The recovery of fertiliser N by sugarcane crops is low in comparison with most other field crops. Application of urea in subsurface bands instead of by broadcasting can greatly reduce loss of fertiliser N due to ammonia volatilisation, but the fertiliser N is still susceptible to loss from leaching or denitrification, which could be affected by soil internal drainage, trash management, or tillage practice. The recovery of fertiliser N in crops and soil from 15N-labelled urea applied as subsurface bands was measured in ratoon crops in southern Queensland and northern New South Wales. Two soil types, with contrasting internal drainage, were used in each region. In Queensland, the cultural practice was either trash burnt with inter-row cultivation or trash retained on the surface ('trash blanket') with no cultivation. In northern New South Wales, where the trash was burnt prior to harvest, the practice was either inter-row cultivation or zero tillage. Crop recovery of fertiliser N was nearly always in the range 20-40% of the amount applied. Residual fertiliser N in the soil at crop maturity ranged from 13 to 42% (average 26%). Total recovery of fertiliser N in the soil-plant system ranged from 35 to 76% (average 52%) at 6 months after application, and from 35 to 96% (average 56%) at crop maturity. Urea fertiliser supplied only 20-40% of the crop N uptake in a given season. Neither crop recovery nor loss of fertiliser N from the soil-plant system were related to the soil type or cultural practice used, indicating that compensatory effects occurred.

Nitrogen (N) fertilizer is being lost from sugarcane soils following application to the crop. This study was conducted to estimate the quantity of N being lost from the soil through biological denitrification and to determine the proportion of gaseous N being emitted either as N2O or as N2. Field studies were conducted on four different soils (humic gley, alluvial massive earth, red earth and gleyed podzolic), and on different crop management systems, by installing plastic (PVC) cylinders (23.5 cm diam., 25 cm long) in the soil to a depth of 20 cm beside the plant row in a ratoon sugarcane crop. 15N-labelled KNO3 was applied as a band across each cylinder to a depth of 2.5 cm at a rate of 160 kg N/ha. After rainfall or irrigation, the cylinders were capped for 3 h intervals and gas in the headspace sampled in the morning and afternoon, for up to 4 days. Denitrification losses from the humic gley ranged from 247 g N/ha.day for cultivated plots to 1673 g N/ha.day for no-till plots. Over the sampling period, this was equivalent to 3.2% and 19.7% of the N applied, respectively. Nitrous oxide accounted for 46% to 78% of the total N lost. For the alluvial, massive earth and the red earth and gleyed podzolic, losses over the sampling period ranged from 25 to 117 g N/h.day and represented <1% of the N applied. Recovery of 15N in the soil ranged from 67% at the first sampling on the red earth soil to 4.9% at the third sampling on the alluvial, massive earth soil. In a glasshouse study, intact soil cores (23.5 cm diam., 20 cm long), taken from the humic gley and the alluvial, massive earth, were waterlogged after band application of 15N-labelled KNO3 at a rate of 160 kg N/ha. Gas samples from the headspace were taken after 3 h, and then morning and afternoon for the next 14 days. Denitrification losses ranged from 13.2 to 38.6% of N applied with the majority of gaseous N loss occurring as N2. Total recoveries after 14 days, including the evolved gases, ranged from 68.7 to 88.2%. We conclude that denitrification is a major cause of fertilizer N loss from fine-textured soils, with nitrous oxide the major gaseous N product when soil nitrate concentrations are high.

