SEED PRODUCTION OF SMOOTH BROME GRASS AS IT IS INFLUENCED BY TIME AND RATE OF APPLICATION OF AMMONIUM SULPHATE Thesis for the Degree of M. S. MICHIGAN STATE COLLEGE Willard N. Crawford I939 ~ 2]; 4 ' '.. £55. I U. I : ‘t" V . _ ,1 . Thy , .' - ‘ , 5 ”J :I' ' ’ ;-_,! , . ' .w‘ I v'w. | I , , 'I h 'u ‘ Jr, 7)" . 5%.. 4* 1 .4 , ‘ V r I I; 4/ I I. l r , , WI 3 L “(WV-a I 7’5“ 7 ‘- "S \ ,- 4.:- “I" M"... g. («a . \ . , : A 3] l by N . ~ 4‘47”!” v . f/ x." ‘1‘”, 1“: . “.3”: I" ,‘I . . I n; a n v_»’ V. ,5 f , I"/‘|,‘ , A I 'r‘ .I "I } £1} l“‘."": ‘A ‘ fl " I 1' a ‘fifi ' S. I I H w . v I . ”Mi. - I :_ ‘ ,- . "'I‘ | C: {(1 $1 77"? 0 “gr *‘Efiflz I ’ , , ,5” if? ‘. {fflk‘l’éfi‘i’iig‘w'fii‘ifmIé Ir: a. ‘ I”? I~~sw:'*;e'”img"vu.=. ,. I A“; ‘- .‘ I’IIILI‘;f:,'u‘VfV-'»L-Iu' ‘J A , ' M539%?“v‘IzIzmhwkzmi - fhiyyil[lfi& we! v... .M . I} 4' “(grid , {);I."I .i'.‘ ‘ . t' I’; I l “I I}? , WI SEED PRODUCTION OF SMOOTH BROII'EE GRASS AS IT IS INFLUENCED BY TIME} AND RATE OF APPLICATION OF AMMONIUM SULPI-DI'I‘E SEED PRODUCTION OF SMOOTH BROME GRASS AS IT IS INFLUENCED BY TIME AND RATE OF APPLICATION OF AMMONIUMZSULPHATE by WILLARD NELSON CRAWFORD M A THESIS Submitted to the Graduate School of Michigan State College of Agriculture and Applied Science in partial fulfilment of the requirements for the degree of MASTER OF SCIENCE Department of Farm Crops 1959 TABLE OF CONTENTS Introduction Literature Review Climate 1958 EXperiment 1959 Experiment Seed and forage yield Tiller production and panicle characteristics Coefficient of fertility Seed quality Residual Influence of the 1958 Treatments on the 1939 crop Relation between seed yield and cost of fertilizer in 1939 Discussion Summary Bibliography Acknowle dgment s IQSTib Page II 12 14 17 19 24 26 31 34 56 SEED PRODUCTION OF SMOOTH BROME GRASS AS IT IS INFLUENCED BY TIME AND RATE OF APPLICATION OF AMMONIUMTSULPHATE Current interest in the utilization of smooth brome grass (gromus inermis Leyss.) for forage purposes has created a demand for a supply of high quality seed. Although it has been established that brome grass seed can be produced in Nuchigan, no infonnation is available on the possibility of increasing the seed yield or improving the seed quality through the use of commercial fertilizers. The experiment here reported was designed to study the influence of the date and rate of application of ammonium sulphate during the first and second seed harvest years. In conjunction with the study, the yield and protein content of the forage at the time of seed harvest; the number of fertile and barren tillers; various panicle characteristics; and the quality of the seed produced were investigated. In August, 1957, a field of moderately fertile, slightly acid, Brookston loam soil on the eXperiment station farm.at East Lansing, Michigan, was fertilized with 400 pounds per acre of an 0-20-20 fertilizer. Bromegrass seed was mixed. with cats and planted in 28-inch rows with a grain drill set to saw two bushels of cats per acre. Approximately two and one-half pounds of brome grass seedper acre were planted. The experiment was designed so that 0 (control plots) 100, 250, 500, 750, and 1000 pounds of ammonium sulphate (81.0%1N) per acre were applied to triplicate 1/140~acre plots in mid-April,tmid-May, and mid-June. The field was divided into three areas, or replications, and each area was further divided into three blocks, one at random.for each date of application. The six rates of application were randomized within each block. This gave, then, a 5 (replica- tions) x 5 (dates of application) X 6 (treatments) plot layout. All of the eXperimental results were subjected to analysis of variance, and comparisons were made with reference to the average of the nine control plots. LITERATURE REVIEW Stapledon and Beddows (14) found that in orchard grass, nitrogen increased the production of seed, fertile tillers, total dry matter, the weight per 1000 seeds and the per cent of germination. Gilbert (7) obtained largest yields of seed of Rhode Island bent grass from those plats on which fertilizers containing large proportions of nitrogen were applied. Evans and Calder (5) found that applications of two art. and five out. of nitrochalk (12.5% N) gave increased seed yields of orchard grass, rye-grass, and timothy; andchcreased seed yields of red fescue and meadow rescue. With orchard grass, early applications were better than late and in all instances seed yield was stimulated comparatively more than forage yield. From their work with pedigree grasses, Evans and Snider further concluded that nitrogen stimulated the produc- tion of fertile tillers in rye-grass, orchard grass, timothy, and meadow foxtail; reduced the number of fertile tillers in red fescue; and on the whole, increased the weight per 1000 seeds; and tended to improve germination of grass seeds in dry years. In a later report, Evans (4) confirmed this earlier work and further concluded that the effect of nitrogen was intensified in a dry year and in crops subsequent to the first harvest year. He found also that nitrogen increased the number of barren tillers in orchard grass, Italian rye- grass, meadow fescue, red fescue, and perennial rye-grass; but in timothy nitrogen "seems to have been efficacious in con- verting a larger number of barren tillers into seed producing tillers." I North and Odland (9) observed a consistent increase in seed yield of Rhode Island Colonial bent grass with increas- ing amount of nitrogen applied, but indicated that there was a limit to the amount that could be safely applied due to the danger of lodging. No consistent differences were apparent in the test weight per bushel due to the fertilizer applica- tions. Under Ohio conditions, Evans (6) obtained increases in yield of both seed and forage of timothy with applications of sodium nitrate. With increasing rates of application there was a gradual decrease in additional amount of hay and seed for each pound of fertilizer used. The applications in May had a greater relative influence on seed yield than on for- age yield. Eighty pounds of sodium nitrate per acre gave an increase in seed yield of 148% over the control, whereas 620 pounds were required to give as great a relative increase in hay yield. In a review of the work of Nilsson-Leissner (8), Jones reported on the differential response to nitrogen fer- tilization between non-rhizomatous and rhizomatous types of grasses. Nitrogen produced marked increases in seed yield in the non-rhizomatous types; but, was not effective in the rhizomatous grasses. In a study of the influence of nitrogen on brome grass seed production in Kansas, Aldous (2) found that 150 and 300 pounds of sodium.nitratc increased the seed yield 78% and 167%, respectively, over the control. Schmitz (12), Osvald (10), Sanders (ll), Zahnley and Duly (18), Ahlgren (l), and others (4, 5, 6, 7, 9, 14) have shown that nitrogen fertilization tends to stimulate herbage production of grasses. Whilson 33. 