Address for Correspondence - University of Hawaii...(Consultant) UNITED KINGDOM Dr. Lalith...

1

Address for Correspondence The Managing Editor, JOSPA School of Agriculture and Food Technology USP Alafua Campus Private Mail Bag Apia SAMOA Telephone : (685) 22372/21671 Fax: (685) 22347 / 22933 Email: [email protected] or [email protected] Sales and Distribution IRETA Publications USP, Alafua Campus Private Mail Bag Apia SAMOA Telephone : (685) 22372/21671 Fax: (685) 22347 / 22933 Email: [email protected] Annual Subscription Free to agricultural workers in USP member countries. US $40.00 (including postage for non USP member countries). Paper Contribution Authors wishing to submit papers are advised to refer to the Guide for Authors on the last pages of this volume.

Layout and Cover Design Aterina Samasoni

Printed by

IRETA Printery USP Alafua Campus

Samoa

ISSN 1016-7774 Copyright © IRETA 2010

JOURNAL OF SOUTH PACIFIC AGRICULTUREJOURNAL OF SOUTH PACIFIC AGRICULTUREJOURNAL OF SOUTH PACIFIC AGRICULTURE The Journal of South Pacific Agriculture (JOSPA) is a joint publication of the Institute for Research, Extension and TrainingThe Journal of South Pacific Agriculture (JOSPA) is a joint publication of the Institute for Research, Extension and TrainingThe Journal of South Pacific Agriculture (JOSPA) is a joint publication of the Institute for Research, Extension and Training iiin n n Agriculture (IRETA) and the School of Agriculture and Food Technology (SAFT) of The University of the South Pacific (USP)Agriculture (IRETA) and the School of Agriculture and Food Technology (SAFT) of The University of the South Pacific (USP)Agriculture (IRETA) and the School of Agriculture and Food Technology (SAFT) of The University of the South Pacific (USP)

i

Managing Editor Dr. Adama Ebenebe

School of Agriculture and Food Technology Alafua Campus

Apia, Samoa

BOARD OF TECHNICAL EDITORS

Associate Professor Mareko Tofinga (Farming Systems/Agronomist) SAFT-USP SAMOA Mr. Mohammed Umar (Agriculture Project Development & Management) IRETA-USP SAMOA Mr. David Hunter (Soil Science) SAFT, USP SAMOA Dr. Joel Miles (Crop Protection) REPUBLIC OF PALAU

Professor E. Martin Aregheore (Animal Science) Marfel Consulting CANADA Professor Umezuruike Linus Opara (Postharvest Technology) Stellenbosch University SOUTH AFRICA Professor Anthony Youdeowei (Consultant) UNITED KINGDOM Dr. Lalith Kumarasingh (MAF Biosecurity) NEW ZEALAND

Referees for this volume Professor E. Martin Aregheore (Canada), Dr. Andrew Daymond (University of Reading, UK), Mr. David Hunter (University of the South Pacific,

Samoa), Dr. Sonny Lameta (University of the South Pacific, Samoa), Dr. Workneh Ayalew, (NARI, Papua New Guinea), Mr. Alan M. Gibbins (New Zealand), Professor Adeyemi Egunyomi (Nigeria), Dr. Abiodun Emmanuel Ayodele (Nigeria), Mr. Mohammed Umar (IRETA-USP, Samoa), Dr.

Michael Bourke (Australian National University, Australia), Professor J. van den Berg (North West University, Potchefstroom, South Africa)

Table of Contents In vitro micro-propagation of Plukenetia conophora Mull.Arg. Olatunde Fajimi and Taiye R. Fasola............................................................................................................1

Development of agrientrepreneurship in Samoa: opportunities and challenges Pradeep K. Singh and Jagdish P. Bhati ........................................................................................................6

Effects of mutant and normal segregant rootstocks of cocoa (Theobroma cacao L.) on growth and yield of hybrid cocoa clones in Papua New Guinea Eremas Tade , Doug A. George , Richard R. Williams, Yoel Efron and Peter Bapiwai.............................14

An ex-ante benefit-cost analysis of the impact of the research, development and extension programme that provided feeding strategies to improve the profitability of village broiler production in Papua New Guinea Ian D. Black and Millicent Yalu ..................................................................................................................23 Effect of replacing 50% of a commercial grower feed with sweet potato silage on the performance of crossbred pigs in Papua New Guinea Michael Dom and Workneh Ayalew.............................................................................................................28

Nitrogen fixation by Casuarina oligodon L. Johnson agroforestry in the Papua New Guinea highlands Johnny Minga Wemin, Rowan Reid and Robert Edis...................................................................................38 Relative impacts of some insecticides on infestation and damage by leaf-eating caterpillars and on non-target arthropod diversity in head cabbage (Brassica oleracea var. capitata) crops in Samoa Adama A. Ebenebe, Kelemeni Navucu and David J. Hunter……………...…………………......…..........53

Guide for authors...........................................................................................................................64

ii

Journal of South Pacific Agriculture, Volume 14: Nos. 1 & 2, 2010

1

In vitro micro-propagation of Plukenetia conophora Mull.Arg.

Olatunde Fajimi1 and Taiye R. Fasola2

ABSTRACT

Plukenetia conophora, an economic plant, is sourced mainly in the wild. In order to promote its rapid and mass propagation, a tissue culture technique was developed. The embryos excised were cultured on MS medium supplemented with naphthalene acetic acid (NAA), benzyl amino purine (BAP) and Kinetine (KIN) at different concentrations. The best combination for the embryo cul-ture was on MS supplemented with 0.3mg KIN and 0.01mg NAA with a shoot length of 3.10 cm, and number of node of 3. The longest root length of 5.50±0.24 was obtained on 0.3mg KIN and 0.05mg NAA. The best result of nodal culture was achieved on MS supplemented with 0.30mg BAP and 0.05 mg NAA with a mean shoot length of 2.13±0.05, root length of 2.00±0.00 and num-ber of node of 2.00±0.00. The growth of Plukenetia conophora was successful in vitro and this will enhance its mass propagation.

Key words: Plukenetia conophora, in vitro culture, embryo, nodal cutting, KIN, BAP, NAA.

1National Centre for Genetic Resources and Biotechnology, Moor plantation, Ibadan, Oyo State, Nigeria. 2Department of Botany and Microbiology, University of Ibadan, Ibadan, Oyo State, Nigeria. Corresponding author e-mail: [email protected]

INTRODUCTION

In Nigeria, deforestation is resulting in substantial loss of plant diversity of indigenous germplasm due to lack of effective management. Plukenetia conophora Mull. Arg. (Family: Euphorbiaceae), formerly known as Tetracarpidium conophorum, is a much branched perennial climber of economic importance as an edible species and used med-icinally (Dalziel, 1937). The climbing shrub found in the moist forest zones of sub-sahara Africa is commonly called the African walnut. Plukenetia conophora is sometimes cultivated principally for its nuts which are cooked and consumed as food and medicine (Oke & Fafunso, 1975). The proximate composition of freshly harvested mature nut of the plant shows that it contains protein, fiber, oil, carbo-hydrate and minerals (Enujiugha, 2003; Enujiugha & Ayodele, 2003). Burkill (1994) stated that the protein content is as high as 45%. Ogunsua & Adebona (1983) and

Nwokolo (1987) also reported the high nutrient potential of the nuts. The oil from the nuts has found other uses in the preparation of wood varnish, vulcanized oil for rubber and leather substitute (Akpuaka & Nwankwo, 2000).

In Southern Nigeria, it is used in the treatment of various diseases and thus has a high potential as an anti microbial plant. Ajaiyeoba & Fadare (2006) reported on the anti bacterial and anti fungal evaluation of the leaf, stem, kernel and root, confirming its use-fulness in folklore in the treatment of dysentery and other diseases. Due to the propagation neglect and the endangered nature of this plant, there is a need to develop a proc-ess of propagation.

Tissue culture is a tool that can be em-ployed as an alternative to the conventional methods of vegetative propagation with the objective of enhancing the rate of multiplica-tion of desired genotypes, as well as for re-aforestation and tree improvement


mailto:[email protected]

2

(Afolayan & Adebola, 2004; Murashige, 1990; Reddy et al., 2001). It was therefore the objective of this study to develop, for the first time, a reliable and prolific shoot multiplication system (protocol) for Plukenetia conophora.

MATERIALS AND METHODS

Embryos and nodal cuttings were used as the explants for the in vitro propagation. Matured fruits were collected from the field gene bank of NACGRAB, Ibadan. The husks were broken open and the seeds removed. Some seeds were planted in washed river bed sand, allowed to germinate and nodal cuttings (explants) obtained from the growing seedlings. For the embryos, the shell of the nut was removed and the cotyledon obtained.

Dis-infection of explants Nodal cuttings (Explants)

The stems of actively growing seedlings were cut with the leaf stripped off and cut into nodal segment. This was washed with liquid detergent under running tap water and disinfected with 70% ethanol for 5 minutes, and 10% sodium hypochlorite (Commercial Jik) for 20 minutes and rinsed thrice in sterile distilled water and ready for culturing.

Embryo The cotyledons obtained from the seeds

were cut into small sizes, washed with liquid soap under running tap water and then disinfected with 70% ethanol for 5 minutes and again disinfected with 10% sodium hypochlorite for 20 minutes and then rinsed three times with sterile distilled water.

Culture conditions The culture media consisted of

Murashige & Skoog (1962) MS basal medium supplemented with vitamins, myo-inositol 100mg, sucrose 30g, casein hydroxylate 200mg and growth regulators. The pH of the medium was adjusted to 5.7 before autoclaving at temperature of 121 0C, pressure of 0.15 psi for a period of 15 minutes. All the cultures were kept at 24±2 0C under cool light fluores-cent lamp for a photoperiod of 16 hours.

Shoot induction from matured embryos Matured excised embryos were cul-

tured on shoot induction medium supple-mented with different concentrations of KIN (0.0 – 0.50) mg/l with NAA /0.05 mg and BAP (0.0 – 0.50) mg/l with NAA 0.05mg. Shoot proliferation from nodal explants

Excised nodal cuttings from actively growing stem were cultured for direct organogenesis on MS medium supplemented with 0.0-0.45mg/l BAP/0.05mg NAA and 0.0- 0.45mg/l KIN/0.05mg NAA for shoot prolif-eration.

RESULTS AND DISCUSSION

The in vitro propagation of Plukenetia conophora (African walnut) regenerated after 12 weeks and the embryos have a healthy appearance after regeneration (Figure 1a-c). Successful in vitro regeneration was achieved on MS basal medium without growth regulators. Embryo of P. conophora cultured on hormone-free basal medium formed a mean shoot length of 4.97±0.03 cm, root length of 3.90±0.05 cm and two nodes (Table 1). How-ever, when basal media were fortified with growth hormone, the best mean result of shoot length of 3.10 cm, and of three nodes were re-corded on MS supplemented with 0.3mg KIN and 0.01mg NAA. The medium supplemented with 0.3mg KIN and 0.05mg NAA gives the longest root length of 5.50±0.24 cm. These concentrations induced a higher percentage of explants with shoots and shoot number per ex-plant than the hormone-free treatment. There-fore, the introduction of growth regulators led to the shoot length and number of nodes incre-ment. An increase in the concentrations of NAA from 0.01mg to 0.05 mg introduced in the media resulted in a decline in the number of shoots formed and an increase in the root length (Table 1).

In the in vitro regeneration of nodal cuttings, nodal culture on MS basal (the hormone-free treatment) showed no significant growth until supplemented with cytokinin (Figure 1d), hence the need for supplementing the MS basal medium with cytokinin. MS medium containing 0.30mg BAP and 0.05mg NAA gave the best mean result of 2.13±0.05

Fajimi & Fasola


3

cm shoot length, number of shoots 2.0, and node number of 2.0. An increase in the concentration of BAP above 0.30mg/l led to a decline in shoot length. Observations of the nodal culture showed that the nodal cutting performed best on MS medium supplemented with BAP (Table 2). The only growth

observed on KIN supplemented medium was on MS supplemented with 0.30mg and 0.05 mg NAA. There was a difference in growth of nodal cuttings on MS containing 0.2mg/l BAP and casein hydroxylate and 0.2mg/l BAP with-out casein hydroxylate as there were differ-ences in their shoot responses also.

a b

c d

Figure 1. In vitro explant regeneration of Plukenetia conophora: (a-c) from embryo, (d) from nodal culture.

Fajimi & Fasola


4

CONCLUSION

Considering the economic importance of Plukenetia conophora, this work has provided a way to explore its potentials for micro-propagation and conservation. A simple effi-cient micro-propagation protocol has been de-

veloped for embryo and nodal cutting of the plant. Enrichment of the culture medium with 0.30mg KIN + 0.01mg NAA resulted in maxi-mum embryo response while the nodal cutting responded best on benzylaminopurine (0.30mg BAP + 0.05mg NAA).

Fajimi & Fasola

REFERENCES AFOLAYAN, A. J. & ADEBOLA, P. O. 2004. In vitro propagation: A biotechnological tool capable of solving the problem of medicinal plants decimation in South Africa. African Journal of Biotechnology, 3 (12):683-687. AJAIYEOBA, E. O. & FADARE, D. A. 2006. Antimicrobial potential of extracts and fractions of the African walnut – Tetracarpidium conophorum. African Journal of Biotechnology, 5 (22):2322-2325. AKPUAKA, M. U. & NWANKWO, E. 2000. Extraction, analysis and utilization of a drying oil from Tetracarpidium conophorum. Bioresource Technology, 73:195-196. BURKILL, H. M. 1994. The useful plants of West Tropical Africa, Vol. 2. Families E-I, 2nd edi-tion. Royal Botanical Gardens, Kew, UK, 636 pp. DALZIEL, J. M. 1937. The useful plants of West Tropical Africa. Whitefriars Press, London, 612 pp.

