1981

Field Crops Research, 4 (1981) 133--145

133

Elsevier Scientific Publishing Company, Amsterdam -- Printed in The Netherlands

EFFECT OF MAIZE + LEGUME INTERCROPPING SYSTEMS AND FERTILIZER NITROGEN ON CROP YIELDS AND RESIDUAL NITROGEN

P.G.E. SEARLE, YUTHAPONG COMUDOM*, D.C. SHEDDEN and R.A. NANCE

Department of Agronomy and Horticultural Science, University of Sydney, N.S. W., 2006 (Australia) *Present address: Highland Agronomy Project, Chiangmai University (Thailand) (Accepted 12 F ebru ary 1981 )

ABSTRACT Searle, P.G.E., Yuthapong Comudom, Shedden, D.C. and Nance, B.A., 1981. Effect of maize + legume intercropping systems and fertilizer nitrogen on crop yields and residual nitrogen. Field Crops Res., 4: 133--145. The study involved two consecutive experiments on the same area at Camden, New South Wales, 34 ° S. In the first, the intercropping experiment, nitrogen was applied at the rate of 0, 25, 50 and 100 kg N ha -1 to maize alone (M), maize + soybean (MS), and maize + peanut (MP) intercropping patterns. In addition, soybeans alone (S) and peanuts alone (P) were grown without the addition of nitrogen, giving a total of 14 treatments. After harvesting the first experiment, above-ground plant material was removed, plots were rotary hoed and residual nitrogen was measured at sowing and after 15 weeks in a crop of wheat. Intercropping treatments gave relative yield totals as high as 1.4 at 0 kg N ha -1 fertilizer nitrogen. Maize grain yield was not affected by legume intercrop, indicating neither competitive depression nor nitrogen transfer from the legume. Intercropping depressed legume dry matter and grain yields at 0 kg N ha -1. In the residual nitrogen experiment, nitrogen uptake by wheat, considered the best criterion of residual nitrogen availability, was affected by cropping pattern. At 0 kg N ha -1 the values were M = 12, MS and MP = 19, S = 46 and P = 54 kg N ha -1, all significantly different at P < 0.05. Exchangeable soil nitrogen at sowing and at anthesis showed similar rankings although those at anthesis were lower than those at sowing. Fertilizer nitrogen had no effect on maize grain yield, but it increased maize total dry matter yield. There was no significant interaction between cropping pattern and fertilizer nitrogen. Fertilizer nitrogen affected nitrogen uptake by wheat at anthesis and exchangeable soil nitrogen at sowing but not at anthesis. The responses in exchangeable soil nitrogen at sowing, and in wheat nitrogen uptake at anthesis, to fertilizer nitrogen in the M treatments were linear, while those in the MS and MP treatments were quadratic, maximum value being attained by about 50 kg N ha 1. Nitrogen applied to intercropped legumes appeared inhibitory to nitrogen fixation, both directly from increased soil nitrogen and indirectly by stimulation of maize growth and shading of intercropped legumes. The data showed that a subsequent crop would benefit as much from following one of the maize + legume intercropping patterns to which no nitrogen had been applied as from following a maize crop to which 100 kg N ha -I had been applied.

