EFFECT OF HARVESTING INTERVAL AND NITROGEN APPLICATION ON TILLER HEIGHT , NUMBER OF GREEN AND DEAD LEAF BLADE PER TILLER IN FIELD SWARDS

Two experimen'td were conducted in July 2000 and in March 2001 at Iwuru, Biase. 15 km from Akamkpa Campus of Crbss' River State University':of Technology, Nigeria to estimate the effect of harvesting intervals (2. 3. 4. 5. 6:8nd 8 weeks) and nitrogen fertilizer application (0. 100, 200, and 300kgNlha) on tiller height, number of green and dead leaf blade per tiller of Northern gamba (Ngg), Guinea grass (Gg) and Star grasses (Sg). Increasing delay in date of harvesting and increasing nitrogen fertilizer application significantly (P < 0.05) exerted a positive effect on tiller height, number of green and dead leaf blades per tiller in fieldswards. Highest height of ,113.6 and 116.1 em were obtained in plots cut every 8 weeks treated with 300kgNtha in 2000 and 2001 respectively. Highest number of green leaf blade on the average occurred in plots cut every 3, 4, 5 and 5 weeks for'Ngg Sg, and Gg respectively compared with other treatments. Applied N retarded leaf death early in the season and accelerated,death and delayed flowering subsequently, later in the season. Delaying cutting reduced number of g:een leaf blades biil increased tiller height. Highest number of dead leaf blades occurred in the 8 weekly cut plots where 30Ckc!'.Jiha was applied and lowest in the 2weekly cut plots where N was not applied. Ngg had the highest number of green leaf blades and dead leaf blades on the average and was significantly (P c 0.05) higher than either those of Gg or Sg during the season. These results are discussed in relation to , . field swards management during the study period.

