Groundnut in Intercropping Systems

M. S. Reddy, C. N. Floyd and R. W. Willey*

In the developing world, groundnuts are com- seasons at ICRISAT Center, to study the growth
monly grown in intercropping systems, espe- patterns and the resource use in this combina-
cially by small farmers who use traditional tion to determine how yield advantages are
combinations often involving up to 5-6 crops. achieved. The first experiment, conducted dur-
Detailed statistics of farming practice are ing the rainy season of 1978, compared sole
difficult to obtain, but it has been estimated that crops with a single intercrop treatment of 1 row
95% of the groundnuts in Nigeria and 56% in millet: 3 rows groundnut. Results have been
Uganda are grown as mixtures with other crops presented in detail elsewhere (Reddy and Willey
(Okigbo and Greenland 1976). In the Northern 1980a) so they are only briefly summarized
Guinea Savanna Zone of Nigeria, Kassam here.
(1976) reported that only about 16% of the total Growth patterns are plotted in Figure 1. Sole
area under groundnut was in sole cropping millet showed a very rapid rate of growth,
while about 70% was in 2-4 crop mixtures. Un- achieving 8134 kg/ha of dry matter in 85 days
derplanting tree crops such as coconut, oilpalm, (Fig. 1b). Sole groundnut growth rate was
and rubber trees with groundnuts in the early somewhat slower, and this crop achieved 4938
years of the plantation is also a common feature kg/ha of dry matter in 105 days (Fig. 1a). Dry
in S.E. Asia (Hardwood and Price 1976) and matter yield of each crop in intercropping is
India (Aiyer 1949). given in comparison with an expected yield, this
This paper considers the intercropping of being the yield that would be achieved if the
groundnut only with other annual crops; it crop experienced the same degree of competi-
deals mainly with the cereal intercrops (millet, tion in intercropping as in sole cropping.
maize, and sorghum), which are by far the most Groundnut growth very closely followed the
important intercrops grown with groundnut. It expected dry matter yield of 75% of its sole crop
also considers briefly a further important group yield, whilst millet produced approximately
— the long-season annuals such as pigeonpea, twice its expected dry matter yield of 25% of its
cotton, castor, and cassava. sole crop yield. In effect, this means that
groundnut produced about the same yield per
plant in intercropping as in sole cropping, while
Intercropping of Groundnut the much more dominant millet approximately
with Cereals doubled its yield per plant in intercropping.
The combined dry matter yield in intercrop-
ping is given in comparison with the yield ex-
Groundnut/Pearl Millet Intercropping pected, if there was no yield advantage (or dis-
The groundnut/millet combination has been advantage) of intercropping, i.e.,of the LER = 1
chosen for special emphasis at ICRISAT be- (LER = Land Equivalent Ratio, or the relative
cause it involves two ICRISAT mandate crops land area required as sole crops to produce the
and the combination is an especially important yields achieved in intercropping). Figure 1c
one on the lighter soils of the semi-arid tropics, shows that with time there was an increasing
notably in West Africa and India. dry matter yield advantage for intercropping; at
A series of crop physiological experiments final harvest the actual LER was 1.29, i.e., an
has been carried out since 1978 in four different advantage of 29% for intercropping. Grain and
pod yields closely followed this pattern and ac-
tual LERs were 0.71 for groundnut and 0.55 for
* Agronomist, Research Intern, and Principal
Agronomist, respectively. Cropping Systems, millet, giving a total LER of 1.26, or an overall
ICRISAT. yield advantage of 26% for intercropping.