0378-4290/81/0000--0000/$02.50

© 1981 Elsevier Scientific Publishing Company

134 INTRODUCTION Among the advantages that intercropping may have over sole cropping is that o f yield increase. Andrews and Kassam (1976} present favourable relative yield total (RYT) values of two-crop mixtures ranging from 1.2 to 1.6, i.e. 20 to 60% increases over sole cropping half the area to one crop and the other half to the other crop. Two beneficial effects may be present when one of the intercrops is a legume: (a) reduced competition for soil nitrogen, and (b) increased residual nitrogen available to a following crop. When a legume is one of the intercrops it is therefore important in the assessment of a crop° ping pattern to measure not only the intercrop yields, b u t also the available residual nitrogen. In this study, two consecutive experiments were carried o u t on the same area. In the first experiment, maize + soybean and maize + peanut row intercropping patterns receiving increasing levels of nitrogen fertilizer were compared with sole cropping of maize (also treated with nitrogen fertilizer), soybeans and peanuts, primarily to see what yield advantages were obtainable from these crop combinations and fertilizer applications. In a second experiment, residual nitrogen from each cropping pattern was examined by measuring total exchangeable soil nitrogen (NO~- N, NO~-N and NH~-N) and nitrogen uptake b y a following cereal crop to see what residual nitrogen was available. The study was carried o u t at latitude 34 ° S in New South Wales. While the study was located outside the tropics, the high summer temperatures during the intercropping experiment make the results relevant to many areas of the tropics in which maize + soybean and maize + peanut cropping patterns are used. MATERIALS AND METHODS

Location and soil The experiments were carried out during November to November of 1978--1979 at Camden, N.S.W. (latitude 34°S). The soil was dark reddish brown (5YR 3/3) light clay classified as a Gn 4.52 on the Northcote (1965) classification system.

Intercropp ing experiment Treatments and design. Nitrogen at the rate of 0, 25, 50 and 100 kg N ha -1 (NO, N25, N50 and N100 treatments) was applied to m o n o c r o p p e d maize (M) and intercropped maize + soybean (MS) and maize + peanut (MP) cropping patterns. In addition soybeans and peanuts were m o n o c r o p p e d (S and P treatments) without the addition of nitrogen. A randomized complete block design was used, with four replications of the 14 treatments.

135 Individual plots measured 7 m by 4 m. Maize was sown in rows 100 cm apart, legume rows were 50 cm apart. Thus there were four rows of maize and eight rows of legumes on plots with both crops; adjacent maize and legume rows were 25 cm apart. Maize (Zea mays L.) cv. XL81 was used because it is recommended for the area. It is medium to tall and high yielding. Inoculated soybean (Glyeine max (L) Merr.) cv. William and peanut (Arachis hypogaea L.) cv. Virginia Bunch were used because o f high yields in earlier experiments.

Land preparation and sowing. The land was prepared and fertilized with 200 kg ha -1 o f superphosphate (9.3% P) and 100 kg ha -1 KC1 (50% K) before sowing. The highest rate of nitrogen (100 kg N ha -1) was the local commercial recommendation for maize. Nitrogen (as urea, 46% N) was applied in a split side dressing: half as a post-emergence application and half when plants began floral initiation at 45 cm height. All species were sown on November 28, 1978, using a cone seeder. Weed control was obtained with Stomp 33OE at 4.5 kg ha -1 and hand removal. The crop was irrigated with overhead sprinklers to maintain rainfall + irrigation at 25 ram/week.

Observations. Time to first flowering was noted. Plant height was measured in the 14th week after sowing when maize growth appeared to have ceased. Solar radiation profiles were measured around midday with a Swissteco linear net radiometer in the 0 and 100 kg N ha -~ plots in the 15th week. Dry nodule weights were determined at first flowering in legumes on ten randomly selected plants from outer guard rows of plots. Twenty weeks after sowing grain yield (at 14% moisture), total aboveground dry matter yield, cobs per plant and 100 seed weight were determined in maize. Branches per plant, pods per branch, seeds per pod, 100 seed weight and grain yield (at 14% moisture) were determined in the legumes: soybeans at 22 weeks, peanuts at 23 weeks after sowing. The date of last harvest was April 9, 1979.

Residual nitrogen experiment Land preparation and sowing. All plant material on the above plots was removed by a forage harvester. Only plant roots and short above-ground stumps remained. All peanuts were removed, including those in guard rows, but all roots were returned and evenly distributed over the plots. All plots were then rotary hoed twice to 20 cm to incorporate plant material. Wheat (Triticum aestivum) cv. Condor at 40 kg ha -1 was sown on May 23, 1980. A basal dressing of 175 kg ha -1 o f superphosphate (9.3% P) and 75 kg ha -1 KC1 (50% K) was applied at sowing. The herbicide Stomp 33OE was also applied at this time and gave excellent weed control.

