Efforts to enhance maize stover utilization for smallholder livestock producers in Malawi

B H Dzowela
Ministry of Agriculture, Chitedze Research Station, P.O. Box 158, Lilongwe

Abstract
Introduction
Materials and methods
Results and discussion
Conclusion
References

Abstract

An attempt to boost the smallholder livestock farmer feed resourses qualitatively and quantitatively through the integration of improved forages in maize crop production systems is discussed. Where the testing of different forage systems undersown in maize was done on smallholder dairy farms, the Rhodes grass-silver-leaf desmodium appeared to make a better contribution to the overall forage production.

On station, the time of sowing/undersowing the forage legumes in a maize crop had no effect on all components of yield except forage legume yield which was depressed by delayed planting. Significant differences were obtained between the different forage systems and maize stripping intensities. There was a tendency for the climbing forage legume species to enhance the crude protein value of the maize stover component.

Introduction

Recent estimates by Mkamanga (1986) on national cereal grain production in Malawi puts the hectarage devoted to maize production at 1,145,100 ha with a total grain production of 1,355,200 tonnes. This production figure is 4.6 times the documented purchases of this crop by the Agricultural Development and Marketing Corporation (ADMARC) (Dzowela 1985). The reason for this discrepancy is that the major portion of the national production is used directly for consumption by the farm families themselves, maize being the stable cereal crop.

Alongside this large cereal grain production, there is an estimated 1.5 million tonnes of maize stover. This maize stover is a readily available feed resource to smallholder livestock producers which is utilized by livestock in situ, after the maize ears are removed, or cut and stall-fed. Maize stover per se as a feedstuff for livestock is inferior when compared to other locally available feed materials (Table 1).

The most limiting factor to animal performance fed on maize stover appears to be that of dietary protein. Protein supplementation alleviates the deficiency but a majority of protein supplements may be too expensive for the smallholder producers to afford on a regular basis. The most logical way of getting round this problem would be to incorporate tropical forages in the maize cropping systems (Dzowela, 1985). After the maize ears have been removed, the forage legume component, especially the climbing types could make an important contribution to maintaining adequate protein levels in the maize stover-based diets. The legume component could act as a source of protein above the 7% threshold and also in promoting increased intake of the associated maize stover.

Table 1. Comparison of maize stover with various other feed stuffs available on smallholder dairy farms in Malawi.

Feedstuff	No. of Samples	%CP	Range	%CF
Madeya (Maize bran)	4	10.7	(10.6 - 10.9)	5.5
Groundnut haulms	5	11.2	(10.9 - 11.6)	39.9
Maize stover with leaves	3	3.2	(2.8 - 4.9)	30.6
Rhodes grass hay	9	4.5	(3.5 - 5.2)	36.2
Maize silage	11	6.6	(5.2 - 9.1)	36.2

Dzowela (1985) outlined some of the chemical-based technologies for increasing the nutrient availability from maize stover. These methods include ammonia treatment and soaking in a solution of sodium hydroxide. This paper focusses on the work currently undertaken by the Pastures Commodity Research Team of the Ministry of Agriculture in attempt to integrate improved forages with maize cropping systems in Malawi.

Materials and methods

Smallholder based work

Using the adaptive on-farm research approach, some smallholder dairy farmers participated in an integrated maize crop-forage improvement programme in the Kasungu Agricultural Development Division. The pasture intervention was based on four systems, namely:

a. Pure Rhodes grass
b. Rhodes grass in association with silverleaf desmodium (Desmodium uncinatum)
c. Rhodes grass in association with Common Centro (Centrosema pubescens)
d. Rhodes grass in association with a "Shotgun" mixture involving Silverleaf desmodium, Common Centrosema, Neonotonia wightii and Joint vetch (Aeschynomene americanum).

All these pasture systems were undersown to maize after the first weeding operation before the maize crop was knee high. A similar approach was adopted in the 1984/85 season, in which the maize crop-improved pasture integration was done on three smallholder farms as a pre-fallow exercise. During the 1985/86 season a quantification of the forage production in both the newly established four pasture systems and those established in the 1984/85 season was done.