The effects of three annual crops (oats, C3, Avena sativa; sorghum, C4, Sorghum bicolor; black gram, C3, Vigna mungo) and a grass pasture (green panic, C4, Panicum maximum) on the content and turnover of organic carbon (0-0.15 m) in Vertisols in south-eastern Queensland were measured. Monocultures of the crops and pasture were grown for 11 years in two blocks (1 and 3) established by cultivating a 44 year old C4 Rhodes grass (Chloris gayana) pasture. The site was previously under C3 brigalow (Acacia harpophylla) scrub.In block 1 on Paleustollic Pellusterts, the organic carbon in both the < 1.6 Mg m-3 and the > 1.6 Mg m-3 fractions decreased linearly with time in the cultivated plots and increased linearly with time in the green panic pasture. In block 3, the organic carbon content of the light fraction (< 1.6 Mg m-3) and the heavy fraction (> 1.6 Mg m-3) in the oats and green panic plots on Typic Pellusterts also showed a linear relationship with time, decreasing in the cultivated plots and increasing under pasture. The organic carbon content of the soil fraction > 1.6 Mg m-3 from the black gram and sorghum plots on Paleustollic Chromusterts showed an initial rapid decline over 2 years with a subsequent slower linear decline.The proportions of organic carbon in the soil due to the original C3 brigalow forest and the subsequent C4 Pasture were calculated from the deltaC-13 values of the heavy soil fractions. In block 1, cultivation resulted in a decline in both the C3 and C4 pools. In block 3, cultivation resulted in a more rapid decline in the older C3 pool than the C4 pool. The rate of decline in block 3 was associated with soil type and, upon cultivation, the C3 carbon in the Chromusterts declined more rapidly than that in the Pellusterts. Organic carbon which was part of the resistant pool under pasture therefore contributed significantly to the labile pool on cultivation.There appeared to be major differences in the mechanisms protecting the organic matter from microbial degradation in the different soil types. In the Chromusterts, a proportion of the older C3 pool which may have been physically protected under pasture was rapidly mineralized on cultivation. The mechanisms protecting the C3 pool in the Pellusterts appeared to be more stable to physical disruption and therefore more resistant to degradation.

Changes in the distribution of nitrate-nitrogen (N) in a clay soil (Pellustert) under oats (Avena sativa cv. Minhafer), sorghum (Sorghum bicolor cv. E57), black gram (Vigna mungo cv. Regur), green panic (Panicum maximum cv. Petrie), and lucerne (Medicago sativa cv. Hunter River), and the uptake of N into plant shoots, were measured at Narayen on the brigalow (Acacia harpophylla) lands of south-eastem Queensland over each cropping season in 1975-85. Nitrate-N accumulated in the subsoil (30-150 cm) under sorghum and black gram, but not under oats. Green panic depleted nitrate-N after 2 years, and lucerne after 1 year. Losses of nitrate-N during 2 wet years reached 300 kg/ha under sorghum and black gram, and 57 kg/ha under oats, but were negligible under green panic and lucerne. Leaching to below 150 cm in the soil was the probable cause. The supply of soil N to oats, sorghum, and black gram was adequate during the 10 years, but the N yield of green panic decreased from 239 kg/ha to 150 kg/ha after 5 years. Accumulation of nitrate-N under sorghum and black gram could be utilised by rotating these crops with green panic or lucerne. This would also improve the productivity of green panic pastures. Rotating the summer crops with oats (winter crop) or with deeprooted crops (e.g. sunflowers) should also be tested. Alternatively, reduction of production of nitrate-N in the soil could be attempted. Zero or reduced tillage could do this, but it may also increase leaching by increasing the entry and movement of water in the soil.

Uptake of nitrogen (N) fertilizer and yields of grain by sorghum on Cununurra clay in north-western Australia have been found previously to be higher under sprinkler than under furrow irrigation. The cause of this was investigated using a 15N balance technique on field microplots. Additional data on sequential changes in the distribution of mineral-N in the soil, and on volatilization of ammonia, were collected from outside the 15N microplots. The improved effectiveness of N fertilizer under sprinkler irrigation was associated with less upward movement of 15N into the top of the ridge and thus out of the rooting zone of the crop. Furrow irrigation caused 22% of the 15N applied to move into the top of the ridge, sprinkler irrigation caused only 2% of the 15N to do this. The urea appeared to move upwards during the first irrigation, and this N became unavailable to the sorghum because of soil dryness, high temperatures, and osmotic effects in the surface soil. Loss of 15N from the soil-plant system was approximately 27% under both systems of irrigation. This loss was attributed to denitrification, because there was no evidence for large losses of N by leaching or volatization of ammonia.