21. (16), working with corn, oats, and barley, and Voorhees (15), working with barnyard millet, found that nitrogen fertilization increased the protein content of the herbage. Wiancko and Walker (17) and Sprague and Hawkins (13) concluded that early applications of nitrogen increased was a gradual decrease in additional amount of hay and seed for each pound of fertilizer used. The applications in May had a greater relative influence on seed yield than on for- age yield. Eighty pounds of sodium nitrate per acre gave an increase in seed yield of 148% over the control, whereas 520 pounds were required to give as great a relative increase in hay yield. In a review of the work of Nilsson-Leissner (8), Jones reported on the differential response to nitrogen fer— tilization between non-rhizomatous and rhizomatous types of grasses. Nitrogen produced.marked increases in seed yield in the non-rhizomatous types; but, was not effective in the rhizomatous grasses. In a study of the influence of nitrogen on brome grass seed production in Kansas, Aldous (2) found that 150 and 300 pounds of sodium nitrate increased the seed yield 78% and 167%, respectively, over the control. Schmitz (12), Osvald (10), Sanders (ll), Zahnley and Duly (18), Ahlgren (l), and others (4, 5, 6, 7, 9, 14) have shown that nitrogen fertilization tends to stimulate herbage production of grasses. Whilson'gfi.‘gl. (16), working with corn, oats, and barley, and Voorhees (15), working with barnyard millet, found that nitrogen fertilization increased the protein content of the herbage. Wiancko and Walker (17) and Sprague and Hawkins (13) concluded that early applications of nitrogen increased the yield of hay, but had little influence on the protein content of timothy at the normal hay harvesting time. The latter two authors found that late applications of nitrogen produced marked increases in the protein content of timothy. Blair (3) obtained results with timothy indicating that, while an application of 150 pounds of sodium nitrate gave increases in the yield of both seed and forage, the rela- tive increase in seed yield was significantly higher than that of the forage yield. CLIMATE The monthly mean.temperature and the total monthly precipitation from.June 1957, to July 1939, are presented in Figures 1 and 2. The mean monthly temperature and total monthly rainfall during the two months just preceding the planting date are included. The daily rainfall during the months in which the fertilizer applications were made is listed in table I. Climatological.data were taken.fram the U.S.D.A, Weather Bureau recording station at East Lansing, Michigan. _ III? .IIJIl cl )I!l _ 1 fill ...II..: I. - ...... .ID‘I.|I , I! Ilia: l'—-—-— —-—V— i I - I I l I v I __ _.._--:-— _—,—v— I I I i l .j..---.. . nIl In. . .1. . I.I ..,. ...I.I....I.. . ..I .I II.I|.|I.I 3.....-‘I I l I I I 1 i t l l A I I l l i I . _i____._.§______-__.__-_4. -_ * Y . l I I I.‘|Y;|M1.v Illll I'll—7' In, ‘ I I (ll: f w l. I: .l I rII‘III-llllll." I I l l ' "“1 l | I 9 .___J-__.. 14....-- _I-_-_ I I i HaoaeoaIAesHoqdom ewe. IQ- I. _ ._ w _ yr! -I- I Ilnllll...IIIIIln II Daily rainfall during months in which ammonium sulphate applications were made** Table I. W T " Ingfies of rainfall 3' April TMay June April may June Day 1938 1938 1938 1939 1939 1939 1 T 0 T- .09 T T 2 T O .01 0 0 .01 3 .30 .07 T 0 0 .13 4 O 0 .30 0 0 0 5 0 O T .11 O 0 6 .01 0 .62 T 0 0 7 0 T .04 .11 T .87 8 .91 .26 O .10 .04 .09 9 0 .02 T T .48 .07 10 0 0 .19 .78 T 1.23 11 O T .12 .11 0 .04 12 T 0 O .01 0 0 13 0 0 O O 0 O 14 0 .65 0 .52 T 0 15 T* .39 .05* O .15* .01* 16 .04 0* T 0 T .02 17 T .81 .03 1.37 T 0 18 T .49 O .25 0 .10 19 .02 .49 O .24 0 .08 20 T T 0 0 .34 .07 21 .03 .02 0 .46 .13 .18 22 0 0 O 0 T .43 23 0 1.05 T O .15 O 24 .03 O .55 0* 0 T 25 0 T .03 0 0 0 26 0 .30 T T T -0 2? 0 .08 O T .34 T 28 .08 1.09 0 0 .44 .36 29 T .01 O O T T 30 0 0 .95 .06 O .08 31 --- 0 --- --- 0 --- ‘Datcs of fertilizer applications: **Data from the U.S.D.A.,‘Weather Bureau, Lansing, Michigan -9- 1958 EXPERIMENT The applications of ammonium sulphate were broad- cast between the rows of each plot. To decrease border effect, one foot at each end and the outside rows of all the plots were out prior to seed harvest. The ultimate plots consisted of four twenty-foot rows. The seed was hand—harvested August 20, dried 30 days at room.temperature, and then thrashed. The per cent of purity was based upon a random three-gram sample of the thresh- ed seed. The seed yield data presented were calculated to a pure seed basis.. Immediately after the seed was harvested, the for- age aftermath.was mowed and the green weight taken. A.repre- sentative sample was dried four days at 600G. for dry matter determination. The protein analyses were made by the Division of Agricultural Chemistry of the Michigan Experiment Station, and the data presented were calculated on a dry matter basis. Table II shows that every rate of application of ammonium sulphate applied in April and May gave a higher yield of seed than the average of the control plots. Because of inconsistencies between replications, however, no increased seed yield was statistically significant when compared to the control. The applications of ammonium.sulphate in June had little or no influence on seed production. Table II. Data from the 1958 experiment showing (1) yield of seed, (2) yield of forage, (3) ratio of per cent increase in seed yield to per cent increase in for- age yield, and (4) per cent of protein in the for- age calculated to a dry matter basis. The tabular values are averages of the three replications. Date and rate of ~ Seed Forage *Ratio, % **Per cent application of yield yield incr. in protein, ammonium sulphate lbs. per lbs. per seed to % oven-dry acre acre incr. in forage forage Control 50 989 ---- 9.23 April 15,1938 100 lbs. /acre 60 1199 .90 ---- 250 64 1106 2.25 9.49 500 3; " 123 1738 1.93 9.75 750 9 " 109 1713 1.62 ---- 1000 9 R 139 2378 1.27 10.16 May11938 100 lbs. /acre 130 1666 2.35 -—-— 250 134 2116 1.48 10.62 500 R " 157 2679 1.85 9.55 750 .. 