Station Based Work

This work was undertaken for a second season at Chitedze Agricultural Research Station during the 1985/86 season. The object was to generate simple technologies geared towards the integration of forage in cereal grain production of smallholder dairy producers. Four pasture systems were tested, namely;

a. Maize - common Centrosema (C. pubescens)
b. Maize - Neonotonia wightii
c. Maize - Centrosema pascuorum
d. Maize - Silverleaf desmodium.

These four forage legume systems were either sown at the same time the maize (MH12) was planted in November, 1985 or undersown after the first weeding of the maize crop when the maize plants were 30 cm high. In both cases the forage legumes were drilled on top of the ridge in between the maize planting stations which were 0.90m apart.

Based upon similar work in the previous year (Dzowela 1985), the maize leaves below the maize cob were stripped at weekly intervals from the silking/anthesis physiological development stage for seven weeks. The stripping of the lower maize leaves which serve no useful purpose in maize grain filling (Loomis 1935; Stickler and Pauli 1961; Hoyt and Bradfield 1962; Chaudhry 1969; Tanaka and Yamaguchi 1972; Soza et al 1975), could provide a cheap feed resource of high nutritive value to dairy animals on smallholder farms (Dzowela 1985). In the current study, results of this work are presented.

Results and discussion

On-farm Research

A quantification of the pasture systems sown the previous (1984/85) season revealed that the silverleaf desmodium - Rhodes grass system produced the largest amounts of forage dry matter (Table 2); being 38 per cent more than the pure Rhodes grass pasture system. The Rhodes-Centrosema system resulted in 11 per cent more forage than the pure Rhodes system. The silverleaf botanically accounted for more than 40 per cent of the sward, based on visual estimates, whereas Centrosema was less than 20 per cent.

Table 2. Second year's forage DM yields from the three pasture systems

Pasture system		DM yields in kg/ha
Pure Rhodes system		3805
Rhodes + Silverleaf		6110
Rhodes + Common Centro		4213
	Mean	4709
	SE	±660*

For the pasture systems established in the 1985/86 season the forage DM yield data are shown in Table 3.

Table 3. Forage DM yields of the pasture intervention systems in 1985/86 season

Pasture Systems	Pasture Components	Forage DM yield in kg/ha
Maize-Rhodes	Maize stover	3834
	Rhodes grass	1430
Maize-Rhodes with silverleaf	Maize stover	4440
	Rhodes grass	1016
	Silverleaf	433
Maize-Rhodes with shotgun mixture	Maize stover	3808
	Rhodes grass	1149
	Legume mixture	152
Maize-Rhodes with Centrosema	Maize stover	2903
	Rhodes grass	2910
	Centrosema	123
SE (Pasture systems)	Stover	±380*
	Rhodes grass	±242**
	Legumes	±225*

The Rhodes grass - Centrosema pasture system was the best intervention from the point of view of total forage production even though the very low maize stover yields in this system was below average. The system contributed 104 per cent more forage over and above the maize stover, but of the forage production only 4 per cent was contributed by the Centrosema component. The Rhodes-silverleaf system which produced comparable forage yields to the pure Rhodes system had a 30 per cent silverleaf contribution to the overall forage production of the pasture intervention over and above maize stover. The Rhodes-shotgun mixture system on the other hand, gave lower forage yields than the pure Rhodes intervention; the legume component in this system contributed 12 per cent to the total forage production over maize stover.

All systems were able to contribute significantly towards feed resources of the smallholder livestock producers over and above the maize stover production. The forage production from the pasture intervention system had a higher crude protein content of well above the minimum recommended value of 7.0 per cent, whereas the maize stover alone is less than 5.0 per cent in crude protein content (Dzowela 1985).

Station Research

Maize grain yield was not affected by the forage legume sowing/undersowing time and both the forage legume system and maize stripping intensity treatments did not have any effect on grain yield either. The fact that there were no significant grain yield reduction as a result of maize leaf stripping confirms what is already known about the physiology of the maize plant (Loomis 1935; Stickler and Pauli 1961; Hoyt and Bradfield 1962; Chaudhry 1969; Tanaka and Yamaguchi 1972; Soza et al. 1975). Maize stover yields were also not affected by legume sowing/undersowing time, forage legume systems or maize leaf stripping intensity (Table 4). With respect to forage dry matter yields as affected by the stripping of the different maize leaves, there was no interaction with time of undersowing. But the forage systems resulted in significant differences in the amounts of stripped leaf forage DM produced. The least amount came from the Neonotonia forage system (P<0.05) as shown in Table 5. There was a significant progressive increase in the amount of forage DM produced the more the plant leaves were stripped (P<0.01).