Recovery, movement and transformations of urea-15N applied to three cultivated Queensland soils were measured as the soil dried out from two initial moisture contents. The soils were a Prairie Soil, a Grey Clay and a Red Earth, providing a range of pH (6.6-8.1) and cation exchange capacity (7-42 C/g). Urea was applied by banding and mixing into the top 2.5 cm layer of soil. Ammonium sulfateJ5N was applied to a second set of soil samples. The soil was incubated at 35C for 21 days. Recovery of urea-15N was 93-103% of the amount applied in the Prairie Soil, 72-92% in the Grey Clay, and 55-83% in the Red Earth. The larger recoveries were for banding urea into dry soil and the smaller for mixing it into moist soil. Some 15N moved into the 2.5 to 5.0 cm layer of soil, the amounts averaged 21%, 18% and 12% of the amount of 15N recovered in the Red Earth, Grey Clay and Prairie Soil respectively. Transformations of urea-N into ammonium-N approached completion after 3 days when urea was mixed into moist soil, but only 55% of completion after 21 days when it was banded into dry soil. The sequential losses of 15N and changes in the ammonium-N content in the soil demonstrated a strong affinity for ammonia by the Prairie Soil, a moderate affinity by the Grey Clay and a weak affinity by the Red Earth. Special precautions needed in the field to achieve efficient use of urea fertilizer therefore increase in importance from the Prairie Soil to the Grey Clay to the Red Earth.

The fate of urea-N during the 14 d after four seasonal applications to a Setaria sphacelata cv. Nandi pasture in south-eastern Queensland was traced. The aim was to explain the variations in the responses by grass, which have been observed after applications of urea. The total losses of urea-15N from the soil-plant system and the fluxes of NH, into the atmosphere were respectively 29 and 12% of the urea-N applied in summer, 45 and 42% in autumn, 23 and 13% in winter, and 20 and 9% in spring. Losses of nitrogen by leaching were probably negligible as mineral-N derived from the fertilizer remained mostly in the 0-2 cm layer of soil. Hydrolysis of urea to NH4+-N was complete after 1 d in summer, 3 d in spring, 6 d in autumn and 7 d in winter. Losses of urea-N were influenced by the water content of the surface soil (0-0.5 cm) at the time of the application and by the subsequent pattern of rainfall. The loss was large when urea was applied to wet soil and rainfall during the next 7 d did not exceed 1 mm, as in autumn. Losses were much smaller when rain (5 mm or more) fell soon after the applications, as in spring and summer. In winter urea was applied to dry soil and fluxes of NH, continued at a moderate rate until rain fell on the sixth day. Losses of 15N were generally about double the losses of NH,. Work on this discrepancy is needed to fully understand the fate of urea-N broadcast onto pastures. Losses of NH, from urea broadcast onto pastures could be minimized in practice by making applications to dry soil just before rain is expected or before irrigation.

Wilson et al. proposed a simple micrometeorological method of estimating the rate of gaseous mass transfer to the atmosphere from a small circular plot which required measurements of time-average species-concentration and horizontal windspeed at a single height. This method has been applied here to estimate the rate of volatilization of ammonia from a 25 m radius plot treated with urea fertilizer. The emission rates thus obtained agreed satisfactorily with estimates based on a mass balance which employed concentration and windspeed measurements at five levels.

A comparison was made between dry combustion in a Leco NP‐28 nitrogen and protein determinator, and acid digestion by the Kjeldahl procedure in the analyses of soil and plant material for total nitrogen. A statistical functional relationship demonstrated that the methods gave results similar in size and reproducibility. Thus the traditional Kjeldahl procedure can be replaced by a simpler and quicker method. Key Words: nitrogen analysiswet digestiondry combustion