Mean result ± standard error.

Table 2. Effect of BAP, KIN and NAA on in vitro regeneration of Plukenetia conophora nodal explants.

Media Shoot length (cm) No. of shoots No. of nodes MS only No significant growth MS + 0.20mg BAP (no casein hydroxylate) 0.57±0.03 1.00±0.00 1.00±0.00 MS + 0.20mg BAP + 0.05mg NAA 0.70±0.08 1.00±0.00 1.00±0.00 MS + 0.30mg BAP + 0.05mg NAA 2.13±0.05 2.00±0.00 2.00±0.00 MS + 0.35mg BAP + 0.05mg NAA 1.10±0.05 2.00±0.00 2.00±0.00 MS + 0.40mg BAP + 0.05mg NAA 0.83±0.07 1.50±0.29 1.67±0.27 MS + 0.45mg BAP + 0.05mg NAA 0.63±0.13 1.30±0.27 1.25±0.25 MS + 0.30mg KIN + 0.05mg NAA 0.53±0.02 1.00 ±0.00 1.00 ±0.00


Table 1. Effect of KIN, BAP and NAA on in vitro regeneration of Plukenetia conophora embryos after 12 weeks.

Media Shoot length (cm) Root length (cm) No. of nodes MS only 4.97±0.03 3.90±0.05 2.00±0.00 MS + 0.30mg KIN + 0.01mg NAA 3.10±0.00 2.00±0.24 3.00±0.00 MS + 0.30mg KIN + 0.05mg NAA 2.00±0.00 5.50±0.24 2.00±0.00 MS + 0.40mg BAP + 0.01mg NAA 2.00±0.00 5.50±0.24 2.00±0.00

Mean result ±standard error.

5

ENUJIUGHA, V. N. 2003. Chemical and functional characteristics of conophor nut. Pakistan Journal of Nutrition, 2:335-338. ENUJIUGHA, V. N. & AYODELE-ONI, O. 2003. Evaluation of nutrient and antinutrients in lesser known under-utilized oil seeds. International Journal of Food Science and Technology, 38:525-528. MURASHIGE, T. & SKOOG, F. 1962. A revised medium for rapid growth and bioassays with tobacco tissue culture. Physiology Plantarum, 15:473-497. MURASHIGE, T. 1990. Plant propagation by tissue culture: A practice with unrealized potential. In: Ammirato, P. V., Evans, D. A., Sharp, W. R. & Bajaj, Y. P. S. (Eds.), Hand book of plant cell culture. Volume 5, Ornamental Species. McGraw Hill publishing company, USA, pp. 3-9. NWOKOLO, E. 1987. Composition and available nutrient in some tropical grains and oil. Nutri-tion Reports International, 36:631-640. OGUNSUA, A. O. & ADEBONA, M. B. 1983. Chemical composition of Tetracarpidium conophorum (Conophor nut). Food Chemistry, 10:173-177. OKE, O. L. & FAFUNSO, M. A. 1975. Lesser known oilseeds: the nutritive value of conophor seeds in in vitro. Nutrition Reports International, 12:41-49. REDDY, P. S., RODRIGUES, R. & RAJASEKHARAN, R. 2001. Shoot organogenesis and mass propagation of Coleus forskohlii leaf derived callus. Plant cell, Tissue and organ culture, 66:183-188.

Fajimi & Fasola


6

Development of agrientrepreneurship in Samoa: opportunities and challenges

Pradeep K. Singh1 and Jagdish P. Bhati2

ABSTRACT

This research aims to have an insight into the state of agricultural entrepreneurship in Samoa and understand the constraints inhibiting its development. Strategies to stimulate agricultural entrepre-neurs and improve their competitiveness have also been suggested. Agriculture is crucial for Samoa’s economy where a large majority of small-holder producers practice labour-intensive mixed farming on customary land. The agribusiness sector of Samoa is tiny, fragmented and ex-periencing challenges of remoteness, high costs of labour, production and transport. Despite these constraints, the country has comparative advantage in selected high value products. To harness this advantage and enable its entrepreneurs to compete successfully, Samoa needs to promote market oriented strategies. Presently, it lacks organized domestic supply chains. It should join other coun-tries in the region in developing Pacific brand and standards to facilitate the exports of niche prod-ucts. An effective association of entrepreneurs supported by agricultural research, extension and training system would help in the adoption of modern farming practices. Targeted investment for removing the marketing constraints is also desired. These measures would reinforce the competi-tiveness of value chains of traditional and non-traditional products. Above stated social capital and networks and knowledge infrastructure would support the transition of present semi-subsistence agriculture to a more remunerative production system and thus facilitate its better integration with domestic and international markets. An advancement of incomplete reforms process, particularly the privatization of state owned bodies and market oriented strategies would create an overall environmental perspective conducive to entrepreneurship and support the growth of agricultural entrepreneurs.

Key words: customary land tenure, knowledge infrastructure, semi-subsistence agriculture, small-holders, supply chains.

1School of Agriculture and Food Technology, The University of the South Pacific, Alafua Campus, Samoa. 2School of Economics, The University of the South Pacific, Suva, Fiji. E-mail: [email protected] Corresponding author e-mail: [email protected]

INTRODUCTION

An agrientrepreneurship is any busi-ness or process of organizing, managing and assessing risks of an enterprise within agricul-tural industry. It involves production, process-ing and distribution of agricultural goods and services including food, fiber, the environ-mental and natural resources. The world of agriculture has changed dramatically since 1982 (World Bank, 2008a), as a result the role

of agricultural entrepreneurs has become more challenging. Dynamic markets, technological and institutional innovations and new roles for the state, private sector, and civil society all characterize new context for agriculture. Glob-alization of agricultural trade is forcing small producers to compete with large agricultural enterprises having better access to business services and abilities to meet quality standards of farm products. Institutions and services in developing countries, however, have not been




7

able to address these weaknesses adequately. The directives of World Trade Organization (WTO) have also resulted in reduced public support services and almost an end to the pref-erential markets for many developing coun-tries, like the small island countries of the Pa-cific region.

Most agricultural producers in develop-ing countries market their produce under con-straints of ineffective institutions without access to reliable and affordable inputs (Penrose-Buckley, 2007). Provisions of market services and value chains, infrastructural facilities inclusive of knowledge infrastructure (Schilden & Verhaar, 2000), institutional support and favourable policy environment would broaden opportunities for private entrepreneurs and improve their competive-ness. Various studies have shown that public policies play an important role in promoting entrepreneurship (Lee & Chan, 1998; Imai & Kawagoe, 2000; Tan & Yadong, 1998; UNDP, 2002). In Samoa, as also in other Pacific island countries (PICs), enough empirical evidences are not available on entrepreneurship and fac-tors influencing entrepreneurs’ behaviour. En-trepreneurial firms of Samoa, particularly the young ones, involve a lot of risks and thus are prone to failures. Objectives

This study has been undertaken with the following specific objectives: i. To highlight important features of

Samoan agricultural economy. ii. To study the challenges experienced by

agrientrepreneurs and the role of public institutions in the development of agrientrepreneurship in Samoa.

iii. To formulate a policy framework for the promotion of entrepreneurship in the agricultural sector.

The study may be helpful in formulat-ing strategies for stimulating agrientrepreneur-ship in Samoa. The findings of the study will be of interest to the policy makers, interna-tional development partners of Samoan Gov-ernment, extension agencies, NGOs and all those who are interested in the development of agrientrepreneurship in Samoa and other small island countries of the Pacific region placed in similar situations. The paper is organized in

six parts. The next part explains the key fea-tures of Samoan economy which have strong impacts on the process of its agricultural de-velopment. The third part discusses the chal-lenges being faced by the agricultural entrepre-neurs in the country. The fourth part highlights the institutional constraints to agrientrepre-neurship development. A policy framework to boost agricultural development by helping the agricultural entrepreneurs in availing the op-portunities for development is presented in part five. The final part provides conclusions and recommendations.

Important features of Samoan economy In Samoa, a small island country in the

Pacific region, a large majority of farm producers are small-holders who experience many constraints. Fortunately, opportunities in the form of new markets for its traditional ex-port commodities and new uses for the staple crops are also emerging. It also has the com-parative advantage in the production of niche organic farm products. The country has a land area of only 2,820 sq km and a sea area/EEZ of 98,500 sq km (Table 1). Among the countries of the region, Samoa has the smallest sea area. It has a population of about 182 thou-sand (2008) with a density of about 65 people per sq km. Seventy eight percent of its popula-tion is rural and earn their livelihood from farming.

In recent years the economic perform-ance of Samoan economy has been mixed with the gross domestic product (GDP) of US$ 523,383,704 (2008) and the gross national in-come (GNI) per caput of US$ 2820. However, the annual growth in GDP during 2008 was minus 3.40 percent. The value addition in the agricultural sector at under eleven percent was also low. Samoan exports as percent of its im-ports have been seven percent (2007), and the imports of food and live animals as proportion of total imports were 19 percent in 2006. Its balance of food trade has worsened in recent years and presently it meets approximately half of the food needs through imports. Samoa spends only 1.8 percent of its total budget on agriculture. Like other countries of the region, it also has the disadvantage of remoteness and distances from major markets and is vulner-able to natural disasters like earthquakes,

Singh & Bhati


8

Table 1. Salient Features of Samoan Economy.

Sources: a Report of the FAO Representative in Samoa, 2003-2008, FAO Sub Regional Office for the Pacific (SAPA), Feb.

2006. b World Bank, 2009b. World Development Indicators, 2009 database. www.worldbank.org/publications. c http://data.worldbank.org/country/samoa d Government of Samoa, Ministry of Commerce, Industry and Labour. (www.mcilsamoa.ws) e Samoa Budget Estimates, 2008/09, Government of Samoa. f Government of Samoa, Situation and Outlook for Samoa Agriculture, Forestry and Fisheries 2004, Ministry of Agri-

culture, Forestry and Fisheries.

Particulars Amount 1. Geographical Area (sq km)a Land area 2,820 Sea area (EEZ) 98,500 2. Population (2008)b Total 182,000 Annual population growth (1990-2008) (%) 0.7 Rural population (% of total) 78 3. Gross Domestic Product b

Total (US$, 2008) 523,383,704 GNI/caput (US$, 2008) 2,820 GDP growth (% annual, 2008)b -3.4 Agricultural sector, % value added (2008)c 10.8 Agricultural sector, value added as % of GDP (2007)c 3 4. Trade Balance Balance (US$)d 167,356,000 Exports as % of imports (2007) 7 Food and live animals imports as % of total imports (2006) 19 5. Balance of Samoa’s Food Trade in Recent Years (%)d 2002 -19.0 2003 -26.4 2004 -29.3 2005 -32.4 2006 -39.6 2007 -49.5 6. Budget Allocation for Agriculture, (2008-09) (US$ million)e 4.95 Share of agriculture in total budget expenditure (%) 1.8 7. Land tenure System (% of land)f

Freehold/leasehold system 2 State/public land 18 Traditional communal system 80

Singh & Bhati

cyclones and tsunamis. Samoa has a hereditary authority struc-

ture and follows “matai” system of village chief. Its rural society observes a strict village setting of reciprocal communal nature of life, with traditional village administration and so-cial governance. Rural people have strong af-filiation with land and practice a high degree of produce sharing. Their livelihood is a mix-ture of subsistence and cash income. Samoan society has about 80 percent of its land under

customary tenure (Table 1); thus gaining se-cure, individual access to such land is difficult. Sharing of wealth within community, a cul-tural practice in Samoa, at times becomes a financial burden (Rees, 2010) on the house-holds. Such cultural obligation also affects the development of enterprise. Though majority of the rural population in Samoa depends on agri-culture for its livelihood, value addition from the sector as percent of its GDP was only about three percent (World Bank, 2009a).


http://www.worldbank.org/publications

http://data.worldbank.org/country/samoa

http://www.mcilsamoa.ws

9

Such a low value addition may be due to the low productivity of communally based farm activity. Samoan industrial sector is small and the public sector still dominates the economy, even after the economic reforms of 1996-97. Since independence, the Government of Samoa has played a major role in production activities and has used trade barriers for import substitution. The private sector in the country is small but has started expanding since the initiation of broad based reforms in areas of tariffs, civil services and privatization of several state owned enterprises (SOEs). Sa-moa’s main merchandise exports are agricul-tural ((copra, coconut oil, cocoa, juice of ‘nonu’ (Morinda citrifolia), fruits, kava, etc.)), which accounts for about 16 percent of her total exports. While auto harnesses and gar-ments are its manufacturing exports, tourism and remittances are the major foreign ex-change earners. Samoa’s growth in formal employment has remained weak despite its relatively better overall economic performance compared to other small island countries in the Pacific region. Challenges faced by agri-entrepreneurs

Agriculture in Samoa comprises a sub-stantial share of its food production activities and exports and employs about two-thirds of the labour force. Farmers practice small scale, labour intensive mixed farming on customary owned lands and produce coconut products, “koko samoa” (local name for cocoa), root crops, fruits and vegetables combined with small scale livestock production. The average farm size in Samoa is about 6.07 hectares and approximately 70 percent of its farmers are small-holder cultivators who operate less than 5 hectares of land each. The agribusiness sec-tor of the country is small, fragmented and suf-fers from high production costs. Only 25 per-cent of its producers are engaged in formal markets (World Bank, 2009a). High costs of trade due to small volume and isolation from major markets are big challenges for entrepre-neurs. They also face difficulties and con-straints due to ineffective institutions and pro-ducers’ organizations; weak technology, infor-mation and management systems; direct gov-ernment intervention in agricultural trade, and recently the competition with few big compa-

nies. There are also regulatory barriers to pri-vate sector participation and investment, and lack of formal security of tenurial arrange-ments on customary lands. Low levels of agri-cultural product and market diversification, and low investment in agribusiness support systems like post-harvest technology, agro-processing and information collection and dis-semination systems also act as barriers to the development of entrepreneurship. Slow growth and progress of agricultural and rural financial markets and weak information delivery are also the weaknesses of Samoan economy. In-vestment in agricultural sector both by public and private sectors is very low. Though the government has formulated strategies for de-velopment (Samoa Development Strategy (SDS) 2003-07 and SDS 2008-12), and is also promoting market-oriented programmes, eco-nomic reforms initiated in mid-nineties have remained incomplete. There is still significant public sector involvement in production and marketing through SOEs like Agriculture Store Corporation, Samoa Trust Estates Corporation (World Bank, ibid), etc. Public-private policy dialogue and coordination with industry are important but have not yet gained momentum in Samoa.