3' as 2835 .46 ---- 1000 " " 107 3465 .46 12.79 June 15,1938 100 lbs. /acre 54 1463 .17 —~-- 250 55 1801 .12 10.02 500 " n 64 2712 I .16 12.25 750 " ” 41 2245 ~--- ---- 1000 R " 61 3185 .18 15.84 Difference P:.05 99 791 required for significance =.Ol 135 1078 *Data was not statistically analysed **Averages of duplicate determinations from 2 replications. Applications of ammonium sulphate were influen- tial in increasing the amount of vegetative growth at the time of seed harvest. The April treatments were least effective in stimulating vegetative growth; while the plots treated in may gave the highest yields of forage. A 1000-pound application .l...‘ .I'L III-.. I!!! 1.11 of ammonium sulphate in April was barely more effective in increasing forage yield than a 250-pound application in either may or June. While all of the ammonium sulphate treatments were associated with increased percentages ofprotein.in the forage at the time of seed harvest, the June applications were most effective, especially on these plots where the rates of ap- plication exceeded 500 pounds per acre. April applications were least effective in increasing the protein content of the forage at the time of seed harvest. The ratios of increase in seed production to the increase in forage production in table 11 show that applica- tions of ammonium.su1phate in April, with the exception of the lOO-pcund rate, tended to stimulate seed production relatively more than forage production. Seed yield was similarly favor- ed by the 100, 250, and 500 pound rates of application in may. Seven.hundred fifty and 1000 pounds of ammonium sulphate applied in.may favored vegetative growth. Applications of ammonium sulphate in June produced marked increases in vegetative growth but had little influence on the seed yields. 1939 EXPERIMENT In order to study the influence of ammonium sulphate on brame grass which was in the second seed year, a new set of plots was established on an adjacent area in the same field. -The plot design and rates of application of ammonium.sulphate -12- were identical with that for the 1938 experiment. £631 _a__n_d_ Forage 31311.3 On.July 20, the mature seed was hand stripped from.the plots, dried four days at 100°F., and then thrashed in a carefully adjusted small nursery thresher. A three—gram sample of the thrashed seed was analysed for the per cent of purity. The yield and protein content of the forage were de- termined as in 1938. Table III shows that the ammoniwm sulphate applied in April or May produced.marked increases in seed and forage yields, and that there was a general increase in forage yield as the rate of application of ammonium sulphate increased. Am- monium.sulphate applied in June had little influence on the seed or forage yields, and these plots were not visibly differ- ent from the untreated plots. Of all the treatments in 1959, the SOC-pound appli- cation of ammonium sulphate in April produced the greatest seed yield. It also gave the greatest increase in seed yield per pound of fertilizer applied. Seed yields resulting from the May applications of ammonium sulphate were generally less than the corresponding applications in April; and a decided decrease in seed yield was apparent on the plots receiving the 1000-pound applications in may. Lodging was evident on all the plots treated with more than 500 pounds per acre of ammonium.sulphate -15- Table III. Data from the 1959 experiment showing (1) yield of seed, (2) yield of forage, (5} ratio of per- cent increase in seed yield to per cent increase in forage yield, and (4) per cent of protein in the air-dry forage. The tabular values are aver- ages of the three replications. Date and rate of *Seed *Forage **Ratios,% **Per cent application of yield yield increase in protein, ammonium.su1phate lbs. per lbs. per seed to % in_ core acre increase in air-dry forage yield forage Control 556 2175 ~--- 5.09 April 24, 1959 100 lbs./acre 456 2607 1.10 4.04 250 " " 525 2917 1.58 ---- 500 " ” 789 5594 1.86 6.56 750 " " 665 5951 1.05 ---- 1000 t " 742 4207 1.16 8.50 may 15, 1959 100 1bs./acre 406 2140 4.87 250 " " 564 2614 2.90 ~--- 500 " " 596 2807 2.51 7.00 750 " " 656 5400 1.50 ~--- 1000 “ " 557 5447 .88 10.25 June 16, 1959 100 " " 527 2060 ---- 4.21 250 " " 484 -2490 2.57 ---- 500 " " 564 2454 .17 6.06 750 " ” 455 2677 1.22 ---- 1000 u " 472 2757 1.27 6.25 Marches P=.05 157 ‘ 980 - . required for Bi nificance P=.01 214 1550 A highly sfgairiaant, remainderiteim, chffioiEEt of correla- tion, r 8 .811, was obtained between seed and forage yields. **Data not statistically analysed. in April or may; and this may account for the decreased seed yields with the heaviest rates of application. There were no Opperent differences in the time of seed maturity between the treated and the untreated plots. -14- The ratios of per cent increase in seed yield to per cent increase in forage yield (table III) indicate that nitrogen stimulated seed production comparatively more than forage yield. The 1000-pound application of ammonium sulphate in may appeared to slightly favor forage production; but, as has been suggested, the seed yield frmm this treatment may have been low because of the excessive lodging. The ratios calculated from the yield of the plots treated in June have little value, as the yields of neither seed nor forage were significantly greater than the yields from the untreated plots. 0n the plots treated in April or may, there was an apparent positive relationship between the amount of nitrogen applied and the per cent of protein in the forage at the time of seed harvest. Applications of ammonium sulphate in June, at the time of heading, were least effective in increasing the per cent of protein in the forage. Tiller Production and Panicle Qharacteristics The average number of fertile and barren tillers within an eight-inch quadrat placed at four predesignated points in each plot was considered a representative sample. One peniole was selected at random from.each placing of the quadrat to determine the number of florets and spikelets per penicle. These counts were made Just previous to seed harvest. Table 17 shows that Ihile nitrogen fertilization had no apparent influence on the number of fertile tillers, the number of barren tillers bore a positive relationship to ooew -15- me.a ma on an Ho.um concoaufinwfle no“ ceasedcn coma xitymo.a HHr an: amp; x, mo.nm eeseAMMMHn smom cm. on mm on a a coca. snow om.e en mm we a a one nmne oo.e mm on an e a com mama mm.e mm mm on a a cam mane om.n me me He enou\.mna ooa mama .ea ease swam ma.o ow mm an e e oooa ombaa oe.o on mm Hm : a one eNAm ow.n we be He : a con meme no.m an we ea 2 a own news no.m on an en ouoe\.mpm one mama ma has ommaa mm.m He noa on = a oooH meoma a¢.m we no we a 2 one manna om.m me an an e a com mama oo.n on me me e a 0mm Hemm as.e me 0e we enee\.mna 00a mama .em Henge amen n¢.