to grain pro- duction on all species, Thurnnan (79) reports that upright varieties of oats produce more forage in the fall than prostrate types but that the total production for the season is about the same, Hubbard and Harper (40) in Oklahoma found that - 15 - severe clipping of small grain produced slightly less for- age and appreciably lower grain yields than did moderate clipping. Yields were affected most by severe clipping in unfavorable growing seasons. The chemical composition of forage from severe and moderately clipped plots was simi- lar. Thurman and Grissom (SO) in Arkansas reported that clipping very young oat plants reduced root growth and forage yields as compared to delayed clipping. The re- growth of oats was reduced when the growing point of the short stems was clipped along with the leaves. Clipping above the terminal growing point and after the plants had a good start more than doubled forage yields. Laude (50) in a greenhouse experiment with soft chess noted that growth was faster from a cut shoot when the terminal bud was retained and the culm continued elongation, than when the bud was removed and regrowth was by tillering. In ad- dition, the tiller was smaller and lighter than the main shoot. Justus and Thurman (45) also report decreased yields of oats by frequent clipping. Burton and Prine (14) found that forage yields of Abruzzi rye, Jouthland oats, and ryegrass were greatly increased by cutting at bi-monthly instead of monthly intervals lato in the season, Early in the season this had little effect on yields. Hoveland (39) in Texas found that oats clipped - 16 - back to one inch whonever they reached three to four inches in height produced 33 percent less forage than where the plants were t)erraitted to grow 10 to 12 inches tall before clipping. Marshall (55), in a Florida trial with dairy- heifers on oats, concluded that increasing the height of forage to about eight to ten inches before grazing as com- pared to six to eight inches was ir; sponsible for obtaining a higher yield of total digestible nutrients. Verbeek (S3) in South Africa also stated the desirability of having oat plants at that height for grazing. He also cautioned that oat and \iheat plants should not be allowed to get much higher than 12 inches because the recovery power of the plants is reduced vrith consequent fewer grazings being ob- tained per season. The effect of cutting sudangrass at two- and six- inch stubble heights was studied by DePeralta (24) in Nebraska. Yields of forage and roots were markedly reduced by close clipping. Seeding rate was also studied but this factor had little effect on yield response under the two cutting treatments. Burger, Jackobs, and Hittle (13), in an Illinois study, found that herbage yields of four sudan- grass varieties were much lower when cut at IS to 20 inches tall than when harvested at hay stage. Little difference existed between varieties when cut at the shorter height. Row spacings may influence the response to cutting - 17 - of oats (73) and pearlmillet (19 )• In general, narrower row spacings favor higher forage yields but defoliation systems may alter this pattern. OATS Materials and Methods All of the experiments in this study were estab- lished on Arredondo loamy fine sand located at the Agronomy Farm, University of Florida, Gainesville, during the win- ters of 1955-56 and 1956-57. Effect of cutting treatments, varieties, and irrigation, winter. 1955-56 A split plot design was employed, consisting of three varieties as main plots, subplots of three dates when clipping stopped in order to produce grain, and sub-sub- plots of six defoliation or cutting treatments. All combi- nations of these were used. Oat varieties were selected on the basis of representative growth habit j decumbent-Arling- ton; intermediate-Floriland; and upright-Serainole. The dates to stop clipping on each variety were based on the usual time of heading at Gainesville, The average heading dates at Gainesville are: Arlington- April 1, Floriland- March 1, and Seminole-February 20, Six cutting heights were selected to obtain a range from very intensive to light defoliation. The treatments were replicated six times. The entire experiment was grown under both irri- gated and non- irrigated conditions, making a total of 646 - Id - - 19 - plots. The iiTigatlon schedule was determined from mer- c\iry tensiometers, placed at depths of three and nine inches. A tensiometer location is shown in Figure 1. After the plants had become established, irrigation water was applied by sprinklers whenever the soil moisture ten- sion reached 500 centimeters at the nine-inch level. Due to the low water-holding capacity of this soil, only one inch of water was applied per irrigation. During the win- ter, a total of 1& inches of water was applied in order to maintain adequate soil moisture. One-half of this water was applied from February 25 to April 20, Very hard winds and blowing sand during the last two weeks of J^rch and the first week of April made it difficult to supply suffi- cient moisture for plant growth. All oat varieties were planted October IS at the rate of four bushels per acre in nine-inch rows using a Planet Jr, 300A planter. The fertilization at planting consisted of! dOO pounds per acre d-8-d, 100 pounds cal- cium sulfate, 40 pounds magnesium sulfate, 10 pounds copper sulfate, 10 pounds manganese sulfate, 10 pounds zinc sul- fate, one pound fertilizer borax, and one-fourth pound sodi- um molybdate. Additional applications of 100 pounds per acre ammonium nitrate were made on November 2S, January 3, February 3, and March 9, making a total of 196 pounds of actual nitrogen per acre for the season. Subsequent - 20 - nitrogen applications were made on the basis of plant tis- sue tests. Both irrigated and unirrigated portions of the experiment received the same fertilization. Each plot consisted of three rows seven feet long; the two outer rows serving as borders and being cut the same as the inner yield row, A 5»S-foot strip of the cen- ter row, 0,0001 acre in area, was harvested and weighed for yield. Cutting was done by hand using paring knives. The proper height of cut (two, five, or nine inches) was main- tained by using as a knife guide a plywood board of the proper height. Nails 5«S feet apart on top of the board indicated the exact length of row. Bamboo stakes driven into the ground at one end of each yield row indicated the height to vdiich the plcmts were to be grown before cutting. The heights of stubble were color coded on both stakes and the wooden cutting boards (red • two inch, white « five inch, blue = nine inch). This feature eliminated much of the potential chance for error in the field. Views of the field experiment and the cutting technique are shown in Figures 2 to 4. The question of when to cut a particular treatment was determined by visual inspection of the plots. Decum- bent oat forage was lifted to an erect position to deter- mine its true height and also for the cutting operation - 21 - itself. Green fbrage was not weighed but only oven dry- weights were recorded in grams. Plot yields were converted to pounds per acre and then subjected to statistical analy- sis. Forage was collected in December, January, and riarch from two replications of each treatment, dried and ground for nitrogen analysis by the Kjeldahl method (4). Grain yields were obtained by harvesting a .0001 acre strip, drying, and threshing with a Vogel machine. Weight of grain per plot was recorded in grams and later converted to pounds per acre* IS * - 22 - Fig, 1, — Mercury tensiometers located in oats» Fig, 2. — Hand cutting oi oats using a two- inch red board. - 23 - Fig» 3« — Hand cutting or oats using a nine-inch oiue boardi Fiff, 4« — The irrigated portion of the 64S plots in the 1955-50 oat clipping study. Bamboo stakes with colored tips indicate the height at which plants are cut. - 24 - Effect of cutting treatments, varieties > and irrigation^ winter 1956-57 In continuation of the clipping studies with oats, a split plot experiment was planted October IB, 1956 using Arlington, Floriland, and Seminole oat varieties as main plots. Subplots consisted of eight clipping treatments as follows: (1) Clipped all winter when six inches tall to leave a two-inch stubble, (2) Clipped all winter when 12 inches tall to leave a two- inch stubble, (3) Clipped all winter when 12 inches tall to leave a five- inch stubble, (4) Clipped all winter when IS inches tall to leave a two- inch stubble, (5) Clipped all winter vrtien 16 inches tall to leave a five-inch stubble, (6) Clipped the first time at six inches to leave a two- inch stubble. Clipped remainder of winter when 12 inches tall to leave a two- inch stubble, (7) Clipped the first time when six inches tall to leave a two-inch stubble. Clipped remainder of winter when 12 inches tall to leave a five-inch stubble, (g) Clipped the first time when 12 inches tall to leave a two-inch stubble. Clipped remainder of the winter when six inches tall to leave a two- inch stubble. - 25 - The experiment was replicated five times and grown imder both irrigated and unirrigated conditions, making a total of 240 plots. Fertilization at the October Id plant- ing date on both irrigated and unirrigated areas consisted of SOO pounds per acre of 6-B-d, 100 pounds calcium sul- fate, 50 pounds magnesium sulfate, and 50 pounds fritted trace elements. In addition, the irrigated experiment re- ceived 200 pounds per acre sodium nitrate on November 6, November 26, December 20, January 9, January 30, and ffeirch 11 making a total of 240 pounds of actual nitrogen per acre for the season. The unirrigated experiment received addi- tional applications of 200 pounds per acre sodium nitrate on November 6, December 20, and January 30 making a total for the season of 144 pounds of actual nitrogen. A total of 11 one-inch applications of water were made to the irrigated plots; the irrigation schedule being determined as in the 1955-56 experiment. Planting and har- vesting operations were carried out as in the previous year. An over-all view of the experiment is shown in Figure 5. In order to obtain information on root production under the various regimes of cutting, two cores six inches deep and four inches in diameter were collected from each plot of the irrigated experiment on January 15. Cores were obtained directly from the row, the top gro^vth being clipped off to the groimd levol. Each plot sample was washed on a - 26 - screen to remove soil and the root material, was dried and then weighed. It was noted early in the winter that freezing in- Jury appeared to be dependent on the stubble height at the time of the freeze. The question was posed as to tempera- ture differences within plots. Consequently, minimum ther- mometers were mounted between the rwrs of two- and five- inch stubble height plots of Floriland oats at heights of three and six inches as shown in Figures 6 and 7. Unfor- tunately, only two moderate freezes occurred after mounting the thermometers on Januaiy 17 so that only limited infor- mation was obtained* - 27 - Fig. 5.