As is shown in Table 1, Samoa is a food deficit country. Over the years its food trade balance has worsened from minus 19 percent in 2002 to over minus 49 percent in 2007. Her food security depends on the capac-ity to produce food and export goods and services to pay for food imports. For the devel-opment, dissemination and adoption of appro-priate technologies, Samoa has to develop a sound system of agricultural research, and extension. Inability of Samoan producers to respond to competition from exporters of ‘taro’ (Fiji) and coconut (Philippines) has adversely affected her exports in recent past. The country has the comparative advantage in high-value, niche organic products like virgin coconut oil, ‘noni’ juice and ‘koko samoa’ which provide good opportunities to small-scale entrepreneurs. The challenges, therefore, are to identify markets and technologies, and the application of standards and development of certified high value organic products. Hav-ing common standards and brand with other Pacific countries may help Samoa in launching

Singh & Bhati


10

its niche, differentiated products in global markets. Its entrepreneurs also have scope for the substitution of costly imported meat products.

Role of public institutions in agrientrepre-neurship development

Main markets for Samoan farm prod-ucts are: (i) international market – for off sea-son fruits, vegetables and their products to niche markets with expatriate Samoan and Polynesian communities in American Samoa, Australia and New Zealand; (ii) domestic tour-ism industry---with demand for meat, milk, fruits and vegetables from hotels and restau-rants; (iii) local markets/super markets--- need-ing low grade produce; and (iv) industries---processing feeds, soaps, etc. The main issue faced by entrepreneurs is lack of organized supply chains to get consistent supply and quality of products at competitive prices. Ex-perience with taro, noni juice and organic co-conut oil shows that small entrepreneurs are responsive to good returns (World Bank, 2009a) however, the producers’ organizations (POs) in Samoa have not been able to make desired impact. It has few organizations like the fruit and vegetable committee, farmers’ association, etc. But these POs, being unstruc-tured, of grass-root type and resource poor, have been unable in linking small entrepre-neurs to markets, and providing the certifica-tion and support. They lack effective coordina-tion and have also not shown desired results. Such organizations, therefore, are not salient to the entrepreneurs.

Land is crucial for agriculture and its access is important for the success and growth of agrientrepreneurship. Samoa follows a sys-tem of customary land tenures where there is lack of individual responsibility, in-secured tenure, low incentive for improvement, and small scale of operations (Johnson, 1990, pp.17-18). The system keeps farm productivity low and is an obstacle to commercialization.

Strategies for the development of agrientre-preneurship

Business environment in Samoa should support the transition of its agricultural entre-preneurs to a more remunerative production system. To improve the competitiveness of

agribusiness and enable the agrientrepreneurs to integrate well in domestic and international markets, value chains of niche products having comparative advantage need to be strength-ened. Effective POs may help in the diagnosis of problems along the producer-consumer chains, identification of market opportunities, and support the development of new products and markets. Public-private sector dialogue is also important for the development of entre-preneurship and Samoa needs a mechanism for effective dialogues between public-private sec-tors. Such networks would motivate entrepre-neurs to increase productivity, production and marketing of surplus produce. The country also has to ensure the quality and safety of its food. Though Samoa had initiated economic reforms in 1996-97, further structural changes are needed to help the switching of her agricul-tural sector from a traditional semi-subsistence production system to a market oriented system.

Samoa should have an agricultural plan with supportive policies and environment for modernizing the entrepreneurship. Such a plan should aim at increasing sales value and pro-ductivity. These aims may be realized by developing quality standards and encouraging the small-holder producers to market quality and differentiated products (World Bank, ibid). An improvement in their competitiveness will also increase their share in domestic and export markets. Samoa should join other coun-tries in the region in developing common stan-dards and Pacific brand to facilitate the marketing of differentiated products in regional and international markets. It should also identify products, technologies and mar-kets to support traditional exports. Agricultural plan should also encourage investment in infrastructure to reduce post-harvest losses; improve the organization and performance of value chains through better coordination of public institutions and POs; and develop labo-ratories, grading standards, market intelligence and research. Emphasis should also be on the development and dissemination of cost-effective technologies to reinforce the capacity of entrepreneurs to respond to opportunities. Development of such pathways would promote commercial farming within the traditional agriculture. Possibilities of introducing cultiva-tion of oil palm, a new commercial crop, are

Singh & Bhati


11

there in Samoa (Anonymous, 2010) and the government has to create a rural business envi-ronment conducive for investment by private companies and participation by small entrepre-neurs. Women entrepreneurs in Samoa also have opportunities but they face many chal-lenges in terms of access to information and knowledge sharing on business related proce-dures and formalities, inability to access prop-erty, credit and even justice. Such barriers in-hibit the realization of their full potential. A more active inclusion of women in public-private dialogue, development of gender-sensitive business entry handbooks and envi-ronment may be helpful (Hedditch & Manuel, 2010; Karin, 2010).

Role of land tenure in the promotion of agricultural entrepreneurship is also an impor-tant policy issue. To ease barriers to land trans-actions and markets and to facilitate for the small entrepreneurs an affordable access to land (World Bank, 2008b), Samoa, like some other PICs, should also consider introducing land reforms. Such reforms should be supple-mented by an investment in infrastructure, and training in improved farming and market intel-ligence. Samoa has to design policies and de-cision making processes to improve the gov-ernance of its agriculture. Such new forms of governance and policy framework can enhance the capacity of entrepreneurs, regulate compe-tition and build public-private partnerships.

CONCLUSIONS AND RECOMMENDA-TIONS

Entrepreneurship is being given a cen-tral role in economic development. Policy makers, development organizations and aca-demics see entrepreneurship as a key mecha-nism for job creation and growth. In develop-ing countries, many programmes have been targeted to promote entrepreneurship. Small-

holder agrientrepreneurs operate in a challeng-ing environment. Dynamic world scenario of-fers opportunities but entrepreneurs need sup-port of institutions and policies to get benefits from these opportunities. Samoa, therefore should have strategies like: prioritization of agriculture in national policies; encouragement to long term investment in basic infrastructure; training to improve farmers’ skills; develop-ment of quality standards and brands as tools for marketing of niche differentiated organic products. Samoa follows a customary system of land tenures. The system has been evolved for subsistence way of life under traditional socio-economic environment, but at times it produces many problems for agrientrepreneurs and is also an obstacle for commercialization. Easing barriers to land transactions and rental markets can improve access to land. Samoan government is avoiding any interference with the allocation, management and record keep-ing rights of customary tenures (AusAID, 2008). But, it may emulate the examples of other Pacific countries who are trying to mod-ify their land tenures systems. A fair land distribution system will improve environment for entrepreneurship. Further, an improvement in investment climate and support for the development of agrientrepreneurs and women-entrepreneurs will create employment and economic growth.

Improving the business environment for entrepreneurs requires political action, commitment, persuasion and coordination. Presently, Samoa lacks an effective mecha-nism for dialogue between public and private sectors. Consequently, the response of private sector to reforms has been poor. Costs of doing business in the country are also high. With further reforms and favourable environment the costs will decline, investment will get pro-motion and private sector will develop.

Singh & Bhati


REFERENCES ANONYMOUS 2010. Samoan Observer, 8 April, P. O. Box 1572, Apia, Samoa, p.5. AusAID 2008. Making Land Work. Vol. One: Reconciling Land and Development in the Pacific, Canberra, 3-4, pp. 113-114.

12

FAO REPRESENTATIVE IN SAMOA 2006. Report 2003-2008. FAO Sub Regional Office for the Pacific (SAPA), Apia, 1 p. GOVERNMENT OF SAMOA 2004. Situation and Outlook for Samoa Agriculture, Forestry and Fisheries, Ministry of Agriculture, Forestry and Fisheries. GOVERNMENT OF SAMOA 2008. Website Archive Articles, Ministry of Commerce, Industry and Labour. (www.mcilsamoa.ws) GOVERNMENT OF SAMOA. 2009. Budget Estimates, 2008/09, Ministry of Finance. http://www.mof.gov.ws HEDDITCH, S. & MANUAL, C. 2010. Samoa- Gender and Investment Climate Reform Assess-ment Report, IFC in partnership with AusAID, 2010 IFC, 2121 Pennsylvania Avenue, NW, Wash-ington DC. www.ifc.org http://data.worldbank.org/country/samoa

IMAI, Y. & KAWAGOE, M. 2000. Business start-ups in Japan: problems and policies. Oxford Review of Economic Policy, 16 (2):114-123. JOHNSON, D. T. 1990. The Business of Farming- A Guide to Farm Business Management in the Tropics, Macmillan Pub., London, pp. 17-18. KARIN, F. 2010. International Finance Corporation (for East Asia and Pacific), in Samoan Ob-server, 15 March, p.5. LEE, J. & CHAN, J. 1998. Chinese entrepreneurship: a study in Singapore. Journal of Manage-ment Development, 17 (2):131-141. PENROSE-BUCKLEY, C. 2007. Producer Organizations: A Guide to Developing Collective Ru-ral Enterprises, Oxfam, Great Britain. REES, P. 2010. The cultural burden of giving. Islands Business, April, pp. 42-43. SCHILDEN, M. VAN DER & VERHAAR, C. H. A. 2000. A concept of modern entrepreneurship in Dutch horticulture. In: Ogier, J. P. (Ed.), Proceedings of the XIV International Symposium on Horticultural Economics. ISHS, Leiden, pp. 439- 447. TAN, J. & LUO, Y. 1998. Competitive strategies under regulatory environment: a study of Chi-nese private entrepreneurs. International Journal of Management, 15 (2):141-151. UNITED NATIONS DEVELOPMENT PROGRAMME 2002. Human Development Report, New York, Oxford University Press, 25 pp. WORLD BANK 2008a. World Development Report On Agriculture, Overview, Washington DC: The World Bank, 9 pp. WORLD BANK 2008b. Rural Development and Agriculture in Pacific Islands, Washington DC: [Online: http://go.worldbank.org/EE551U7XW1] WORLD BANK 2009a. Samoa Agriculture Competitiveness Enhancement Project, P. I. D. Report No. AB 4890, Washington DC: [Online: http://go.worldbank.org/K2CBHVB7HO]

Singh & Bhati


http://www.mcilsamoa.ws

http://www.mof.gov.ws

http://www.ifc.org

http://data.worldbank.org/country/samoa

http://go.worldbank.org/EE551U7XW1

http://go.worldbank.org/K2CBHVB7HO

13

WORLD BANK 2009b. World Development Indicators, 2009 database. Washington DC: www.worldbank.org/publications.

Singh & Bhati


http://www.worldbank.org/publications

14

Effects of mutant and normal segregant rootstocks of cocoa (Theobroma cacao L.) on growth and yield of hybrid cocoa clones in Papua New Guinea

Eremas Tade1, Doug A. George2, Richard R. Williams2, Yoel Efron3 and Peter Bapiwai1

ABSTRACT

This study was designed to improve the understanding of the dwarf cocoa mutant and to investi-gate the feasibility of using a dwarf mutant of cocoa as a commercial rootstock for various hybrid cocoa clones in Papua New Guinea. The results indicated that the mutant rootstock significantly reduced tree vigour at least in the early years. When used as a rootstock, the mutant had a signifi-cant dwarfing effect on scion growth of cocoa clones in the nursery and persisted in the field 12 months after planting. After that, the cocoa clones grafted on mutant rootstock reverted to normal growth. The major advantages of reducing tree vigour are to improve the harvest index, ease of harvesting, pruning and overall management of the trees. This work has also demonstrated for the first time that cocoa clones grafted on mutant rootstock can produce similar yields to clones grafted on normal or commercial rootstocks. Moreover, this study has also established that cocoa clones derived from orthotropic scions can yield the same as plagiotropic scions when grafted onto either mutant or normal rootstocks.