136

Nitrogen Exchangeable soft nitrogen (NH~-N + NO~-N + NO]-N) was determined at sowing and anthesis of the crop at 19 weeks (October 3, 1979) after sowing within a 1-m 2 quadrat kept plant-free in the centre o f each plot. Invasion of this area by lateral roots from surrounding plants was controlled by digging vertically with a spade to a depth of 30 cm around the perimeter of the quadrat. Ten soil core samples (each 2.5 cm in diameter and 15 cm deep) were taken at random from each quadrat at each sampling time. These were bulked. A 40-g subsample was then placed in 200 ml o f 2MKC1 for the determination of exchangeable nitrogen according to the m e t h o d of Bremner (1965). Dry matter yield o f wheat was obtained 19 weeks after sowing, (during anthesis), when nitrogen uptake would be near maximum. A sample was analysed for nitrogen by the Kjeldahl m e t h o d (Yoshida et al., 1972) to obtain nitrogen uptake values for each plot. RESULTS

Intercropping experiment Climate. The highest daily temperature during the season was 38 ° C, attained in the seventh week after sowing; other daily maxima were 34 ° C or below. Mean m o n t h l y m a x i m u m and minimum temperatures were in the range 30.5 ° C to 19.5 ° C and 17.3 ° C to 7.8 ° C, respectively. Solar radiation values were high, exceeding 3500 mW h cm -2 for the first 4 m o n t h s o f growth and decreasing to 1750 mW h cm -2 in the last m o n t h of growth. Rain fell in every m o n t h , the minimum being 126 mm per month. Plant population and development There were no significant treatment effects on plant population, plant height or time to first flowering. The final plant population counts were: maize 47 900, soybeans 106 600 and peanuts 100 400. Final plant heights were: maize 210 cm, soybeans 69 cm and peanuts 42 cm. The times to first flowering were: maize 8 weeks, and both legumes 5 weeks. Solar radiation reaching the ground was significantly reduced by each canopy and the reduction was greater where nitrogen had been applied (Table I).

Effect of cropping pattern and nitrogen on maize Maize grain yield, cobs per plant, and 100 seed weight were not affected by treatment. The overall grain yield was 7770 kg ha -1, the number o f cobs per plant 1.0 and 100 seed weight 29.0 g.

137 TABLE I Solar r a d i a t i o n m e a s u r e d at d i f f e r e n t levels in t h e maize + legume p l o t s as a f f e c t e d b y n i t r o g e n level. Values are m e a n s for m a i z e + s o y b e a n a n d maize + p e a n u t plots. L i g h t m e a s u r e d at t h e t o p o f t h e maize c a n o p y was t a k e n as 100% Level in c r o p

N i t r o g e n level

A b o v e maize c r o p A b o v e legume c r o p At ground

NO

N 100

100 a 43 b 26 d

100 a 32 c 18 e

N u m b e r s followed b y the same l e t t e r are n o t significantly d i f f e r e n t a t t h e 5% level according to D u n c a n ' s m u l t i p l e range test.

Total dry matter yield was affected (at P < 0.01) by cropping pattern and nitrogen (Table II) but the interaction effect was not significant. Only the M and M + P treatments differed significantly, while nitrogen increased yields up to the highest level of application. T A B L E II Main effects o f c r o p p i n g p a t t e r n a n d n i t r o g e n o n t o t a l d r y m a t t e r yield of maize tops Cropping pattern

Yield ( t o n n e s h a - ' )

N i t r o g e n (kg h a ' )

Yield ( t o n n e s h a - ' )