The total forage yields from all the leaves below the cob indicate the potential of using such a feed resource for ruminant feeding which is otherwise a useless component in maize grain filling processes. Use of these leaves below the cob for forage purposes is a practice in Guatemala, Egypt and Kenya (Soza 1975; Abate et al, 1985). It is a feed resource of high quality with a crude protein content of no less than 12 per cent (Dzowela 1985). Delay in sowing/undersowing of the forage legume from the time of planting maize to after the first maize weeding operation resulted in a significant reduction in forage yields from the pasture systems (P<0.01) as shown in Table 6. This reduction is a reflection of the smothering effect of the maize canopy on the early development of the forage legumes.

The forage legume component contribution to overall forage production in the different interventions was not affected by time of sowing/undersowing. The least amount of forage was contributed by Neonotonia which for two years now has indicated to be slow in establishment and hence most smothered in association with a maize crop (Dzowela 1985). The largest amount of forage came from the Pascuorum followed by the Silverleaf system (Table 7). There was a tendency for the legume forage DM to increase with maize leaf stripping intensity as a manifestation of improving light relationship underneath the maize canopy.

Table 4. Maize grain and stover yields from the four forage legume systems and maize leaf stripping intensities.

Pasture System		Maize leaf stripping
		1	2	3	4	5	6	7	Means
Maize-Common Centrosema:	Grain	6415	7760	7857	7387	6761	6257	6027	6922
	Stover	8504	9309	9230	8633	7771	7955	7711	8445
Maize-Neonotonia:	Grain	5992	5503	6490	5284	6566	6340	6868	6149
	Stover	8188	7951	9359	8235	8463	7481	8005	8249
Maize-Pascuorun Centrosema:	Grain	6424	5206	7117	5593	5832	6581	4818	5939
	Stover	8649	6888	8116	7847	8463	8879	7279	8017
Maize-Silverleaf	Grain	6680	7370	6553	7416	6882	6689	6125	6816
	Stover	7150	8952	7847	8163	8561	8232	7109	8002
Means:	Grain	6378	6460	7004	6420	6510	6465	5960
	Stover	8123	8275	8638	8220	8314	8137	7526
S.E. Pasture	Systems:
	Grain	±202NS
	Stover	±288NS
S.E. Stripping	Intensity:
	Grain	±355NS
	Stover	±342NS

Table 5. Stripped maize leaves and forage DM yields from the different Pasture systems and position of the stripped leaf below the cob

Pasture System	Leaf Position							Total Means	Forage kg/ha
	1	2	3	4	5	6	7
Common Centrosema	243	226	317	477	515	481	544	400	2803
Neonotonia	162	223	326	384	523	471	512	372	2601
Pusecuorum Centrosema	178	232	351	497	514	569	533	411	2876
Silverleaf	178	199	306	437	488	534	522	381	2664
Means	190	220	325	449	510	514	528
S.E. of Pasture systems means	±8.2*
S.E. of leaf position means	±173***

Table 6. Forage legume DM yield in kg/ha in response to time of sowing per undersowing

Forage System	Stripping Intensity							Means
	1	2	3	4	5	6	7
Maize-Common Centrosema	T1 324	448	681	377	253	200	380	353
	T2 146	274	271	207	221	163	188	210
Maize-Neonotonia	T1 23	211	98	59	130	86	71	97
	T2 23	41	22	45	46	67	32	39
Maize-Pascuorum	T1 738	1044	739	576	740	656	866	764
	T2 222	364	392	350	497	340	475	377
Maize-Silverleaf	T1 801	478	563	653	522	719	318	579
	T2 248	280	363	224	465	358	402	334
Means	T1 472	545	518	417	535	428	364	468
	T2 159	240	262	207	308	232	274	240
S.E. (T. Means)	±85***

Table 7. Forage DM Yield in kg/ha of the legume components in the different systems