Key words: cocoa, rootstocks, clones, mutant.

1Cocoa Coconut Institute of PNG, P.O. Box 1846, Rabaul Papua New Guinea. 2The University of Queensland, School of Land, Food and Crop Sciences, Gatton Campus, Gatton, QLD 4343, Austra-lia. 348 Deganiot St., Tivon, Israel 36054. Corresponding author e-mail: [email protected]

INTRODUCTION

Cocoa (Theobroma cacao L.) is an important cash crop in the coastal areas of Papua New Guinea (PNG). Papua New Guinea contributes about 1.1% of cocoa to the total world production with cocoa contributing about 14% of the total agricultural exports to the national economy of PNG. Cocoa is an important export revenue earner for the coun-try and ranks third as a major source of export revenue after palm oil and coffee.

In cocoa, selection of scion is com-monly practiced and usually selection for high yielding, disease and pest resistant clones are chosen. There is less interest in developing or selection of cocoa rootstocks. Studies on root-stock-scion relationship in cocoa revealed that

vigorous rootstocks increased yields and clones on dwarfing rootstock decreased yields (Murray & Cope, 1959). Yin (2004) recently confirmed that higher yields were associated with the most vigorous rootstocks; however, yield components (bean weight and number of beans per pod) and uniformity of trees were not influenced by rootstock. Rootstocks in cocoa have still been poorly studied; more understanding is required on how dwarfing rootstocks can be developed in cocoa.

In 2000, the dwarf mutant, MJ 12-226, with abnormal growth characteristics was identified at the Cocoa and Coconut Institute (CCI) of PNG. The mutant, found among progenies of the cross SCA 12 x NA 149, was characterized by short stature, small and nar-row leaves, a small root system and multiple



15

stems with strong branching habit (Efron et al., 2002, 2003a, 2003b). The parents Scavina 12 originated from Peru and Nanay 149 from Ec-uador (Pound, 1948).

Dwarfing rootstock is considered the best means to control the size of some fruit tree species. Size-controlling rootstocks are more common in apple cultivation than any other deciduous tree fruit production. M9 is the most common dwarfing rootstock used world wide due to its capability to reduce the vigour of the scion, induce precocity and pro-duce efficient cropping and large fruit size (Webster, 2001). Recent studies on M9 dwarf apple provided strong evidence that the root system has specific effect on tree size regard-less of genotype (Lauri et al., 2006). In cocoa, however, there is limited information regard-ing this subject.

The objective of this study is to investi-gate the feasibility of using a dwarf cocoa mu-tant as a commercial rootstock to control vig-our of hybrid cocoa clones, and to determine whether these lead to higher yield. Trials were established to compare the growth of the mu-tant dwarf with that of the normal cocoa, and to evaluate the effect of cocoa dwarf mutant rootstock on vigour and productivity of cocoa clones under PNG environmental conditions.


The experiments were conducted at the Cocoa and Coconut Institute of PNG, Rabaul, Papua New Guinea. The Institute is located on the coast, at the north-eastern tip of the island

of New Britain (4° 15’ S 152° 10’ E). The soils are highly fertile, young volcanic ash de-rived from basalt. In this location, the average rainfall is 2500 mm spread over 240 rain days per annum and mean maximum temperature of 32°C and minimum of 20°C with average rela-tive humidity of 90% and mean sunshine hours of 5.6 h/day.

Seed preparation Seeds were obtained from open polli-

nated MJ 12-226 mutant cocoa clones grown at CCI. Seeds were harvested and treated with fungicide (10 g of Metalaxyl, Ridomil Plus 72) mixed in 1 litre of water and then later rinsed with clean water. The seeds were then sown in black nursery bags (18 cm x 40 cm) filled with topsoil. The seedlings were raised in the nurs-ery under 25% shade cloth for four months prior to planting out in the field. Two months after sowing, the seedlings were counted when segregation into mutant and normal plants was distinguishable.

Bud grafting Grafting was done at CCI by patch

budding on 2-3 month old seedlings when mu-tant and normal segregants were clearly distin-guishable. The primary lateral branches were removed from the mutant seedlings prior to bud grafting. Budding tapes were removed two weeks after budding and sprouting was counted thereafter at two day intervals. These budded cocoa clones were then raised in the nursery under 25% shade cloth for four months before planting out in the field.

Treatments

1. Cocoa clones

Cocoa clone Pedigree Vegetative vigour of clones 17-3/1 K82 x KEE 43 Less vigorous 33-55/1 KA2-101 x KEE 12 Intermediate 37-13/1 KA2-106 x KEE 43 Highly vigorous

K82, KA2-101 and KA2-106 are female parents derived from Trinitario cocoa varieties and KEE 43, KEE 12 are male parents derived from Upper Amazonian cocoa varieties.

Tade, George, Williams, Efron & Bapiwai


16

The cocoa clones 17-3/1, 33-15/1 and 37-13/1 were used as scions to study the effect of the mutant as a dwarfing rootstock. Pla-giotropic and orthotropic scions of these cocoa clones were used in this experiment. Experimental design

The field trial was established four months after grafting in April 2003 at Block 5A, Tavilo plantation under Gliricidia sepium leguminous shade trees (at 12 m spacing). The cocoa clones were planted at a planting density of 833 trees per hectare (4 m x 3 m spacing). Planting was a split-split-plot design with three clones as main plots, two scions as sub-plots and two rootstocks as sub-sub-plots with three replications. Six plants were planted in each sub-subplot giving a total of 24 trees in a main plot. The trial was surrounded by perimeter guard rows with clones of similar vigour.

Management of field experiment Management practices corresponded to

those of commercial cocoa on CCI plantations. A formation pruning of the clones was per-formed after the first vegetative data was col-lected (six months after planting). Formation pruning involved cutting the tip of plants and removal of lower branches up to 30 cm from ground level. Maintenance prunings were con-ducted twice yearly in order to remove exces-sive vegetation on the canopy. Basal chupon removal and sanitary pruning (removal of dis-eased or dead branches) were also carried out at regular intervals. The shade tree canopy was pruned every three months to keep shade cover to a minimum (25% shade), and weed control was carried out at two month intervals.

Cocoa pod borer (CPB) pruning The incidence of cocoa pod borer

(Conopomorpha cramerella) (CPB) was alerted to CCI in April 2006. CPB is the most destructive cocoa pest in Asia. Its presence in Rabaul had become a major concern for the Cocoa Industry and was treated as an emer-gency by the government of PNG. The first stage of heavy canopy pruning due to CPB in-festation was carried out on this trial in May 2006. Prior to canopy pruning, all mature and immature cocoa pods were counted and re-corded and then removed from the tree trunk and branches. The second phase of CPB eradi-cation was conducted in September, 2006, which only involved removal of new flowers, young pods and mature pods from the cocoa trees. Yield data and vegetative data collection from this trial ceased on May 2006 and recom-menced in May 2007 to allow trees to recover from heavy canopy pruning due to CPB. This resulted in one year loss of vegetative and yield data from this experiment.

Vegetative data collection Tree height was measured with a

graduated measuring stick and stem diameter was measured at 15 cm above ground with a vernier caliper six months after field planting and continued at six month intervals for a pe-riod of 30 months. Vegetative data collection was carried out between October 2003 (6 months after field planting) and April 2006. After CPB pruning, vegetative data was col-lected only once in May 2007.

2. Scions Scion type Source of scion Plagiotropic Plagiotropic budwood Orthotropic Orthotropic budwood

3. Rootstocks Rootstock type Source of seeds Mutant segregant Open pollinated mutant MJ 12-226 Normal segregant Open pollinated mutant MJ 12-226



17

Yield data collection Pod harvesting began in March, 2005

and was carried out monthly until April, 2006 before CPB infestation and then recommenced in May 2007 after heavy pruning due to CPB, with collections every four weeks until Janu-ary 2008. At each harvest, the total number of ripe healthy pods was recorded. Fresh bean weight was also recorded for each plot. Con-version ratios of weight of wet beans to dry bean yields were obtained using Lockwood & Edwards’ (1980) method.

Statistical analysis Data were analysed with GENSTAT

Release 7.2 – Discovery Edition 3 statistical analysis software (Lawes Agricultural Trust, Rothamsted Experimental Station, UK). Vege-tative and yield data were subjected to ANOVA using split-split-plot and split-plot design, respectively, and means were separated by LSD (P = 0.05).

RESULTS

Effect of Rootstock The mutant rootstock significantly (P <

0.001) decreased stem diameter of seedlings at six months, but thereafter there was no signifi-cant difference (Table 1). It also significantly (P < 0.001) reduced plant height at six months and also at 12 months (P < 0.05) after field planting (Table 2), but was not significant after 12 months. The statistical analysis showed that the effect of rootstock on cumulative yield was not significant (Table 3). This study gives some clear indications that in cocoa the root-stock influences the vegetative growth of scion at least in the early years (Figures 1 and 2). The most apparent effect in this study was the reduction of plant height on the mutant root-stock by about 23% and 26% at 6 months and 12 months, respectively, after field planting.

Table 2. Effects of rootstock on plant height (cm) over 30 months after field planting.

6 58.5a 75.4b 66.9 5.8 12 96.8b 130.4b 113.9 10.1 18 183.2 186.6 184.9 NS 24 214.2 216.8 215.5 NS

Means followed by the same letters are not significantly different; NS indicates no significance.

30 279.1 282.5 280.8 NS

Months Mutant Normal Mean LSD<0.05

Rootstock



Table 1. Effects of rootstock on stem diameter (cm) over 30 months after field planting.

Months Mutant Normal Mean LSD<0.05

12 19.0 21.2 20.1 NS 18 36.3 38.9 37.6 NS 24 56.3 56.9 56.6 NS 30 77.0 77.2 77.1 NS

Means followed by the same letters are not significantly different; NS indicates no significance.

6 8.7a 10.9b 9.8 0.9

Rootstock

18

Table 3. Effects of the mutant and normal segregant rootstocks on cumulative (2 years) yield of cocoa.

Mutant 1,840 Normal 1,800 Mean 1,825

NS indicates no significance. LSD<0.05 NS

Rootstock Dry bean (kg/ha)

Figure 1. Growth characteristics of orthotropic scion (left) and plagiotropic scion (right) of the hybrid cocoa clone 17-3/1 grafted on mutant rootstock 30 months after field planting.

Mutant Rootstock Normal Rootstock

Figure 2. Cocoa variety 73-14/1 bud grafted onto mutant (left) and normal (right) rootstocks at 12 months after field planting in the planting density trial.



19

Effect of scion There was no apparent significant ef-

fect of plagiotropic scion and orthotropic scion on vegetative growth (stem diameter and plant height) over 30 months after field planting (Table 4). The results also show that there was no significant effect of plagiotropic and orthotropic scions on cumulative yield of co-coa across both rootstock types and cocoa va-rieties (Table 4). Scion effect on vegetative growth and cumulative yield were not statisti-cally significant. It was expected that orthotropic scion with more branches would be more vigorous and therefore yield higher than the plagiotropic scion. The results show that this was not the case, probably because the plants were still at the early stage of growth. The continuity of this experiment for a longer series of years until the mature tree yield pe-riod is reached (from five to eight years) would allow the confirmation of such results.

Effect of clones

The three clones tested did not show significant differences in plant growth over the two and half year period (Table 5).

The cocoa clone 17-3/1 tended to produce higher cumulative yields than 37-13/1 and 33-15/1 and was statistically different (P < 0.05)

across different rootstocks and scion types. The cocoa clone 17-3/1 tended to confirm its precocity and superior yielding ability com-pared with clones 37-13/1 and 33-15/1, and clone 37-13/1 usually has a more vigorous vegetative growth than the other two varieties (Efron, 2002). Its vigorous nature was not ex-pressed probably because the plants were still at their immature stage.

DISCUSSION

The mutant rootstock reduced plant height only up to 12 months after field plant-ing and then reverted to normal growth there-after. In contrast, apple rootstocks are clones (i.e. one genotype) and usually produced as layers. It is not the case for cocoa trees in this study, where the rootstocks are not one clone, but mutant open-pollinated progenies of a dwarf clone MJ 12-226. This means that all rootstocks used are in fact different genotypes (even if they display the same dwarf pheno-type). The variability of rootstock might thus explain the results obtained in this study re-garding the feasibility of using a dwarf mutant of cocoa as a commercial rootstock for various hybrid cocoa clones in PNG.

The mutant rootstock effect of cocoa observed in this trial supports results observed in other fruit tree species. In citrus, a similar effect was observed on sweet orange scion grafted on South African Dwarf rootstock and was found to be unsuitable for dwarfing com-binations since the trees were similar to or lar-ger than the standard rootstock (Phillips & Castle, 1977). Similarly, there were varying effects of some dwarfing apple rootstocks on the vigour of apple cultivars. Trees of ‘Gala’ apple on M.27 EMLA, B.491, P.16 and P.22 produced the smallest trees whereas V.1, M.26 EMLA, RN29 and Pajam 2 produced the larg-est trees (Hirst, 2001). The growth vigour of apple cultivars ‘Golden Delicious’, ‘Gloster’ and ‘Melrose’ was highest on Pajam 2 with dwarfing rootstocks M9, M27, Jork 9, and Pajam 1 (Kosina, 2002). Szczygiel & Czync-zyk (2002) reported that the effect of dwarfing rootstock on tree vigour varied with apple cul-tivars in relation to rootstocks and soil condi-tions.