Star grass (Cynodon polystarchyus) are adopted to a wide developed clayey, mottled and occasionally conceretionary sub-soil (FA0 1966).The area was previously cropped with range of soils in the Savanna Zone of Nigeria.Their effective cassava, foll&,ed by a five year fallow.in which for grazing and has been due to their high productivity M ~~, ~~~ (&inea grass) was the dominant fallow spec,es, and quality.Generally, studies have been conducted with these and other temperate species to estimates the influence The site was manually cleared with matchet, stumped, and allowed to dry for some days then gathered together and of cutting frequencies and nitrogen lev& on yield and yield removed, components (Omaliko 1980;Pearse and Wilman 1984;Wilman and Peaise 1984;Woosehouse 1992;Ubi and Table 1 : Rainfall values from 1999 to 2001 Ornaliko 2004, Ubi and lque 2005) Information regarding Months 199912000 200012001 cutting frequencies and nitrogen levels on tiller growth and green leaves and dead leaf census had not been sufficient.-L 1 weeks.There were 4 N levels of urea fe4lizer applied at the Data was subjdcted to analysis of variance (ANOVA) .andrates of 0, 100, 200 and 300kg~lha-'.
means compared with Fisher's Least Significant difference The trial was conducted in a 3 x 4 x 6 split-split-plot in a (LSD) at 5% leve!. randomized complete block (RCB) design, replicated three times.The main plot treatment consisted of three grass RESULTS species, the sub-plots were the fertilizer levels and the subsub-plots were the intervals between harvests.
The results of species x harvesting interval on tiller height (cm) .fnere were 9 main plots (species).36.sub-plots (fehlizer (mean of 4N levels) is presented in table 2, Hamsting rates) and 216 sub-sub plots intervals).Plot size frequencies affected tiller height such that tiller height on the "was 4m 48m, sub-p'ots 'Ize was 3m 12m and sub-sub-plot average, significantly (P c 0.05), increased with increasing size was 3m x 6% harvesting interval from 28.3cm for B 2-weeklv interval to Tiller heights (cm) were taken by measuring heights of three tillers, randomly selected from within each plot to be harvested.Measurement was in-situ from tiller base to the leaf blade tip, and sampling was done along the diagonals of each plot to be hawesled.
A set of ten tillers were randomly harvested and separated into green leaf blades and dead leaf blades.A leaf was regarded green when it had over 50% of its length green while blades with less green tissue were regarded as dead leaf blades.
94.4cm for a 8weekly interval in 2000.~u r i k g the 2001 planting season, tiller height (cm) increased from 28.5cm for a 2-weekly interval to 96.2 (cm) for a 8 -weekly interval.There was a 122.5% (36.0-80.1)cm unit increase in 2000 and a 120.2% (35.7-78.6)cmunit increase in 2001 when cutting intervals increased from a 2. 3, and 4 weeks to a 5, 6 and 8 weeks.Doubling the interval between harvests from 2, 3, and 4weekly cuts to 5, 6, and 8 weekly cuts significantly doubled tiller height in the three species.This indicates that growth was more restricted in frequently cut dots but was fully expressed in the prolonged intervals.throughout the season.he Sp x N interaction in respect of tiller height (cm) was significant (PC 0.05) Table 4).There was .a8.6% (55.7-60.5)cmunit increase in tiller height in 2000, and a 11.9%The interaction of species x harvesting interval in respect of number of green leaf blade per tiller is presented in Table 5.
Increasing intervals between harvest significantly (Pc0.05)influenced the number of green leaf blade per tiller.The three species differ significantly (Pc0.05) in terms of number of green leaf blade, but the species produced their maximum green leaf at different harvesting intervals.The highest number of green leaf blade was obtained by cutting Ngg every 3 weeks and was significantly (P c 0.05) higher than:,that obtained under any of the other intervals within the three species.Cutting Ngg every 3 weeks, Sg every 4 weeks and Gg, every 5 weeks gave the highest yield of green leaf blade.The response of the three species to harvest intervals x nitrogen in terms of number of green leaf per tiller is presented in Table 6.In 2000, highest number of green leaf blade (8.9) was obtained from a 5weekly ciii plot where 200kgNlha was   applied and was significantly (Pc0.05)higher than 3.6 obtained from a weekly cut plot were no N was applied.
During the 2001 growing season, highest number of green leaf blade 98.8) was applied from a 5 -':weekly cut plot where 200kgNlha was obtained and was significantly (P ~0.05) higher than 3.7 obtained from plots cut every 2 weeks where no N was applied.On the average, green ieaf blade obtained -in plots that received 200kgNlha (6.7 in 2000 and 6.6 in 2061) was significantly higher than that obtained under any of the other treat me^^ (Taljle 6).   he interaction of harvest interval x nitrogen in respect of number of dead leaf blades is presented in Table 8, The highest number of dead leaf (7.1 in 2000 and 6.6 in 2001) occurred in plot cut every 8 weeks where 300kgN/ha was applied and this was significantly (P < 0.05) higher than that obtained under any of the other intervals and Nrates.The number of dead leaf blades was significantly (P < 0.05) increased by every addition of 100kgN/ha, throughout the study period.The death of leaf blade was reduced by applied N. 'early in the season and subsequently accelerated by applied N, later in the season.

DISCUSSION.
Tiller height, and number of green leaf blade are some of,the measures of crop growth which usually influence the amount of dry matter yield.Increased tiller height in this study was dub to prolonged harvesting interval.In a similar.study, (Wool House 1992) reported an increase in tiller height due to 33.0% increase in average tiller height from 37.5cm t o 50.0cm by harvesting cynodon strain at six weeks rather than four weeks interval, and is in agreement with the findings in this study.The result of this study showed that nitrogen fertilizer.application increased tiller height and is consistent with the results of (Wilman and Pearse 1984;Bearse and Wilman 1984;Chheda and Akintola 1971), that N application exerts a significant effect on tiller growth and development.
The early response of species to treatment combinations in the present experiment was particularly striking.The i.ositive effect of applied N on number of green leaf blade coupled with its retarding senescence in the early stage after application, led to a significant effect in net gain in number of green leaf blade early in the season.In addition, applied N had increased the size of green leaf blades within a few days of application, in line with the reports of (Wilman and Pearse 1984; Ryle 1984) .The area of photosynthesis per unit area of leaf would also be increased by applied N in the early stages of regrowth.
The present results suggest that applied N may tend to have the.reverseeffect in larger, older tillers.The positive effect of applied N on the number of green leaf blade that emerged per tiller in this study is in agreement with the earlier results of (Wilman and Wright 1993).
From this study of senescence and dead tissue tended to be retarded by applied N, and subsequently tended to increase later in the season.The change from N reducing death of leaf blades earlier in the season to N increasing death of leaf blades occurred at different times as reported by Wilman et a1 (1976); Wilaman and Martius, (1977);and Oyenuga (1960) Taking the average of 8 weeks, period of study, applied N increased the amount of dead leaf blade, and when there was succession of N application and harvest during the season, the total area of leaf blade per hectare which died was certainly increased by applied N.
Evidently, Ngg has a faster rate of leaf turn over, than Gg and Sg, and rather more leaves per tiller, this definitely might have accounted for Ngg having the greatest number of green leaves and dead leaves.
It is suggested that in leaf aging, the life-potential of the in'djvidual cells is not exhausted, but that changes are brought about uqder the influence of stimuli coming from other parts of the plant.By this implication, senescence in Ndeficient grass is to some extent premature and N is withdrawn from the older leaves even before full expansion when N is ample, the whole rate of plant activity, including leaf turn over is accelerated and this effect soon overrides the effect N can have in retarding senescence especially when tpe effect of N in increasing leaf size has been carried through the dying leaves.
Green leaves are very important either as a life plant or as herbage.The number and size of green leaf blade per tiller at any given time contributes to the level of the nutritional quality and yield.N fertilizer application and a suitable management system can help improve this quality and yield.In this study, it has been found that harvesting Ngg very 3 weeks, Sg every 4 weeks and Gg every 5 weeks with the application of 2 0 0 k ~~l h a -' is considered ideal for greater economic returns to the farmer and improvement of the livestock sector, noted by Omaliko (1980).