M MS MP

22.1 19.0 20.8

0 25 50 100

17,6 19,6 21.1 24,3 0.79 2.3 3.1

SE LSD ( 5 % ) LSD ( 1 % )

0.68 2.0 2.6

Effect of cropping pattern and nitrogen on legumes Total dry matter yield and grain yield were higher in monocropped legume treatments than in intercropped treatments, which did not differ among themselves (Table III). There were significant treatment effects on soybean nodule weights per plant (P < 0.05). The S treatment (0.2 g plant -~) was significantly higher than the MS treatments (0.09 g plant -~). Nitrogen level did not significantly affect MS treatments. There were no significant treatment effects on peanut nodule weight per plant. Intercropping significantly (P < 0.01) reduced branches per plant (4.1 to 1.6 in soybean, 7.8 to 4.8 in peanuts), pods per branch (16.6 to 9.7 in soy-

138

bean, 1.9--1.3 in peanut), and 100 seed weight (24.5 to 23.1 g in soybean, 62.5 to 39.9 g in peanut). There was no effect of cropping pattern on seeds per pod in either legume. Increasing fertilizer nitrogen level had significant effects only on pods per branch in peanuts (1.3 to 0.9) and 100 seed weight in soybean (24.2 to 22.1 g) and peanuts (44.0 to 36.9 g). Significant linear correlations, (all P < 0.001) were obtained for both legumes between grain yield and branches per plant and pods per branch. No significant correlation was obtained between grain yield and seeds per pod for either legume or with seed weight for soybean. However, for peanut grain yield and seed weight were highly and linearly correlated ( P < 0.01).

Relative yield totals. RYT values, calculated by summing the ratios of the yield of each crop in mixture to that in monoculture, were highest in the intercrop at the lowest level of applied nitrogen and lowest in the monocrop (Table IV). Application of nitrogen tended to decrease the values below that of the maize + legume value at zero level of applied nitrogen. TABLE III Effect of cropping pattern and nitrogen on total top and grain yield of legumes Cropping pattern and nitrogen level

S, P MS, MS, MS, MS,

MP MP MP MP

at at at at

NO N25 N50 N100

SE

Total

Grain yield ÷x

DM yield of tops (kg ha -1) (kg ha -1) Soybean

Peanut

5,125 1,051 1,100 1,242 1,170

6,664 2,708 2,400 2,367 2,311

a b b b b

310 xxx

a b b b b

314 xxx

Soybean

Peanut

2,677 534 563 569 532

1,741 530 447 418 378

a b b b b

159xxx

a b b b b

98xxx

Numbers with the same letter are not significantly different at the 5% level according to Duncan's multiple range test. +Grain Yield at 14% moisture. xMain effect of nitrogen on maize grain yield: NO, 7108; N25, 7569; N50, 7783; N100, 8532 kg ha -1. Differences not significant. TABLE IV Relative yield totals of intercropping patterns Cropping system

maize + soybean

maize + peanut

M, S, P MS, MP at MS, MP at MS, MP at MS, MP at

1.00 1.36 1.03 1.01 1.09

1.00 1.37 1.12 1.22 1.29

NO N25 N50 N100

139

Residual nitrogen experiment Climate and plant population. Mean monthly m a x i m u m and minimum temperatures varied during the May--October growing period over the range 19 to 24 ° C and 3.5 to 10.5 e C, respectively. Between 25 and 43 mm of rain fell each month, dry periods being supplemented with irrigation. There was no effect of treatment on plant population which was 530 000 plants ha -1. Exchangeable soil nitrogen and nitrogen uptake. The effect of cropping pattern and fertilizer nitrogen on exchangeable nitrogen at sowing (0 weeks) is shown in Fig.1. There were highly significant main effects of cropping pattern (P < 0.01) and fertilizer nitrogen (P < 0.01). The relationships between exchangeable nitrogen and fertilizer nitrogen were quadratic for the MS (r 2 = 0.98, P < 0.01) and MP cropping patterns (r 2 = 0.99, P < 0.01) and linear for the M cropping pattern (r 2 = 0.97, P < 0.01). Exchangeable nitrogen increased only to about 50 kg N ha -1 in the maize-legume cropping patterns, but to 100 kg N ha -1 in the M cropping pattern. The situation at anthesis (19 weeks) was as shown in Fig.2. The main ef-