133
Figure 1. Sole crop yields and actual and expected intercrop yields of groundnut and millet.

Resource use was of particular interest in this level of efficiency as the sole crops, was apprec-
combination. Considering moisture use first, iably higher than the measured amount inter-
the amounts of water transpired through the cepted (Table 1). Calculation showed that the
sole crops and the intercrop are presented in intercrop appeared to use light with 28% grea-
Table 1. (The amount for the intercrop could not ter efficiency. This agrees very closely with the
be apportioned between the crops.) For the LERs given earlier, suggesting that the yield ad-
combined intercrop, an expected moisture use vantages of intercropping were due very largely
was also estimated by calculating for each to more efficient use of light. In fact, during the
component the amount of moisture which period of maximum leaf area, the intercrop
would have been used if dry matter had been supported a leaf area that was appro-
produced at the same efficiency as the respec- ximately 30% greater than the sole crops. Thus
tive sole crops. It can beseenthatthiscaiculated the greater efficiency of light use may at least
moisture use was very similar to the actual partly have been because light was more evenly
moisture use, thus there was no evidence that distributed over more leaves. It could also have
intercropping was able to produce more dry been partly due to the combination of a C4 crop
matter per unit of water transpired through the in the upper canopy layers and a C3 one in the
crop. lower canopy layers.
Light interception patterns are presented in An important feature of this first experiment,
Figure 2. Sole millet showed a particularly rapid however, was that it was conducted at a rela-
development of light interception, but the sole tively high level of fertilization (80 kg N/ha and
groundnut was rather slower. The combined 50 kg P20s/ha) and the season turned out to be
intercrop was intermediate to the two sole particularly wet with rainfall well above aver-
crops in the early stages, but by about 60 days it age. Thus it was considered that a major reason
was similar to both the sole crops; thereafter it why the higher intercropping yields appeared
declined because of senescence and removal of to be especially associated with increased ef-
t h e m i l l e t and then senescence of the ficiency of light use could have been because
groundnut. Light use by the individual compo- nutrients and water were not limiting. A main
nents in intercropping could not be distin- objective of subsequent experiments was to
guished. But the estimated amount of light re-examine the relative importance of this light
energy which would have been needed to pro- factor in situations where the below-ground re-
duce the intercrop yields, assuming the same sources were more limiting. Results have been

134
Table 1. Efficiency of resource use in pearl millet/groundnut intercropping.

Millet Groundnut

Water use
Sole cropping
Dry matter (kg/ha) 8134.00 4938.00
Water used (transpiration, cm) 15.86 19.63
Water-use efficiency (kg/cm) 513.00 252.00
Intercropping
Dry matter (kg/ha) 4129.00 3821.00
Water used at sole-crop 8.05 15.19
efficiencies (cm) 23.24
Expected water-use efficiency (kg/ha) 342.00
Actual water used (cm) 22.79
Actual water-use efficiency 349.00

Light-energy conversion
Sole cropping
Dry matter (kg/ha) 8134.00 4938.00
Total light intercepted (kcals/cm2} 14.26 19.25
Efficiency of conversion (mg/kcal) 5.70 2.57
Intercropping
Dry matter (kg/ha) 4129.00 3821.00
Energy required at sole crop 7.24 14.90
conversion rate (kcals/cm2) 22.14
Expected conversion efficiency (mg/kcal) 3.59
Actual interception (kcals/cm2) 17.25
Actual conversion rate (mg/kcal) 4.60

presented in detail elsewhere (Reddy and

Willey (1980b), so again they are only briefly
summarized here.
During the postrainy season of 1978, an ex-
periment was conducted to study the effect of
no-stress and stress moisture regimes (Table2).
The pattern of intercrop results in no-stress was
similar to that reported in the previous experi-
ment and the reproductive yield advantage was
25%. Under stress the reproductive yield ad-
vantage was rather higher at 29%. The ef-
ficiency with which light energy was converted
into dry matter was calculated as in the previous
experiment; in no-stress the intercrop was 2 1 %
more efficient than expected, while in stress it
was only 7% more efficient. Thus the results
suggest that when moisture is more limiting,
the efficiency of light use may be a less impor-
tant factor in determining the yield advantage of
Light interception by sole crops and this particular crop combination.
Figure 2.
an intercrop of pearl millet and During the rainy season of 1979, an experi-
groundnut. ment was carried out to study the effect of two

135
Table 2. Grain or pod yields and land equivalent ratios in pearl millet/groundnut Intercropping
under two different moisture regimes (1978 postralny season).