In contrast, when the mutant scion was

Table 5. Effects of three cocoa clones on cumulative (2 years) yield of cocoa.

17-3/1 2,149a 33-15/1 1,613b 37-13/1 1,712ab Mean 1,825

Means followed by the same letters are not signifi-LSD<0.05 530

Cocoa clone Dry bean (kg/ha)



Table 4. Effects of the plagiotropic and orthotropic scions on cumulative (2 years) yield of cocoa.

Plagiotropic 1,882 Orthotropic 1,767 Mean 1,825 LSD<0.05 NS

NS indicates no significance.

Scion Dry bean (kg/ha)

20

grafted onto normal rootstock it did not induce normal growth of the mutant scion, and also mutant scion on mutant rootstock remained mutant (Efron, unpublished). There was no change over time. The grafted seedlings showed dwarfing characteristics with high branching habit and small narrow leaves. This suggests that the reduced growth in the dwarf genotype may not be due to reduced produc-tion of endogenous gibberellic acid (GA) since hormonal substances usually move through grafting. It also suggests that regulation is oc-curring through the shoots and not the roots in contrast to normal scion grafted onto mutant rootstock. Phinney & Katsumi (1967) found that approach grafting between d1 dwarf mu-tant and normal plant of maize induced growth response to the dwarf plant. McComb & McComb (1970) also reported that a solution of GA3 poured around the roots of tall root-stocks carrying dwarf scions of pea brought about a marked increase in the rate of shoot growth demonstrating the passage of GA across the graft union.

On the other hand, grafting studies us-ing rms mutants of pea (Beveridge et al., 1994, 1997) were the first to show that wild-type (normal) rootstocks can completely restore the branching phenotype of mutant shoots to that of the wild type. Similarly, the mutant petunia (dad-1) grafted onto wild type rootstock was converted to a phenotype resembling the wild type (Napoli, 1996). If the mutant genotype was caused by the interruption of GA biosyn-thesis, the grafting of mutant scion onto nor-mal rootstock would induce a growth response of the mutant scion. The results of the present study found that grafting normal scions onto mutant rootstocks reduced the growth of scion at least in the first 12 months but thereafter the scion reverted to normal growth.

No significant interaction effect was observed between rootstock x scion, rootstock x clone, scion x clone, and rootstock x scion x clone on vegetative growth over 30 months after field planting. Similarly, the interaction effects on cumulative yield were also not sta-tistically significant, probably due to the age of the plants. There was no significant effect of rootstock, the interaction of rootstock with scion and clone on vegetative growth and the cumulative yield of cocoa probably because

the trees were still in the immature stage. Con-versely, in apples, results from eight years of data showed significant interaction between the five rootstocks and three cultivars (scions) in tree vigour in a study by Szczygiel & Czynczyk (2002). Studies of rootstock-cultivar interactions in apples have also shown that the rootstock controls total growth, while scion controls distribution of growth, such as short vs. long shoots (Feree et al., 2001a, b).

CONCLUSIONS

Whilst the prime objective of dwarfing mutant rootstock effect in field experiments was not achieved, the work has clearly demon-strated that rootstock in cocoa can influence vegetative growth of scions at least in the early years of plant growth. This was apparent when normal scion was bud grafted on to the mutant rootstock where it has an effect on the scion even though it was temporary. This result sug-gests that the accumulation of endogenous growth substances, especially auxin and cyto-kinin ratio, and cambial activity between the scion (normal) and the rootstock (mutant) may overcome the dwarfing effect of the mutant genotype after 12 months in the field. These findings further indicate that the dwarfing of the mutant cocoa is regulated from the shoots and not the root system. This would explain why the mutant scion grafted on normal root-stock remained mutant and was not restored to normal growth habit.

This work has also demonstrated for the first time that cocoa clones grafted on mu-tant rootstock can produce similar yields as clones grafted on normal or commercial root-stocks. However, further studies on mutant rootstocks are required before it is considered for use in cocoa farming in the future. More-over, this study has also established that cocoa clones derived from orthotropic scions can yield the same as plagiotropic scions when grafted onto either mutant or normal root-stocks. The current bud grafting practice in cocoa farming is the use of plagiotropic scions, however, based on results of the present study it is proposed that orthotropic scions should also be incorporated into cocoa farming in the future. Orthotropic growth is more familiar to most cocoa growers in PNG than plagiotropic



21

growth. In order to develop the mutant root-

stock commercially, rootstock development for cocoa will require research commitment inputs similar to cocoa breeding. The abundance of literature devoted to rootstocks for other fruit trees contains a plethora of information that will assist the formation of a cocoa rootstock programme. In particular, apple and orange growers of the world have moved headlong into intensively managed orchards of small trees during the late 20th century. In some re-spects, ideas and practices developed for dwarf varieties of these fruit trees can be applied in-directly to the development of cocoa trees of reduced vigour. In particular, the importance of integrating appropriate rootstocks to be inte-grated into management system for cocoa

farming cannot be underestimated. There are three major areas that require immediate atten-tion for the success of long term breeding pro-gramme for cocoa rootstock; (1) search for rootstocks that reduce the vigour of cocoa; (2) rootstocks that impart precocious flower bud induction; (3) rootstocks that control vigour and retain adequate productivity.

ACKNOWLEDGEMENTS

This work is part of a PhD thesis sub-mitted by the first author to The University of Queensland, Australia. We thank the Austra-lian Government, through AusAid for funding this work. This paper is published with the per-mission of the Chief Executive Officer, Cocoa Coconut Institute of PNG.

REFERENCES BEVERIDGE, C. A., ROSS, J. J. & MURFET, I. C. 1994. The rms-1 mutant of pea has elevated indole-3-acetic acid levels and reduced root-sap zeatin riboside content but increased branching controlled by graft-transmissible signal(s). Plant Physiology, 104:953-959. BEVERIDGE, C. A., SYMONNS, G. M., MURFET, I. C., ROSS, J. J. & RAMEAU, C. 1997. Branching mutant rms-2 in Pisum sativum. Plant Physiology, 115:1251-1258. EFRON, Y., NIDESON, D., EPAINA, P. & FAURE, M. 2002. A cacao growth mutant with dwarfing effect as rootstock. INGENIC Newsletter, 7:16-19. EFRON, Y., TADE, E. & EPAINA, P. 2003a. Accelerated rate of bud sprouting on the cacao growth mutant rootstock. INGENIC Newsletter, 8:8-12. EFRON, Y., NIDESON, D., EPAINA, P. & TADE, E. 2003b. Inheritance of a cacao growth mu-tant, MJ12-226: A possible interaction between nuclear and cytoplasmic genes. INGENIC News-letter, 8:12-15. FERREE, D., ERB, W. A. & MORRISON, F. D. 2001a. Influence of four apple cultivars on five dwarfing rootstocks on morphology of two year-old limb sections. Fruit Varieties Journal, 53:159-165. FERREE, D., ERB, W. A. & MORRISON, F. D. 2001b. Influence of four apple cultivars on five dwarfing rootstocks on spur quality. Fruit Varieties Journal, 53:166-172. HIRST, P. M. 2001. Early performance of Gala on 18 dwarf and 4 semi-dwarf rootstocks growing at 24 sites in North America. Acta Horticulturae, 557:199-205. KOSINA, J. 2002. Evaluation of some dwarf apple rootstocks. Horticultural Science (Prague), 29 (1):23-25.



22

LAURI, P., MAGUYLO, K. & TROTTIER, C. 2006. Architecture and size relations: an essay on the apple (Malus X domestica, Rosaceae) tree. American Journal of Botany, 93 (3):357-368. LOCKWOOD, G. & EDWARDS, D. F. 1980. Determination of pod and bean characters in prog-eny trials with cocoa. Tropical Agriculture, 57:289-300. MCCOMB, A. J. & MCCOMB, J. A. 1970. Growth substances and the relation between pheno-type and genotype in Pisum sativum. Planta, 91:235-245. MURRAY, D. B. & COPE, F. W. 1959. A stock-scion experiment with cacao, III. The Regional Research Centre of th British Caribbean at the Imperial College of Tropical Agriculture Trinidad, W.I. Annual Report on Cacao Research, 1957-58, 29-35. NAPOLI, C. 1996. Highly branched phenotype of Petunia dad 1-1 mutant is reversed by grafting. Plant Physiology, 111:27-37. PHINNEY, B. O. & KATSUMI, M. 1967. Genetic control of gibberellin production. Chemical Regulation in Plants, 2:79-83. PHILLIPS, R. L. & CASTLE, W. S. 1977. Evaluation of twelve rootstocks for dwarfing citrus. Journal of the American Society for Horticultural Science, 102:526-528. POUND, F. J. 1948. Studies on the fruitfulness in cacao.II. First Annual Report on Cacao Re-search (Trinidad):26-28. SZCZYGIEL, A. & CZYNCZYK, A. 2002. Suitability of some semidwarf and dwarf rootstocks to three apple cultivars in the SubCarpathian region. Journal of Fruit and Ornamental Plant Re-search, 10:85-93. WEBSTER, A. D. 2001. Rootstocks for temperate fruit crops: current uses, future potential and alternative strategies. Acta Horticulturae, 557:25-33. YIN, J. P. T. 2004. Rootstock effects on cocoa in Sabah, Malaysia. Experimental Agriculture, 40:445-452.



23

An ex-ante benefit-cost analysis of the impact of the research, development and extension programme that provided feeding strategies to improve the profitability of village broiler pro-duction in Papua New Guinea Ian D. Black1 and Millicent Yalu2

ABSTRACT

An ex-ante benefit-cost analysis was carried out. It assessed the economic impact of the research, development and extension programme that provided feeding strategies to improve the profitabil-ity of village broiler production in Papua New Guinea. The research and development aspects of the programme have developed a mini-mill approach for the provision to villagers of a high pro-tein concentrate based on local feed sources, to be combined with a local carbohydrate source such as sweet potato in the highlands and cassava in the lowlands. Other options used in the analysis include a commercial feed high protein concentrate plus local carbohydrate source and a dilution strategy of half normal commercial feed plus local carbohydrate source. Our estimate of the pro-gramme costs was 3.3 M kina in mid financial year 2008/2009 values. On our core assumption of 35% market penetration of these new technologies, the analysis showed an internal rate of return to the programme of 25% p.a., a net present value of 45.6 M kina and a benefit-cost ratio of 15:1. The benefits were attributed to a producer surplus of 16.3 M kina and a consumer surplus of 32.7 M kina. The analysis was robust in that the programme was of good value in terms of a low-end market penetration of 20% and a break-even market penetration of <1 %.

Key words: village, broilers, benefit-cost analysis, local feed sources.

1South Australian Research and Development Institute, GPO Box 397, Adelaide, SA, 5001. Present address: 32 Devonshire St, Walkerville, SA 5081, Australia. Ph. 61 8 8269 4120. [email protected] 2National Agricultural Research Institute, PO Box 4415, Lae 411, Morobe Province, Papua New Guinea. Corresponding author email: [email protected]

INTRODUCTION

As a general statement, it is probably true to say that much of the Papua New Guinea (PNG) human population has low food protein intake. Over the past 30 years, a village broiler industry has evolved that partially addresses the issue of this low protein consumption. The basis of the industry is the purchase of a batch of 52 day-old meat chicks together with com-mercial feed based on imported grain. In the past 10 years, growth of the industry is per-ceived to have slowed or even plateaued be-cause of the depreciation of the kina and more

recently the increased world price of grain, thus making commercial feed considerably more expensive.

The research, development and exten-sion (R, D & E) programme objective was to use local feed resources - fish, copra and palm kernel meals - to produce high protein concen-trate mixes using mini-mills. Mini-mills con-sist of relatively cheap equipment that com-bined together, and in conjunction with the basic feed ingredients, result in an animal feed. The concentrate was then used on a 50:50 ba-sis with a local carbohydrate source such as sweet potato in the highlands and cassava in



24

the lowlands. The programme was funded by the Australian Centre for International Agri-cultural Research (ACIAR), together with the partner PNG organisations3, in order to assist in reducing the feed costs of village broiler production.

An initial finding was that a village broiler industry was not well established on outlying islands. Day-old chicks have to be flown in from Lae at considerable expense and high chick mortality (Quartermain, Moxon, pers. comm.).

The economic analyses associated with the R, D & E programme had three outcomes: 1) Comparative costing of locally milled

(“mini mill”) high protein concentrate based on local feedstocks compared to the price of the commercially available alternative in the highlands and lowlands of PNG.

2) Comparative village operation feeding costs (feed plus labour) of rations based on:

(a) a mini mill concentrate plus local carbohydrate source - sweet potato (highlands) or cassava (lowlands), (b) a commercial feed high protein

concentrate plus local carbohy-drate source, and

(c) a dilution strategy of half nor mal commercial feed plus local carbohydrate source, compared to commercial feed.

3) An ex-ante benefit-cost analysis of the programme.

This paper presents results of the ex-ante benefit-cost analysis associated with the programme (3). The analysis draws on (1) and (2), above, the results of which are summa-rised in Table 1.