Forage System	Maize leaf stripping intensity
	1	2	3	4	5	6	7	Means
Maize-Centrosema	235	260	476	293	188	207	194	269
Maize-Neonotonia	22	125	60	52	88	76	61	68
Maize-Pascuorum	480	704	560	463	619	498	670	571
Maize-Silverleaf	525	379	463	439	494	538	360	457
Means	315	392	390	312	470	330	319
S.E. (Pasture systems)	±61 **
S.E. (Stripping intensity)	±42 ***

Table 8. Crude protein percentages (on DM basis) of the maize stover from the four pasture systems

Forage System	Maize leaf stripping intensity							Means
	1	2	3	4	5	6	7
Maize-Centrosema	7.07	8.24	7.16	6.34	6.42	6.88	5.99	6.91
Maize-Neonotonia	6.97	7.88	6.88	5.92	6.81	5.54	5.04	6.43
Maize-Pascuorum Centrosema	6.46	7.94	6.91	5.94	4.69	4.88	4.13	5.85
Maize-Silverleaf	5.25	5.48	4.94	4.00	4.92	4.50	3.71	4.69
Means	6.44	7.46	6.47	5.55	5.71	5.45	4.72
S.E. Pasture System Means	±0.202 **
S.E. Leaf Stripping Intensity Means	±0.351 ***

Qualitatively, maize stover crude protein content was not appreciably affected by the forage legume sowing/undersowing time. The inclusion of the climbing forage legumes (Centrosema Neonotonia and Pascuorum) resulted in a maize/legume and stover product higher in crude protein content than the silverleaf-based stover (P<0.01) (Table 8). The climbing forage legumes were intact on to the stover at the time of harvesting the stover and contributed to the high CP contents. The silverleaf desmodium, on the other hand being determinate in its growth habit was not attached to the maize stover at the time of cutting.

There was a tendency for the maize stover crude protein content to decrease with the maize stripping intensity (Table 6). Maize stripping as such is a physical removal of the component of the maize plant that is more nutritive than the stem. However, the fact that the stripped leaf material when stripped before senescence is a product of high quality (Dzowela, 1985) which could be fed to animals fresh or dried, may more than compensate for the reduction in the CP content of stover per se.

Conclusion

These sets of data appear to demonstrate the potential for integrating improved forage technology in maize cropping systems of smallholder livestock producers. It helps to improve the feed resources of such smallholder livestock producers quantitively and qualitatively. Once adopted they could have an enhancing effect on the utilization of crop by-products such as maize stover and some of the otherwise wasted products from the maize plant.

The potential of climbing forage legume species ought to be explored further, especially those that stay green longer after the maize crop has dried up.

References

Abate A., Kayongo-Male H. and Wanyoike M. 1985. Fodder for high potential areas of Kenya. To be published in the proceedings of the Pasture Network for Eastern and Southern Africa Workshop on "Animal Feed Resources for Small-scale Live stock Producers", November 11-15, Nairobi, Kenya.

Boza R.F. Violic A.D. and Claure V. 1975. Maize forage defoliation. A paper presented at the XXI PCCMCA Meeting, El Salvador.

Chaudry A.R. 1969. Effect of defoliation treatments on yield in maize. Proceedings of the Sixth Inter-Asian Corn Improvement Workshop, India, p.37-39.

Dzowela B.H. 1985. Maize stover improvement with Legume forages. To be published in the proceedings of the Pasture Network for Eastern and Southern Africa Workshop on "Animal Feed Resources for Small-scale Livestock Producers", November 11-15, Nairobi, Kenya.

Hoyt P. and Bradfield R. 1962. Effect of varying leaf area by partial defoliation treatments applied at various stages of development. Agronomy Journal 54: 523-525.

Loomis W.E. 1933. The translation of Carbohydrates in maize. Journal of Science 9: 509-520.

Mkamanga G.Y. 1986. Cereal Crops Research Achievements and Production Problems in Malawi. A paper presented at the Annual Research Extension Workshop, 17-21 March, Mangochi, Malawi.

Stickler F.C. and Pauli A.W. 1961. Leaf removal in grain siorghum. 1. Effect of certain defoliation treatments on yield and components of yield. Agronomy Journal 53: 99-102.

Tanaka A. and Yamaguchi J. 1972. Dry matter production, yield components and grain yield of the maize plant. Journal of the Faculty of Agriculture, Hokkaido University.