Millet Groundnut
grain yields pod yields
Millet Groundnut Total
Treatments (kg/ha) LER (kg/ha) LER LER

NO STRESS
(Irrigated every
10 days)
Sole crop 2674 - 2441 - -
1 : 3 Intercrop 1220 0.46 1928 0.70 1.25
STRESS
(Irrigated every
20 days)
Sole crop 2114 - 2040 - -
1 : 3 Intercrop 937 0.44 1734 0.85 1.29

LSD (0.05) within

a moisture regime 109 146 0.09
LSD (0.05) across
moisture regimes 133 217 0.08
CV (%) Main plots 3.26 4.95 2.57
CV (%) Split plots 3.60 4.03 4.38

different nitrogen levels on the millet (Table 3). phosphate levels are also varied so that nutritional
The pattern of results was again similar to the stress also applies to the groundnut.
previous experiments in that at a high level of
nitrogen (Nso) the reproductive yield advantage Groundnut/Maize Intercropping
was 2 1 % but this increased under stress (nil N)
to 32%. Dry matter yield advantages were even Groundnut is very commonly intercropped with
higher (Table 3). The efficiency of light energy maize in Southeast Asia and Africa. Mutsaers
conversion of the intercrop compared with the (1978) reported that in western Cameroon, the
sole crops was calculated as in the earlier exper- farmer grows groundnut as the main crop with
iments. At Nso, the intercrop was only 14% more maize interplanted at a fairly low density. Exper-
efficient, which was a rather smaller effect than iments carried out during three seasons in the
in the previous experiments. At nil N, however, Yaound'e area, Cameroon, to evaluate ground-
the improved light use efficiency of the inter- nut/maize mixtures, gaveyield advantages over
crop was even higher, being 2 1 % . At first, this pure stands ranging from 6-16%. Evans (1960)
effect at nil N is rather surprising, as it seems to obtained yield advantages ranging from 9-54%
contradict the earlier suggestion from the mois- from five different experiments conducted at
ture regime experiment that when a factor other two different locations in Tanzania during 1957
than light is more limiting, the efficiency of light and 1958. In Ghana, Azab (1968) studied
use is less important. But the results may simply groundnut/maize intercropping by varying the
indicate some essential differences between sowing time of each crop. He observed that the
the moisture stress and nitrogen stress situa- mean yield of groundnuts was significantly
tions which were created. One notable differ- higher when sown 4 weeks earlier than
ence of course was that the moisture stress maize. The traditional practice of sowing both
appiied to both component crops, whereas the crops at the same time gave an intermediate
nitrogen differences applied only to millet Cur- yield. Koli (1975) reported that the yields of
rent studies are examining situations where groundnuts in mixed cropping treatments were

136
Table 3. Grain or pod yields and land equivalent ratios in pearl millet/groundnut Intercropping
under two different levels of nitrogen applied to the millet (1979 rainy season).

Pearl millet Groundnut

grain yields Pearl pod yields
Groundnut Total
millet
Treatments (kg/ha) LER (kg/ha) LER LER

Sole groundnut - - 2998 - -

Sole pearl millet
(0 kg N/ha) 1968 - - -
Sole pearl millet
(80 kg N/ha) 2872 - - - -
1:3 Intercrop
(0 kg N/ha) 1063 0.54 2345 0.78 1.32
1:3 Intercrop
(80 kg N/ha) 1436 0.50 2131 0.71 1.21