~A portion of the irrigated 1956-57 oat clipping experiment. Fig, 6# — riiniinum thermometer with shields mounted on pole at five-foot level. - 23 - Fig, 7» — Minimum thermometers mounted at heights of three and six inches in oat plots. - 29 - Effect of sheep grazing treatments^ vfinter 1956-57 In order to ascertain yields of forage and effects on the plants under different intensities of grazing, a small paddock sheep grazing study was initiated where the animals were used only as a tool in removing the forage* Animal gains were not considered in this study* Arlington oats were seeded with a grain drill on a one-fourth acre area after fertilizing with SOO pounds per acre d-S-S, 100 pounds calcium sulfate, 50 poimds magnesium sulfate, and 500 pounds per acre fritted trace elements. Additional nitrogen applications of 200 pounds per acre of sodium nitrate were made on November 6, November 26, and December 2d, making a total of 160 pounds of actual nitro- gen for the season* Eight inches of irrigation water were applied during the winter. A randomized block design was onployed with five grazing treatments and four replications, thus dividing the area into 20 paddocks* The treatments were as follows: (1) Grazed when the plants reached six inches, leaving a two- inch stubble* (2) Grazed when the plants reached 12 inches, leaving a two- inch stubble* (3) Grazed when the plants reached 12 inches, leaving a five- inch stubble* (4) Grazed the first time when plants were six inches tall - 30 - with a two- inch stubble, thereafter when plants were 12 inches tall, leaving a two-inch stubble. (5) Grazed the first time when plants were 12 inches tall leaving a two-inch stubble, thereafter when plants were six inches tall leaving a two-inch stubble. The estimated yields of forage were calculated for each of the treatments using data from the clipping experi- ment of the previous year and the size of the paddock ad- justed so that it could be grazed down within a day by one sheep. It was later discovered that one sheep would not stay and graze well alone, so two animals per paddock were used, making it possible to graze down the area to the de- sired height in about two hours. Woven wire fence three feet high and creosoted posts were used in construction of the paddocks. The one- half acre reserve area of oats near the paddocks was fenced with woven wire four feet high to discourage fence jumping. In one comer of this area was constructed a small shelter of sheet metal closed on the north and west as protection against inclement weather. The eight native breed sheep were shut up in the shelter each night as protection against dogs. Figures & and 9 show views of the experiment. Before grazing, yields were taken by cutting to the desired stubble height three quadrats (0.0001 acre each) at random in each paddock. The three quadrat yields were - 31 - averaged together and the statistical analysis determined on the four replications for each treatment. After clip- ping, the sheep were turned into the paddock and allowed to graze it down to the stubble height of the clipped quad- rats. It was found that the sheep grazed the two- inch stubble paddocks somewhat closer than desired, usually one to two inches, vrtiile the five-inch stubble paddocks were grazed to leave a stubble of four to six inches. - 32 - Fig, S» — Sheep grazing in the holding area of oat management study. i''ig. y. — ir'adaocKs or oat management stuay. Two sheep were used in each of the four replications of a par- ticular grazing treatment. - 33 - Results Effect of cutting treatments, varieties, and Irrigation. winter 1955-56 In Tables 1 and 2 are shovm the average yields per treatment combination. The statistical analysis suniniary in Table 3 was calculated from the original plot yields in grams oven dry forage per 0.0001 acre. Results of the over-all statistical analysis shovm in Table 4 indicate that the highest forage yield for the winter was obtained with the Arlington variety, followed by Floriland, and then Seminole. All t^ree varieties generally responded much the ssune to irrigation, the yield increases ranging from 44 to 4^ percent. As shown by Figure 10, there was little difference in the earliness of forage production by the three varie- ties. The Seminole variety is usually classed as an early forage producer but this is probably due to its upright habit of growth, giving an appearance of greater forage production than the more decumbent types. Tables 1 and 2 indicate that Seminole made most of its production late in the growing season. Continued high production throughout the winter was found to be better with Arlington and Flori- land than with the Seminole variety, regardless of the cut- ting treatment used. The comparison of dates when clipping was stopped J4 - n 8! OS n H O o I CO o 0^ I ■CO CM I to (4 ca i N H ooo HOnO MOW • « • 5 X ■H n bO •H o. P4 •H iH O OOO MCMTv >ooto • fap NiH M •o •9 « oi 0) 0 ^ ^ Q} NO-vOOO - o « • p ^J^ WH «H 0) o ^ /3 CD « X M • s s totovo « * «» f^WH OOO vOiHW AAA ifS<*\W OOO vO»AH iHOOn AAA ITVCAW OOO AAA OOO tOvOH oto-* AAA -itWH OOO OvOH AAA f\WH H •H C-d 43 O C Q> ■p Sin § ^na fcO) HOB •H P4 OOO AAA ^r^W OOO WNWH u>OnO AAA OOO tOHt^ AAA OOO AAA OOO f^tOO tOH-* AAA O C « 43 SrH hCiH O f^ O S TJ CO O C © bOH O aH O S Ut-i O C o o • A 43 g g o I o 43 a> o 43 (0 O JhH d « o g 1 4:> to « iH 0) ^ ^ u> cu 1 p* to •H H i-» O >4 t M 0 1 hO to « H ^1 O 1 u^ A 1 CM e H ;< u -< 1 u, 1 iH § O CM 04 1 vO o UN 8 o CM CM I « CM CMrH OQO f^OtO HOvO * % • r\CMH OOO tO-d-ON ■HCMtO « • » C*\CM»H il) OOO Q> OOO +> OOH +3 OHO -§ U>-;JaO ^ » » * rHiH CMCMH bO bO a a t3 TJ (0 (tf Q> OOO O OOO X t0-4T^ ^ ;4^CMiA -4-nOH O-d-WN > {> k OOO C0 StOON (MOH « ONOvO s lAOH h » • • A * o HrH o CMCMiH § ingto rilan inole ■P ingto rilan inole s. r-i O B Ur-i Q) a HOE f^H O (X p. «< O o 00 H H « •H H O o c © +3 «H t4)H O HOE ijH 0) (4 0 +> O CO o I p< O •a o +3 • CO o tiH JO.O p :3 (4 .p CO u o » § ^ o n 4^ •H faO 4 O - 36 - n +5 m 0) +5 J52 3 v cow CO o u 3 cf COOT * &, o u o CO •M- « » ♦ « « * « * * p v\H"Srrv tOOvO OOH • • » • • • • '•'••• -^■tooi t0-4"O ifNC>~CMNO sOf^NO OvOlTk HtO-*CNJ « * • « to ow «AH«nH to s H c^^ \0\0-* if\»rviHHr^ IN HifXO ^ <*\\OvO MO»A -* « «k * .« « m • « * « • A HvOOi too-o VN-d-lAOi-* H Cv. H HI C^r-»f*\CMaN ^ H -* to « « * » « # » * ~*«CM tOvOw OMHtOP"- • • • • • • • • • • • «y wM^ C^tOQ mvOvO Nt^OtOtfN vOOtO n^VN^. * * « * m ■ m * « * HnOH H-4- H^<»>H ON C-- CM fH M MC4M vOWMA OCMfHt»- if\ VOONt^ O-tNN HVNOrHtO NO OHr«- VOOitO CMUMACMOv CM • .« .* « * * « ^ « . A « * • •t H ■00 CM to WtOM nOC^OC^O -* 2 Of-i -*iHM O -*■*»><»% CM WN > t*\ ir\ rH H m H VSCtO W-^O WNOOO VN <*\ s. 1-4 <*\ rHrHCMCM CM CM rr\ •0 2 § o s ■p «l « 43 8: i O •H •H ■p «M a H « «H o O •« g > I' •H ca-— a — * n +* ..'"' •H •• 4> a> c8 (B ^ ■P « ** CO 09 ©"H— ' J5 «-' o *^_ ^^ * ■P O -P •• >^ iH 5^°«». ♦ O f-l Q) h n > ^4 fX r-l H-ri O ■P 0) o JO +> K X « p fi4 P*U U Q> <0 (-) o ■P X ^ s •p ^) iH « )U CQ dt-iHE-i t* H PS «> w •H CO QO» 1 •§ CO O t J O CJ W - 37 - TABLE 4 SUMMARY OF OAT FORAGE YISLDS AS I?IFLUENCED BY IRRIGATION, VARIETY, DATE WHEN CLIPPING CEASED, AND CLIPPING TREATMENT, WINTER 1955-56 Poiinds per kcre Oven Dry Forage Irrigated Unirrlgated Varieties: Arlington 3,^90 Floriland 2,720 Seminole 2,040 L.S.D. at 5% 290" Date i^en clipping ceased in order to produce grain! Clipped all season 3,910 Clipped until 2 weeks before average heading datea 3,020 Clipped until 6 weeks be- fore average heading date 1,720 L.S.D. at 5% 170 Clipping treatments t 6 in. tall, 2 in. stubble 2,340 12 in. tall, 2 in. stubble 2,990 12 in. tall, 5 in. stubble 2,980 18 in. tall, 2 in. stubble 3,380 18 in. tall, 5 in. stubble 3,230 18 in. tall, 9 in. stubble 2,370 L.S.D. at 5% 200 2,660 1,870 1,410 110 2,830 2,170 930 110 2,480 2,090 2,160 1,850 1,370 170 ^Average heading dates at Gainesville t Arlington - April 1 Floriland - I^larch 10 Seminole - February 20 . 3d - 4» > (0 1 •H $4 I 43 > U * O $^ (0 O O (h « (0 0) iH CO « o n o o « 2 •H ^ Eci 8 1-4 4>rH 81,: u h • •HnO ir\ «H I O irv if\ CON Oi-I •H ^ a> n •a 30VM0d A«a AVa/3H0V/S8n 9AV d a 0 a> CO CO 1

&:< -P •&•*=' X I ^ - 42 - CO I •< O o < I OlA CEJrH M UK] CQ 1-404 ^ MM o g ^ CO iSi o C •H s CO o +> B o u coco 2 2 CO «a O U COCO CO coco Q O CO S t^ ITN l>- NO e». H f^ -* <> r- «rv t"- ON «*\ & H CM ■CO C^ O • * «k «k iH M M c^ CM CM -* « « * « »A f^ to »A cr\ H (M •H to CM vO vO >r\ CM ^ :! i« • » rH v\ O H CM CM « « » # -t CM rH r-- O -* CM H vO o NO H H CM «n H r^ H ir\ O ON « * •t WN vO r- H H -4- lA U\ w\ ir\ CM r\ n CO C +> o C •H i s-s •H ^ « u H H •P u ? ■« cd s^ « « o H Eh 43 t O •H CO - 52 - - 43 - ^> « o t a n (5 u s. n pt. ■p I ■P o o Q q M O H • H o « 0> » • 0 '^ iH i-< H H H H IfN Xi ^ ^ X> ^ ^ ;o Xt ^ Xi Xi ^ 43 3 s a S 3 d m 43 43 43 43 43 43 01 as 09 to a CO 4 Xi 4:3 Xi x: £ x: • u 0 0 0 0 0 CO ^ 5 a a a 5 1 • h4 1 1 (M 1 1 CM 1 ON 1 CM • rH * • H H •;3 •a i-l C0 -^ -S +3 49 43 43 43 43 (0 tt m 9i CO «B 0> Q) o iH iH i-i •H H - 44 - With 12-inch plants, leaving a five-inch stubble did not significantly increase the forage yield of any variety. The efficacy of irrigation is shovm in Figure 12 where the distribution of forage throughout the season is graphed for the Arlington variety when cut at 12 inches to leave a five -inch stubble. The major contribution of ir- rigation was in furnishing early forage during December and January. The value of irrigation was erased by im- proper defoliation management. Figure 13 illustrated this fact in the case of Seminole oats but similar results oc- curred with the other two varieties. Close and frequent cutting resulted in low production despite adequate mois- ture being present. The sharp production dips during Janu- ary on all of these graphs are due to freezes which seri- ously limited forage growth. The crude protein content of forage from most of the clipping treatments was high, as shown in Table 7. The protein content remained highest throughout the season with six-inch plants clipped back to two inches. Plants 16 inches tall with a two- inch stubble were lowest in protein. In the case of 12-inch plants, those having a five-inch stubble were somewhat higher in crude protein during the month of March than plants with a two- inch stubble. Grain yield data are summarized in Table S. Very low grain yields were obtained as shown in Tables 9, 10, - 45 - and 11, particularly with the Arlington and Floriland vari- eties where losses due to bird damage were appreciable. Ir- rigation increased the yield of Seminole oats by 50 percent but even so, the yield was low. As expected, grain yields decreased when the forage was removed as late as two weeks before the average heading date. However, the yield re- duction was much less pronounced where moisture was not limiting. Under unirrigated conditions, clipping treatments had little effect on the yield of grain except in the case of closely defoliated plants six inches tall with two-inch stubble where the yield was reduced. With adequate mois- ture, grain yields were greatly affected by defoliation. Plants clipped when 12 inches tall to leave a five-inch stubble produced considerably more grain than any other treatment. - 46 - o til »- S < K * 3 O (0 o 3evuod A»a Ava/3uov/ sen 8av I u • ITS 05 «H a> CD O eS Vi bOO) •HJB OH 4>4:» «H 4» «^ 3 W O !c • (D • a tl043 S!8 g - 47 - O I §»n O^ "iu S5 CD * tiO iH to n d « I no ■p (0 C Q) °^ ♦» O «-P bO •H 0} (D d ••S n • *3 bOcO •H O Q> Q 39vtiod Aua Ava/3tiov/sen 9av I « CO - 4^ - 3 § c : 1 IfXtOOtOOrW 0\ V\-^O^Cv r>.