24--

20

°

J f

• .i

j

E

&

g E 12

12

w

t 0

25

t 50

Fertilizer nitrogen (kg N ha-1)

I 100

I 0 Fertilizer

I 25

I 50

nitrogen (kg N ha -1)

Fig. 1. E f f e c t o f fertilizer nitrogen and cropping s y s t e m o n e x c h a n g e a b l e soil nitrogen at sowing. • Maize + peanut, • m a i z e + s o y b e a n , • maize alone cropping patterns. Fig. 2. E f f e c t o f fertilizer nitrogen and cropping s y s t e m o n e x c h a n g e a b l e soil nitrogen at anthesis. • m a i z e + peanut, • m a i z e + s o y b e a n , A m a i z e alone cropping patterns.

I 100

140 fect of cropping pattern was highly significant (P < 0.01), MS and MP patterns both having higher (P < 0.01) nitrogen than the M pattern but not differing significantly among themselves. There was no effect of fertilizer nitrogen. Comparison o f the regressions fitted to the M pattern data in Figs. 1 and 2, shows that the lack of a significant response to fertilizer nitrogen at anthesis was due to b o t h an increase in exchangeable nitrogen at 0 kg N ha -] and a decrease in exchangeable nitrogen at 100 kg N ha -1. The lack of significant response to fertilizer nitrogen at anthesis in the maize-legume patterns was largely due to decreased exchangeable nitrogen values at higher levels o f fertilizer nitrogen. Nitrogen uptake b y wheat at anthesis was affected b y b o t h cropping pattern (P < 0.01) and fertilizer nitrogen (P < 0.01) (Fig.3). The response in the M pattern to nitrogen was linear (r ~ = 0.99, P < 0.01) and differed in slope from b o t h the MS and MP patterns, which did not differ between themselves. A single quadratic has been fitted to the intercropping patterns (r 2 = 0.99, P < 0.01). The response to nitrogen continued up to 100 kg N ha -1 in the M pattern, while in the intercropped patterns, maximum response was attained b y 50 kg N ha -1. The values for both maize intercropped with legumes without nitrogen and maize alone fertilized with 100 kg N ha -1 are similar. Table V shows that at 0 weeks, exchangeable nitrogen values were lowest in the M pattern, highest in the m o n o c r o p p e d legumes (which did not differ

2 z c~

215

I

5O

I

100

Fertilizer nitrogen ( kg N ha-1)

Fig.3. E f f e c t o f fertilizer n i t r o g e n a n d c r o p p i n g s y s t e m o n n i t r o g e n u p t a k e b y w h e a t at anthesis. • maize + peanut, • maize + soybeans, • maize alone cropping patterns.

141 TABLE V E f f e c t of p r i o r c r o p p i n g p a t t e r n (all w i t h o u t a d d e d fertilizer n i t r o g e n ) o n residual exc h a n g e a b l e soil n i t r o g e n at 0 a n d at 19 weeks a f t e r sowing a n d u p t a k e o f n i t r o g e n b y following wheat crop Cropping

Maize Soybean Peanut Maize + s o y b e a n Maize + p e a n u t

Exchangeable nitrogen (p.p.m. N)

Nitrogen uptake (kg N h a -1 )

(0 weeks x )

( 1 9 weeks +)

( 1 9 weeks*)