AbstractNitrogen balance studies in pastures have shown unexplained losses of applied N with much of the loss assumed to be denitrification and NH3 volatilization. This project studied soil, plant, and microclimate effects on diurnal and seasonal fluctuations of gaseous NH3 flux in a subtropical pasture fertilized with urea. Soil and microclimate measurements were taken concurrently with NH3 flux density determinations to relate these factors to magnitude and direction of NH3 transport. Average daily NH3 volatilization varied with seasonal soil and microclimate conditions and time after area application. A short period of large NH3, efflux was observed after urea application during warmer seasons whereas smaller efflux for longer duration was observed during cooler seasons. Soil water content and rainfall after urea application modified seasonal NH3 losses. Diurnal NH3 cycling was apparent, with large efflux occurring during daytime and small efflux or influx at night. Prior to each urea application, average diurnal NH3 transport was into the soil‐plant system possibly due to low soil mineral and plant N. Brief periods of NH, influx were common throughout all seasons, particularly around sunset and sunrise. Soil surface temperature was the most highly correlated factor influencing NH3 flux density during the summer season. During the remainder of the year evapotranspiration had the highest correlation, although the increased midday fluxes were probably due to paralleled increase in soil temperature and windspeed. All of the major influencing factors are. interrelated through their dependence on solar radiation. Rainfall distribution and amount after urea application appeared to control the total NH3 losses from applied urea. The rainfall influence was probably the result of rainfall dispersing urea which prevented high concentrations of NH3 and NH4+ from building up around urea‐prills.

Ammonia volatilization from artificially applied cattle urine was measured in a fertilized (374 kg nitrogen ha-1 year-1) and grazed (4 beef steers ha-1) pasture of Nandi setaria (Setaria sphacelata var. sericea) in south-eastern Queensland. Microplots (0.5 by 0.5 m) of pasture were treated with 1.4 1. of urine (= 37-48 g nitrogen m-2) and evolved ammonia measured with a flow-through chamber system that maintained near ambient temperatures and air flows. Calculated volatilization over a 14-day period was 18.8 % of the applied urine nitrogen in June, 14.4 % in November-December and 28.4 % in February-March. More than half of the volatilization occurred within 48 h of applying the urine. Measurements of soil mineral nitrogen in February-March showed that more than 80% of the urinary urea was hydrolysed within 2 h and only traces remained unhydrolysed after 24 h. Soil ammonium declined from 700 to 1000 g nitrogen cm-3 in the surface 2.5 cm on the first day to near-background levels on the 14th day. Accumulation of nitrate was rapid after the 2nd day, and accounted for one-third of the applied nitrogen by the 14th day. Nitrite reached a maximum of 2.3 % of the applied nitrogen on the 7th day and had virtually disappeared by the 14th day. Only traces of nitrite were detected below 2.5 cm depth.

Pretreatment with water of samples of a heavy clay soil sieved to 12 mm enhanced extraction of mineral nitrogen with 2N KC1. The pretreatment gave results similar to those given by the much slower method of sieving the samples to 2 mm. Further time saving resulted from the improved filtering behaviour of the water pretreated extracts.

The importance of surface run-off water, leaching and evolution of gases on losses of nitrogen fertilizer from a Rhodes grass pasture in south-eastern Queensland were assessed. Field microplots encased in steel tubes 21 cm in diameter and 60 cm deep were equipped to collect surface run-off, fertilized with 15NH415NO3 prills at the rate of 150 kg nitrogen ha-1 and destructively sampled at 4, 8, 12, 16 and 40 weeks after fertilizing. The recovery of 15N in the soil-plant system, losses of 15N in surface run-off and movements of 15N down the soil profile were measured. Open pasture plots were fertilized with NH4NO3 at rates of 0 and 150 kg nitrogen ha-1 and harvested at the same times as the microplots. The results were used to calculate the apparent recovery of fertilizer nitrogen by the plant tops. Pasture cores of 11.5 cm diam. and 12.0 cm deep were given the same fertilizer treatment as the microplots, placed in gas-tight growth chambers for periods of 4 weeks starting at 0, 4, 8, 12 and 16 weeks after fertilizing, and used to measure gaseous losses of 15N. The effects of soil water content ranging from field capacity to waterlogged on these losses were studied on a second series of cores. The apparent recovery of fertilizer nitrogen and the recovery of 15N in plant tops were usually well below 20%, and the recovery of 15N in the soil-plant system of the microplots was always below 50% of the amount applied. Most of the loss of 15N occurred during the first 4 weeks. A large part of the 15N lost from the field microplots was not traced, but the results demonstrated that surface run-off and leachate should not be ignored during nitrogen balance studies on pastures in south-eastern Queensland. Surface run-off generally removed less than 5% of the 15N, but the loss was 40% from one microplot. Losses due to leaching were not quantified, but a small significant excess of 15N in soil layers below 60 cm suggested that they did occur. Gaseous losses of 15N from waterlogged pasture cores reached 27%, but they were small or absent from cores with a soil water content at or below field capacity. Detailed work in the gas-tight growth chambers to define the soil conditions associated with gaseous losses of nitrogen are needed to relate laboratory findings to field conditions.