-*avM%o«o •d itf • •••«• • 1 • i • 4 • •*«•« « ® CMOtO r\iAC^ to 1 ^-M<^u^ HC«-ONf^-*NO « g •^- P4 « «> • * • 4 4 « • < • • • « • • • • • • H M •p Hi-l\OC*-vO*r\ ■00 C^vO C^tO -4* OOOvOC«-vO Oi •-9 O 1 1 r^r^N WCMM CMC4CMMMC4 r^i*\ n X|\0 g ^ voto <*^oo^o^ o oc^ot^ C^f^^HHtO o> k« U> & • ••«•• • t • • • • «••*'•« n < 1 O « -*CMONMtOO CM 1 H WNOtO M^Mu\-COH ^ »rv u U t*\r>-\NCM WC^ r^ <*NMCM ^ . > •P H o 3 •H O 3 O -P WNW\CMU\ON MMVTiCaiAON OCM WtOtOtO vOWMtOtOtO NOWCM^OtOtO se; rHiHHHiH i-^r-iHHr-i r-i r-t t-i r-t H Q> C H •HH « Eoo •H 12 S ^<^| m 85 & 2 ca 6 •d «i < o P !> 5 0) •H $4 H •H t f P > in Si • § - 49 - MM to OK WN fig* 00 a ^ go CT^ 1 to H u> to H « n +3 to a rH +> « 1 W) c ITk •H 1 04 C4 u< H •H H 0 CM 1 d H CM vO -dr^O tOHCM -4^00 H vOCM to-:*' I CM o to-*o to I 1 u c^oo O HvOvO O H OS X •J3 CTnvOCM vOOItO OCJ^r-l r^ocM I ChVTN I CMvO (D OCMO W) ^ OCMCM O vOtO VN *H f^C»-tO «> ja n M 9 B HOB ^•H « u 0) I 5 I o •a « o « « o »^H O^ O d a (4 a -p o as -P 0 - 50 - ON g! H»H ^ I Sun OcJ tf M H «^ S £3 (0 +5 bO a •H o. •H H O to VA I -CO C4 I to CM H CM I CM H 1 P4 H 0 6 1 •H H Arl Flo Sera H 0 0 u 8 n 4* o c ■s m a> o o .p ■a •H » ja o • 0} o tnH ©,Q ^^ « 3 ti -P m i '^ o -p n +} 5 I - 51 - TABLE 10 SUMMARY OF GRAIN YIELDS OF OATS AS AFFECTED BY DATE WHEN CLIPPING CEASED, VARIETY, AND CLIPPING TREATP4ENT, WINTER 1955-56 Pounds per Acre of Grain Irrigated Unirrigated Varieties: Arlington Floriland Seminole 490 5^0 1,040 210 L.S .D. at 5lo 110 80 Date when clipping ceased: Clipped until six weeks be- fore average heading date 790 Clipped until two weeks be- fore average heading date 610 L.S*D« at % 30 Effect of clipping treatments: 6 inches tall, 2- inch stubble 760 12 inches tall, 2-inch stubble 700 12 inches tall, 5- inch stubble 950 \t inches tall, 2- inch stubble 540 18 inches tall, 5 -inch stubble 560 18 inches tall, 9-inch stubble 720 L.S.D, at % 110 604 281 80 310 400 450 450 510 520 90 52 - a ^ * « * « « « « 1 sS -4-^ CO ITkrH >Ow\ -*r^r*\-* Q •P * « g 5 t- O •rt e I: ^ •H tot»-c\i OCMO <0r».-4O w> «0 13 (Q • • • • • • t • • • • • S (h 0) rH«AO ONCVO* -*O«0CM-4- iH 22« Is ^8 ^J4« ^I^otc?^ ^ M .« «k . .« .* « . .«>«.«.* « <% gs 3 o* U\r-i HHH Mr^H-c»-iH -* CO CO iH H H «r\ * tt So * it It * :<»-< r^\0-* rHOc^ >iau • • • • • • • • • • • »< -4-TvO QrHtO tOtAQOtO E^M 0^'^0^ ^f^OOiiH ^lA TJ S O* r-ir-ir-t vO-d-SrHiH Oo O CO m . .* * . .* « . * +> cr\ rMH HrH iH »ss: S) r-t Cfj irj vQ •iH toPuN i r^wr- HOM m^A ■COQvO r^e^^v%t>* CM CO ^^M^ CQc^ONCM S « ^B^ . .* * . .* t. , .« .» >• .« » « a> 3 & Or-i r^CV tO-*-*HC«- ^ > s**! CO CO evi H HH Q) H !>^E-^^^ si •P 6 • 0) og fe if\(\ro HCMO u^ou^o«^ ir\ 0) • H . O -H G) ti •• > 05 > U ■N- +9 © P O -P X ^ J3 *3 X H X p O e (D 0) ^ ?H ^ (0 10 <0 e •P43 +>4J 0) m to U U ^•H •H-H«H MUNMUN (M CM I I I I CMCM I I CVJCNJtOtO I I CM \OHiHrHrHvOvOH O C»N O O CM ^ as 43 tdvO vr\0-4- r^r«N C^nO wn i s ♦i-p 0} 4:) 43 4^ Cir-i <» U U-ri ^••H S 4J-PH O g MfH Q) >> ttO bO'4 (^ CO •p q c •H 43p t>oo (4 0) (3 u +5 o •a « O ■P • V^45 23 U ^\OvO>0 CM OOOOOOQO NO CM -* !>. OmH TO vO vO C^nO -4- C^ O C^ v\ rH rH H W rH H fH (H OOQOOOOO : CV 0) AX! A3 +>+>+5 CMU> +j 4i +>+» 0) 01 m t:^ h ti'H ^?:J'^ IU> OJ CM 1 1 1 1 (V0( 1 1 MCV-tOtO 1 1 CM vOrHrHHHvONOH I CM O (M ■CO o CM o CM > (1) rH 4^ n a oo CM CM n s a> S +> fi •• a> G e 0) n -P 0) 0) Ih »ta • ^^ • ^ +3 CO Js43 • (tf d K^l >o JO 1 I s O I. •H O • ^ .o s> u S *^

!x| tt ■4--4-M .«» « -*oto to ■K- « « mo "^ t/\too NO -CO.* .« .« « . * • iH»AH o^w 0^ ^^& On C^ VNo^vO o .« >* * «» r«-OrH ITv NO-*-* f^ (M ^ m O H CO I u oo w > Q> H ■P C •p « -p § o •H & •H CO ♦ u •p tt 1 o •H «H •H % •H CO * « •P 5 - 59 - a ^ 3 Of cow CO n 10 COCO CO n <0 COCO c^ 5 -* o^ c^ »r\ <0 r^ r^ •eo to CNJ o CM c<\ o H N r^ Ov »> • to O^ »f\ NO ev to to »f\ On -* vO »rv « * to CM rH H -4- O -t to to -* o to CM -* r-l c^ to CM n a O (0 'H •P 43 c « (1> O s •H +3 H <0 g ci H u o vO UN • to to • (9 O P P 9 Q> g g •p 4 § 1§ P § o •H •H CO CO * P s - ou - irrigated oat plots January 15 are shown in Table 16, Close frequent clipping, as with the six-inch plants cut back to two inches, seriously decreased root and crown weights. Allowing the plants to grow 12 inches tall before beginning the intensive defoliation treatment of cutting when six inches tall did not increase the root production. Differences between varieties were highly significant as shown in Table 17» Floriland produced a much larger quan- tity of roots as compared to the other two varieties. The reason for this is probably twofold. First, Arlington growth was severely retarded by the fungal infestation. Secondly, the growing point of an erect growing variety like Seminole is more easily destroyed by clipping, thus disturbing root development. The data for Seminole root yields indicate that root growth was decreased to a greater extent under two than under five-inch stubble for both 12- and Id- inch plants. This suggests the likelihood that re- tardation of root growth may be more serious with close clipping of erect growing than decumbent oat varieties. Freeze damage to irrigated Floriland oats was se- vere in early January on plots vAich had recently been cut to leave a two- inch stubble. Minimum daily temperatures in Table IS indicate that under very cold conditions the air between plants with high stubble may be two or three degrees warmer than where stubble is short. Unfortunately, no more S \0 >A>5 »AP<-tOTOTO 0) • •••••*• k X rH 0* (M CM M W N H 0) H o 0\ o t^ Q> rr\ H -* C^ -* t3 NO C ^ll • ••••••• • 8 g 0«HiHHNCM»HrH H >^ • S PQ CO Q 1 3 ;r\vo o to O ■00 ^ iV •ri • ••••••• • L d "■C X M r^-^r^r^-i--**^ r^ ^ 0> S3^ O H CC4 H b* f co>0 »r> >rk»r\ CO ISC^ s H^ 1 ^ • • oo 5 ^ Ov CN »ri »r\ H M vO H 1 »HWCMW>rv»AOM ON S^ • ••••••• rH CM CM CM Ol iH M H r4 5 H O • 2 1 m tx •-) |V) !2; n o ^ §^ s H •• 43 4 OE- d •H CO O S^ E J3 SB (^ h ;< B m 43 0 S^ o o» o 43 • ?5 ^ « < « • t*H S*^ ££S Q &§> s 0) 0) (1) • ® ft 2| « o 5«5 CO U43 • (d 0 u 1 1 1 ^ >o 55 ta CMCMnO 1 IH t Ht-i • 1 1 III *First e height o *H*H 5 4J CM WNCMvrv CM S 4J CM 1 1 1 1 CMCM 1 B 1 (MCMtO-CO 1 1 CM sOHrHHHVC»OH ^ n O s Ti •zvd. • • • OOO • • • cMr^o vor-^ Vf> r-\- •H o* £3 !g -*MtO o\ (0 +3 O a. s. d o •H CO—* 43 • 0) H I u 0 OOOOtOtO.*>ONOOr^OO>ONO • • ••••••••••••••• MO CM r^tO-i-M OnM r'Ao^O »A<7\C>i M^ CM-* NO vO UN VN ir\ »A u^vO vO w> >r\ UMA VMA -4"0 rH-CO-«0-*-ONO-4"-*>Or^NOCM-4-CM • • •••••••••••• OH rHOiHOOOOOCMiHOH CMtO CM O -too -00 \0 NO CM NO-* CM to too-* NO-4-HtOCMCM r^CM-d-NOMf^-* irNUNUM/MTNNONO U^ UN UMTV «N »r\ CMr^ QSnD to >rv^ to rH NO O to CM to On CM NO NO CM O to NO CM CM e<. CM H ON r-» c^nO fH CM {v^ VO *r\UNVNVMfMf\NONO UMTVUMTvUMrv toto OtoOtoONto CMtoONOtotoo OnOn rHtONOCMf-JC^iHiHCMCMCMNOrHHr-l \0 ITv JTv UM/MA UNNO NO UN U^ »/\ U> UMTN -4- O CM -* r- H IS WN C^ On -* On iTi t*w rH • •••••••••>»♦• O OOOOOOOOOHrHOfH OWNCMVN OOtOCMNONO.^CMtOtOC^OtOOtO C^On-*0 f-i -*CMCM-* tONO u o o 0) OtOCMr^ jLivo • • • » » VNvO iH ON O O -d-CMCMo^SNO C3nnO<*NCM t^ CM cm <^ cm fntO CM r^r^ ir\UMr\uMr\ir\\ONO um/njtvunwmtv ONOtoo^c^^-^-^^cJNu^r^f*^<*^^s. • ••••••••••••• CNNOCMiHt^HH C^r-{ (^vOrH CM CM v\ »r\ umtwmtvnO NO WNUNUMfMrvvrv C^tO (JNO r-i CM rrN_*lANO t^to OnO iH iHfHiHCMCMCMCMCMCMCMCMCMCMr^r^iHCMo^.4-»r\ co93(O(tf3333S3Sc0((S<0a) t3 • ••••••• ••• HrHiHOr-IHOO OOO ^tOOC^OCV»«>.rH OWsO ooc^iu^^-CMOr^ o o to tO<0«Hr>~\-4-HOiH OnO O^r-i fi iTvtOvO • ••••••• OrHOOfHOOO 00 OCM>0O0vCMirv\0 iH-!tO -4-tOM OtOONOvOCNJlVvO vooto retool OONM • • • OOH CVlOOOvOWO^ vOOON • ••••••• ••• On O M U^vO Oi rH CV -d-CJxM -4- u\sO irv vwo -* ^ -d" • ••••••• ••• to OiH -*vr\fH O r^ O r'^tO iH -4--d-\0 trwwo -tr^SS r^o^-4- O fH CM r^ -4- i^vO C^ SO C^tO OfH iH «H rH r-l iH iH rH 5 ON O 0) a> t3 0) EO O I o •H 4J 4:> H to I «0 I tJO O 25 O is o M I O o a: o H O H O O O O O O O r>^ NO >0 ^0 ^ ^ vO O ON o O o O O O O O r^ fN c«^ r^ r^ r^ (*> o UN «f\ vr\ ITN IfN lf\ O <»N iH i-l fH H rH iH CM Of •P CTn o >rs t-i E-* M -;*• CM C^ iH CM •w CM 43 +> C » S jj j3 m 1 1 •-D 1 < •< (0 o CO - al - Three cutting treatments were imposed on each of the nitrogen levels; (1) Cut when 12 inches tall to leave a four- inch stubble. (2) Cut when 30 inches tall to leave a foiur-inch stubble* (3) Cut when 30 inches tall to leave a ten-inch stubble. The experiment was planted May 2 and the stand was destroyed by a sandstorm. Another planting made May B was washed out by \musually heavy rain,, A successful planting was made May 26 in another area. The Starr variety was planted at ten pounds per acre in 19- inch rows. Each plot was nine feet long and consisted of four rows, of which 0,0005 acre was harvested from the two center rows for yield data. Harvesting methods were similar to those em- ployed in preceeding experiments. Forage samples were col- lected from two replications on each harvest date, dried, ground in a Wiley mill, and analyzed for total nitrogen by the KJeldahl method (4). Crude protein was expressed by multiplying the nitrogen content by 6,25. Changes in the stand of plants were studied by counting the numbers of live plants in a one-foot strip of row. One count was made in each plot on three different dates, A crude measure of the effect of nitrogen level and cutting treatment on root production was obtained by taking a core four inches in diameter and six inches deep from - 62 - each plot on June 2& and August !• liach core was washed on a screen to remove the soil and the root material was dried and then weighed. Trash and rocks were carefully removed. Cores were obtained immediately adjacent to the row of plants rather than directly below the row. Previous ex- perience had shown that it was difficult to decide between crowns and roots so this problem was circumvented by samp- ling adjacent to the row of plants and excluding the crowns. Effect of planting date. 1956 and 1957. Pearlmillet is usually considered to be a plant which makes its production over a very short period of time. The question arose as to whether this growing period is influenced by the date of planting, stimulating the fol- lowing experiments during 1956 and 1957 • The 1956 experiment originally consisted of six dates of planting at monthly intervals from March to August arranged in a randomized block design with four replica- tions. Severe sandstorms destroyed both the March and April plantings; consequently, the yield data are based only on four planting dates. The Starr variety was planted at ten pounds per acre in 3S-inch rows. Each plot was 16 feet long and consisted of three rows, of which 0.001 acre of the center row was harvested for yield. Plants were clipped by hand to leave a four-inch stubble whenever they reached 30 inches. At each planting, the - 83 - following fertilizers were applied: 400 pounds per acre B-d-df 50 pounds magnesium sulfate, and 50 pounds fritted trace elements. Additional nitrogen was supplied by am- monium nitrate whenever the plots were cut as shown in Table 22. In 1957 the experiment was modified, the changes being based on results from the 1956 season* The Starr variety again was planted, this time in 19- inch rows. In- dividual plots consisted of six rows nine feet long, of which 0,0005 acre of the two center rows was harvested for yield data. Plantings were made monthly from March through July, using a randomized block design with five replica- tions. Plants were clipped when they reached a