Peter J. thome*

Livestock Feeds Project, Ministry of Agriculture and Lands, PO Box G13, Honiara,Solomon Islands

Received June 1991; revised May 1992

A concentrated protein supplement based on locally available fish meal and wheat mill run was evaluatedas a means of improving the efficiency and profitability of smallholder pig production in the SolomonIslands. In comparison with a low nutrient-dense diet based on cassava and coconut alone, supplementationwith small amounts of concentrate [20% of dry matter (DM) intake] resulted in a much improvedgrowth rate (P < 0.001), total DM intake (P < 0.001), and feed conversion ratio (P < 0.001) ofgrowing pigs between 6 kg and 55 kg live weight. Restricting DM intake of pigs fed low nutrient-dense feeds and concentrate had little adverse effect on growth rate (P = 0.380) as the animalscompensated for reduced feed intake with improved feed conversion ratio (P = 0.050). A financialappraisal of these data suggested that the improvements in production parameters observed withthe use of concentrate would translate into an enhanced financial performance for the smallholderpig keeper albeit at an increased risk. Returns to labour for this system were greater than for otherforms of land use currently practised in the Solomon Islands.

Keywords: Pig feeds; Solomon Islands; Cassava; Low nutrient-dense pig diets; Concentrated supplement

in the diets of pigs kept by smallholders. It canalso be of value in ensuring a reliable supply ofessential minerals and vitamins and for the applicationof anti-parasitic medications.

Previous studies have shown that feeding ofconcentrates with low nutrient-dense feeds canproduce growth rates similar to those observed withcompound feeds under commercial conditions (Calder,1960; Gomez, 1977; Watt, 1973; Thome, 1992).However, little attention has been paid to the interactingsocioeconomic factors which are ultimately likelyto be most important in determining adoption ofthis type of feeding system by farmers.

The feeding trial described here was designed togenerate physical data that could be used in a financialappraisal of the use of concentrates to supplementthe low nutrient-dense feeds (cassava and coconut)that are widely available in the villages of the SolomonIslands.

Materials and methods

Dietary treatments

Three dietary treatments based on low nutrient-denseraw materials were tested in the experiment:Treatment A -fresh chopped cassava offered adlibitum with a daily allowance of fresh coconut.This treatment was designed to simulate the tra-

Rural markets in the Solomon Islands exhibit aconsiderable demand for pigs because of theimportance of pork in fulfilling the social obligationsof the customary feast. However, poor productivityfrom smallholder livestock enterprises, due mainlyto poor nutrition, means that it may be difficult tomeet this demand locally.

Distribution of complete compound pig feeds froma central manufacturer could improve standards ofnutrition considerably but shipping is complicatedwithin the scattered archipelago. Furthermore, theuse of a complete compound feed does not allowfor optimizing the use of those feed resources suchas household wastes and crop residues which maybe available as a result of the other activities ofthe smallholder farmer.

The use of "Concentrated feeds formulated spe-cifically to complement available, low nutrient-denseraw materials is a possible alternative. Transportdifficulties are reduced as the volume of concentraterequired to support a given level of production canbe as little as one fifth of the equivalent amountof compound feed. Using concentrates, a reasona-bly balanced diet could be supplied to the villagepig despite less than optimum conditions ofproduction. A concentrate is required principally tomake up the shortfall of protein usually occurring

.Present address: Natural Resources Institute, CentralAvenue, Chatham Maritime, Kent ME4 4TB, U.K.

@ 1993 Trap. Agric. ([rinidad)Trop. Agric. (Trinidad) Vol. 70 No.1 January 199378

Pig production in Solomon Islands: P.J. Thorne

Table 1 Raw material composition of feeds,

sess statistically the differences between treatmentsB and C. .,

ConcentrateComposition %

Components

Mill run'Fish mealSaltLysine-HOPremixMecadox growth

promotor 1.

Analysis

DI)I matterCrude proteinEther extractCrude fibreAshNitrogen-free

extract ~

CalciumPhosphorousLysineMethionine +

cystine-'Mixed wheat-bran fraction

Cassava%

Coconut% Animals and procedure

At the start of the trial, a total of 18 male and18 female cross-bred pigs of between 6 and 10kg live weight (one to two weeks post-weaning) wereallocated randomly to treatment. The exact geneticconstitution of the animals was uncertain as theydid not result from a carefully controlled breedingprogramme. They would have contained genes invarying proportions from Large White, Tamworth,Berkshire, and native pigs. A weight-based dailyallowance (Table 2) for the concentrate and coconutswas calculated using the asymptotic model of voluntaryfeed intake (ARC, 1981). Cassava was fed ad libitumto the animals on treatment B and to a restrictedscale (also calculated using the asymptotic equation)to those on treatment C. All feed intakes were record-ed daily throughout the experimental period. Animalswere weighed weekly and, on reaching a live weightin excess of 55 kg, were fasted for 24 hours beforeslaughter. The final starved live weight and theweight of the eviscerated and trimmed carcass withthe head attached were used in the calculation ofdressing percentage.