LSD (0.05) 233 117 0.12

CV(%) 8 4 6.71

one-third to one-half the yields obtained from gen. This finding agrees with the general trend
sole crops, but yield of maize was not reduced observed in the groundnut/millet experiment
to the same extent. The general observation in referred to above (Table 3) and it has important
all reports on the maize/groundnut combina- implications in practice because it suggests that
tion is that groundnut yield is readily depressed intercropping may be more advantageous in
by competition from the maize. low fertility situations.
A groundnut/maize experiment was con- This groundnut/maize experiment was fol-
ducted on an Alfisol at ICRISAT in the rainy sea- lowed by a post rainy season crop of sorghum to
son of 1978 to study whether there was any study the residual effect of sole versus inter-
beneficial transfer of fixed nitrogen from the cropped groundnut. The results showed that if
legume to the cereal. Treatments consisted of no nitrogen were applied to the groundnut/
maize at 0, 50, 100, and 150 kg/ha of applied maize intercrop, there was a beneficial residual
nitrogen, and with and without a groundnut in- effect on the following sorghum. Where nitro-
tercrop. With no applied nitrogen, maize growth gen was applied to the maize, however, the
was very poor and obviously nitrogen deficient, groundnut growth was suppressed and the re-
and there was no visual evidence of growth sidual benefit rapidly diminished (Rao et al.
being any better if the groundnut intercrop were 1979).
present. This observation was supported
by maize grain yields which were un-
Groundnut/Sorghum Intercropping
affected by the groundnut at any level of nit-
rogen. The relative yield advantage of inter- In India and Africa, groundnut is very com-
cropping compared with sole cropping was monly intercropped with sorghum. Some re-
44% at zero nitrogen level but this decreased ports have emphasized that significant yield re-
with increase in applied nitrogen and it was zero ductions of groundnuts have been obtained
at the highest nitrogen level (Rao et al. 1979). when they have been intercropped with sor-
Since there was no evidence that these differ- ghum. John et al. (1943) reported that sorghum
ences in yield advantage could be due to differ- depressed the yield of groundnut by about 50%
ences in nitrogen transfer, it is possible that and Bodade(1964) obtained reductions of 52%.
they occurred because intercropping was more But despite reductions in groundnut yields,
efficient in using soil nitrogen, an effect that was there are many reports of overall benefits when
more evident at lower levels of applied nitro- the yields of both crops are considered.

137
 Bodade (1964) reported that mixed cropping ater in intercropping because of possible in-
of sorghum and groundnut gave higher yields teractions with the associated cereal crops. It
than sole cropping and two rows of sorghum has also been emphasized that for crops grow-
with eight rows of groundnut was one of the ing with a more dominant associated crop,
best treatments. Lingegouda et al. (1972) re- there may be particular need for identification
ported that three rows of groundnut and one and selection of genotypes within the actual
row of sorghum was more profitable (Rs. 3918/- intercrop situation because genotype perfor-
per ha) than pure sorghum (Rs. 3123/-) or pure mance in intercropping may not be very closely
groundnut (Rs. 2672/-). A positive benefit was related to genotype performance in sole crop-
shown in almost all experimental combinations ping (Willey 1979).
of groundnuts with sorghum in East Africa At ICRISAT, studies on the identification of
(Evans 1960). Experiments conducted at groundnut genotypes for intercropping with
ICRISAT with this combination have given yield pearl millet have been carried out since 1977. To
advantages as high as 38% (Rao and Willey date, results are only available for a relatively
1980) while Tarhalkar and Rao (1979) have few genotypes of groundnut, and these have
reported yield advantages up to 57%. been examined in combination with only a few
pearl millet genotypes (Table 4). All studies
were in simple replacement series treatments
Groundnut Genotypes of 3 groundnut rows: 1 pearl millet row. Results
for Groundnut/Cereal have indicated that with increasing groundnut
Intercropping maturity, and the associated change from
bunch to runner habit, the groundnut contribu-
As in sole c r o p p i n g , it seems likely that tion in intercropping (i.e. groundnut LER) tends
groundnut performance in intercropping could to increase (Table 4). This is probably because
be improved by identification of suitable of the increasing time for compensation of the
genotypes. Indeed it can be argued that the po- groundnut after cereal harvest.
tential for genotype improvement could be gre- However, this increasing groundnut con-

Table 4. The affact of groundnut ganotypa and ganaral typa of millat ganotypa on groundnut LER
and total LER in groundnut /pearl millat Intarcropping.