height of 30 inches, leaving a ten- inch stubble. The initial fer- tilizer application at each planting date was the same as in the 1956 experiment. Nitrogen applications are shown in Table 23. Samples were collected at a number of harvests, dried, ground, and analyzed for nitrogen by the Kjeldahl method (4)* . d4 - TABLE 22 POUNDS PER ACRE OF NITROGEN APPLIED TO STARR MILLET DATE OF PLANTING TEST, 1956 Date Fertilizer Planting Dates Applied May 1 J\me 4 July 6 August 3 Initial planting application 32 32 32 32 June 21 66 July 3 66 July 6 66 August 10 66 66 66 Total 164 164 9d 9d - 65 - < Ha M n 1 UN rH 1 1 4 1 0* 1 4 1 1 o; N •H H •H t Q 0) •H H CM M ugh the season than plants clipped less often. There was a small decline in protein content toward the end of the growing season, the decrease being more pronounced in the less frequently clipped plants. The protein contents were affected only slightly by the height of the stubble. The protein content - 90 - o o a. bJ CO o O 00 IT o 5 H H I 39vaoj AMQ Ava/aaov/sen 3av §s e^ i bO ©•H o o A •d +> O EcH a. o t-\ o o »^ 0) 0) « > Q) « CO ! § • C\i M CV -H • -P bO^ ^ hO En* -H 0) xs, *i a (8 cu TABLE 26 EFFECT OF CUTTING TREATMENT ON THE CRUDE PROTEIN CONTENT OF IRRIGATED C(M40N PEARLMILLET, 1955 Height of Forage at Cutting Height of Cut (Stubble Height) Percent Crude Protein* Inches Inches June July- August 12 12 16 IB IB 30 30 30 30 54 54 54 54 10 i 10 Id 10 It 2«.l 25.6 23.6 27.2 30.4 25.7 24.9 26.0 23.6 27.1 25.7 23.5 24. d 25.2 20.4 21.6 22.5 14.4 22.4 20.4 14.3 22.7 20.4 13.0 23.1 23.3 16.6 l«.l 16.2 14.4 19.4 19.4 11.7 19.2 17.0 12.6 — 17.2 14.1 ^Each figure is an average of the analyses of two replications from the irrigated plots and expressed on the basis of oven dry forage. - 92 - was largely dependent upon the height of the plant prior to cutting and showed a surprising uniformity when only the height of the stubble varied. The only exception was the 30- inch plants cut with an IS- inch stubble which did have a sli^tly higher protein content* Since forage yields for 30- inch plants with Id- inch stubble were as high or higher than shorter stubble heights, it suggests that yields of protein per acre may have been increased by leaving a high stubble. The laborious task of collecting root samples yielded little information as to the effect of clipping on root weights. Table 27 gives an indication that the short, more frequently clipped plants produced less roots. How- ever, as shown in Table 28, the extremely large variation between samples resulted in none of these differences being statistically significant. Effect of cutting treatments, varieties, row spacings. and irrigation. 1956 A heavy infestation of crabgrass was encountered in the area in vdiich the plots were located. This, in ad- dition to daily rains during June and July, mads it virtu- ally impossible to obtain any semblance of weed control. Plots having plants clipped frequently to leave a four- inch stubble permitted more light penetration, resulting in heavier growth of crabgrass. The heavy growth of crabgrass '3 - TABLE 27 EFFECT OF CUTTING TREATMENTS ON ROOT YIELD IN IRRIGATED COMMON PEARLMILLcIT, 1955 Hei^t of Height Grams of Oven Dry Root Plants of per Core, Average of When Cut Stubble Three Replications Inches Inches 0-6 Inches 12 i 3.53 12 2.79 IB i 7.61 Id 5.60 16 10 2.12 30 4 6.56 30 6 6.74 30 10 d.17 30 16 6.91 54 ^ 5.63 54 7.55 54 10 6.05 54 16 6.76 L.S.D. at 5^ n.s. n,£ '•Not significant at five percent level. - 94 - TABLE 2^ ANALYSIS OF VARIAHCEj EFFECT OF CLIPPING TREATMENTS ON ROOT PRODUCTION OF" IRRIGATED COMON PEARLMILLET^ 1955 Source Degree of Freedom Sun of Squares Mean Square Replications Treatments Error 2 6.93 3.46 12 176,60 14.72 24 253.12 10.55 Total 38 436.65 ')'^ was probably responsible for the shortened life of the pearlmillet stand in 1956* The summary of yield data in Table 29 does not per- mit simple conclusions due to the presence of highly sig- nificant interactions as shown in Table 30. Individual treatment yields listed in Table 31 indicate the nature of the interactions. In every treatment combination, a row spacing of 19 inches was more productive than one of 3^ inches* How- ever, an even closer row spacing of seven inches did not generally result in an increased yield of forage. With 30-inch plants cut to leave an iS-inch stubble there was a considerable reduction in yield with a seven- inch row spacing, particularly with the Starr variety. Close and frequent clipping, as in the case of the 12- and 18- inch plants, resulted in a severe decline in the yield of forage. Quite predictably, the 54- inch cut- ting treatment yielded the largest quantity of dry matter although this material was coarse and stemmy. With common pearlmillet in seven- or 19- inch rows more forage was ob- tained from 30-inch plants cut to leave a four- inch stub- ble. This did not seem to be the case when the plants were grown in 3 S- inch rows. The Starr variety presents a some- what different picture. The four- inch stubble height favored the seven- inch row spacing but for plants in 19- or . 96 - TABLE 29 SimikRY OF IRRIGATED PEARLMILLET FORAGE YIELDS AS AFFECTED BY ROW SPACING, VARIETY, AND CUTTING TREATMENT, 1956 Pounds per Acre Oven Dry Forage Varieties! Starr Common L.S.D, at 5% 7i510 7,220 240 Row S pacing st 7-inch 19-inch 3 8- inch L.S.D. at 5% 8^200 7,810 6,100 1,040 Cutting Treatments? 54 inches 30 inches 30 inches 30 inches 16 inches 12 inches tall', talli talli talli talli tall. 4- 4- IC 1^ 4- 4- inch inch l-inch t-inch inch inch stubble stubble I stubble I stubble stubble stubble L.S.D. at 5% lli700 8^060 ?•% 4i750 4,320 480 - 97 - TABLE 30 ANALYSIS OF VARIANCE x FORAGE YIELDS OF IRRIGATED PEARLMILLET AS AFFECTED BY ROW SPACING, VARIETY, AND CUTTING TREATMENT, 1956 Source °^Sree of Sum of Mean Freedom Squares Square Main Plots: Replications Varieties Error (a) Subplots I Row Spacing Row Spacing x Varieties Error (b) Sub-subplotst Cutting Treatments Cutting Treatments x Varieties Cutting Treatments x Row Spacing Cutting Treatments x Row Spacing x Varieties Error (c) Total 143 1,121.26 3 1 3 1.94 3.09 0.61 0.65 3.09* .20 2 2 12 119.^1 0.29 6.52 59.90** 0.14 0.54 5 655.24 171.05** 5 21.79 4.36** 10 30.30 3.03** 10 90 19.22 62.45 1.92** 0.69 *Significant at five percent level. ♦♦Significant at one percent level. - 9B - -to^ c^r^vo >-( 0 . .A .«k • .■» .» * (3 PQ 0) f^ to top- tO«0vO 0) vO bO ■s Q»r\ C0 0 ^^ CO o fe 0) CihEh* >» -4- 000 -4-OCM ^WCM -g fcS ;:< 1 -OOOitO t^p-to .s <^ Q 0 . « . -t» \OvOvO 0) 3 1 vO-d-o^ -4CNJVO Xi ■< « 0 Oi ,« .«k m . .* ,A » CJ>H Ok r-i irs^ry u^-4•CM 0 HH +3 ga CO 0) H Cd Ur-i << OXi Cr, > «HX) o Si +3 CO n 9 $1 (i> a So+> >H •H . ,11 ,gg 0 s ^ g«« X g«K CO +3 o •H jcjx: D J3^ •ri bO fc ^^ jCOO "H J300 fe'H ♦H 1 1 > -H 1 1 <0 0 X C{ 1 ON-00 1 OstO 0) 0 r^Hr^ +> E n) 0 *> CO 0 CO •H - 99 - 3d- inch rows the higher stubble heights of 10 or 16 inches were more productive. Some light is shed on this problem by examining Table 32, Here it will be seen that 30- inch plants of the Starr variety cut to four inches yielded the same number of clippings under all row spacings. However, 30-inch plants with 10- inch stubble yielded one less cut- ting and the plants with Id- inch stubble yielded three less cuttings when comparing the seven- inch row spacing with the other two row spacings. From this it may be concluded that leaving a high stubble on closely planted Starr millet either decreased the rate of growth or appreciably shortened the productive season. High stubble appeared to favor the production of Starr pearlmillet in 19- or 3d- inch rows. In order to test the significance of this statement, a separate analysis of variance was calculated in Table 33 on the Starr variety in 19- or 3d- inch rows. The summary shown in Table 34 indi- cates that the higher stubble heists definitely favored growth of the Starr variety in 19- or 3d- inch rows. When plants were clipped at 12 or Id inches in height and grown in 19- or 3d- inch rows, both Starr and com- mon varieties made a surge of growth in late May and then declined to a low level for the remainder of the season. The productive life of the planting was approximately two months. However, with a seven- inch row spacing the - 100 - OO (0 I <© ftJ I o o O 0) (0 iH 0) ©CO-— 35 fi 0) O4O ■CO vO -4- vO >» 0 r-i (g S . •T> 5 0) «) s >. r^ •^ ■p > >» 0) 0 H •H OJ 3 U •^ & • •H 0) fc ck S cd iH •^ +3 CO >> > iH 0 3 •-3 a ® 1 •-D > >» 0 H Pi S •-3 i 0) 1 •CO g r^ -3 >, 43 Q> > >» •H a "3 OS ►•3 t> • c ^ 0) g ON s g rH '^ o o >. ^ rH 0 :3 • •-3 5 1 0) C^ ►-3 . •■— > 43«hH • hOOXS C CO'^ , 4i— . 43CH q • txoo n G c L,-- 1 ^ ITN 0 f-- r-i •n • • • • • 1 0) r^ t^ 0 ON 0 i to Oi q m CM CM CM CM CM CM tJ H c^ rH -d- 9 ♦ '• 1 • • 1 r^ NO 1 0^ to 1 a W iH r-i H c 5 rA r-i -* CM -d- c^ ■p '• • '♦ • • • «s 0 CJN -4- vO C> NO 0 r-\ CM CM CM CNi H •H H O^ CM -4- -* £ • t '• • • 1 rH 1 NO to to 1 CM rH H rH $ CM ITv to CM -* ITN . • • • • • • NO r-i H CM r-i rH rH r-i i -* vO -* lA H NO • • t ■• • • ^ 1 •CO C^ to C>- UN C0 U CM CM CM CM H 0 CM 1 • CM • 0 • ON • CO U vO 1 -* C^ u> 0 •HTJ r-t rH H r-i H tf • • ON • NO • UN • • S)0 £v. ir\ »r\ «r\ C^ -d- •H CM CM CM CM CM fH O, *i V S ce: o tv 1^ » « r^ UA VA vO H H • • • vO CM CM CM O o^ c^ CM o Cn. r-i to • • • • o O CM CM H iH r^ xO rH H CM o Eh H > 0) a 0) a o (0 U CO * - 113 - ui (D OD 3 fO X u z « <0 00 (9 <0 _l o QC * UI u. >- < (M N O CO '*' ^ CVI HlMOUSab JO S3H3NI III CO X u z <0 <9 00 Z Q. <0 -I U tc Ml li. < >- < -i«0 o Ui -J m n i- w X u - (0 QL a. -J o CO o e z u ac - ca CO Jl O 00 (D ^ CM HlM0d93U dO SBHONI - 114 - Fig. 26, — Thirty- inch plants of Starr millet clipped back to four and 10-inch stubble heights, 1957< Fig. 27i— Regrowth of Starr millet as affected by stubble heights, six days after cutting, 1957. - 115 - Regrowth of individual plants was slower with closer row spacings as shown in Figure 25. Apparently, closer row spacing resulted in nwre total forage per acre but less forage being produced per plant. This is also at- tested to by the reduced number of clippings obtained with the seven- inch row spacing as shown in Table 41 • Seasonal production of the row spacings were similar but there was a tendency for 30-inch plants clipped to a four- inch stub- ble to cease growing somewhat earlier than high stubble plants as illustrated in Figure 2d« The reverse was true for comiBon pearlmillet in 1955, indicating that varieties may differ in response to clipping* The crude protein content of the forage (Table 42) remained remarkably high for the duration of the season under all clipping treatments except where plants were cut when 54 inches tall* Protein contents diminished slightly with reduced width of rows. The data suggest that frequent applications of nitrogen fertilizer were capable of main- taining a high crude protein content in Starr millet, even during September when plants were heading, thus overcoming the expected decline in protein with increasing maturity. - 116 - TABLE 41 NUMBER OF CLIPPINGS OBTAINED FROM STARR MILLET AS INFLUENCED BY CUTTING- TREATMENTS AND RCM SPACING^, 1957 Cutting Treatments* Row Spacing in Inches 38 19 12-4 ia-4 30-4 30-10 30-ld 54-4 15 16 12 11 11 d 5 4 4 9 t 6 12 11 6 2 2 2 *First number refers to hei^t of plant in inches when cut; second number is the height of the stubble # . 117 - ?,?><-. o ley/HOd kUQ Ava/3b3v/'sai oav a ■P ■P /J (B