9.0 29.9 32.7 16.7 15.3

13.1 23.1 23.4 14.8 17.9

12 46 54 19 19

c a a b b

c a a bc ab

d b a c c

Figures in t h e same c o l u m n are n o t significantly d i f f e r e n t at t h e 5% level if f o l l o w e d b y t h e same l e t t e r a c c o r d i n g t o D u n c a n ' s m u l t i p l e r a n g e test. x : a t sowing ÷: a t a n t h e s i s

significantly from each other) and intermediate in the maize + legume patterns (also not significantly different). At 19 weeks the same general ranking was evident but the overall range was smaller owing to both a higher M pattern value and lower monocropped legume values. The P pattern value had decreased so t h a t it was not significantly different from the MP pattern, while the M pattern was no longer significantly different from the MS pattern. The nitrogen uptake values reflect the same ranking as exchangeable nitrogen at 0 weeks but the P pattern was significantly greater than the S pattern. Significant linear and positive correlations were obtained between nitrogen uptake by wheat at 19 weeks and both total dry matter yield (r 2 = 0.99, P < 0.05 in soybean; r 2 = 0.98, P < 0.05 in peanut) and grain yield (r 2 = 0.99, P < 0.05 in soybean; r 2 = 0.97, P < 0.05 in peanut). DISCUSSION

Effect of fertilizer nitrogen The lack of a significant response in maize grain yield to applied nitrogen was unexpected in view of the fact that grain yield increases o f over 2 tonnes ha -1 were obtained to nitrogen in the M and MP treatments. Two factors were probably mainly responsible for a lack of significant response: (a) the high grain yield at zero level of applied nitrogen (6.7 tonnes ha -1 for the MNO treatment), indicating that available soil nitrogen was not particularly low, and (b) the variability in maize grain yield between similar treatments in different blocks (coefficient of variation 17.8%). Nitrogen had apparently

142 remained in the soil from a previous crop of tomatoes. The variability was found, on checking the history of the site, to be largely attributable to a flood some years previously which had scoured depressions later refilled with sandy soil. In contrast there was a total dry matter response to nitrogen (Table II). The maize was able to compete easily with soybeans in the uptake of nitrogen because of its shallower root system (Beets, 1977), resulting in better growth and greater shading at higher levels of nitrogen (Table I). Increased shading probably accounts for the reductions in branches per plant, pods per branch and 100 seed weight in both soybeans and peanuts, even though total dry matter yield, grain yield (Table III) and plant height of both legumes were not affected. Reduced nodulation in soybean due to nitrogen was probably due also to increased shading, as well as direct suppression of nodulation b y increased soil nitrogen levels. Had nodulation in peanuts been measured at maturity instead o f at flowering, significant treatment effects may also have been found. Fertilizer nitrogen applied to the first experiment carried through to affect residual nitrogen available to the following crop. As the highest fertilizer rate was not excessively high, it being a normal commercial rate, it was likely that most of the applied nitrogen was taken up b y the crops, particularly the maize as mentioned above. It is therefore likely that the significant differences within both exchangeable soil nitrogen at sowing (Fig.l) and nitrogen uptake by wheat (Fig.3) were due to differences in the quantities mineralized from the crop residues. To have such differences observable just 4 weeks after harvesting the first experiment is quite consistent with the finding o f Bartholomew (1965) that mineralization rates within the first few weeks can be quite high due to the activity of microflora on the most nitrogen-rich residues. The linear response in nitrogen uptake to fertilizer nitrogen in the M pattern (Fig. 3), indicates that luxury levels of nitrogen had not been reached by the plants. They were therefore able to respond to additional inputs of relations obtained between nitrogen uptake b y wheat and total dry matter and grain yields of the legumes confirms the input o f nitrogen from this source. It appears from Figs.1 and 3 that there was little or no benefit from fixed nitrogen above 50 kg N ha -1, presumably n o t due to the attainment of luxury levels of nitrogen uptake b u t rather to the suppression of nitrogen fixation at higher levels of fertilizer nitrogen. The lack o f a significant difference in exchangeable nitrogen at 19 weeks between any t w o cropping patterns (Fig.2) is not altogether unexpected. The first flush of mineralization had probably ceased because the most readily decomposable organic matter had been mineralized already and because mean daily temperatures had fallen from those at the time of sowing. Exchangeable nitrogen values were probably lower at the second sampling at 19 weeks because of leaching by rain and irrigation water, and perhaps also because of denitrification. Crop uptake o f course cannot be invoked here as the sampling area was kept plant-free. The nitrogen contribution by the legumes is still evident.