Groundnut Genotypes

1. Chico 2. MH2 3. TMV2 4. R33-1 5. MK374 6. M-13

Spanish Valencia Spanish Virginia Virginia Virginia Means

bunch dwarf bunch semi- semi- runner (Genotypes
Pearl millet genotypes 85 days 95 days 100 days spreading spreading 130-140 3-6)
110 days 125 days days

GAM73/GAM75 g nut LER 0.51* 0.63d 0.72d 0.80c 0.81 d 0.74

(dwarf, late) Total LER 1.13 1.25 1.22 1.27 1.33 1.27
BK560/WC-C75 g nut LER 0.48s 0.48a 0.61e 0.636 0.80c 0.80e 0.71
(medium/medium) Total LER 1.03 1.17 1.27 1.23 1.25 1.39 1.29
PHB-14/IVSAX75 g nut LER 0.67* 0.70* 0.68* 0.74* 0.70
(tall/medium) Total LER 1.09 1.18 1.01 1.28 1.14
Ex-Bornu g nut LER 0.90* 0.90* 0.80a 0.90* 0.88
(all, late) Total LER 1.25 1.22 1.15 1.28 1.23
Means g nut LER 0.48 0.50 0.70 0.74 0.77 0.81
Total LER 1.03 1.15 1.22 1.21 1.17 1.32

a. Mean of 1 trial b. Maan of 2 trials c Mean of 3 trials d. Moan of 4 trials a. Mean of 5 trials

138
 tribution is not so clearly reflected in increasing occurs. The groundnuts can give reasonably
yield advantages for the combined effect of efficient use of resources during the early
both crops (i.e. total LER); although the latest period when the long season annuals are slow
maturing groundnut M-13 (130-140 days) was to establish; after groundnut harvest, the long
associated with the highest mean value for total season annuals are able to make use of later
LER, there were no real differences in total LER resources, especially of the residual soil mois-
observable between the three genotypes TMV2 ture.
(100 days), Robut33-1 (110 days), and MK-374
(125 days). There was also little difference in
groundnut or total LER for the different millet Groundnut/Pigeonpea Intercropping
genotypes, t h o u g h the range of millet
This combination is particularly prevalent on
genotypes was admittedly limited.
red soils of the southern States of India. A
In these initial stages of identification, simul- common practice here is that if rains commence
taneous screening of genotypes of both crops at the normal time a groundnut/sorghum or
was carried out because there appeared to be groundnut/millet intercrop is grown, but if rains
scope for selecting more suitable genotypes of are delayed groundnut/pigeonpea is grown.
both crops. No marked interaction between Pigeonpea rows are usually wide-spaced up to
genotypes of the two crops has been observed 5 m apart with up to 8-10 groundnut rows in
so work is now concentrating on examining a between. This traditional practice helps to ob-
larger number of genotypes of each crop tain high yields of the groundnut cash crop but
against a standard genotype of the other crop. the overall advantage of intercropping may not
With the groundnuts, a more detailed study is be high because pigeonpea is too sparsely dis-
also being carried outtodeterminethe extentto tributed to make efficient use of late season
which the better intercrop performance of the resources and produce a worthwhile yield con-
longer maturing genotypes is due to greater tribution. Most studies have examined this pre-
time for compensation after cereal harvest or to dominantly groundnut situation.
some other characters which allow better John et al. (1943) reported from a 3-year study
growth and production in the dominated inter- that groundnut/pigeonpea in 8:1 proportion
crop situation. In the summer season of 1980, was 43% more profitable than sole groundnut.
groundnut genotypes were grown with a stan- Similar results were reported from studies at
dard cereal (Sorghum CSH-8); the duration of Tindivanam over a 7-year period during 1942-
cereal competition was examined by removing 49 (Seshadri et al. 1956). Veeraswamy et al.
thesorghum at differenttimes, and the intensity (1974) and Appadurai et al. (1974) showed that
of cereal competition was examined by means the arrangement of 6 groundnut: 1 pigeonpea
of a treatment in which alternate pairs of sor- was more economical than 8 : 1 ; groundnut
ghum leaves were removed. First results gave 99% of its sole crop yield and pigeonpea
suggest that increased groundnut contribution 37% of its sole crop yield, totaling an advan-
with reduced cereal duration was of the same tage of 36%.
order for all groundnut genotypes and both
At the other extreme, an alternate row ar-
levels of competition. Differences in groundnut
rangement at ICRISAT gave an LER of 1.53
performance were small at a given cereal dura-
comprising 95% pigeonpea and 58% ground-
tion, though there was a tendency for the bunch
nut (Rao and Willey 1980). This may not be
types to do less well than the late runner types.
ideal economically because of the reduced
groundnut contribution, but it illustrates that
higher yield advantages can be obtained with
Groundnut Intercropped with higher proportions of pigeonpea.
Long Season Annual Crops A good compromise situation is indicated by
some studies on five Alfisol locations within IC-
No growth studies have been reported for com- RISAT in 1979-80. Pigeonpea was grown in 135
binations of groundnuts with any of the long cm rows with five very close-spaced rows of
season annuals. However, it is evident from the groundnut between. The population of each
general growth patterns of the crops that con- crop was equivalent to its sole crop optimum.
siderable temporal complementarity of growth Intercrop yields averaged 82% of groundnut