•HiH IS 03 - 118 - -4- PQ ,0 s ■p u CO 0) . * 1 . . • ooto ... 17.2 19.1 17.8 . • . HOn-CO CMHH . • • NOJCM ■CO CMS- r^H too 1 r^H 1 to S 1 C^ 1 - r^H o H § -3 On vO H CM CM too vO-*-4- HtOtO CVHH CM NO to CM CM tO-d-CM ^»•c^H CM CM CM toor*- CM (0 c o •H 05 U «> O O u u 0 o 4 n c T3 O ®, CO O •H H W •H 0) CO h (0 =»^ H U {X X 0) (4 Q> C O O 0) CO +> o a CO ^ 0 '^ 5 s a 0) Xi ■p ■p s «H o p •a «H 0) J3 O P CO <1> Ur-i 0) Xi - to to CM H iHtO-t -*c^to CMCMrH tOOC^ 43 CO I -d--* Oto • • • OCM-* ONCMvO » t * CMCMH tOOvC^ C^ ■P CO vO c^ to NO >A 1 f^ iH 1 Q) 1 -Q +3 1 e (0 1 Q) 3> IK S 1 » ■P Ur-i «S UJ^ (1) -* « 0) ^ 1 ^N,0 E-t o .o ^ f^ ►^hH bO o 5 NO •P +5 -* s 1 o CM H O H +3 1 c O o •P r^ t4M « OM a> -4- < 0) f^ 1 ^s.ffi H O •^ > c<\ h4«H bO 0^=° 5 r^ •P +> -* 5$ 1 O M M O ^ H +> 1 C O S t e +5 ^.M <0 OJ«{ p -* < 0) t^ "^2 H O ^ ^ m •H M 45 ■P -* 3 O •P (0 a U «B (0 a Q -00 H M <*> O CM o ■CO <*N CM iH-d-f^ Q $3 O»A0 MvO «0 CM NO f^ -* MOON -^P^ON ONtO c^ to cn r^f^CM CMCMCNJ M(M CM CM CM CM CM O CM M NO CM M to On M CT» •cooNO M<0^ NO o r>Mrv>rv r\ir\NO »A 0^ NO ^ NO CM CM CM CM CM CM CM CM CM CM CM M CM NO M CM CM to to ONO-* CM On to NO NO (JN r*-r^r^ Mr^NO M O if\ CM v\ CM CM CM CM CM CM CM CM CM CM CM nOOnO MCMCM M r^O if\tO MOn MMMCMCM 4 r^U^tO-*ONNOO^ MMCMCM - 131 - fH H • • -CO H rH Ol s n ^ o C o • • o 0) 5d CM C •H to •• >o ITV O. o • • n -* O «i-< « • • o X r^ t^ u iH r-i n 5 5 ^^ -* >■ d •o • • c •p o (^ «rs (Q s a iH H a ^ ■2, 43 +> a W o C4 irv ^e, •§) • ^ -CO • tt ^ «) • CM s^ 0} o • tJ O ^4 a> • o o Q>jc: OS H +3 a) •H Xi n ^ 0) h 0) § 2 a. Q) « >< d CO s Q) a - 132 - the last harvest. Total nitrogen production per acre was calculated from two replications of the forage yield data and nitrogen analyses at each clipping. The total nitrogen yields for clipping treatment means, shown in Table 4^, indicate that 30- inch plants with high stubble were most productive. However, Table 49 notes a clipping treatment x nitrogen level interaction. Examination of the individual treatment yields shows that at the lowest level of fertilization there was little difference between clipping treatments but as the nitrogen level was increased, the 30- inch plants were more productive. A still further increase in nitrogen production occurred at the highest level of fertiliisation when 30-inch plants were cut to leave a 10-inch stubble. Nitrogen deficiency s3nnptom3 were prominent in 30- inch plants grown at the lowest level of fertilization. Yellow- ing was much less noticeable in 12-inch plants at this rate of nitrogen application. Numbers of live shoots declined under all managerial treatments as shown in Tables 50 and 51. The August 11 count indicated that the decline in numbers was greater where plants were clipped to a four- inch stubble. Toward the end of the growing season there was a very sharp decrease in live shoots on plots receiving the highest nitrogen fer- tilization, irrespective of cutting treatment. - 133 - TABLE 4d POUNDS PER ACRE NITROGEN COOTAINED IN STARE MILLET UNDER THREE CLIPPING TREATI-IENTS AND THREE NITROGEN FERTILIZER LEVELS, 1956 Pounds per Acre Nitro- gen Biweekly Total Lbs, per Acre Nitrogen Applied Cutting Treatments 'WT W^ Mean 15 120 115.6 93.2 105.2 104.7 30 210 139.2 163.0 200,6 174.3 60 390 156.3 206.3 271.6 212.1 Mean 137.0 161.5 192.5 L.S.D. between: At 5% level: Nitrogen Clipping means: treatment means: 22.4 27.1 *First number refers to height of the plant in inches when cut; second number is the height of the stubble. - 134 - TABLE 49 ANALYSIS OF VARIANCE: POUNDS PER ACRE NITROGEN CONTAINED IN STARR MILLET UNDER THREE CLIPPING TREATMENTS AND THREE NITROGEN FERTILIZER LEVELS, 195^ Source D.F. Sum of Squares Mean Square Main plots: Replications Nitrogen levels Error (a) 1 2 2 1,753.2 35,620.9 162.3 1^753.2* 17,310.4** 31.2 Subplots: Cutting treatments CT X N Error (b) 2 9i26l.3 d,596.0 2,201.4 4i630.3** 2,U9.0* 366.0 Total 17 57,600.1 ''Significant at five percent level. Significant at one percent level. ** - 135 - TABLE 50 NUMBER OF LIVE STARR MILIET SHOOTS PER ONE FOOT OF ROW AS AFFECTED BY NITROGEN AI^D CUTTING TREATIffiNTS, 1958 Date of Cutting Biweekly Nitrogen Application Tlbs/A) Cutting Treatments 12-4 30-4 30-10 Mean July 18 Mean August 11 Mean August 29 Mean 15 30 60 15 30 60 15 30 60 n 43 46 41 43 46 46 44 44 44 43 45 31 25 22 26 26 22 29 30 31 29 27 25 26 25 30 22 20 15 23 19 10 23 21 12 23 20 13 19 IB 19 L.S«D. between! Nltros«>n means; Cutting treatment means j At 5^ level} July 18 August 11 August 29 n.s, n.s. n.3, 3 n.s, *First number refers to height of plant in inches when cut; second number is the height of the stubble. - 136 - pa IS CM H (Q •00 * ooo vO H I lAOi o 000 O TV-CO • • • Oco,-* O-t-* « 00 r^ lAirvON • • • or-o HOtO tOOsOi O w^^'^ ocat^ • • • -d-tOfN NHtO to r^CM M o o CM W (V-*0 vrv CM 49 a ' n a> 0) B (0 > « a 0 Or-i 2 •H ■— +9 -^ •p c: « .« «^ •• (0 0)^. hO '— (0 0 hO flK , +> H & P &,-P P •• •H h 0 0} ■P X 5 «H +J +5^ ^ O. 0 iH Jh 0 3E-« C CEJSSW iH 00 W 5 (0 -9 S CO 0) o +> c8 o •H •H CO ^iw^'rr (T»M":jr" ti^vim-''' ttPi'«i?r'*^''^^wp!TfW'» ■" - 137 - Root production was drastically reduced early in the growing season by close frequent clipping as shown in Tables 52 and 53 • Although mean root yields for nitrogen levels indicated no significant differences, the August 1 sampling showed a highly significant interaction between cutting treatments and nitrogen levels. Root weights of 30- inch plants with a four-inch stubble sharply declined with the higher levels of nitrogen whereas the 12-inch plants showed a modest gain in roots as the nitrogen level was increased. Effect of planting date. 1956 and 1957 The loss of the March and April plantings of Starr millet by blowing sand in 1956 point out a hazard in early planting. The losses incurred in this experiment wei^ probably greater than would be normally expected under farm conditions because the small plot area was entirely surrounded by large open fields of tilled loose soil. The yield data of Tables 54 and 55 show that early planting of Starr millet resulted in a larger number of harvests and a larger total yield for the season. A highly significant reduction in yield resulted from planting later in the season as evidenced by Tables 56 and 57. The seasonal distribution of production in 1957 shown in Figure 32 indicates that the length of the produc- tive season was about as long for the early as for the - 136 - UN 9 1 a> S O 01 r-t r-i +3 1 c o +3 0) c^ 01 B §* 9) -* «4 U, 1 Eh o bO 5 *i *> -J- 3 1 o CM H S a> a (0 O 03 H ■P 1 •CO c O Oi Q) c^ 43 -4- 0 1 O (M H tj (u a> U-H OH -< O, P. u< fXC >. OH to bO JM iind tro wee ©•H^ II a< %ca 1 to CVi to CM o H CM CNJ -4- CM CM CM O CM 8 CM CM CM UN CM ON CM CM CM CM c^ CM o CM CM »rN O O H r^ vO 0} IT* 4^ m S3 « 0) S 43 0 CM a> CM o +> a> 4:* • CM Q H • a 01 g CO d43 3 (0 0) 0) S ^ 43 bOC £5 4343 •H 3 ao •vO 010 fiO CO CM 43 0> 1 43 n Q IS 43 G JC • 0) OfH 43 J3 ^ 01 :i U4i 0) 01 V^ (0 JC 43 SI •ri U 0) x> - 139 - UN W i gto 1-3 fri CO 3 C COCO 0> CO n o 1^ coco o c 0) 49 3 b o « C0{> 0) o s r^UNO iHQO OOO • • • OOO CMc^O CMOC^ OOO • • • OOO UNCMO n 0) W !> C 0) Or-J •H --» •P C O v\0 VOf-CJN f^OrH OOO CNITNO HHOi • • • • • • OOO OOO « « NvOr-l r^r^ON ^8& §88 OOO • • • • • • OOO OOO vOHCV OWN to VOQH r-IOH • • • OOO n •P o H CO M-*0 ^ +3 •p -— oo w NO NO -4- r-i 0) o c CO 5 5 - 140 - TABLE 54 POUNDS PER ACRE OF OVEN DRY STARR MILLET FORAGE AS AFFECTED BY PLANTING DATE, 1956 Harvest Date Planting Date May 1 June 4 July 6 August 3 June 21 2,350 July 5 3,520 July 19 3,190 August 3 2,070 1,540 2,050 August 19 1,220 September 7 2,100 September 11 1,900 Total 7,940 6,630 3,270 1,900 L.S.D. between: At 5% level: Planting date totals s 640 pounds - 141 - TABLE 55 POUNDS PER ACRE OF OVEN DRY STARR MILLET FORAGE AS AFFECTED BY DATE OF PLANTING, 1957 HftmroQ+' Fija^o Date of Planting I^iarch 14 April 16 May 14 June 14 July 15 May 14 1,560 May 2B 2,560 May 30 2,4^ June 19 2.400 2,020 June 21 2,060 July 9 1,S40 July 17 1,900 2,020 July 25 1,040 August 9 1.860 August 15 940 1,320 1,860 August 27 1,520 September 6 1,060 1,380 1,800 September 12 680 September 20 1,560 September 26 360 11,680 120 8,600 300 7,620 300 3,960 Total 3,080 h 3,0, between: At ?;. level: Plant in Lg date totals 1,400 pounds - 142 - TABLE 56 ANALYSIS OF VARIANCE: POUNDS PER ACRE OF OVEN DRY STARR MILLET FORAGE AS AFFECTED BT PLANTING DATE, 1956 Souxxje D*F. Sum of Squares Mean Square Replications 3 0.32 0.11 Dates of planting 3 93.45 32.82** Error 9 1.43 0.16 Total 15 100.25 ♦^Significant at one percent level. - 143 - TABLE 57 ANALYSIS OF VARIANCE: POUNDS PER ACRE OF OVEN DRY STARR MILLET FORAGE AS AFFECTED BY PLANTING DATE, 1957 Source D.F. Sum of Squares Mean Square Replications 4 1.07 0.26 Dates of planting 4 61.62 15.40** Error 16 4.97 0.31 Total 24 67.66 ♦♦Significant at one percent level. - 144 - o <0 o CM O O 4> <0 U u t» •p o o u (0 •H I ■p s o w I I 39VM0J AWa AVa/3«0V/S91 3AV .8 - U5 - later plantings. However, in 1956 the early plantings did not remain productive as long. This may be due to the low stubble cut used in 1956 as contrasted to the higher 10- inch stubble of 1957 • Another reason for the extended life of early plantings in 1957 may be due to the more frequent smaller applications of nitrogen used thiroughout the sea- son. Nitrogen fertilization was not applied after July on the May planting in 1956 and this may have been responsible for declining vegetative growth. Crude protein content, as shown in Table 5^* re- mained much the same throughout the growing season for all dates of planting. With the application of nitrogen at bi- monthly inteiTvals, it was possible to maintain the crude protein content of the forage at a high level fix>m March until September. - 146 - TABLE 53 PERCENT CRUDE PROTEIN* OF STARR PEARLMILLET PLANTJD AT FIVE DIFFERENT DATES, 1957 Dat e Planted Dats Samples March 14 April 16 May 14 June 14 July 15 I4ay 28 13.1 May 30 17.2 June 19 18,8 18.8 June 21 15.3 July 9 15.6 July 17 16.9 13.1 10.6 July 25 15.6 August 9 15.9 August 15 20,3 15.6 U.l August 27 16.2 September 6 lif.4 18.8 15.6 September 12 11.6 Mean 16.9 14.4 15.2 15. S 16,2 ®Each figure listed is an average of the analyses of two replications and expressed on the basis of oven dry forage . DISCUSSION Oats Cutting treatments had less effect on forage yields of oats than on pearlmlllet (Tables 4» 12, and 20) « This Is probably due to low temperature being a much more Impor- tant factor In limiting the growth of oats during the win- ter months. The relationship of time when grazing or clip- ping occurs to the time of a hard freeze Is likely to be much more Important than the Intensity of the defoliation alone* Forage yields of all three oat varieties tested were sharply reduced by clipping or grazing when the plants reached a height of six Inches (Tables 1, 2, 12, 13, and 20), The reduction In yield was nearly as severe with Ar- lington, a decumbent variety, as with Seminole, which has an erect growth habit* Apparently, when forage of Arling- ton, a decixmbent variety, was lifted up to remove a pro- portion of the forage similar to Seminole, an erect growing type, the reduction in yield was severe with both varie- ties. Root development was seriously curtailed, regardless of the