METHODS

(1), (2) The costs of feeding broilers in PNG village operations: a comparison of different feed sources The price of mini-mill produced high protein

concentrate4 This was derived from first principles,

given that such operations barely exist. The three steps involved (detailed in Black & Yalu, 2010) were: • Fixed asset annual overhead cost: land;

fence; sheds; compacted entry and hard-standing; mini-mill – hammer mill (including electric or diesel motor), dry-ing/roasting cauldron, mixer (including motor), pelleter (including motor), heat-ing fire box using local fuels, scales, power supply and connection to machin-ery, cutting and handling tools, storage and materials movement equipment; of-fice equipment, storage containers, sec-ond hand vehicle. These assets were am-ortised over a number of years depend-ing on type.

• Annual running costs: fuel, municipal rates, power, phone, security, office sup-plies, pest control, skilled manual la-bourer, repairs and maintenance.

• Raw material feed costs: high protein – fishmeal, copra meal, palm kernel meal; high energy – coconut oil; low cost by-product – mill run (bran/pollard); premix (additional vitamins and essential amino acids).

Commercial feed and commercial concentrate Commercial feed prices are from ex-

ample retail outlets in the highlands and low-lands. It should be noted that the cost of a commercial concentrate (Table 1) is very ap-proximate (commercial concentrate is not yet produced, so no direct valuation was avail-able). The commercial concentrate is based on imported grain legumes, whose prices are gen-erally 50% or more than feed wheat prices. In addition, there is an assumed higher cost for the smaller mill batch runs in commercial mills for commercial concentrate compared with the normal commercial feed.

3Research and Development: National Agricultural Research Institute; Economic services & student education: Unitech; Extension: Lutheran Development Service, Christian Leaders Trainers College, Ok Tedi Development Foun-dation, Salvation Army. 4All costing shown in this paper are based on the value of the kina in approximately mid 2008/09 financial year.

Black & Yalu


25

Black & Yalu


PNG broiler feeding system Kina % Lowlands mini mill HPHEC + cassava 276 64 Lowlands commercial concentrate + cassava 318 74 Lowlands diluted commercial feed 334 77 Lowlands commercial feed 431 100 Highlands mini mill HPLEC + sweet potato 311 65 Highlands commercial concentrate + sweet potato 350 73 Highlands diluted commercial feed 359 75 Highlands commercial feed 481 100 HPHEC = High protein, high energy concentrate; HPLEC = High protein, low energy concentrate.

Table 1. Village feeding costs per batch of 52 chickens and cost comparison.

(3) An ex-ante benefit-cost analysis of the economic impact of the R, D & E pro-gramme

Analysis was carried out using the DREAM programme V3.0.0 (Wood et al., 2001). The programme was run in the “closed economy” mode, as all village grown broilers are con-sumed within village or town communities, often through sale in local markets. Model Assumptions • Approximately 6 million broilers are

produced in village production in PNG (Quartermain, 2000; Glatz, 2006). There are no later estimates than these to our knowledge, so we have used this esti-mate as the basis for the analysis. A con-servative sale price of 20 kina/bird was also chosen. Current indications are that the price is now a little higher than this, depending on location.

• Market penetration: Although there is considerable interest at village demon-stration trial levels in this programme (Glatz et al., 2010; P. Glatz, pers. comm.), we have chosen a market pene-tration of 35% for these new technolo-gies, resulting from the R, D & E pro-gramme. We assume a sigmoid adoption rate over 9 years commencing one year after cessation of the extension of the R&D by the extension NGOs involved. We suggest these figures are conserva-tive.

• Sensitivity Analysis: We used assump-tions of 20% and 50% market penetra-tion.

• Break-even analysis: We tested the con-cept of absolute minimum market pene-

tration to check that the programme was worthwhile in financial terms.

• Costs: R, D & E - Costs are in PNG kina at year 2008 values. They include the full ACIAR costs as well as the full fi-nancial and in-kind costs of the PNG contributor organisations. Adoption costs – these are detailed in Black & Yalu (2010). The carbohydrate sources (cassava in the lowlands and sweet po-tato in the highlands) were costed at market rates (Fresh Produce Develop-ment Agency 2006-2008).

• Benefits: Table 1 forms the basis of the benefits, together with the estimated 6 M p.a. village broiler production and a sale price of 20 kina/bird. Table 1 and the associated discussion indicate that we have little prior knowledge of which of the new technologies is likely to pre-dominate. Although the mini-mill tech-nology is notably cheaper than the other technologies, it is by no means certain that the PNG mainland will be covered by a viable network of mini-mills pro-ducing poultry feed. Hence we used the mean of a 29% reduction in feeding costs of village broilers from the three new feeding regimes.

Table 1 shows a comparison of feeding costs per chicken batch. Apart from differ-ences in village labour costs between full com-mercial feed and the concentrate or dilution strategy (cooking the carbohydrate source re-quires extra labour), noted above, the differ-ences between highlands and lowlands costs are due to differences in the market price of cassava in the lowlands (60 toia/kg) and sweet potato in the highlands (70 toia/kg) as well as

26

A break-even analysis showed that a market penetration of <1% would allow the cost of the R, D & E programme to be covered by the value of the benefits.

DISCUSSION

Table 1 shows that all diets where a local carbohydrate source is substituted for all or some commercial feed has a distinct cost advantage for village broiler operations. In particular, the combination of a mini mill con-centrate based on local ingredients plus a local carbohydrate source appears very attractive.

It needs noting that while the cost ad-vantages for feeding regimes utilising local feed sources look considerable, it does not

necessarily imply complete or near complete adoption of the new technologies. Sweet po-tato and cassava are grown using hand labour and little or no mechanisation in a de-enervating climate where malaria, particularly in the lowlands, and other diseases are en-demic. Therefore the marginal social cost of growing sweet potato or cassava may be sig-nificantly higher than the market cost used here. In contrast, commercial feed can be used in village operations with little associated la-bour. These social considerations contra-indicate rapid and complete adoption of the new technologies.

Clearly the conservative core estimates of market penetration, as well as the break-even analysis, indicate that the R, D & E pro-

Black & Yalu


differences in the bulk ingredient transport costs, as reflected in retail prices at point of sale to villagers. A transport cost of 250 kina/t has been used for the highlands and 50 kina/t in the lowlands, reflecting current information to hand. Table 1 shows the relative costs, using commercial feed as the denominator. • A discount rate of 5% p.a. was used. • Following Alston et al. (1995, pp. 20-

322), we used a price/production elastic-ity of 1.0 and a price/consumption elas-ticity of 0.5 in the DREAM programme.

• Given this stimulus of reduced feed costs to the market for village broilers, we have assumed that the market will grow by 1% p.a., also taking into considera-

tion the convenience and changing tastes towards poultry meat within PNG village communities.

• A benefit stream of 25 years was used, with no dis-adoption of the technologies, except in favour of possible superior technologies in the future. That is, newer technologies will improve on these tech-nologies under consideration, in terms of further productivity gains.

RESULTS

Table 2 shows the results from the core assumptions and sensitivity analyses.

Table 2. Benefit-cost analysis results.

Consumer Internal rate of Benefit-cost Net present Producer benefits benefits return (%) ratio (:1) value (M kina) (M kina) (M kina) Core assumption: 35% market penetration 25 15 45.596 16.282 32.685

Sensitivity analysis: 20% market penetration 20 9 24.488 9.246 18.492

Sensitivity analysis: 50% market penetration 27 22 66.965 23.405 46.810 Programme costs: These were estimated at 3.251 M kina.

27

grammes have been well worthwhile. It re-mains to be seen which technology makes the greatest contribution. While our calculations indicate that the mini-mill technology is supe-rior, as stated above, it is not clear that the PNG mainland will be adequately covered with a set of mills to meet village require-ments. A range of small business skills is re-quired to run such operations. In addition, con-tinuity of local high protein supplies such as copra and palm kernel meal, particularly in the highlands, remains problematic. Also, as stated above, the convenience of purchase of fully formulated rations mitigates against a rapid and complete adoption of the new tech-nologies.

W. Ayalew (pers. comm.) has stated that these technologies are already available in many third world countries. However, this work offers a template for similar sorts of adoptions in other Pacific Island nations. In

addition, it offers a template for other avian and porcine feeding systems in the Pacific. Of interest, also in this context, is the suggestion that for localities where high protein copra meal or palm kernel meal is readily available but local sources of carbohydrate are relatively expensive, it may be feasible to transport high energy concentrate to villages as a dilution for these meals (Glatz et al., 2010; P. Glatz, pers. comm.).

ACKNOWLEDGEMENTS

We thank ACIAR and NARI for finan-cial support. Thanks also to D. Maip, along with his supervisors P. Manus and W. Kerua, at Unitech, Lae, for provision of some initial feed sources market data and analysis. Also thanks to G. Denn for providing the costing of mini mills.

REFERENCES ALSTON, J. M., NORTON, G. W. & PARDEY, P. G. 1995. Science under Scarcity: Principles and Practice for Agricultural Research and Priority Setting. Cornell University Press, Ithaca & London. BLACK, I. D. & YALU, M. 2010. Economic analyses associated with the ACIAR project LPS/2005/094 "Improving the profitability of village broiler production in PNG" and AS2 2001 077 “Poultry feeding systems in PNG”: 1. The costs of feeding broilers in PNG village operations: a comparison of different feed sources. 2. An ex-ante benefit-cost analysis of the economic impact of the two programmes. Report to P. Glatz for inclusion in the final report to ACIAR on project LPS/2005/094. Available from the senior author on request. FRESH PRODUCE DEVELOPMENT AGENCY, 2006-2008. Spreadsheet aggregated monthly data for major centres in PNG. Databases available from the senior author on request. GLATZ, P. C. 2006. Poultry feeding systems in PNG. Final Report to The Australian Centre for International Agricultural Research. http://www.aciar.gov.au/project/LPS/2001/077. GLATZ, P. C., BLACK, I. D., AYALEW, W., PANDI, J. K., HUGHES, R. J., MIAO, Z. H., WA-HANUI, J., JANSEN, T., MANU, V. & RODDA, B. K. 2010. Opportunities and sustainability of smallholder poultry production in the South Pacific Region. Australian Poultry Science Sympo-sium, 21:145-152. QUARTERMAIN, A. R. 2000. Non-commercial Poultry Production in Papua New Guinea. Asian-Australasian Journal of Animal Science, 13, Supplement July 2000, C: 304-307. WOOD, S., YOU, L. & BAITX, W. 2001. DREAM Version 3. IFFPRI, Washington, DC.

Black & Yalu


http://www.aciar.gov.au/project/LPS/2001/077

28

Effect of replacing 50% of a commercial grower feed with sweet potato silage on the performance of crossbred pigs in Papua New Guinea

Michael Dom1 and Workneh Ayalew2

1&2National Agricultural Research Institute, Livestock Station, Labu, PO Box 1639, LAE, Morobe Province 411, Papua New Guinea.

Corresponding author email: [email protected]

ABSTRACT

Sweet potato (Ipomoea batatas L. Lam.) tuber is fed to growing and fattening pigs either raw or cooked, together with the foliage, usually as one large daily offer, a feeding regimen predominant in Papua New Guinea where the crop is a staple food. Sweet potato tuber is low (6%) in crude pro-tein content, and highly digestible regardless of the mode of preparation. The foliage is much higher in crude protein content but has low dry matter intake. Obtaining a suitable protein source in growing pig rations continues to be a hindrance to the productivity of smallholder commercial farmers. In this evaluation, it was observed that sweet potato silage (made from tuber and vine on a 1:1 ratio), fed ad libitum, can substitute 50% of a commercial feed (16% crude protein) while maintaining good daily weight gain, efficient feed conversion and acceptable back-fat thickness but lower dry matter intake. An overall improved digestibility of nutrients in sweet potato silage is advanced as the likely explanation; however, other factors such as the energy density and genetics warrant some attention. The returns from carcass sale were noted and point to an economic incen-tive to using ensiling technology, the benefits of which include long term storage and less use of fuel wood.

Key words: sweet potato, ensiling, silage, growing pigs, Papua New Guinea.

INTRODUCTION

In many tropical countries, sweet po-tato (Ipomoea batatas L. Lam.) tuber is fed to growing and fattening pigs either raw or cooked, together with the foliage, usually as one large daily offer. This pig feeding regimen is common in many Asian and South East Asian rural farming communities where sweet potato is sometimes the only feed available, either seasonally or on a daily basis; other feed crops are not grown and commercial or agri-cultural by-product feed ingredients are unaf-fordable or difficult to obtain. The feeding regimen is also predominant in Papua New Guinea (PNG) where sweet potato (SP) is a staple food crop to 60% of the rural population (Hanson et al., 2001); and consequently a ma-

jor source of income for many farmers in rural villages and peri-urban settlements producing for domestic consumption. Apart from the noteworthy socio-cultural value of pigs in PNG, they are an important income earner as pork meat is in high demand in much of the country.

In PNG, SP is planted all year round, having a growing season of four to six months or longer depending on altitude. In the major SP producing highland provinces alone where competition between humans and pigs for the crop is greater than in other provinces, SP pro-duction was estimated at some 2.8 million tones per annum (Bourke & Vlassak, 2004). SP tuber for pigs may be 50% of fresh total production since only half the amount was determined to go to human consumption



29

(Hanson et al., 2001) and there is no other demand for SP tuber or vine. With such de-pendence on this crop, making better use of available SP tuber and vine forage in post-harvest processing and storage is vital, to lessen the competition for it as food/feed resource and rationalize use of labour for con-tinuous production.