143

Effect of cropping pattern Intercropping the maize with legumes has shown an obvious benefit in increased yields, as high as 36.5% for b o t h legumes (Table IV). Since there was no depression in maize grain yield with intercropping there was a clear bonus o f a high protein grain of at least 532 kg ha-' in soybean and 378 kg ha -1 in peanuts (Table III). This RYT value is lower than the 60% improvement in yield recorded by Andrews and Kassam (1976) b u t better than the 20--25% range of Singh (1977) f o u n d in the results o f the All-India Co-ordinated Experiments in which soybeans, peanuts and other legumes had been used in sorghum-based intercrop systems. The fact that maize grain yield was not depressed in any intercropped treatment here was probably due to the lack of competition for nutrients (P, K, Ca and S were in the basal dressing and N levels appeared adequate for good yields) and water (which was always supplied). Beets (1977) has also noted that when cereals and legumes are grown together, it is usually the cereal which is least affected by the interaction. There appears to have been no contribution of fixed nitrogen from the legumes to the maize. This would have been possible only if the legume had senesced well before the maize plants matured (Henzell and Vallis, 1976). Maize introduced competition primarily for light (Table I), to the disadvantage o f the legumes in terms o f branches per plant and pods per branch in both soybeans and peanuts, and 100 seed weight in peanuts and nodule weight in soybeans. Similar effects on yield components of intercropped legumes have been found b y Johnston et al. (1969) and Mann and Jaworski (1970), and on nodulation in soybeans b y R e d d y and Chatterjee (1973) and Wahua and Miller (1978). Field observation showed that intercropped legumes were markedly etiolated, thus making them as tall as monocropped, b u t heavier and higher yielding plants (Table III). While the proportional reduction in yield of soybeans due to intercropping was greater than in peanuts, it should be noted that the absolute grain yields of the intercropped legumes were substantially the same. Nitrogen uptake values for the wheat crop may be considered the most accurate index o f residual nitrogen available to a following crop. Uptake data in Fig.3 show that there was a significant second benefit from intercropping with legumes: as much nitrogen was available to the second crop following a maize + legume pattern to which no nitrogen had been applied as there was following a M pattern to which 100 kg N ha -1 had been applied. The exchangeable nitrogen values in Fig.1 reflect a similar pattern as Fig.3. Fig.1 legume values at 0 kg N ha -~ are lower than those at 100 kg N ha -~ applied to the M pattern probably due to the slower rate of mineralization o f nitrogen-poor residues in these treatments. At 19 weeks the contribution o f nitrogen from the legumes was still evident in the exchangeable nitrogen