139
and 85% of pigeonpeas, i.e. 67% total advan- rogen from nitrogen fixation. This ground-
tage. nut/cassava combination gave around double
the net income compared with the sole cassava
planting. Contrary to this, the Department of
Groundnut/Cotton Intercropping
Agriculture, Tanganyika (1959) reported that
Joshi and Joshi (1965) reported that a combina- when early sown groundnuts were intercrop-
tion of 2-3 rows of groundnut between cotton ped with late-planted cassava, the yield of
rows spaced 6 feet apart gave significantly groundnuts was not seriously affected, but the
higher monetary returns compared to either yields of cassava were reduced to less than
sole crop. Varma and Kanke (1969) reported that one-fifth of the sole crop. Potti and Thomas
growing cotton with groundnut was much more (1978) reported that trials conducted in the far-
remunerative than growing it alone; yields of mers' fields in Kerala, India gave an average of
groundnuts were additional to the cotton yields 1263 kg/ha of groundnut in addition to the cas-
usually obtained. Similar intercropping of sava yield.
cotton and groundnut has been recommended
for the northern districts of Madras by Narayan
Reddy (1961). In the Sudan, Anthony and Wil- Conclusions
mott (1957) also found higher yields from
groundnut and cotton intercropped together. There is good evidence that groundnut/cereal
intercropping can give worthwhile yield advan-
tages over sole cropping. The ICRISAT studies
Groundnut/Castor Intercropping suggest that these advantages can be due partly
Reddy et al. (1965) reported that growing castor to more efficient use of light, but further re-
mixed with groundnut was better than raising a search is needed to determine the importance
pure crop of castor, and monetary returns were of this light factor when below-ground re-
61.9% higher than pure castor. They also re- sources are limiting. The more rapid early
ported that the yield of castor was more when it growth of the cereals, and the later maturity of
was grown mixed with groundnut compared to groundnut compared with the early cereals,
castor grown mixed with greengram, cowpea, may also be an important factor giving some
Setaria, millet or sorghum. In East Africa, Evans complementarity between the crops and allow-
and Sreedharan (1962) showed that there was a ing better use of resources.
clear increase in production when castorbean Other ICRISAT studies have shown that the
and groundnuts were planted together com- later maturing, semi-spreading or runner types
pared to sole cropping. Tarhalkar and Rao of groundnut have given the highest groundnut
(1975) reported that intercropping of castor/ yields in intercropping, but this has not always
groundnut gave monetary returns up to Rs 4394 resulted in improved yield benefits from the
per hectare compared with Rs 3317 per hectare whole system.
obtained from a pure castor crop. Although there has been little detailed work
on the intercropping of groundnuts with the
long season annuals, pigeonpea, cotton, castor,
Groundnut/Cassava Intercropping and cassava, there is good agronomic evidence
Introducing an additional crop like groundnut that these systems can give very substantial
between the traditionally wide-spaced cassava yield advantages. The general growth patterns
plantings would increase the production ef- of these crops suggest that the main factor re-
ficiency of cassava-planted land as well as con- sponsible for these advantages is that the use of
serving soil moisture and fertility. An experi- early resources by the groundnut complements
ment conducted at Khon Kaen University, the use of late resources by the longer season
Thailand in 1977, produced higher yields of cas- crops.
sava (26 756 kg/ha) when intercropped with
groundnuts compared to sole crop of cassava
(24 538 kg/ha). The experiment indicated that References
presumably intercropped groundnut increased AIYER, A. K. Y. N. 1949. Mixed cropping in India. Indian
the yield of cassava by supplying additional nit- Journal of Agricultural Science 19: 439-543.