variety (Table 16). Yields of dry matter in the irrigated plots were in most cases doubled by allowing plants to grow 12 Inches - 147 - - us - tall before cutting. These results substantiate previous studies with oats by other workers (40, SO, 39, 45 )• Under unirrigated conditions the beneficial effects of allowing plants to attain a height of 12 inches before clipping were markedly decreased. This may possibly be due to higher transpirational losses of water by oat plants having more leaf tissue as is the case of plants cut when 12 inches tall. Close clipping of six- inch plants the first time followed the remainder of the season by cutting when 12 inches tall did not change total yields appreciably from that of plants clipped all season at 12 inches. Likewise, permitting plants to grow 12 inches tall the first time and subsequently cutting when six inches tall, resulted in yields similar to cutting at six inches all season. An explanation for the failure of this initial clipping to have much effect on subsequent yields is suggested by the root data in the 1956-57 experiment (Table 16). Allowing plants to grow 12 inches tall before the first defoliation did not permit adequate development of roots to compensate for the intensive cutting at six inches the remainder of the season. Clipping at twelve inches in height, in addition to increasing total yields of dry matter, also provided more forage during the midwinter period (Figure 11). This - 149 - treatment also continued to be somev/faat more productive late in the growing season than other cutting systems. Clipping when 12 inches tall was not frequent enough to reduce root growth but may have stimulated more tillering with consequent increased vegetative growth. Allowing oat plants to grow IS inches tall before cutting resulted in little or no yield increases except in the case of Seminole, a variety having an erect habit of growth. In addition, this cutting treatment resulted in the productive season beginning considerably later in the winter. Observations made of individual plants indicated that the growing p)olnt was decapitated sooner in IS- inch plants than with more intensive defoliation. It has been shown in other work (S3) that the recovery power of these tall plants is quickly reduced, resulting in fewer clip- pings or grazings. Stubble height after clipping had little or no di- rect effect on forage yield. There was an indication in the 1956-57 experiment that a five-inch stubble may enco\xr- age more root growth in the Seminole variety (Table 16), This would be reasonable to expect in view of its erect habit of growth, thus making the growing point more vulner- able to cutting. Indirectly, close clipping or grazing Just prior to heavy frost was responsible for serious stand losses by - 150 - freezing, I^aintenance of a higher stubble during critical cold periods may be a factor in protection against freezes # During succeeding freezes, temperatures in high stubble were observed to be several degrees higher than in low stubble (Table IS) but this was probably not a factor of major importance. The very small amount of freeze injury on the tops of high stubble plants suggested the possi- bility that there was an inherent difference in cold resis- tance between plants of the two stubble heights. The winters of 1955-56 and 1956-57 were somev^at drier than noiroal, thus accentuating the response to irri- gation. In both winters the major contribution of irriga- tion was in stimulating growth during December and January (Figure 12), Although the yield per acre of forage pro- duced during this period was not large, this amount of green feed might be quite important in a dairy pasture pro- gram. The value of irrigation for forage production early in the winter was almost cancelled by clipping plants when they were only six inches tall. The necessity of maintain- ing proper plant height under irrigated grazing conditions is obvious. Grain yields of oats resulting from early cessation of clipping were extremely low and the results of question- able value (Table 10), Under the conditions of this test, the value of forage produced during the pre-heading and - 151 - heading period would undoubtedly far exceed the value of the grain* It is interesting to note, however, that under irrigated conditions clipping at a hei^t of 12 inches to leave a five- inch stubble was beneficial to grain yields. Under low moisture conditions, clipping treatments had little effect on grain yields except in the case of plants cut at a height of six inches where the yield was reduced. A five-inch stubble height was beneficial only to the erect growing Seminole variety when clipping ceased six weeks before the average heading date (Table d). When clipping was continued until two weeks before heading, the value of higher stubble for 12- inch plants was apparent with all varieties, regardless of growth habit. This would siiggest that where grain production is expected after a season of grazing, it would be well to maintain a higher stubble height later in the season to avoid destroying the grofwing point. Pearlmillet Total dry matter production of pearlmillet was chiefly dependent upon the height of the plants ^Aien they were defoliated. Common pearlmillet, at all row spacings tested, was most productive when allowed to grow 54 inches tall (Tables 24 and 31 )• However, 30- inch plants of the Starr variety, grown in seven- or 19- inch ro%rs and cut - 152 - to leave a four- or ten- inch stubble, resulted in yields approaching those of 54- inch plants (Table 3B)» Both com- mon and Starr millet cut at a height of 12 inches yielded about half that of 30- inch plants although the reduction was less severe in seven-inch rows (Tables 24, 31, 3^, and 43 )• Likewise, yields weire reduced by cutting plants when 16 inches tall but to a lesser degree than with 12- inch plants. The forage yield reduction from frequent close clipping is most probably a result of reduced root growth. Root production of 12-inch plants was only about half that of 30- inch plants one month after planting as shown by data from the 195^ experiment (Table 52). Stubble height had a minor effect on total dry mat- ter production of pearlmillet. Yields of 54- inch plants were reduced with an Id- inch stubble (Table 24). Thirty- inch plants generally were indifferent in yield response to various stubble heights. However, the 1956 experiment indicated that the Starr variety in 19- or 3 6- inch rows may respond favorably to a hi^er stubble than common pearlmil- let. High stubble was found to be more productive under very heavy nitrogen fertilization (Table 43). As pointed out in previous work (20), this may be due to a shorter period of root growth stoppage with high stubble than with close clipping, thus permitting greater utilization of the - 153 - additional nitrogen* Cutting to leave a 10- or IS- inch stubble on 30- inch plants extended the productive season of Starr millet in the 1956 and 1957 experiments. This may also be the reason the 1957 date of planting test, with a 10- inch stubble, remained productive longer than the 1956 test, cut to leave a four- inch stubble. It is interesting to note that in 1955, common pearlmillet remained productive longer with a short stubble, indicating that varieties do not re- spond the same, Regrowth on 30- inch plants was much more rapid with a 10- or Id- inch stubble than with a four- inch stubble as evidenced by the 1957 results (Figures 3^, 39, and 40). This was mainly due to extension of new leaf tissue on high stubble plants vdiereas closely clipped pearlmillet plants, having lost their growing points, were forced to develop new leaves from buds at the crown. Nevertheless, the lat- ter did admirably well as indicated by the yields. Com- parable results have been reported by other workers (80, 50), This immediately suggests that the productivity might be further enhanced by cutting each time Just above the growing point, thus raising the height of the stubble at intervals throughout the season. This might result in less disruption of root development with consequent uniform for- age growth. - 154 - The additional productivity potential of an im- proved variety such as Starr may be negated by improper de- foliation. Both varieties gave similar low yields when clipped at a height of 12 inches. Data from the 1956 ex- periment (Table 34) suggest that the Starr variety may re- spond favorably to a higher stubble hei^t, although this was not fully substantiated in the 1957 and 195^ tests ex- cept under high nitrogen fertilization (Table 43 )• Pearlraillet row spacings of seven and 19 inches were definitely superior to 3^ inches, thus supporting pre- vious work in Georgia (19)» This might well be expected due to the erect growth habit of this species. Differences between the seven- and 19-inch spacings were not signifi- cant (Tables 29 and 3B) but the latter was more desirable as it permitted cultivation for weed control. Furthermore, the seven- inch row spacings generally did not remain pro- ductive as long as wider spacings. The erect, stemmy ap- pearance and sparse number of leaves on plants in seven- inch rows suggested that competition for light may be great- er, thus encouraging premature lengthening of the growing point. If extension of the growing point was greater v/ith a seven- inch ro^-^r spacing, then it would not be difficult to see how close clipping might shorten the length of the pro- ductive period. Although soil tensioraeters indicated a moisture - 155 - deficiency during part of the grovdng period, no increase in yield was noted from irrigation during 1955 and 1956* The value of irrigation for pearlmillet probably resides in its ability to aid in seed germination during a temporary dry period, thus insuring early stands • It is probable that the large differences noted in yield between dates of planting (Tables 54 and 55) w)uld not have been so pa?o- nounced except for the use of irrigation at planting. The date of planting was an important factor in determining pearlmillet yields (Tables 54 and 55 )• Total dry matter production was highest from March and April seedings with a steady decline in yield for each month of delay in plant- ing. One hazard of r^Iarch or April plantings in north Florida is the possibility of stand loss by sandstorms, a difficulty encountered in the 1956 experiment. All plant- ing dates showed the same seasonal pattern of a large initial surge of grovrth followed by a gradual decline in productivity. However, early dates of planting generally remained productive as long as those planted in midsummer. This may not necessarily be true under a lovrer level of nitrogen fertilization than was used in these tests. The decline in production is associated with a decrease in the numbers of live shoots, irrespective of cutting treatment, as shown in the 195^ experiment (Table 50). The reduction in numbers of shoots was greatest under hi^ nitrogen - 156 - fertilization. In examining the graphs showing distribution of forage growth over the growing season (Figures 22, 23, 24, 2d, 29, 30, 31, and 32), it is quite apparent that they all show a similar pattern of development — a sharp peak in pro- duction early in the life of the plant followed by a steady decline in the remainder of the season. This general