¢ mm on be Honpsou son mo eeHeMHMm sop yo soy no unease m nom nom causesm «cocoa m Hem memosfi m Hem opsmmasm seweosse eponoan uponoau hem mneaaap .msoaaap uo soapwodamme Hogans senses epoaoxamm senses eaflpueu no open one open .xbommHsonm‘ .bmwswhd Honssz Honmsz nepfipzr. .r t L .ueowpeofiamon cones one no newesesm on» one menace neaspsp any .309 no memos“ sumac mom mpenoau so senses aspop eopmaseaeo one “av use upoaomamu non eponoau Ho senses amp new moaoflcsm Hem mpoaexfimm Ho season on» Any «son no mesons enema nee .mneaaap mosses no season one «my use .nnoaaap caspaem ho menses one .Hv medsoem pneawnogae mama on» Sony even .bH enema seem twe.a ma. _ on mm H .nm coneowoawwaa non causeway {beam ,ELLthotflL L AH! mm , L amir.rr.mmwnm‘!; oeseanmHn ewes om.n on we on a a coca. snow om.e en mm we a a one name oo.e on on mm a 2 con mums mm.e mm mm on : a 0mm mama om.n mm me He onoe\.ena ooa mama .ea ease mmam ma.o oe mm en a a oooa ombaa oc.o mm am an z a one «NHm oe.m me be as a e com memo no.m an we 6e 2 a 0mm name ao.m on an we oeou\.mnw ooH mama ma has ommaa mm.m He moa on a s oooH meoma oe.n me mm as c a one manna on.n me an an e : com mama oo.n on no me a a 0mm Heme as.e me 0e we enee\.wna ooa meme .em Hangs amen ne.e mm mm be Honpsoe son Ho peaomamw son no soy no names“ m nom nom eaoaeom mesoea w Hem memos“ m Hem opemmasm.esfisosse uponoan uponoau mom mnoaaap .msoaaap no compeoaamge menses Aegean upoaosamm senses oaapnen Ho span was even jeomflsflemw .mmmnesa mosses nonasz sweeps; t, , .osowpeOHamen comes one no sameness one one menace nuance» one .309 no memos“ sumac Hem eponoao no menses Hepop eepsaseaeo on» “my use "pedomamn pom mposoau mo menses on» key Moaowsem nee upeaexfimm no season on» .nv “son no mesons sumac nee .maoaaae seamen Ho posses one ”my use .nsoaaflp oaapnom no menace on» Adv wnflsoem psesflnomue mama on» song even .bH capes -16- the amount of fertilizer applied,:regardless of the date of application. The density of the stand, as measured by the total number of calms per eight inches of row, was increased by every rate of application greater than.lOO pounds per acre. The number of spikelets per panicle did not vary widely between'treatments. The average number of spikelets per pan- icle on the plots treated in April was higher than the average of the controls; the average on the May treated plots closely approximated the controls; and the average number of spikelets per panicle on the plots treated in June was below that of the controls. A statistical analysis showed that the average number of spikelets per panicle on the April-treated plots massignificantly higher than the comparable average on the plots treated in June, but that neither differed significantly from the average number of spikelets per panicle on the un- treated plots. All of the applications of ammonium.sulphate in April, or May increased the number of florets per spikelet over the average of the untreated plots, whereas, applications of ammonium sulphate in June produced no consistent increases in the number of florets per spikelet. The 750 and 1000- pcund rates of application in May appear to have been slight- ly more effective in increasing the nuflbcr of florets per spikelet than the corresponding treatments in April. With the exception of the loo-pound may treatment, every rate of application of ammonium sulphate in April or ., .._ .— t. . ' . f C { .t . . , . . I c i A I -17... may increased the total number of florets per eight inches of row. The greatest number of florets were produced on those plots receiving 500 pounds of ammonium.sulphate in April. The plots treated in June, in general, produced a. slightly smaller total number of florets per eight inches of row than were produced on the untreated plots. Coefficient 2; Fertility The coefficient of fertility is the ratio of the actual yield of seed to the calculated yield of seed eXpressed in per cent, where the calculated yield is the product of the number of fertile tillers, the number of florets per panicle, and the weight per individual seed. These values are a mea- sure of the numerical per cent of the florets actually con- taining seed. The calculated number of florets which produced seed was obtained by multiplying the total number of florets per eight inches of row by the coefficient of fertility. Table V shows that the rates of application had little influence on the coefficient of fertility; but that the June treatments resulted in coefficients of fertility consistently higher than the corresponding April or May treatments. A com— parison between the data on the number of florets per spikelet (table IV) and the calculated coefficients of fertility infers that the fever florets per spikelet, the greater the percentage of floret fertility. No determinations of the actual number of seeds per spikelet were made. -18- mnoa m.e e.mH we.w m.ma Ho.um ooeuoaummumm. you confindcn Ass m.n n.HH am.m n.wH mo.um A cmmenouufia OMAN A.mb e.an mw.p n.mm a e coca uaom n.0s m.mm 00.5 d.em c a one need m.nb H.H¢ mm.m ”.ma e a com mama a.¢> o.mn mm.s n.mm : a 0mm Dena m.em m.mn mm.m «.ma cucuxupa cod mama .ea ones aomm m.ms m.on mm.m H.wm : a ooou mmow H.0e m.mm Hm.oa ¢.a¢ : a one mmsm n.ms $.0n eo.m n.mm : a can seem m.em H.Hn ma.m b.an : a 0mm emma H.ms o.mm sm.w H.nm cacs\.cnw ooa awed ma hes mono 0.0b m.on oo.mH n.me a e OOOH naom m.np o.mm Nb.oa m.me : a one meow m.ms o.mm as.ma ¢.e¢ c a can noem H.ee o.mm o¢.w m.mn : s 0mm Hose o.ms o.mm mo.s m.mm onosxuna ooa mama .em Hausa mesa N.Hm 0.9m we.m o.mm Honanoo wees muflosu Hcfincpcs “my son «0 son uonm .mcgosw m ccnmmampqs hpaawpneu meson“ m Hem no mesosfl w cpenmasm anemones 9cm mpenoau uemmanm no macaw ma 9mm macaw ea no sofipcofiammc no Menage on» ma deem eacwoaumcoo cacfiz.vccm eacflm deem no once was even woncavoaco vamp Hem emmmmhd Hespcd ucpcadoacb it! .tl .mcofipcowamcn omen» esp mo mcwmncee cup mac menace ncaspcp one .ccom ecosecnm means 309 no memos“ pnmfle neg mpenoau mo senses cepcadoaac can any can "seem conmcnnps: eemmfinpmuoscn esp sfi deem mo pace use can flew ”haw adapncm we pccfloanucoe esp haw "son no mcncsfi unmflc hem came» comm deduce on» lav use .eaowa some eopaasoauo men may weapons encasemawo snag on» acne upon .