28.5567.730.570.15200

()()

91.041.87.92.6

15.8

56.03.7

31.71.51.0

32.00.80.41.30.9

11.7

4.002.363.15

1.19

18.1

0.010.030.10

0.05

28.5

0.060.070.03

0.01

Chemical analyses

Full proximate analyses of the feed samples wereundertaken according to the methods of AOAC (1984).Amino acids were estimated using an auto-analyzermethod based on that of Spackman et 01. (1958).

Results and discussion

No effects of pen were evident in any of the para-meters measured.

Physical performance

The physical performance data are summarizl-:J inTable 3.

Pigs fed treatment A were characterized by theirslow growth rates and low voluntary intakes of afeed which was inefficiently converted into live weight.This dietary treatment was designed to simulate, as

Table 2 Amount of concentrate, cassava, and coconutfed to animals

ditional extensive system practised currently by thevillage pig farmer which uses feeds available on-farm. The most significant characteristic of this systemfrom a nutritional point of view is the serious limitationon protein supply.Treatment B -fresh chopped cassava offered adlibitum with daily allowances of fresh coconut andconcentrate. This may be described as a semi-intensivesystem as the basic feeds used are the same asfor treatment A, but a concentrated supplement whichwould in practice be brought in from off-farm isused to counteract the limitations imposed by aninadequate protein supply.Treatment C -fresh chopped cassava offeredrestricted with daily allowances of fresh coconut andconcentrate (semi-intensive system). This treatmentmay also be described as semi-intensive, but it wasdesigned to simulate the restriction on feed intakewhich is likely to occur under practical conditionsbecause of limited labour availability for feeding ormisconceptions on the part of the farmer as to theamount of feed that pigs are capable of consumingwhen their diets are properly balanced for essentialnutrients. The degree of restriction of cassava inthis treatment was calculated to impose a limitationof approximately 20% on total dry matter (OM)intake.

The ingredient composition of the concentrateand chemica! analyses of all feeds are shown inTable 1.

Cassava'(kg pig-J day-I)

1.11.62.02.42.83.13.43.63.8

Weight of pig Concentrate

(kg) (g pig-I day-I)-< 15 160

15 -20 23020-25 29025-30 35030-35 40035-40 45040 -45 49045-50 53050-55 560

Coconut(kg pig-I day

1.01,02.02.03.03.04.04.04.0

Experimental design and statistical analysis

The experiment was designed so that each of thethree dietary treatments was replicated by three pensof four (two male and two female) pigs. A two-way analysis of variance to isolate pen and treatmenteffects was applied to all data. Student's i-test forpaired samples was used, where appropriate, to as- *r reatment Conly

Trop. Agric. (Trinidad) Vol. 70 No.1 January 1993 79

Pig production in Solomon Islands: P.J. Thorne

Table 3 Physical performance of pigs under the three treatments

DressingpercentageTreatment

,,;

Weightgain

(g pig-I day-I)

A 26"

B 424b 1432b 3.3Sb 81.7"

c 40gb 1131C 2.76C 80.8a

d.f. 11 2 2

SE 9.1 50 0.11 1.5

P <0.001 <0.001-

Means for each parameter associated with the same superscript did not differ significan~

<0.001 0.590

on treatment B (Table 4). However, the feed conversionratio of the animals fed treatment C was improved(P = 0.05) such that there was little difference(P = 0.38) between the two treatments in growthrates observed. These results suggest that, up tothe observed 20% restriction on total OM intake,the effects of restrictions on cassava intake whichmight occur, in practice, could be beneficial ratherthan detrimental to overall productive efficiency.

The poor growth rates of the pigs on treatmentA prevented them reaching slaughter weight duringthe experiment. Dressing percentages observed withthe other two treatments were similar (P = 0.59).