144

values (Fig.2) even though the effect of nitrogen had disappeared. Nitrogen uptake values in all cropping patterns at 0 kg N ha -1 in Table V reveal the additional information that greatest residual nitrogen inputs were made by the monocropped legumes, particularly peanuts. (This is also reflected in the exchangeable nitrogen values at 0 and 19 weeks.) However, while the residual nitrogen values were highest for the S and P cropping patterns, the total grain yield was considerably lower than for the intercropped treatment at 0 kg N ha -1 (Table IV). Table V also reveals that residual nitrogen measured by nitrogen uptake on the intercropped plots (19 kg N ha -l ) was not much greater than for the monocropped maize (12 kg N ha -1 ). However, here while the difference in fixed nitrogen may appear small (and is also reflected in exchangeable nitrogen values), there was a considerable (at least 36%) increase in yield due to intercropping. The reason residual nitrogen values were lower in the maize + legume intercropping pattern than in the monocropped legume patterns (Table V) was probably twofold. Firstly, a greater depletion of soil nitrogen due to removal in the intercropped maize, and secondly, a suppression of the legume and its capacity to nodulate and fix nitrogen due to shading by maize. Higher inputs of residual nitrogen from legumes could have been obtained if the tops had also been returned to the plots. The results obtained here have shown clear benefits from intercropping maize with legume in terms of both total grain yield and nitrogen inputs. This information is unlikely to modify local maize cropping practices as the crop is wholly mechanized and damage to the legume intercrop would occur during harvesting. However, these findings in a hot, sub-humid environment should be relevant to tropical areas and to similar climatic regimes at higher latitudes where intercropping is practised. ACKNOWLEDGEMENTS

The Authors are grateful to Mr. D. King and Mr. P. Nixon of the University Farms, Camden, N.S.W., for technical assistance. The second author is also grateful for a Colombo Plan Award. The provision of maize seed by Mr. P. Rothwell, De Kalb Shand, Tamworth, N.S.W., is acknowledged. This investigation was part of a study on cropping systems initiated by the East-West Resource Systems Institute of the East-West Centre, Honolulu, HI, U.S.A.

REFERENCES Andrews, D.J. and Kassam, A.H., 1976. The importance of multiple cropping in increasing world food supplies. In: R.I. Papendick, P.A. Sanchez and G.B. Triplett (Editors), Multiple Cropping Symposium. Am. Soc. Agron., Madison, WI, pp. 1--10. Bartholomew, W.V., 1965. Mineralization and immobilization of nitrogen in the decomposition of plant and animal residues. In: W.V. Bartholomew and F.E. Clark (Editors), Soil Nitrogen. Am. Soc. Agron., pp. 285--306.

145 Beets, W.C., 1977. Multiple cropping o f maize and soybeans. Neth. J. Agric. Sci., 25: 95--102. Bremner, J.M., 1965. Inorganic forms of nitrogen. In: C.A. Black (Editor), Methods of Soil Analysis. Am. Soc. Agron., Madison, WI, pp. 1179--1232. Henzeil, E.F. and Vallis, I., 1976. Transfer of nitrogen between legumes and other crops. In: A. Ayanaba and P.J. Dart (Editors), Biological Nitrogen Fixation in Farming Systems in the Tropics. Wiley, New York, N.Y., pp. 73--88. Johnston, T.K., Pendleton, J.W., Peters, D.B. and Hicks, D.R., 1969.Influence of supplemental light on apparent photosynthesis, yield and yield components of soybeans. Crop Sci., 9: 577--581. Mann, J.D. and Jaworski, E.D., 1970. Comparison of stresses which may limit soybean yields. Crop Sci., 10: 620--624. Northcote, K.H., 1965. A factual key for the recognition of Australian soils. 2nd Edition, CSIRO Aust. Div. Soils, Divl. Rept. 2/65, Adelaide, 124 pp. Reddy, M. and Chatterjee, B.N., 1973. Nodulation in soybean (Glycine max) grown as a pure and mixed crop. Indian J. Agron., 18: 410--415. Singh, S.P., 1977. Intercropping and double-cropping studies in grain sorghum. In: International Sorghum Workshop, held at the International Crops Research Institute for the Semi Arid Tropics, Hyberabad, India, 6--12 March, 1977. Wahua, T.A.T. and Miller, D.A., 1978. Effects of intercropping on soybean nitrogen fixation and plant composition on associated sorghum and soybeans. Agron. J., 70: 292--295. Yoshida, S., Forno, D., Cook, J.H. and Gomez, K.A., 1972. Laboratory Manual for Physiological Studies of Rice. Second edition. I.R.R.I. Philippines, 83 pp.