trend seemed to be little affected by the date of planting, vari- ety, nitrogen level, or system of defoliation. Although it was quite possible to change the total forage production by various management practices, the general shape of the production curve remained the same. It is likely that a study of the phasic developnent of the pearlmillet plant would be needed to explain this phenomenon. Crude pix>tein percentage of the forage generally declined as the plants were permitted to grow taller before cutting (Tables 26, 35, 42, and 47) • Twelve- and Id- inch plants contained far more protein than could be economically utilized by livestock, suggesting that this would be a very wasteful grazing system, aside from its low yield of dry matter. Forage from the plants cut with an Id- inch stubble was somewhat higher in crude protein than where a lower stubble was used, due largely to the liurger proportion of leaves harvested. However, in terras of nitrogen produced - 157 - per acre, there was little difference between cutting treatments at the lowest level of nitrogen fertilization (Table 4^)* With additional increments of nitrogen ferti- lizer, the nitrogen returned in the forage increased more rapidly with 30-inch plants cut to leave a high stubble than for other treatments, due largely to the higher yield of dry matter. In application of these results to pasture prac- tices, it would appear highly desirable to allow pearlmil- let to attain a height of 24 to 30 inches before grazing is begun. It is obviously impossible to graze these plants down to a four- inch stubble and this apparently is not necessary as a higher stubble results in as much or, under certain conditions, more forage. It would seem likely that a system where livestock are grazed frequently on taller pearlmillet in rotational fashion, as in strip grazing, would result in more high quality forage being produced* SUMMARY AND CONCLUSIONS Field studies were conducted during the four year period 1955-5^ at Gainesville, Florida to determine the be- havior of pearlmillet (Pennisetum glaucum (L.)) under dif- ferential cutting treatments. Plants were cut when 54» 30, 1^, or 12 inches tall leaving stubble heights of IS, 10, six, and four inches. The effects of rxjw spacings, irri- gation, planting date, and nitrogen levels were studied to a more limited extent* Three varieties of oats (A vena sativa L, ) also were subjected to various cutting treatmaits under both irrigated and unirrigated conditions during the winters of 1955-56 and 1956-57. Plants were cut when they reached heights of 18, 12 and six inches in combination with stub- ble heists of nine, five, and two inches. Clipping treatments had less effect on forage yields of oats than pearlmillet. However, forage yields of three oat varieties having decumbent, intermediate, and erect growth habits were sharply reduced by clipping at a height of six inches. Dry matter yields under both irriga- ted and unirrigated conditions were generally doubled by allowing plants to groyr 12 inches tall before cutting. Root development was also seriously curtailed by close, - 158 - - 159 - frequent defoliation. Clipping at 12 inches in height pro- vided more forage during the midwinter period. Permitting the plants to grow 16 inches tall before clipping did not increase total dry matter yields for the Arlington and Floriland varieties but was beneficial to Seminole, an up- right growing variety. Stubble height after clipping did not directly af- fect forage yields of oats, A two- inch stubble resulted in yields similar to that obtained with cutting at five inches; however, heavy stand losses occurred where oats had been clipped or grazed to leave a one- or two-inch stubble just prior to a freeze. Grain yields of oats resulting from early cessation of clipping were extremely low. Under irrigation, leaving a five-inch stubble after clipping 12-inch plants increased grain yields. Clipping plants at a height of six inches reduced grain yields under unirrigated conditions but not where additional water was applied. Irrigation did not increase yields of pearlmillet but had a beneficial effect on oat forage production. Ir- rigation was most useful in stimulating production early in the winter. The value of irrigation early in the winter was erased by clipping oat plants vrtien they were only six inches tall. The dry matter production of pearlmillet, as affec- - 160 - ted by clipping, was generally in inverse proportion to the frequency of defoliation. Plants cut when 54 inches tall produced the most dry matter when a 3^- inch row spacing was used. However, when a closer row spacing of seven or 19 inches was used, the dry matter yields of 30-inch plants of the Starr variety were nearly as large. Yields of both common and Starr millet cut at a height of 12 inches were about half that of 30-inch plants although the reduction was less severe in seven- inch rows. Yields were also re- duced by cutting plants when IS inches tall but to a lesser degree than with 12- inch plants. Root production of 12- inch plants was only about half that of 30- inch plants one month after planting. The hei^t of the plant when cut, reflecting stage of maturity, determined the protein content of the forage. Accordingly, the crude protein content of 12-inch plants was twice that of plants cut vrtien 54 inches tall. Closer rov; spacings generally resulted in a slightly lower crude protein content. Stubble height had little or no effect on total dry matter production of pearlmillet plants 30 inches or less in height at lower levels of nitrogen fertilisation. However, the crude protein content of forage from 30- inch plants with a 10- or IS- inch stubble was somewhat higher than for plants with a four- inch stubble. Thirty-inch plants with a stubble - 161 - height of 10 inches fertilized at a high rate of nitrogen produced considerably more forage and total nitrogen than did plants vdth a four- inch stubble, Pearlmillet row spacings of seven and 19 inches were definitely superior to 38 inches. Seven- inch row spacings generally did not remain productive as long as wider spacings. The date of planting was one of the most important factors in determining forage yields of pearlmillet. Total dry matter production was highest from March and April seedings with a steady decline in yield for each month of delay in planting. Each planting date showed a peak in production duiring the first one or two clippings, followed by a gradual decline. Early dates of planting generally remained productive as long as those planted in midsummer. Numbers of live shoots on pearlmillet decreased during the growing season, irrespective of cutting treat- ments. The reduction in numbers of shoots was greatest under hl^ nitrogen fertilization. LITERATURE CITED 1. Ahlgren, H. L, Effect of fertilization, cutting treat- ments, and irrigation on yield of forage, and chem- ical composition of the rhizomes of Kentucky blue- grass. Jour, Amer, Soc» Agron, 30:663-691. 193^. 2. Ahlgren, H, L,, and - Burcalowi F, V, Bromegrass and alfalfa for hay, pasture, or silage. Wis. Agr. Col. Ext, Cir, 344. rev. 1950. 3. Aldous, A.E. Effect of different clipping treatments on the yield and vigor of prairie grass vegetation. Ecology 11:752-759. 1930. 4. AnonyTnouiS. Association of Official Agricultural Chem- ists— Methods of Analysis, Sth ed. Washington, D, C. 1955. 5. Archibald, E, S, Pasture improvement in eastern Cana- da, Canada Dept, Agr, Farmers Bui. 51. 193 S. 6. Baker, H, K, Studies on the root development of herbage plants, II, The effect of cutting on the root and stubble development, and herbage production, of spaced perennial ryegrass plants. Jour, Brit. Grassl, Soc. 12:116-126, 7. Bird, J, N, Stage of cutting studies: I, Grasses, Jour, Amer. Soc, Agron. 35:^45-661, 1943. 6. Biswell, H, H, and Weaver, J, E, Effect of frequent clipping on the development of roots and tops of grasses in prairie sod. Ecology 14:368-390, 1933. 9, Blaisdell, J, P. and Pechanec, J, F, Effects of herb- age removal at various dates on vigor of bluebunch wheatgrass and arrowleaf balsamroot. Ecology 30: 298-305. 1949. 10, Brougham, R. VJ. A study of pasture growth, Austr. Jour. Agr. Res. 6:804-612. 1955. 11, Brougham, R. W. Effect of intensity of defoliation on regrowth of pasture. Austr, Jour, Agr. Res. 7: 377-387, 1956, - 162 - - 163 - 12. Brown, E, M, Seasonal variations in the growth and chemical composition of Kentucky bluegrass. Mo. Agr. Exp. 3ta. Res. Bui. 36O. 1943. 13. Burger. A. W., Jackobs, J. A., and Hittle, C. N. Yield of sudangrass varieties as affected by time and frequency of cutting. Agron. Jour. 50:37-39« 195a. 14* Burton^ G. Wi, and Prine, G. M. Forage production of rye, oats, and ryegrass as influenced by fertiliza- tion and management. Agron. Jour. 50:260-262. 195s. 15. Canfield- R. H. The effect of intensity and frequency of clipping on density and vield of black grama and tobosa grass, U.S.D.A. Tech. Bui. 6^1. 1939. 16. Carrier, L,, and Oakley, ?.. A. The management of bluegrass pastures. Va. Agr. Exp. Sta. Bui. 204* 1914. 17. Cooper, C, S. The effect of time and height of cut- ting on the yield, crude protein content, and vegetative composition of a native flood meadow in eastern Oregon. Agron. Jour. 4S;257-25S. 1956. 18. Cotton, J. S. Improvement of pastures in eastern New York and the New England states. U. S. Bur. Plant Indus. Cir, 49. 1910. 19. Cragmiles, J. P., Elrod, J. M., and Luttrell, E. S, The influence of method of seeding and seed treat- ment on stands and yields of sudangrass and millet. Ga. Agr. Exp. Sta. Mimeo. Ser. N. S. 47. 195^. 20. Crider, ?• J. Root growth stoppage resulting from defoliation of grass. U.S.D.A. Tech. Bui. 1102. 1955. 21. Crozier, A. A, Forage plants and wheat. Mich. Agr. Exp. Sta. Bia. 141:130-132. 1897. 22. Cutler, G. H., Pavea, S. D., and Mulvey, R. R. The effects of clipping to stimulate pasturing winter wheat on the growth and quality of the crop. Agron. Jour. 41:169-173. 1949. 23. Davies, G. M. The effects of winter and spring de- foliation on the subsequent yield of grain and - 164 straw of S 147 oats. Jour. Agr# Sci» 47:363-366, 1956. 24* DePeralta, F, Some principles of competition as il- lustrated by Sudan grass (Holcus sorghum sudanensis (Piper) Hitch.). Ecol. Mohogr, 5:355-404. 1935. 25. Dexter, S, T. Response of quackgrass to defoliation and fertilization. Plant Physiol. 11:843-^51. 1936. 26. Dibbem, J. C. Vegetative responses of Bromis inermis to certain variations in environment. Bot. Ga?.. 109:44-5^. 1947. 27. Ellett, W. B., and Carrier, L, The effect of frequent clippings on total yield and composition of grasses. Jour. Amer. Soc. Agron. 7:35-^7. 1915. 28. Fagan, T. W. The nutritive value of pasture grass as influenced by management. Welsh Jour. Agr. 5: 99- 109. 1929. 29. Gemert, W. B. Native grass behavior as affected by periodic clipping. Jour. Amer. Soc. Agron. 2S: 447-456. 1936. 30. Graber, L. F, Food reserves in relation to other fac- tors in limiting the groT^h of grasses. Plant Physiol. 6:43-71. 1931. 31. Graber, L. F. Competitive efficiency and productivity of bluegrass (Poa pratensis L.) vrith partial de- foliations at two levels of cutting. Jour, Amer. Soc. Agron. 25:32^-333. 1933. 32. Graber, L, F., and Ream, H, W, Growth of bluegrass with various defoliations and abundant nitrogen supply. Jour. Amer. Soc. Agron. 