140
ANTHONY, K. R. M., and WILLIMOTT, S. G. 1957. Cotton with maize and groundnuts. Netherlands Journal of
interplanting experiments in the South-West Sudan. Agricultural Science 26: 344-353.
Empire Journal of Experimental Agriculture
25: 29-36. NARAYAN REDDY, M. R. 1961. Economics of groundnut
and cotton mixtures versus pure groundnut crop in
APPADURAI, R., and SELVARAJ, K. V. 1974. A note on the northern track of Andhra Pradesh. Andhra Ag-
groundnut — redgram mixture in lower Bhawani ricultural Journal 8: 66.
Project area. Madras Agricultural Journal
6 1 : 803-804.
OKIGBO, B. N., and GREENLAND, D. J. 1976. Intercrop-
ping systems in tropical Africa. Pages 66-101 in
AZAB, Y. E. A. 1968. Pages 5-8 in Applied agronomic
Multiple cropping. American Society of Agronomy
research on field food crops in Northern Ghana
Special publication No. 27, Madison, Wisconsin,
(FAO. No. TA 2596).
USA.
DEPARTMENT OF AGRICULTURE, TANGANYIKA. 1959. In-
tercropping. Report of the Department of Agricul- PoTTI, V. S. S., and THOMAS, A. 1.1978. Groundnut as a
ture, Tanganyika. 11pp. companion crop in tapioca. In Proceedings of the
national symposium on intercropping of pulse
EVANS, A. C. 1960. Studies of intercropping. I. Maize or crops, IARI, New Delhi, July 17-19.
sorghum with groundnuts. East African Agricultural
and Forestry Journal 26: 1-10. RAO, M. R., AHMED, S., GUNASENA, H. P. M„ and ALCAN-
TARA, A. P. 1979. Multi-locational evaluation of pro-
EVANS, A. C , and SREEDHARAN, A. 1962. Studies of ductivity and stability of some cereal-legume inter-
intercropping. II. Castorbeans with groundnuts or cropping systems: A review of inputs trial III. In
soyabean. East African Agricultural and Forestry Proceedings of the Final Inputs Review Meeting,
Journal 28(1): 7 - 8 . held at Honolulu, Hawaii, Aug 20-24.

HARDWOOD, R. R., and PRICE, E. C. 1976. Multiple crop- RAO, M. R., and WILLEY, R. W. 1980. Preliminary
ping in tropical Asia. Pages 11-40 in Multiple crop- studies of intercropping combinations based on
ping. Society of Agronomy, Special Publication No. pigeonpea or sorghum. Experimental Agriculture
27, Madison, Wisconsin, USA. 16: 29-39.

JOSHI, S. N., and JOSHI, H. V. 1965. Mixed cropping of REDDY, G. P., RAO, S. C , and REDDY, R. 1965. Mixed
groundnut/cotton under irrigated conditions in cropping in castor. Indian Oilseeds Journal
Saurashtra. Indian Oilseeds Journal 9(4): 244-248. 9(4): 310-316.

JOHN, C. M., SESHADRI, C. R., and BHAVANI SHANKER REDDY, M. S., and WILLEY, R. W. 1980. Growth and
RAO, M. 1943. Mixed cropping of groundnut. Madras resource use studies in an intercrop pearl millet/
Agricultural Journal 31: 191-200. groundnut. Field Crops Research (In press).

K A S S A M , A . H. 1976.Pages41-48 in Crops of the West REDDY, M. S., and WILLEY, R. W. 1980. The relative
African Semi-Arid Tropics. ICRISAT, Hyderabad, importance of above and below ground resource
India. use in determining yield advantages in pearl millet/
groundnut intercropping. Paper presented at the
KHON KAEN UNIVERSITY/FORD CROPPING SYSTEMS PROJECT second symposium on intercropping in semi-arid
ANNUAL REPORT. 1977. Pages 25-39 in Intercropping of areas. Held at University of Dar-es-Salaam, Moro-
cassava with field crops. goro, Tanzania, Aug 4-7.