CONCLUSION
The results of this study have shown that establishing s-rds of different grass species with nitrogen application dould guarantee extended harvesting periods.The essence of grass sward management is to strike a balance between quantity and quality and harvesting each species every 6 weeks,with the application of 200kgNlha was found to be cost effective with good economic returns to the farmer.
and 5* 14' N and 5' 18'N latitude with a rainfall of over 2000mm in the rain forest veaetation of the basement " Grasses sucti as Northern gamba grass (Andropogon complex soil.The soil consists mainly of schrists, quartzites/ gayanus), Guinea grass C.", S q j 2 (Panicurn maximum) contain most of the plant protein and minerals, have January most of its highly digestible dry matter and are low in fiber in 86.6 comparison to stem.The number of green leaf blade per tiller February in a particular grass species places high value on this species March 65system and N-rates could be used to improve I the leaf quality and quantity.Wilman and Pearse (1 984) noted May 'that an increase in the interval between harvest allowed .alargzr positive response of applied N to develop in respect of .tiller andleaf,growth and development.this study are to determine.theharvesting August frequency and also N-fertilizer rate that would give the highest September 71.2 number of gree~n leaf blades when the tiller attains.8certain height, for good quality herbage throughout the growing October 34Two experiment& were conducted in July 2000 and March -Mean -, 50.1 !+---33.5~-'

Table 2 :
Effeclol SpeUeS and hawesling lnlervals on t~lier height (CI~)) (luiean of 4N Levels)Harvesting of intervals (wccks) the three species every 8 weeks gave the highest tiller height during the season.The data on harvest x nit~ogen interaction presented in table 3 clearly show the significant effect of the treatments on tiller height.Highest heights of 11 3.6 (cm) and 116.1 (cm) for 2000 and 2001 were obtained in plots cut every 2 weeks, where no N was applied.
(54.3 -60.3)crn unit increase in 2000 and 17.7% (60.8 -71.6) unit increase in 2001 when Nrate was increased from 200kgNlha to 300kgNlha.Height increases of tillers was accelerated by high levels of Napplication, such that the three species obtained their maximum heights where 3OOkgNlha was applied compared with other treatments.
species x harvesting intervals on the number of deaf leaf blade per tiller is presented in Table7.Delaying cutting significantly (P < 0.05) increased number of deed leaf blade.There was.a 25.9% (3.6 .-4.5) unit increase in ,dead leaf b'l'ade each for 2000 and 2001 when cutting iniervals increased from 3 2, 3, and 4 weeks average to a 5, 6 and 8 weeks average.The three species differed significantly (P < 0.05) in terms of nurnbdr of dead leaf blade per.tiller.The 'iesults showed that Ngg produced.equal numbei of blades (14.7) in 2000 and 2001 and was significantly higher .thanthese of either Gg or Sg.
d speck.a d hrwntlnO Intowah on number of dead loavosllillor) (Moan of I N Levels) . . . . . . .Hawesting d in(.w.lS ( W W ~S ) . =I.(1971)  found that seasonal variations in tiller was less pronounced in plots harvested more often with a