Financial appraisal

The two principal variables affecting the appraisalof the data from this experiment were identifiedas:

1 Price of pigs, affected chiefly by location. Atthe time of the experiment, the normal rangein pig price was between $2.50 kS-l live weight[Solomon Island dollars; £1.00 = $3.74 (March,1990)) where pigs are sold intermittently toneighbouring villagers and a maximum of$6.50 kg-l live weight in areas of concentrateddemand [towns, industrial sites (in these situationsthere is a waged population with no land rightsin the vicinty to allow them to pursue their ownagricultural activities)].

closely as practicable under the conditions of theexperiment, the nutritional status of village pigsfed in the traditional manner. In practice, the villagediet would probably be slightly more diverse in itscomposition which would also vary from day to daydepending on the availability of particular feeds (deFredrick, 1971). However, the growth rates observedin animals fed treatment A were of the same orderof magnitude as a mean figure of 56 g day-l calculatedfrom survey data for village pigs at three locationsin the Solomon Islands (P.J. Thome, unpublisheddata), and this treatment was therefore felt to representan appropriate control for the experiment.

Use of the concentrate with the same low nutrient-dense ingredients resulted in disproportionateimprovements in all of the physical parametersmeasured, clearly illustrating the effects of improvingoverall dietary nutrient balance on voluntary feedintake and gross efficiency of feed utilization. In treat-ment 8, the provision by the concentrate of thenutrients which were in short supply in treatmentA resulted in a twofold (when corrected for unequallive weight) increase in the ad libitum consumptionof cassava.

The degree of restriction of cassava intake imposedin treatment C reduced its consumption by 33%in comparison with the pigs fed cassava ad libitum

Table 4

Treatment2, Opportunity cost of cassava, affected by similar

factors to pig price. Cost ranges effectively fromzero to around $2.30 kg-10M where formalmarkets are established (Anon., 1989).300bB 26Sa 867b

288b26za 581C The price of copra ($0.40) is controlled country-wide by the Commodities Export and MarketingAuthority and may vary in either direction with timedepending on world market prices. The price ofconcentrate could be expected to rise over time withincreases in raw material prices. At the time of theexperiment these had been stable for more than12 months. Effects of exchange rate fluctuations

d.f.

SE

2 2 2

3625

p 0.027 <0.001

Means for each parameter associated with the same superscript didnot differ significantly

80 Trop. Agric. (Trinidad) Vol. 70 No.1 January 1993

Pig production in Solomon Islands: P.J. Thorne

10 I

~

-15

L-20

0 0.5 I 15

price 01 cassava ($)

Figure 1 Effect of price of pigs and cassava on marginsover feed costs for extensive and semi-intensive pig feeding.Selling price of pigs Give weight) and dietary treatment are(-*) $2.50 kg-I (A); (-8) $2.50 kg-I (C); f+) $4.50 kg-I (A);(~) $4.50 kg-l (C); f*-) $6.50 kg-I (A); i;e) $6.50 kg-I (C)

.I 6 I

~

~1'>,

G"CGE~.."t..

'E)

'(>

-20 ~

Figure 2 Effect of price of pigs and cassava on returnsman-day-1 from semi-intensive pig feeding. Selling price ofpigs live weight are (-8) $2.50 kg-I; ~) $4.50 kg-I; (-6-)$6.50 kg-I

influx into the capital of pigs from the provincesalthough these would have to compete with increasingnumbers of animals which are being reared by largerproducers using cheap compound feeds based onlocally available feeds. However, the concept ofgrowing a crop, not for direct consumption or salebut as a raw material in another agricultural under-taking, is probably alien to many Solomon Islanderswho are perhaps more concerned with short-termreturns and risk minimization. Adoption of semi-intensive feeding systems for pigs will also dependon a reliable distribution network for concentrate

would be minimal as imported ingredients formedonly a small part of the concentrate's formulation.Returns from the semi-intensive feeding system willalso vary with yields of cassava and coconut. Forthe purposes of this appraisal yields were fixed atlevels (cassava, 6.11 t OM ha-I yrIj coconut, 0.9t OM ha-I yrI) reasonable for smallholders in theSolomon Islands (Patten and Aeming, 1988).

In Rgure 1, margins over feed costs arecompared for the extensive and semi-intensivesystems. In both cases, margins were responsiveto changes in pig price and. particularly, in cassavaprice. The effects of varying opportunity cost ofcassava were less pronounced in the pigs fed treat-ment C suggesting that, in practical terms, the in-herent stability of returns under the semi-intensivesystem would be improved considerably in compari-son with the traditional system. This must, however,be assessed in relation to the fact that with treatmentA, operating with negative margins represents lostopportunity whereas with treatment C, it is morelikely to represent an actual loss of money. Risk,therefore, is greater when the semi-intensive systemis employed. The two factors most important inavoiding this risk are likely to be adequate man-agement skills to maintain pig performance andthe reliability of the market for pigs.