> canes -19- A.question may be raised as to the reliability of the data on the tiller and floret counts from such small samples as were taken (Pg. 14). If the values obtained from the sample counts were representative of the reapective plots, there should exist a positive relationship between the actual and the calculated seed yield. The data was statistically analysed and a coefficient of correlation, r = .52, signifi- cant at the two per cent point, was obtained. The seed, as it was stripped in the field at the time of harvest, contained, for the most part, only sterile florets as impurity. The per cent of seed in this stripped material should be somewhat complementary to the coefficient of fer- tility. Table v shows that, as was true with the coefficient of fertility, the June-treated plots were associated with floret fertility, higher than that of either the April- or may- treated plots. The table likewise shows that a close positive relationship existed between the actual seed yield and the cal- culated number of florets which produced seed. £633 Quality Five hundred seeds were counted at random for each lot of seed used for purity analysis and weighed on a fine balance to determine the weight per 1000 seeds. Because of the bulkiness of broms grass seed in rela- tion to its weight, a modification of the standard test weight procedure was necessary. With the funnel span and in place twcfiinehes above the top of the kettle, the seed was passed -19- A.question.may be raised as to the reliability of the data on the tiller and floret counts from such small samples as were taken (pg. 14). If the values obtained from the sample counts were representative of the respective plots, there should exist a positive relationship between the actual and the calculated seed yield. The data was statistically analysed and a coefficient of correlation, r = .52, signifi- cant at the two per cent point, was obtained. The seed, as it was stripped in the field at the time of harvest, contained, for the most part, only sterile florets as impurity. The per cent of seed in this stripped material should be somewhat complementary to the coefficient of fer- tility. Table v shows that, as was true with the coefficient of fertility, the June-treated plots were associated with floret fertility, higher than that of either the April- or may- treated plots. The table likewise shows that a close positive relationship existed between the actual seed yield and the cal— culated number of florets which produced seed. §_e_c__d Quality Five hundred seeds were counted at random for each lot of seed used for purity analysis and weighed on a fine balance ts determine the weight per 1000 seeds. Because of the bulkiness of brome grass seed in rela» ties to its weight, a modification of the standard test weight procedure was necessary. With the funnel span and in place tlehinshes above the top of the kettle, the seed was passed -19- A question may be raised as to the reliability of the data on the tiller and floret counts from such small samples as were taken (pg. 14). If the values obtained from the sample counts were representative of the resPective plots, there should exist a positive relationship between the actual and the calculated seed yield. The data was statistically analysed and a coefficient of correlation, r = .58, signifi- cant at the two per cent point, was obtained. The seed, as it was stripped in the field at the time of harvest, contained, for the most part, only sterile florets as impurity. The per cent of seed in this stripped material should be somewhat complementary to the coefficient of fer- tility. Table V shows that, as was true with the coefficient of fertility, the June-treated plots were associated with floret fertility, higher than that of either the April- or may- treated plots. The table likewise shows that a close positive relationship existed between the actual seed yield and the cal- culated number of florets which produced seed. §gg§,guality Five hundred seeds were counted at random for each lot of seed used for purity analysis and weighed on a fine balance to determine the weight per 1000 seeds. Because of the bulkiness of brome grass seed in rela» tion to its weight, a modification of the standard test weight procedure was necessary. With the funnel open and in place two inches above the top of the kettle, the seed was passed through a one-half inch mesh screen placed inside the funnel five inches above the outlet. The kettle was filled until the excess seed accumulated above the kettle to the funnel outlet. The seed in the kettle was then leveled by slowly passing a piece of rigid sheet metal horizontally across the surface of the kettle; at the same time raking the sur- plus seed onto the metal sheet with a serrated blade longer than the width of the test weight kettle. This technique permitted of an accuracy within 0.5 of a pound per bushel between duplicate tests. Since the purity of the samples was not uniform the test weights were corrected to a 100 per cent pure basis by the formula, T I-P I i : C1: where: T = Test weight of the impure sample P 2 Per cent purity of the sample I = Test weight of the impurity in the sample D Per cent of impurity in the sample Ct Calculated test weight of pure seed in the sample In.the calculations to adjust the test weight to a 106 per cent pure basis, the impurity was assumed to be uniform in all the samples, and to be similar to that which was re- moved.in the cleaning of the seed. When impurities obtained from.the cleaning process was added to a sample of pure seed, trials showed that the test weight“cf the seed varied inversely with the amount of impurity -21- in.the sample (Table VI). Frem.the test weight of the sample ot'pure seed and the test weight of a sample of known purity, the test weight of the impurity at the percentage level of the sample of known purity was calculated. ‘Values for the test weight of the impurity in table VI were obtained by solving for "I" in the test weight adjustment formula. The test weight of the impurity in all of the seed yield samples was interpolated from a curve plotted.from.values of ”I" aelculated from a uniform.sample of seed adjusted to a known percentage of purity by adding the required.amount of impurity. Table VI. The test weight of brome grass seed, and the cal- culated test weight of the impurity in a sample; as they vary with known percentages of puritya‘ "Per cent purity 1*Ayerage test‘ Calculated test of the sample weight per bushel weight of impurity LI) 100 16.1 ~--- 99 15.75 5.0 98 15.4 4.9 97 15.0 4.7 96 14.6 4.5 95 14.2 4.4 94 13.9 4.4 93 13.6 4.4 93 13.3 4.4 91 13.0 4.4 90 12.7 4.4 80 10.3 4.1 70 8.7 4.8 50 6.2 3.8 85 4.7 3.8 0 3.6 3.6 [hieracss or duplicate test weight determinations. The per cent of germination was based on duplicate samples of 100 seeds, germinated at 35° C. for nine days, between.moist blotters in a controlled-humidity germinator. Adequate facilities for the germination of grass seed were not available, consequently the values obtained were low. Since the tests