23:93^-946, 1931. 33. Hanson, W, R,, and Stoddart, L. A. Effects of grazing upon bunch wheatgrass. Jour. Amer. Soc. Agron. 32: 27^-289. 1940. 34. Harrison, C. M. Effect of cutting and fertilizer ap- plications on grass development. Plant Physiol. 6: 669-6^4. 1931. 35. Harrison, C. M, Responses of Kentucky bluegrass to variations in temperature, light, cutting, and - 165 - fertilizing. Plant Physiol. 9:^3-106. 193h» 36. Harrison, C. M. , and Hodgson, C, W. Response of cer- tain perennial grasses to cutting treatments. Jour. Amer. Soc. Agron. 31:413-430. 1939. 37. Heady, H. 7, The raeasurement and value of plant height in the study of herbaceous vegetation. Tocology 3^:313-319. 1957. 33, Holscher, C. E, The effects of clipping bluestem, wheatgrass, and blue grama at different heights and frequencies. Ecology 26:143-156. 1945 • 39. Hoveland, C. S, Effect of early clipping on the forage production of oats. Texas Agr. Sxp. Sta. Prog. Rept. 1784. 1955. 40. Hubbard, J. C, and Harper, H. J, Effect of clipping small grains on yield and composition of forage and grain. Agron. Jour. 41:85-92, 1949. 41. Hudson, A. V.\ Investigations in pasture production. VI. Studies of effect of sulphate of ammonia and frequency of cutting on production of pasture under system of mowing only. New Zealand Jour, of Agr. 48:17-24. 1934. 42. Hutcheson, T. B., and Ifolfe, T, X, Thn fertilization and management of bluegrass pastures. Va. Agr. Exp. Sta. Ann. Rept. 1917-19:39-44. 1919. 43. Jantti, A., and Heinonen, ^., Effect of defoliation and soil moisture on grassland regrowth. Jour. Brit. Grassland Soc. 12:56-61. 1956. 44. Johnson, A. A., and Dexter, 3. T. The response of quackgrass to variations in height of cutting and rates of application of nitrogen. Jour. Amer. Soc. Agron. 31:67-76. 1939. 45. Justus, N., and Thurman, R. L. The effect of clipping and grazing on the subsequent growth of winter oats. Agron. Jour. 47:82-83. 1955. 46. Kennedy, W. Ki, and Russell, K. B. Relationship of top growth, root growth, and apparent specific grav- ity of the soil under different clipping treatments of a Kentucky bluegrass-;d.ld white clover pasture. Jour. Amer. Soc. Agron. 40:535-540. 1948. - 166 - 47« Kohler, G, C, The effect of stage of growth on the chemistry of grasses. Jour. Biol. Chem. 152: 215-223. 1944. 4^. Laird, A. S. A study of the root systems of some important sod- forming grasses. Fla. Agr. Exp. Sta. Bui. 211. 1930. 49. Lang, R., and Barnes, 0. K. Range forage production in relation to time and frequency of harvesting, ^o. Agr. Exp. Sta. Bui. 253. 1942. 50. Laude, H. M. Growth of the annual grass plant in response to herbage removal. Jour. Range Mangt. 10:37-39. 1957. 51. Leukel, W. A., and Bamette, R. M. Cutting experi- ments with bahiagrass grown in lysimeters. Fla. Agr. Exp. Sta. Tech. Bui. 286. 1935. 52. Leukel, W. A., Camp. J. P., and Coleman. J. M. Effect of frequent cutting and nitrate fertilissation on the growth behavior and relative composition of pasture grasses. Fla. Agr. Exp. Sta. Bui. 269. 1934. 53. Lowom, R. L.- The effects of fertilization, species competition, and cutting treatments on the behavior of dallis-grass, Paspalum dilatatum Poir., and carpetgrass, Axonopus af finis Chase. Jour. Amer. Soc. Agron. 36:590-600. 1944. 54. Lowom, R. L. The effect of defoliation, soil fer- tility, temperature, and length of day on the growth of some perennial grasses. Jour. Amer. Soc. Agron. 371570-5^2. 1945. 55. Marshall, S. P. Value of oat pasture for dairy cattle. Fla. Agr. Exp. Sta. Bui. 5^4. 1957. 56. Marshall, S. P., Sanchez, A. B., Somers, H. L., and Dix Arnold, P. T. Value of pearlmillet pasture for dairy cattle. Fla. Agr. Exp. Sta. Bui. 527. 1953. 57. Mcllvanie, S. K. Carbohydrate and nitrogen- trends in bluebunch wheatgrass, Agropyron spicatum. with special reference to grazing influence. Plant Physiol. 7:540-557. 1932. 58. Morey, D. D., Chapman, W. H. , and Earfiart, R. W. - 167 - Qrowlng oats In Florida. Fla. Agr. Exp. Sta. Bui. 523. 1953. 59. Morris, H. D., and Gardner, F. P. The effect of nitrogen fertilization and duration of clipping ' period on forage and grain yields of oats, ii^eat, and rye. Agron. Jour. 50:454-457. 195 S. 60. Mortimer, G. B.,-and Ahlgren^ H. L. Influence of fertilization, irrigation, and stage and height of cutting on yield and composition of Kentucky blue- grass TPoa pratensis L.). Jour. Amer. Soc. Agron. 2^:515-533. 1936. 61. Mott, G.O, Effectiveness of fertilization and manage- ment in increasing yields of pastures in Indiana. Soil Sci. Soc. Amer. Proc. S:276-2dl. 1944. 62. Myers, D. J. Grazing management of oversown pastures. Jour, of Agr. (Victoria Dept. Agr., Austr. ) 55: 247-248. 1957. 63. Newell, L, C, and Keim, F. D. Effects of mowing fre- quency on the yield and protein content of several grasses gz*own in pure stands. Nebr. Agr. Exp. Sta. Res. Bui. 150. 1947. 64. Parker, K. W., and Sampson, A. W. Influence of leaf- age removal on anatomical structure of roots of Stipa pulchra and Bromus hordaceus. Plant Physiol. 5:543-553. 1930. 65. Paterson, D. D. The - influence of time of cutting on the growth, yield, and composition of tropical fodder grasses. I. Elephant grass (Pennisetum purpureum). Jour. Agr. Sci. 23:6l5-6/iXi 1933. 66. Paterson, D. D. The cropping qualities of certain tropical fodder grasses. Empire Jour. Exp. Agr. 4:6-16. 1936. 67. Prine, G. M. , and Burton, G. W. The effect of nitro- gen rate and clipping frequency upon the yield, protein content, and certain morphological charac- - teristics of Coastal Bermudagrass ( Cynodon dactylon . (1) Pers.). Agron. Jour. 4^:296-301^ 1956. 68. Robertson, J. H. Effect of frequent clipping on the development of certain grass seedlings. Plant Phy- siol. 8:425-447. 1933. - 168 - 69« Robinson, D, H, The spring grazing of autumn-sovm com. Jour. Min, Agr» TEnglandT. 1951» 70. Robinson, R. R., and Sprague, V, G, The clover popu- lations and yields of a Kentucky bluegrass sod as affected by nitrogen fertilization, clipping treat- ments, and irrigation. Jour. Amer, Soc. Agron, 39! 107-116. 1947. 71. Sampson, A. W., and Malmsten, H. E, Grazing periods and forage production on national forests, U.S.D.A. Bui, 1405. 1926, 72. Sarvis, J, T, Effects of different systems and in- tensities of grazing on the native vegetation at the Northern Great Plains Field Station, U,S,D.A, Bui, 1170. 1923. 73. Schwartzbeck, R. A, Effect of row width and rate of seeding on the forage production of irrigated oats, Texas Agr, Exp, Sta, Prog, Rept. 1878. 1956. 74. Stapledon, R, G, , and Beddows, A. R. The quantitative and qualitative response of cocksfoot (Dactylis glome rata ) to sodium nitrate and to supe rpho s phat e , Welsh Jour. Agr. 2:103-113. 1926. 75. Stapledon, R. G., and Davies, ¥. Experiments to test the yield and other properties of various species and strains of herbage plants under different meth- ods of management. Welsh Plant- Breeding Sta., Aberystwyth Bui. Ser. H, No. 10, 1930. 76. Stapledon, R. G., and Milton, W. E, J. The effect of different cutting and manurial treatments on the tiller and root development of cocksfoot. Welsh Jour. Agr. 6:166-174. 1930. 77. Stoddart, L. A. Some physical and chemical responses of Agropyr^an spicatum to herbage removal at vari- ous seasonsT Utah Agr. Exp. Sta. Bui. 324. 1946. 78. Sturckie, D. G. The influence of various top-cutting treatments on rootstocks of Johnsongrass (Sorghum halapense). Jour. Amer. Soc. Agron. 22:82-93. 1930^ 79. Thurman, R, L, Small grain management experiments in Arkansas, Ark, Agr, Exp, Sta, Bui, 566. 1956. - 169 - do, Thiirman, R. L,, and Grissom, P» Relationship of top and root growth of oats. Agron. Jour. 46:474- 475, 1954, Si. Tin eke r. M. A. H, The effect of cutting to the ground level upon the grovrtih of established plants of Dac- tylis /ylomerata and Phleum pratense. Welsh Jour. Agr. 6:182-198. 82. Torstensson, G. Preliminary investigation by Swedish Grassland Association. Influence of the interval of time between cuts and height of stubble on yield and root development in grasses. Imp. Bur. Pas- tures and Forage Crops Vdjneo, Pub. 2. 1936» 83. Verbeek, W. A. Oats and udieat as grazing for high producing cows. Farming in South Africa 21:545- 547. 1946. 84* Wagner |» R. £. Yields and botanical composition of four grass-legume mixtures imder differential cut- ting. U.S.D.A. Tech. Bui. 1063. 1952. 85. Washko, J. B. The effects of grazing winter small grains. Jour. Amer. Soc. Agron. 39:659-666. 1947. 86. Washko, J. 6.. Edge, J* L., and Haskins. A. L. Graz- ing of small grain. Supp. No, 3, Bui. 540, 64th Ann. Rept., Penn. Agr. Exp. Sta. 1952. 87. Waters, H, J. Studies of the timothy plant. (Part I). Mo. Agr. Exp. Sta. Res. Bui. 19. 1915. 6d, Williams, T. E., and Baker, H. K. Studies on the root development of herbage plants. I. Techniques of herbage root investigations. Jour. Brit. Grassl. Soc. 12:49-55. 1957. 89. Wilsie, C. P., Akamine, E. K., and Takahashi , ■ M. Ef* feet of frequency of cutting on the growth, yield, and composition of napier grass. Jour. Amer. Soc. Agron. 32:266-273. 1940. 90. Wilsie, C. P., Peterson, M. L., and Hughes, H. D. Bromegrass in Iowa. Iowa Agr. Exp. Sta. Bui. P75. 1945. BIOGRAPHICAL SKETCH Carl 3, Hoveland was bom October 25, 1927 on a dairy farm near Sand Creek, Wisconsin, His elementary edu- cation was obtained in a local one-room rural school and high school vmrk in the nearby towns of Chetek and Colfax, Wisconsin, Upon graduation from high school, he enrolled at the University of Wisconsin in June 1945 and completed three semesters before entering the U, S, Marine Corps in August 1946. In June 194^ he reentered the University of Wisconsin, majoring in soils* The Bachelor of Science de- gree with honors was granted in June 1950 and after fur- ther graduate work, the Master of Science degree with a major in soils was bestowed in February 1952. From March 1952 until April 1955 he served as as- sistant agronomist at Texas Agricultural Experiment Sta- tion, Substation 19 » Crystal City, Texas, The work over this three-year period was mainly concerned with research on irrigated pastures. An interim assistantship in agron- omy at the University of Florida from June 1955 to Febru- ary 1957 was followed by a research assistantship in agronomy and later a teaching assistantship in the botany department. During this period of time, vrork vras undertaken - 170 - - 171 - tovard a doctor of philosophy degree \f±th a major in agron- omy and a minor in animal nutrition. He is a member of Alpha Zeta, Sigma Xi, and Gamma Sigma Delta honorary fraternities. Professional societies in which he holds membership are: American Society of Agronomy, American Society of Range I^Ianagement , American Geographical Society, and American Society of Plant Physi- ologists, He is married to the former Dorothy Anderson of Dodgeville, Wisconsin, This dissertation was prepared under the direction of the chairman of the candidate's supervisory committee and has been approved by all members of that committee* It was submitted to the Dean of the College of Agriculture and to the Graduate Coiincil, and was approved as partial fulfillment of the requirements for the degree of Doctor of Philosophy, January 31, 1959 Dean, College of Agriculture SUPERVISORY COMMITTEE: Dean, Graduate School (3 1 Ww^ V Uv:T-Aa 2 97 5S *^