Kou, S. E. 1975. Pure cropping and mixed cropping of SESHADRI, C. R., AIYADURAI, J. G., and SRINIVASALU, N.
maize and groundnuts in Ghana. Ghana Journal of 1956. Groundnut-mixed cropping experiment. Mad-
Agricultural Science 8: 23-30. ras Agricultural Journal 43: 496-504.

LING EGO woA, B. K„ SHANTAVEERABADRAIAH, S. M., IN- TARHALKAR, P. P., and RAO, N. G. P. 1975. Changing
AMDAR, S. S., PRITHVIRAJ, and KRISHNAMURTHY, K. concepts and practices of cropping systems. Indian
1972. Studies on mixed cropping of groundnut hy- Farming, June 3-7.
brid and hybrid sorghum. Indian Journal of Ag-
ronomy 17(1): 27-29. VARMA, M. P., and KANKE, M. S. S. 1969. Selection of
intercrops for cotton in India. Experimental Agricul-
MUTSAERS, J. A. 1978. Mixed cropping experiments ture 5: 223-230.

141
VEERASWAMY, R., RATHNASWAMY, R„ and PALANIS- WILLEY, R. W. 1979. Intercropping — Its importance
WAMY, G. A. 1974. Studies in mixed cropping of red - and research needs. I. Competition and yield advan-
gram and groundnut under irrigation. Madras Ag- tages and II. Research approaches. Field Crop
ricultural Journal 6 1 : 801-802. Abstracts 32: 2-10 and 7 3 - 8 1 .

142
 Session 5 — Crop Nutrition and Agronomy

Discussion

Microbiology P. T. C. Nambiar
We are doing such trials but we have no
J. S. Saini results yet.
In a trial in the Punjab we found that when
winter wheat was planted either after J. C. Wynne
groundnuts, hybrid maize, or local maize we We are cooperating in these trials with
got better wheat yields after the local maize. ICRISAT. We do not have results yet, but I would
This was surprising. What explanation can be suspect that the combinations would be
offered? specific to sites. A lot would depend on the
variety and the photosynthetic activity of the
P. T. C. Nambiar variety in the different environments.
It is difficult to generalize on this. One likely
explanation is that in this instance, nitrogen S. N. Nigam
was not limiting. At ICRISAT we have obtained In one of the slides that showed analysis of sev-
a 30% yield increase in pearl millet when it eral characters, there was no significance for
followed groundnuts. nodule number for the host cultivars. I would
have expected that there was a large amount
N. D. Desai of variability for nodule number, unless very
When do nodules form and when does fixa- few genotypes were in the study. Secondly,
tion commence? regarding the use of plant color in evaluating
fixation by different strains of Rhizobium, the
P. T. C. Nambiar different botanical types of groundnuts them-
Initiation varies from season to season. In the selves vary in leaf color. Could we say in
rainy season they form as soon as 11 days general that the Virginia types have better
after planting. In the postrainy season, they nodulation than Valencia and Spanish types?
may not form until 18 days after planting.
Nitrogenase activity commences 20 days after J. C. Wynne
planting in the rainy season. The nodule varies with genotype, and if
enough genotypes are used then significant
P. J. Dart differences are recorded. The results shown
There are large nitrogen reserves in the seed were limited and were not for all the experi-
and therefore a shorter dose of nitrogen may ments we have conducted. We would not use
not be needed. Water also limits nodulation leaf color for selection purposes. When we
and the uptake of nitrate. evaluate strains we use nitrogen-free soils and
we remove the cotyledons also. Color is then a
D. J. Nevill useful parameter for comparison of strains on
We heard a lot about host and Rhizobium a single cultivar. Generally in North Carolina,
strain interactions. What about higher order we find that Virginia cultivars are the best
interactions such as strain x host x environ- nodulators followed by Spanish and then
ment interactions? In other words, does a Valencia botanical types.
successful combination of strain and host
behave the same way in North Carolina as it R. 0. Hammons
does in India? Nonnodulating lines were found by Dr. G or bet

143