were performed under similar conditions, the values were considered comparative. Duplicate germina- tion tests on a composite sample of the seed when tested by the Michigan state seed analyst, germinated 97 per cent. Table VII shows that, in this experiment, every ap~ plication of ammonium.sulphate produced weights per 1000 seeds greater than those from.the control plots. The various applications of ammonium sulphate in each month responded similarly; but, in general, the June treatments were most ef- fective in increasing the weight per 1000 seeds. Applications of 100, 250, 500, 750, and 1000 pounds of the fertilizer ap- plied in June produced respectively, five, seven, nine, 13 and 11 per cent increase in seed weight, as compared to the con- trol. In.May and June, the 1000 pound rates of application of ammonium.sulphate were less effective than the 750 pound applications. ‘ Ammonium.sulphate applied in April or Mhy had little influence on the test weight per bushel of the seed produced. an the other hand, ammonium sulphate applied in June produced marked increases in the test weight or the seed. All of the applications is June greater than 100 pounds per acre were -23— about equally effective in increasing the test weight per buShel 0 Data from the 1959 eXperiment showing (1) the weight per 1000 seeds, (2) the test weight per bushel, (5) the per cent of germination, and (4) the per cent of purity of the seed. The tabular values are averagesr of the three replications. Table VII 9 fi"te and rate of?’ We1ght’ Test weight Per cent Per Bent application of per 1000 per bushel of of ammonium sulphate seeds._gm. pounds germination puritz_ Control 5.35 14.7 75.8 95.8 April 24, 1959 190 lbs./acre 5.56 15.4 68.7 94.5 850 7 7 5.52 14.7 67.5 96.3 500 7 " 5.48 14.5 75.4 96.4 750 5 n 5.58 14.1 72.0 96.7 1000 7 " 5.65 15.1 75.0 95.8 m15 1959 190 its. /acre 5.56 15. 75.7 96.5 850 " 5.43 14.8 77.2 96.6 500 7 9 5.55 14.5 75.8 97.5 750 9 ” 5.52 15.0 79.0 97.1 1000 " " 5.47 14. 75.2 95.0 June 16, 1959 100 1bs./acre 3.41 15.7 72.5 96.5 850 7 ” 5.49 16.8 80.0 97.4 500 " 7 5.55 16.7 78.5 96.5 750 7 " 5.66 17.1 79.5 96.8 1000 7 " 5.60 17.0 77.2 97.7 153T5renpe P=. . 1.2’ 7?5"‘ l ‘1.8 required for gflifiomce P1300]- 016 107 1000 2.5 The rate or application of ammonium sulphate had little influence on the per cent of germination; although there ap- peared to be a slight general increase in germination with the may and June fertilizer treatments. The per cent of purity was generally improved with the applications of ammonium sulphate, especially when the fertilizer was applied in May or June. -24- RESIDUAL INFLUfiNCE OF THE 1938 TREATMENTS ON THE 1939 CROP The plots which were treated in 1958 were maintained to study the residual influence of nitrogen applications upon the subsequent season's growth. These plots were kept free from.weeds, and no further treatment was applied in 1959. All of the methods of analysis in this emperiment were iden- tical with those in the 1959 eXperiment. The data in table VIII shows that the plots which were treated with 750 and 1000 pounds of ammonium.sulphate per acre in June 1958 were the only ones which produced seed yields materially above the average yield from the untreated plots. Throughout the growing period in 1959, these plots were visibly taller and had a more dense stand (i.e., more culms per unit length of row) than the other plots. The yields of forage in 1959 were significantly in- creased by the 1000 pound application of ammonium sulphate in April, 1958, the 750 and 1000 pound applications in may, 1958, and the three highest rates of application in June, 1958. The data on the protein content of the forage in table VIII. while not conclusive, suggest an inverse relationship between the rate of application of ammonium sulphate in 1958 and the protein content of the forage at the time of seed harvest in 1959. While the rate of application of ammonium.sulphate in 1958 had little influence on the test weight in 1959, the test weight a} the seed from.the plots treated in June 1958 was .cooe use seed on» no mamhasse acowpeapspm 02L m.ma ma. n.a soc ems ao.nm moasomwaswau you umnaodcn \m.HH ”no. 09H + Maori, mm mo.um, cooenmeHn o.en 50.9 H.¢H mo.m swam owe a a OOOH m.no ma.n e.ea tat: omen nae : a one m.om ma.n o.ea Hm.e ooam men s a con 9.00 ma.n e.ea cunt omea Haw com e.mo oa.m a.ea 90.0 band mam onoc\. ema ooa mama .nH cash o.no ow.m ¢.ea on.» omnm men a z coca m.mo Hm.m m. «a tuna beam on» e a on» n.ae ma.n b.ea Hm.e mama mom 2 a con 9.90 mH.n m.ea tut: pupa How x can m.se he.» H. ma oo.m some can onoa\. mna cod mama .oa has m.nm nm.m o.¢a sw.e nmam wan z a coca .m n.bo ma.» H.ma 3cm: cued new a a one am 0.90 ma.» H.¢H Ho.w mama won a a con o.oo om.n o.ea otto puma com com o.se pa.n n.ca oo.n uses «an enou\.ama ooh whoa ea Hausa o.mm mm.» m.ea ao.m coma dam achpsoo soapcoHSEow .msw.mooom .sp nomw, owcnom enos\.mpa onow\emna eponmaom_asaeoaau no oooa Hem wnwwce knounww ea cacah cacflh Ho soapsoflammd memo mom pnmwmm unmet swouonmtfiuHr omsemhw wamellllmw open www.mpwm .mnoapcofiamou connp one no some chops one menace amazed» one .oofipeoaanom go peso yam flog use .uocee oooa mom pnwaeh.nmv .ocem no Henson Hem pnwaee peep Adv owenow one mo poopsoo saoponm “my caoah owdnou any cacah doom “av wuaeonm mono mama esp no space spoon» ucsfiaapneu coca on» yo camouflage Hoocauen one no heave one song open .HHH> canoe ~26- generally lower than that of the seed from the corresponding plots treated in April or way. With one exception, applications of ammonium sulphate in 1958 had no material influence on the weight per 1000 seeds or the per cent of germination of the seed harvested in 1959. Both the weight per 1000 seeds and the per cent of germination of the seed from the plots receiving 1000 pounds of ammonium sulphate in June were significantly lower than the average of the controls. Relationship Between Seed Yield and Cost 2: Fertilizer *“ 12m "" The data on the seed yield from.the 1959 experiment in table IX show that, on an acre basis, fertilization with ammonium sulphate increased the value of the seed yield from $55.60 on the untreated plots to as much as $78.90 on the plots receiving 500 pounds of the fertilizer per acre in midrlpril; while the cost of the fertilizer was but $17.00 for the 1000 pound rate of application. Every application of ammonium sulphate in April or May in 1959 was profitable; but the June treatments did not in general pay for the fer- tilizer applied. .oop Hem oo.emw as pmoo aesfiaapucm on» “venom gem eoa as omsacp new does one» r--- oo.sa oe.aa was use a . coca --¢: ms.ma om.m am new a a one :.4. on.m