Figure 2 illustrates that with the semi-intensivesystem, high pig prices are necessary to sustain positivereturns man-day-I when the opportunity cost of cassavais high. In areas where pig prices are lower, thepossibility of negative returns arises. However, asboth pig meat and cassava are food staples, marketdemand and consequently prices for each are unlikelyto be entirely independent. Where there is littledemand for pig meat and therefore a low marketprice, there is also likely to be a relatively smallmarket and low price offered for cassava. In practice,this situation would operate to reduce the risk inpig production below the levels suggested by casualperusal of Rgure 2 and assist in sustaining positivereturns from the enterprise.

In Table 5, the returns from both pig feedingsystems are compared with those for other small-holder activities commonly undertaken in the SolomonIslands. This table also serves to illustrate the importanceof considering the appropriate socioeconomic para-meters when evaluating production systems. On thebasis of gross margin ha-l, extensive pig keepingappears to represent a drain on the farmer's resourceswhereas a positive, though, small return to labouris observed. In most rural areas of the SolomonIslands, pressure on land is not great and it is notgenerally possible to assign costs to labour, sinceit is family or co-operative labour. In this situation,returns to labour are generally more reliable in thefinancial appraisal of a system than returns unit-area-I.

Pig raising using the semi-intensive system describedhas potential as an effective land use system forthe smallholder which is capable of adding signifi-cantly to cash income. In comparison with otherenterprises, the relative benefits of engaging in itwould appear sufficient to stimulate widespreadinvolvement. Widespread adoption of semi-intensivefeeding practices might also stimulate an increased

Trop. Agric. (Trinidad) Vol. 70 No.1 January 1993 81

Pig production in Solomon Islands: P.J. Thorne

Table 5 Comparison of gross margins and returns man-day-1 from cash-generating smallholder enterprises

Technical Co-operation Programme of the BritishGovernment's Overseas Development Administra-tion is gratefully acknowledged as is the co-opera-tion of staff of the Solomon Islands Government,Ministry of Agriculture and Lands.

~

1

References

Copra 23500 5.00Cocoa 52760 5.11Sweet potato 800.00 8.88Chillies 2132.70 3.47Peanuts 990.00 13.07String bean 670.00 3.57

Based on information from DBSI (1983); Patten and Aeming (1988);P.S. Mullard, (personal communication).It was calculated that 1 ha of land planted to cassava and coconutsin the appropriate proportions could be used to support 13 pigsunder the extensive system or 6 pigs under the semi-intensive

feeds. This might be accomplished using the existingnetwork of copra-buying points operated by theSolomon Islands Commodities Export and MarketingAuthority.

Semi-intensive feeding systems for pigs are perhapsbest seen as-part of an overall strategy for improvingthe productivity of smallholder pig production en-terprises, reducing reliance on expensive importedcompound feeds for pig feeding, and ultimatelygenerating a supply of locally produced meat thatis cheap and sufficiently reliable to substitute forimported meat and meat products.

AcknowledgementsThe author thanks Mrs P.M. Carlaw of the NaturalResources Institute for chemical analyses of theexperimental diets. The financial support of the

Agricultural Research Council (1981) The NutrientRequirements of Pigs, London, Commonwealth Ag-ricultural Bureau

Anon. (1989) Socio-economic Survey of Small-holderFarming Systems in Solomon Islands, Honiara,Solomon Islands, Ministry of Agriculture and Lands

Association of Official Analytical Chemists (1984)Official Methods of Analysis of the Association ofOfficial Analytical Chemists, Washington, AOAC

Calder, A. (1960) Sweet potatoes for pigs, RhodesianAgric. J. 57 293-298

de Fredrick, D.F. (1971) Pig Raising in the BritishSolomon Islands Protectorate, Hong Kong, CathayPress

Development Bank of the Solomon Islands (1983) Field&se Manual (2nd edition), Honiara, SolomonIslands, DBSI, Agricultural Promotions Division

Gomez, G. (1977) Life cycle swine feeding systems withcassava, in: Cassava as Animal Feed, Proceedingsof a Workshop held at the University of Guelph, Canada,19-20 April 1977, Ottawa, Canada, IDRC-095e, pp.65-71

Patten,L.H. and Aeming,EM.(1988)Farm ManagementHandbook for Solomon Islands, South Pacific Small-holder Project, University of New England

Spackman, G.W., Stein, W.H. and Moore, S. (1958)Automatic recording apparatus for use in the chroma-tography of amino acids, Analy. Chem. 30 1190-1206

Thome, P.J. (1992) Alternatives to imported compoundfeeds for growing pigs in Solomon Islands, Trop.Agric. (Trinidad), 69 141-144

Watt, I.R. (1973) Feeding sweet potatoes to pigs, Harvest2 138-141

Trop. Agric. [Trinidad) Vol. 70 No.1 January 199382