One such post-harvest process is ensil-ing. Ensiling techniques are documented and used on a large scale in many developed coun-tries. Research and development interests in ensiling technology are on the rise in PNG. The key benefits of making SP silage are pro-longed storage of perishable forage and its simplicity for on-farm practice. It requires less continuous labour in its preparation as pig feed, and reduces using other resources such as fuel wood for cooking.

The inadequacy of protein sources in growing pig rations continues to be a hindrance to the productivity of smallholder and village farmers who do not have the re-sources or are unable to obtain commercial feeds regularly. SP is a very suitable energy source for growing and fattening pigs although it has relatively low crude protein (CP) con-tent. Fresh SP tubers contain 29% dry matter (DM), gross energy (GE) of 17 MJ/kg DM, and 6% CP, while SP vine contains 15% DM, gross energy of 14MJ/kg DM and 18% to 20% CP content (Godoy & Elliot, 1981; Noblet et al., 1990; Dominguez, 1990; in Pérez, 1997; An, 2004).

The digestibility of SP in raw, cooked and ensiled rations by pigs summarized by Perez (1997) show that except for CP that was lower for silages, DM, OM, GE, ME, and CP values were similar. However, Dom & Ayalew (2009a) offered SP silage to growing pigs, in a mixture of 60% tuber and 40% vine, together with a commercial pig grower ration at 1:1 ra-tio by weight and found that the effective di-gestibilities of DM, CP, NFE, and ash and ni-trogen retention were high. Sweet potato silage therefore offers one option for partially replac-ing more expensive commercial feeds for growing pigs.

Additionally, in measurements of growth performance of SP silage fed pigs in metabolic cages, Dom & Ayalew (2009b) re-ported favourable daily live weight gain when

compared against expected rates for growing pigs in the tropics (Durrance, 1971, cited in Payne, 1990). Furthermore, daily gain and feed conversion efficiency were considerably better than in similar rations of SP tuber and vine si-lage where the silage content was 40% and 60% of the total diet (Giang et al., 2004).

However, growing pigs of either sex kept by smallholders are usually fed in groups or pairs, free to roam in a pen or larger area than a metabolic cage. The objective of this feeding trial was to evaluate growth performance of pigs on the same SP-based diets under pen manage-ment with paired animals kept in one feeding pen. It was hypothesized that growth rates and feed conversion efficiency were not different for growing pigs fed either a diet of sweet po-tato silage or cooked tubers and fresh vine combined with standard commercial ration at half the dry matter offer of a complete stan-dard treatment.


Location The trial was conducted at the Labu

Livestock Research Station, Lae, Morobe Prov-ince of the National Agricultural Research Insti-tute (NARI) of PNG. The site is located about 12 km from the Lae town on the Wau-Bulolo Road (Lat. 6o 40’ 27” S Lon. 146 o 54’ 33” E). The climate at this station is typically warm and wet with an average temperature of 32°C and humidity of 85% to 90%. Sweet potato silage making

Sweet potato silage was made entirely from purchased SP tuber and vines. Sixty to eighty kilogram bags of tubers were bought from Lae Main Market at current prices rang-ing from K1.00 to K2.00 per kilo depending on market supply. SP vine was sourced from local gardens and paid for at an agreed rate of K1.00 per kilo. The two-weekly phased ensil-ing was planned in order to reduce the use of ensiling equipment (using only 10 bins), mate-rials and labour, and limit spending. Sweet potato silage was made using an adapted tech-nique (Dom, 2007).

The ensiling of fresh sweet potato for-age was done on a single day and silage batches were stored for over two weeks before

Dom & Ayalew


30

1K stands for Kina, the legal tender of PNG, and K1.00 exchanged for US$0.3580 to US$0.3770 from April to July 2008 (BPNG, 2008). 2Sweet potato vines are not gathered for sale in PNG. Some SP vine is usually left to rot after harvesting of tubers and such waste of valuable forage has not been estimated.

Composition of the treatment diets Dry matter content of the ingredients

was determined on-station using a Labec® forced air draft oven, drying at 85 to 105°C. Crude protein in the standard commercial pig grower and SP material are shown in Table 2. Feed energy and other nutrients were not measured.

The control diet, as well as 50% by weight of the other two diets, was Flame Stock feed pig grower pellets (Associated Mills Ltd), a standard commercial feed bought in a single batch from a local supplier. Ingredients in the commercial pig grower were cereal grains, wheat by-products, vegetable and animal pro-tein meals and fats, vitamins and minerals. No ingredient composition values were indicated.

The diets were composed of the pig grower and: a) cooked tuber and fresh vine named SPctfv; b) sweet potato silage named SPS, and; c) or the sole commercial pig grower as reference, STD. Pig grower commercial standard had a crude protein content of 16% (Table 2). Offer weights for STD diet started at 4000g supplying 640g CP/offer/pen (for paired pigs). Half the weight of the standard pig grower was replaced in the test diets by silage (SPS diet) or boiled tuber fed with additional 1000g of fresh vine (SPctfy). In the mixed di-ets SPS and SPctfv commercial pellet feed weight started at 2000g (320g CP/offer/pen). The pig grower was increased in all feed offered from day 24 after observed higher intake. SPS and SPctfv were increased to 2250g (360g CP/

nd means no data, records were lost. *Fresh weights of the forage material before processing. **Moisture loss from the bulk mass after sun-drying measured as a difference in per cent before and after sun-drying; it also included some loss of material during handling. *** The weight of combined semi-dry tuber and vine forage. †Major water losses as when storing SP vine forage overnight.

Table 1. Sweet potato silage formulations for the feeding trial.

Date made

No. of silos

Bulk weight of forage (kg)*

Moisture loss after sun-drying (tuber) or wilt-ing (vines) (%) **

Mixed forage weight (kg)***

Salt (0.5%)

Total silage weight (kg) Tuber Vine Tuber Vine

03/04/08 1 74.0 63.0 48.5 4.0 69.0 0.345 69.3 04/04/08 1 69.0 90.0 47.7 10.2 67.0 0.335 67.3 05/04/08 2 157.0 53.0 35.3 28.3† 132.0 0.660 132.7 09/05/08 1 77.0 56.0 34.6 10.3 73.0 0.365 73.4 16/05/08 1 nd nd 34.9 10.2 80.0 0.400 80.4 29/05/08 2 nd nd 37.1 6.4 95.0 0.475 95.5 13/06/08 2 nd nd 47.3 9.4 87.0 0.435 87.4

Dom & Ayalew


use (Table 1). All SP forage was washed clean and cut or shredded by a mechanical chopper. The freshly chopped SP was weighed then dried under direct sunlight. Tuber was dried for about two to four hours till damp to the touch and juice released only when hand squeezed. Vine and leaves were dried until wilted (30 to 60 minutes). Moisture loss was difficult to control when sun-drying in ex-

tremely variable weather. Equal weights of the semi-dry tuber and vine were collected and thoroughly mixed together while adding salt. The silage was immediately packed, com-pressed by hand into large 80L polyethylene bags placed inside 75L plastic bins. The plastic bags were tied off using lengths of soft twine string. Surface microbial growth on the silage was not visibly obvious.

31

Parameter SP tuber and vine silage* Cooked tuber Pig grower† Chopped SP

vine (Young)‡ Grated SP tuber

(Mature)‡

DM (%) 31.9** 35.1** 89.7** 13.6 35.5 CP (%) - - 16.0 2.2 1.5

Diets STD SPS SPctfv

Composition 100% pig grower 50% pig grower + SP silage ad libitum

50% pig grower + cooked tuber ad libitum + 1000g fresh vine

M%* 89.7 60.8 76.2 CP%** 16.0 8.0 8.0

Table 3. Dry matter (DM%) and crude protein (CP%) composition of the experimental diets.

* Calculated from the DM% of sweet potato silage, cooked tuber, pig grower and chopped SP vine. ** Calculated from the CP% of pig grower only.

Table 2. Dry matter (DM%) and crude protein (CP%) composition of the diet components.

* Mixed (50/50) sweet potato tuber and vine silage for SPS diet. ** Determined by oven drying three 1kg samples over 24 hours at 85 to 105ºC. † Ingredients noted on Flame Stock feed pig grower from Associated Mills Ltd. ‡ Analysis of freshly processed sweet potato tuber and vine at National Analytical Laboratories UDC, Lae, PNG.

Animals and experimental design Six male and six female weanling

crossbred pigs [mixed parentage of some Large White, Landrace, Duroc, Tamworth and native pig (Sus crofa papuensis)] used in the trial were purchased from a local smallholder farmer. The pigs were three to four weeks of age (early weaning) and had body weights that ranged from 12kg to 19kg. Post-adoption weights after one week were between 15kg and 23kg. Live fasted body weights were checked at seven day intervals on a 100kg Bilda hanging scale (±0.5kg).

The design of the experiment was a 2x2x3 randomized complete block with three

treatment diets, two sides of an experimental shed taken as blocks and two replications. Average paired pig body weights and sex were balanced out when placing pigs into pens. The batch made silage was not considered as a variable in this experiment. A standard ensil-ing technique was used by the same two work-ers using one semi-dry weight formulation of SP forage and salt additive.

Mixing feeds Daily feed offers were in the area of

4000g to 5000g for STD diet, 6000g to 7250g for SPS and for SPctfv 8000g to 10,250g daily for each paired pen. The offers were adjusted

Dom & Ayalew


offer/pen). Similarly, STD pig grower was in-creased to 4500g (720g CP/offer/pen). In the last ten days pig grower in STD was increased to 5000g (800g CP/offer/pen) but this was not the case in the SP diets. The DM and CP com-positions of the treatment diets are shown in Table 3.

Fresh SP vine for daily offer in SPctfv

was harvested every morning from Labu Sta-tion forage garden. SP tubers for cooking were brought fresh in 60 to 80 kilogram bags, washed and stored in a cool dry environment. The unpeeled tubers were boiled in a large stainless steel pot (20L capacity) in the eve-ning, drained and cooled overnight for weigh-ing offer the next day (up to 15kg wet weight).

32

on a daily basis for decreased or increased in-take, which ensured that no additional offer was necessary at any time during the trial. The three diets offered were mixed at 7am every day. Cooked SP tubers were macerated by hand with a large plastic spoon and mixed with commercial feed; SP silage was measured out and mixed by gloved hand. Fresh SP vines were weighed to 1000g bundles and offered whole together with the cooked tuber and com-mercial mixture. The average intake of stan-dard feed in the sole commercial ration (STD) during the adaptation period was used to deter-mine the amount of feed offered in the subse-quent observation periods of the three diets. Weighing offer and refusal

Feed was offered to paired pigs in large concrete troughs with wire separators allowing for access points and limiting wastage from messing. Feeding took place at 9am on weigh-ing days but at 8am every other day. Refusals were collected at 7am each day before washing down and refilling. Offer and refusal were weighed by electronic balance with accuracy of ±1 gram.

Data collection and analysis Pig performance was measured by feed

intake, growth and back-fat thickness. The ex-periment commenced on 5 May 2008 and lasted 54 days. Analysis of variance was done using GenStat® (Lawes Agricultural Trust, 2005). Paired t-Tests of feed intake and feed conversion ratio were done in MS Excel 2003©. The final data elements, calculated from raw data of feed offered and refusal and live body weight, were fresh weight feed intake, dry matter intake, total live weight gain, average daily body weight gain and feed conversion ratio.


Dry matter feed intake and feed conversion ratio

The mean feed intake values are shown in Table 4. The DM intake for STD was much higher than both mixed SP diets. The mean DM intake of ensiled sweet potato (SPS) was lower and very different from the cooked tuber and fresh vine offer (SPctfv). The average feed con-

version ratio of the SPS was much better than SPctfv or STD (Table 4). Calculated overall feed conversion ratios were 3.09, 2.21 and 2.33 respectively for SPctfv, SPS and STD mixed diets.

Growth performance The initial body weights of the twelve

pigs ranged from 15.0 to 24.0kg and their final weights ranged from 39.0 to 61.0kg. The result-ing weight gain (LWG) and average daily gain (ADG) for three treatment diets were signifi-cantly different (P = 0.002) (Table 5). The mean ADG on commercial standard pig grower (STD) was significantly higher (730g/day) than the two mixed SP diets. The mixed silage diet had a daily weight gain (540g/day) significantly better (P < 0.05) than the cooked tuber and fresh vine diet (502g/day). Observations on pigs

The runt of the litter that weighed 12 kg at the start performed extremely well on STD diet and ended as the second largest pig at 57 kg, a total gain of 45 kilograms in less than two months. The four pigs fed solely on the pig grower weighed close to 60 kg, two of which appeared lethargic and were unable to move about freely (in fact with obvious diffi-culty) in contrast to pigs on the SP diets, that were energetically playful.

Pigs fed the cooked and fresh SP diet ate with some selectivity, although the feed was mashed up, one pair preferred the com-mercial pellet feed over the cooked tuber but the other pair was surprisingly the opposite. SP vine were relished by all pigs and very little, if any, was refused. However, in the SPS diet more pellets were consumed along with the silage. The subject pigs reached sexual matur-ity during the course of the trial and some mat-ing behaviour was observed in the last week of feeding.

Measure of pig back-fat thickness Three of the largest male pigs from each

diet were selected for measurement of their back-fat thickness after sale and slaughter at a certi-fied abattoir in Lae. The depth of fat was meas-ured using a steel ruler on three cuts made at the centre of shoulder joint, on the rump above the back leg and on the back above the last rib. The

Dom & Ayalew


33

Parameter

Treatment means by pen* (mean ± s.e.)