Regrowths of Setaria sphacelata CV. Nandi and of Chloris gayana CV. Pioneer were wilted for various periods before laboratory ensilage. Dry matter contents at ensiling were from 17 to 42 per cent in the Nandi setaria and from 22 to 68 per cent in the Rhodes grass. The unwilted silages were unstable during storage, but decomposition was not extensive, and losses of dry matter and nitrogen were not heavy. Wilting to dry matter contents of near 40 per cent and above stable silage with very small losses of dry matter and nitrogen. Milder wilting to dry matter contents of 30-35 per cent did not produce stable silage. The water activity of the plant material fell during wilting and during the early stages of ensilage. Levels in the wilted silages were low enough to prevent the production of butyric acid, and in the heavily wilted silage to also retard production of lactic acid. Contents of acetic acid and volatile bases fell as the dry matter contents at ensiling increased. The laboratory silos were air-tight. The results of this experiment would therefore not apply in practice unless air was excluded from the silage.

Phaseolus atropurpureus, Desmodium intortm, and Lotononis bainesii were ensiled in laboratory silos. Molasses was added to the first two legumes at 0, 2, 4, and 8 per cent of their wet weight. Changes in pH and water activity (Aw) and in contents of lactic acid, volatile acids, and volatile bases were followed during storage. When no molasses was added, P. atropurpureus made poorly preserved silage, D. intortum had some of the features of well preserved silage, and L. bainesii made excellent silage. Only 15 per cent of the nitrogen in the D. intortm silage was present as volatile bases, even though the pH remained above 5.0. No explanation was found for this, but the effects of a rapid release of acetic acid during the initial stages of ensilage deserve further study. The preservation of L. bainesii was probably not due solely to the development of acid conditions (pH <.4.2) as the pH at five days was as high as 4.9. P. atropurpureus and D. intortum with 8.0 per cent of molasses made well preserved lactic-acid silage but the smaller additions usually gave rise to butyric-acid silage. Measurements of water activity (Aw) suggested that low values during the initial stages of ensilage were a prerequisite for the production of stable lactic-acid silage.

A study was made of the chemical composition and silage fermentation of Setaria sphacelata CV. Nandi. The grass was ensiled at three periods of the growing season, with two maturities at harvest, and grown with two levels of nitrogen fertilizer. On each harvest day cuts were taken in the morning, at noon, and in the afternoon. The object was to measure the variation that occurred under field conditions in south-eastern Queensland. The water soluble carbohydrate content of the grass was always below 6 per cent of its dry weight, and the silage had very low contents of lactic acid, high contents of volatile acids and high pH values. Volatile base contents were usually well below 20 per cent of the total nitrogen, showing that degradation of protein was not extensive. By traditional standards of silage quality the S. sphacelata was poorly preserved. On the other hand a degree of preservation was achieved as shown by the moderate amount of protein breakdown and small losses of dry matter and nitrogen during storage. Variation in the extent of fermentation changes, and of losses during ensilage, demonstrated that grass cut at five weeks was more effectively preserved than grass cut at eight weeks. Also, on any one harvest day, grass cut during the morning had lower dry matter contents, and was not as well preserved as cuts made later during the day.

The effectiveness of molasses as an aid to the preservation of Setaria sphacelata (var. Nandi) was tested by ensiling two harvests of the grass with 0, 1, 2, and 4 per cent of added molasses. When the grass was cut at a vegetative stage none of the molasses additions stimulated a satisfactory lactic acid fermentation ; but, in spite of high pH values, the silages were relatively stable during storage. Second harvest material (numerous tillers in full head) with 4 per cent molasses produced stable lactic acid silage, but with smaller additions the silages were unstable and lactic acid was destroyed during storage. During the first five days in the silos substantial amounts of sugar were lost at the expense of lactic acid formation, and 4 per cent of molasses appeared to be a minimal dressing for S. sphacelata silage.