om. m sea a : oom mm.o em.e om.ma ems ewe u a one ::.a os.a g--- n--- has oncs\.una ooH mesa .ea ones oa.a oo.sH oH.eH Hes sun . a coca mm.sa ms.ma oo.on can one a a one om.na on.o oo.em can can a s con mm.oa mm.e oo.om mom «on a a one on.» os.a oo.m on woe enos\.ana cos snag .ea has oo.am oo.ss on.mn men use a 2 coca no.sa ne.ma os.om son new a a one om.cn om.m oe.ne was mes . u can ne.ma mm.e oa.ea sea new a a on» 09.0 os.H oo.m cm was onoe\.apa ooa mama .em Henge -uiu. -u-v u--- u--- one Honpsoo eyes see, choc once hem, enoc\.mna once sponmase assesses pacepsoap mom eases eaewh ace .mna go soapsofiammo ncsaaflpnem pace ocmsmnosw coca ea cacfiw no open one even seem amnonm aonHHmnwmu mo cede» omsonomm ,wemm ri r .uaowpseflamen mouse on» no news these one menace neasnop one smmma ea neuaaapnon oucanSu.asasoaso no snows tooaamme on» song msapdsmoh once pom uneaaoo ea aauonm can use eaofih use. one .HH canoe -28... DISCUSSION Ammonium sulphate applied during the first seed year in mid-April or mid-may of 1958 appeared to have a slightly more beneficial effect on the seed yield than did similar applications in mid-June; but no rate of application was significantly more effective than another. The May-treated plots in general produced slightly more seed than the cor- responding plots treated in April. The treatments in April 1958 were least effective in stimulating the yield and pro- tein content of the forage; therefore, it is suggested that perhaps much of the stimulating effect of the nitrogen in these treatments was dissipated in the drainage water, since the fertilizer was applied too early in the season for the seedling grass plants to take up or utilize the soluble nit- rogen. On the plots established in 1959, when the brome grass plants were in the second seed year, all of the applications of ammonium sulphate in April or may greater than 100 pounds per acre resulted in seed yields significantly greater than the control. The fertilizer applications in April were gen- erally more effective than in May, while the treatments in June resulted in seed yields only slightly above the average of the untreated plots. In 1959 the plants had developed root systems large enough so that the fertilizer put on in April was taken up to a much greater extent than it was in 1958, when the grass plants were in the seedling stage of development. -29- Regardless of the date or rate of application of ammonium sulphate in 1959, a close positive relationship existed be- tween the seed and forage yields. (See footnote, table 111). Plants must possess photosynthetic tissue in order for food storage to take place. In general when all of the green tissue is removed from perennial plants, the "recovery growth” temporarily causes an exhaustion of the previously stored foods; while the greatest food storage occurs when com- paratively large amounts of chlorophyll-bearing tissue are present. Seed production in grasses is a form of food storage and consequently does not take place until food materials, in excess of vegetative growth requirements, are manufactured by the photosynthetic tissue. The data from this eXperiment in- dicate that the yield of seed depended upon the amount of chlor- ophyll-bearing tissue produced throughout the growing season. The ammonium sulphate applications had little influence on the number of fertile tillers, whereas it did result in marked increases in the number and size of the barren tillers. A comparison of the data from the tiller counts with the data on the forage yields, (table III) indicates that size was more important than the total number of tillers, in determin- ing the yield of forage at the time of seed harvest. This relationship suggests that not only the carbohydrates manufac— tured by the increased area of green tissue on the seed bear- ing culms, but also a portion of that manufactured in the bar- ren stems was available for storage in the seed producing culms. ~50- While it was true that the plots which received the heaviest applications of ammonium.sulphate produced the greatest amount of foliar tissue, they did not necessarily produce the great- est seed yields. Extreme lodging resulted from the rank veg- etative growth on these plots, and this probably interfered with.the normal development of seed. The yield of seed bore a close positive relationship to the calculated number of florets which produced seed, so it was evident that the nitrogen fertilization either directly or indirectly stimulated the development of florets which did not develop without the addition of nitrogen. The plots re- ceiving 500 pounds of ammonium.sulphate in April produced the greatest number of florets per eight inches of row, the great- est number of florets which produced seed, and the greatest seed yield. The total number of florets per eight inches of row on the plots treated in June was generally less than that on the untreated plots, while the seed yield, with one exception, was slightly higher. The June applications of ammonium sulphate were made at heading time when normal food storage had begun. The nitrogen applied stimulated vegetativeness at the expense of reserve carbohydrates and resulted in a partial depletion of the carbohydrate food reserves. As a result, some of the floret primordia which would have developed failed to do so because of this return to a vegetative condition of the plants. The higher coefficient of fertility, the increased weight per -51- 1000 seeds, and the comparatively high test weight per bushel or'secd obtained from.the plots treated in June probably cem- pensated for the decrease in the number of florets produced when compared to the controls. From.the study of the residual influence of ammonium sulphate on the succeeding season’s growth, it was observed that only the heaviest rates of application in 1958 resulted in seed and forage yields in 1939 materially greater than the yields from the untreated plots. It is inferred that part of the fertilizer remained in the soil throughout the season, or was stored in the plants in the fall, to become utilized by the plants after the inception of growth in 1939. The pro- longed period of drought during the fall of 1938 (fig. 2) fav- ored such a condition. SUMMARY Three sets of plots of smooth brcme grass were fertil- ised with ammonium.aulphate at five different rates of appli- cation: one set of plots was fertilized in April, a second in may, and a third.in June. The experiment was carried out dur- ing the first seed harvest year, and repeated on a new set of [plots during the second seed harvest year. Ammonium.sulphate applied in April or May of the first seed year resulted in seed yields greater than the controls; ‘whercas the same fertilizer applied in June did not consistently -52- stimulate seed yields. In the second seed year, applications of ammonium.sulphate in April resulted in marked increases in seed yield when compared to the control. The may appli- cations were generally not as effective as those in April, while applications in June resulted in seed yields only slightly greater than the control. Lodging was evident with the highest rates of appli- cation of nitrogen in June in the first seed year, and in April and may in the second seed year. There were no apparent differences in the time of maturity of the seed between the treated and the untreated plots. Forage production was stimulated.most by the applica- tions of ammonium sulphate in May in the first seed year, where- as April applications resulted in the greatest increases in forage yield in the second seed year. The number of fertile tillers and spikelets per panicle were only slightly influenced by the applications of ammonium. sulphate: whereas, the number of barren tillers and florets per spikelet were significantly increased. The applications in June were most effective in improv- ing the quality of the seed, as measured by the weight per 1000 seeds, the test weight per bushel, and the per cent of germination. In both years of the experiment the protein.analyses showed that the protein content of the forage at the time of -33- seed harvest consistently increased as the rate of applica- tion of ammonium sulphate increased. June was the most effec- tive date of application in the first seed year: whereas, in the second seed year, the percentage of protein in the forage was highest from those plots treated in April and May. Heavy applications of nitrogen in May or June in the first seed year were associated with increased seed and for- age yields in the second seed year when no further treatment was applied. The protein content of the forage, the weight per 1000 seeds, and the per cent of germination were slightly Areduced in the second seed year on those plots receiving the heaviest rates of application in June of the first seed year. No residual influence of the 100 or 250 pound rates of appli- cation in the first seed year was apparent in the second seed year. -33- seed harvest consistently increased as the rate of applica- tion of ammonium.sulphate increased. June was the most effec- tive date of application in the first seed year: whereas, in the second seed year, the percentage of protein in the forage was highest from those plots treated in April and may. Heavy applications of nitrogen in May or June in the first seed year were associated with increased seed and for- age yields in the second seed year when no further treatment was applied. The protein content of the forage, the weight per 1000 seeds, and the per cent of germination were slightly .reduced in the second seed year on those plots receiving the heaviest rates of application in June of the first seed year. No residual influence of the 100 or 850 pound rates of appli- cation in the first seed year was apparent in the second seed year. (l) (3) (3) (4) (5) (6} (7) (8) (9) (10} (ll) (13) ~54. LITERATURE CITED Ahlgren,H.L. Effect of fertilization, cutting treatments, and irrigation on yield of forage and chemical composition of the rhizomes of Kentucky bluegrass. Jour. Amer. Soc. Agron. 50:685-691. 1958. Aldous,A.E. Unpublished data from the Kansas emperiment station, Manhattan, Kansas. 1958. Blair,W.S. Dom. Empt. Sta. Kentville, N.S. Rept. Supt. for 1925:55-56. 1925. Bvans,G. Seed yields of pedigree and commercial grass strains. Welsh Jour. agr. Vol. X. pp.151-142. 1954. and Calder,R.A. Manuring pedigree grasses for seed production. welsh Jour. Agr. Vol. VII. pp.l95-208. 1951. Evans,Morgan W. Effects of application of nitrate of soda upon the yields of timothy hay and seed. Jour. Amer. Soc. ngron. 26:255-240. 1954. Gilbert,B.E. Forty—third Ann. Rept. Dir. Rhode Island Agr. EXpt. Sta. 45rd Ann. Rept. pg. 56. 1951. Nilsson-Leissner,G. Nagra Forsok med kvavegédsling till rodsvingel och angsgréefroodlingar. (Experiments in nit- rogenous manuring of seed plots of Festuca rubra and Fee pretensis) Svensk Frotidning 6:29-5I. 1957. {ReviewEE‘ by R.Peter Jones in Herbage Reviews 5:91-95. 1957.) North,H.F.A. and Odland,T.E. Seed yields of Rhode Island Colonial bent as influenced by the kind of fertilizer applied. Jour. Amer. Soc. ngron. 26:959-945. 1954. Osvald,H. EXperiments with nitrogenous fertilizers on pasture land. (Transl. title, Swedish) Svenska Mosskul- turfor. Tidskr. 44:67-84. 1950. German abs. pp. 85-84 (Abs. Expt. Sta. Rec. 65:118. 1950.) Sanders, K.B. Composition of pasture grass as influenced by soil type and fertilizer treatment. Thesis for degree of M.S. at Michigan State College. 1951. Sohmdtz, Nickolas. Thmothy fertilization and culture. Md. Agr. Expt. Sta. Bul. 803. 1917. ( . n o . . n u . v ' l A | . .' '- . . . . a c a q u a n u u a ' .- a . o a , a e . v . n u n o s n . . . u e r ’ < a . . . c A — ‘ I e n 4 a A. . ‘ f I O I I C . . 1 n ‘ — w . . ' n e a l I ( 9 l ' I e u A a a o . ‘ . u o ‘ o u n n ' I a . . . e c u c u C u u A n l a ' - ' - -35- (15) Sprague,H.B. and Hawkins,A. Increasing the protein con- tent of timothy, without sacrificing yield, by delayed applications of nitrogenous fertilizers. N.J. Agr. Expt. Sta. Bul. 644. 1958. (14) Stapledon, R.G. and Beddows,A.R. The qualitative and quantitative response of cooksfoot to sodium nitrate and to super phOSphate. Welsh Jour. Agr. Vol. II. Pp. 105-115. 1926. (15) Voorhees,E.B. Field experiments with nitrate of soda on forage crops and on market garden crops. N.J. Agr. Expt. Sta. Bul. 164. 1905. (16) Whilson,A.R., Wells,F.T., and Vivian,A. Influence of soil on the protein content of crops. Wise. Agr. Expt. Sta. Ann. Rept. 1902:192-209. 1902. (17) Wiancko,A.T., Walker,G.P., and Mulver,R.R. Nitrogenous fertilizers for top dressing field crops. Ind.(Purdue) Agr. Expt. Sta. Bul. 586. 1955. (18) Zahnley,J.W. and Duley,F.L. The effect of nitrogenous fertilizers on the growth of lawn grasses. Jour. Amer. e e .- i e l i a I I a .. .9 x C v I, a 1 I i n b 1 I i I I I i. T l c a . O u, I 1 i I a V I I v e e . .- s l 9 f I a 6 m D d a I 4 i I A I I I h, .1 I ACKNOWLEDGMENTS The writer is indebted to Dr. C. M. Harrison and Mr. H. E. Brown for their assistance in out- lining this project, and for their helpful sug- gestions throughout the investigation. The writer is also grateful to Dr. C. M. Harrison, Dr. C. R. wages and Mr. C. W. Hodgson for their criticism of the manuscript. M)l)fi))l)))))))W“)(Elfilll's