Paired t-Test results

SPctfv SPS STD SPctfv-SPS

SPctfv-STD

SPS-STD

DMI 3106±171.19 2380±174.93 3388±198.33 0.0000 0.0023 0.0000

FCR 4.0±0.78 2.7±0.28 2.6±0.21 0.0017 0.0009 0.2877

Parameter Treatments s.e. LSD(5%) P

SPctfv SPS STD

LWG (kg) 28.1a 30.2a 40.9b 1.90 6.09 0.002

ADG g/day) 502a 540a 730b 34.0 108.8 0.002

Table 4. Dry matter intake (DMI) and feed conversion ratio (FCR) mean values and their paired comparison by diet.

*Feed intake from each pen was for two pigs, where pens were initially balanced by total male and female weight for all six pens.

carcass weight and sales prices were also noted, with the unit prices determined upon carcass evaluation after slaughter (Table 6).

The acceptable limit of back-fat thick-ness at the purchasing supermarket for lean pig carcasses was 10 mm. Local butchers have com-mented that pigs with excessive fat made very little prime meat cuts, therefore leaner pigs were preferred as there was less wastage going into

sausages, a lower value end product. The pig raised on either of the SP diets had carcass weights with the less wastage, only 14 kg, whereas for STD pigs the wastage was 20 kg. Moreover, SP silage diet produced pigs with acceptable back-fat thickness and gave a higher rate of return on the carcass weight, Kina 0.60 per kilo more than those on the commercial diet.

Diet

Final live weight (kg)

Carcass weight (kg)

Back-fat depth (mm)

Sale value (Kina)

Unit price (Kina/kg)

STD 68.0 48.0 22.28 322.00 4.74

SPS 59.0 45.0 10.20 315.00 5.34

SPctfv 51.0 37.0 13.19 259.00 5.08

Table 6. Comparison of final live weight, carcass weight, back-fat thickness, sales value and unit price of pork sold from three sample pigs.

NOTE: Carcass measures were made at the Livestock Development Corporation abattoir in Lae, Morobe Province.

Dom & Ayalew

Table 5. Comparison of mean live weight (LWG) and daily gain (ADG) during the observation period by diet.

Means of a row with different superscripts are statistically significantly different at least at P = 0.05.


DISCUSSION

The standard commercial pig grower supplied at least eight percent of the crude protein re-quirements in mixed diets (Table 3) and sweet

potato is an excellent source of readily digesti-ble starch for energy. The average daily gain (ADG) and calculated feed conversion ratios (FCR) of the three diets were comparable with the literature. Dry matter intake of SPS (by pen

34

for two pigs) was much less than either SPctfv or STD but the daily gain and feed efficiency were competitive.

It has been shown that dry matter intake and average daily gain decreased when cooked sweet potato progressively replaced maize diets, with no significant change in feed conversion (Corring & Rettagliati, 1969; cited in Pérez, 1997). When castrate males and female growing pigs (30-60 kg) were fed fresh sweet potato foli-age ad libitum with a supplement containing 249 g protein (13.2% CP) the weight gain was 597 g/day and feed conversion rate of 2.89 kg/kg (González et al., 2003) indicated better per-formances at higher levels of protein inclusion. A similar result was reported by Malynicz & Nad (1973) for PNG native pigs (Sus crofa papuensis).

Dom & Ayalew (2009b) fed crossbred (Large White sow x Landrace-Duroc boar) growing pigs SP silage at 50% of a standard diet (14% CP) and found significantly lower dry matter intake while daily gain and efficiency was not affected. Giang et al. (2004) suggested that reduced DM intakes of SP silage diets could be due to the bulkiness as well as the high fibre content of the feeds. In their rigorous trials where crossbred (Mongcai sow x Yorkshire boar) pigs were raised from 15 kg to 50 kg on SP silage diets (with 40% and 60% of tuber and vine content) substituting 50% of a balanced standard diet, the average daily gains were 423 g to 328 g (Giang et al., 2004), compared to a daily gain of 540 g/day in the present study (Table 5). Their mixed silage diets, which con-sisted of maize based standard diet (17% CP), resulted in a poorer rate of feed conversion to growth than our experimental mixed diet used with commercial pellet feed (16% CP) as a stan-dard. A slightly higher CP content in the diet with silage did not lead to better growth per-formance so there is some level beyond which foliage inclusion in a diet will not improve body weight gains.

There are several environmental, physio-logical and dietary factors that influence volun-tary intake in pigs; these include genetics, num-ber of pigs per pen, energy density and feed processing (NRC, 1998) which appear pertinent for understanding SP silage as fed to local cross-bred pigs. An important factor for growing pigs fed ad libitum is the amount of energy in the

feed (ARC, 1981; Cole, 1984; Chiba et al., 1991), where intake is usually 3 to 4 times that required for maintenance (NRC, 1998). Sweet potato silage is high in digestible energy and hence its high gross energy value (Tomita et al., 1985; Pérez, 1997). Several authors noted that the digestible energy of sweet potato was high regardless of the mode of its presentation - raw or cooked, as silage or as chips (Canope et al., 1977; Rose & White, 1980; Dominguez, 1992; Tomita et al., 1985; Noblet et al., 1990; in Pérez, 1997). Tomita et al. (1985), Perez (1997) and Dom & Ayalew (2009a) found that apparent DM digestibility of silage diets were as high as 90%, and this was much higher than the 70.9% reported by Giang et al. (2004). Fresh SP vine on the other hand is constrained by reduced di-gestibility when fed at levels over 10% DM (Dominguez, 1992). Adversely the digestibility of crude protein in ensiled sweet potato tubers is low; two estimates are 32% (Tomita et al., 1985; in Pérez, 1997) and 47% (Giang et al., 2004).

However, in this trial the lower DM in-take on mixed SP silage diet (SPS) was not an apparent disadvantage since daily gain was sig-nificantly better than SPctfv (P < 0.002) and calculated feed conversion (FCR=2.21) was bet-ter than those of both the traditional cooked tu-ber and fresh vine diet (FCR=3.09) and sole fed commercial pig grower (FCR 2.33). The better FCR despite lower DM intake of SP silage di-ets in comparison to cooked and fresh SP also meant better overall performance. This cou-pled with the better carcass quality ensured better unit price of the end product (Table 6). Peters et al. (2001) and Giang et al. (2004) both concluded that the cost of feed per kg weight gain were lower for the SP silage diets. The incremental cost of this ensiling technique in SP silage production was estimated by Dom & Ayalew (2009b) at K1.34 per kilogram while the commercial feed costs K1.45 per kilogram. Using these two values the resulting cost for raising a single pig on a mixed sweet potato silage and pig grower diet for sixty days (at the reported feed intake) would be Kina 99.58, providing a net return on feed cost of Kina 215.42.

Equally important are the practicalities to the smallholder pig farmers of preparing the SP silage at home in suitable quantities and

Dom & Ayalew


35

making the mixed diets on daily basis. The time and labour required to continually make the SP silage and mix the daily diets are not too de-manding, and the daily chores can be handled by those members of the family who mostly stay at home.

CONCLUSION AND RECOMMENDA-TIONS

The feed conversion ratio of the sweet potato silage diet was similar to the standard diet and this resulted in better performances than the cooked sweet potato and fresh vine diet that has been the traditional pig feeding regime for generations of smallholder and vil-lage farmers. The quality of carcass was also better than of the sole commercial diet, al-though the live weight gained was less than what was achieved on the sole commercial ra-tion. Substituting half of the commercial ra-tions with SP silage does not adversely affect pig grower performance. It was suggested that improved digestibility of sweet potato silage encouraged performance on mixed diets. In fact the carcass quality and marginal profits can be improved. In the face of rising cost of commercial pig rations, smallholder pig farm-ers can effectively utilize these rations as sources of digestible protein by bulking these rations up with more fibrous but digestible and cheaper SP silage.

A unique strategic advantage associated with SP silage making is the opportunity to pre-serve highly digestible fibrous diet for several months without loss of its nutritional qualities thereby allowing pig farmers to maintain good plane of nutrition long after the SP harvest sea-son. The scale of piggery operation will deter-mine the size of SP crop and the volume of en-siling. Management implications arise in the preparation, handling and storage of silage when feeds are offered to maintain maximum growth

at the least cost, which is the aim of all commer-cial pig growers regardless of their level of man-agement. A comprehensive cost-benefit and perhaps socioeconomic assessment of this technology may be justified by its broad rele-vance to numerous smallholder pig and sweet potato farmers. Furthermore, capitalizing on the usefulness of ensiling sweet potato will require further research into the development and integration of simple mechanical milling (shredding/chopping) machines and adapted storage units (silos). Naturally, sweet potato cultivars vary in their suitability as well as pro-ductivity of SP silage, and hence research is warranted in the agronomic and breeding as-pects of the crop.

Related follow up studies should in-clude investigation of alternative local and/or seasonal protein sources from industrial by-products. For example, poultry concentrates, fish meal and copra meal from the food indus-tries can provide the critical digestible proteins to improve the nutritive quality of forage-based diets.

Many commercial-oriented pig farmers tend to grow out crossbred weaner pigs on for-age-based diets. Generally, the local crossbred pigs perform better on mixed forage diets than either native pigs or exotic breeds, however, in PNG the genetic potential of such crosses have not been fully investigated and is an opportu-nity for further research and development.

ACKNOWLEDGEMENTS

This feed trial was fully funded by NARI (Project U10001). The authors would like to thank Ku Kobila and Stanley Amben for their patience and cooperation in its execu-tion. Particular recognition is afforded to John Levi for his tireless work in the SP ensiling, and conscientious care of the experimental ani-mals throughout the trial days and nights.

Dom & Ayalew


REFERENCES

AGRICULTURAL RESEARCH COUNCIL (ARC) 1981. The Nutrient Requirements of Pigs; Technical Review Rev. ed. Slough, England, Commonwealth Agricultural Bureaux xxii. 307 pp. AN, L.V. 2005. Sweet potato leaves for growing pigs. Doctoral dissertation. Dept. of Animal Nu-trition and Management, SLU. Acta Universitatis agriculturae Suecia. Agraria vol. 470.

36

BOURKE, R. M., VLASSAK, V. & Australian National University 2004. Estimates of food crop production in Papua New Guinea. Land Management Group, Dept. of Human Geography, Re-search School of Pacific Studies, Australian National University, Canberra. CANOPE, J., LE DIVIDICH, J., HEDREVILLE, F. & DESPOIS, F. 1977. Influence d'un traite-ment technologique sur l'efficacité alimentaire des principaux produits amylacés tropicaux: patate douce et banane dans l'alimentation du proc. Nouvelles agronomiques des Antilles et de la Guyane, 3 (3/4):310–322 (as cited in Pérez 1997). CHIBA, L. I., LEWIS, A. J., & PEO, E. R. Jr. 1991. Amino acid and energy interrelationship in pigs weighing 20 to 50 kilograms; I. Rate and efficiency of weight gain, Journal of Animal Sci-ence, 69:694-707. COLE, D. J. A. 1984. The nutrient density of pig diets – allowances and appetites. In: Wiseman, J. (Ed.), Fats in Animal Nutrition. London, Butterworth, pp. 301-312. CORRING, T. & RETTAGLIATI, J. 1969. Utilisation de la patate douce dans l'alimentation du porc en Guadaloupe. Journées de Recherche Porcine, pp. 105-107 (as cited in Pérez, R. 1997). DOM, M. 2007. Guidelines storing sweet potato as silage for feeding pigs, NARI Livestock Research Programme (Unpublished). DOM, M. & AYALEW, W. 2009a. Adaptation and testing of ensiling sweet potato tuber and vine for feeding pigs: Digestibility and nitrogen retention on mixed silage diets. Journal of the Institute of Chemists PNG, 2:77-88. DOM, M. & AYALEW, W. 2009b. Adaptation and testing of ensiling sweet potato tuber and vine for feeding pigs: On-station growth performance on mixed silage diets. PNG Journal of Science, Research and Technology, 1:86-96. DOMINGUEZ, P. L. 1992. Feeding of sweet potato in monogastrics. In: Machin, D. & Nyvold, S. (Eds.), Roots, tubers, plantains and bananas in animal feeding. Animal Production and Health pa-per No. 95, FAO, Rome. pp. 217-233. http://www.fao.org/ag/aga/agap/frg/AHPP95/95-217.pdf DURRANCE, K. L. 1971. Basic information for swine production. University of Florida, Gains-ville. In: An Introduction to Animal Husbandry in the Tropics, (Ed. Payne W.J.A.), Fourth Edition, ELBS, Longman Singapore. GIANG, H. H., LE, V. L. & OGLE, B. 2004. Digestibility of dried and ensiled sweet potato roots and vines and their effect on the performance and economic efficiency of F1 crossbred fattening pigs. Livestock Research for Rural Development, [Online]. Vol. 16, Art. #50. Available at: http://www.cipav.org.co/lrrd/lrrd16/7/gian16050.htm [Accessed June 2, 2009]. GODOY, R. & ELLIOT, R. 1981. Efecto de cinco forrajes tropicales sobre algunos parámetros de la función ruminal y flujo de nutrientes al deudeno de bovinos alimentados a base de melaza-urea. Prod. Anim. Trop. 6 (2):177–184. (as cited in Pérez 1997). GONZÁLEZ, C., DÍAZ, I., VECCHIONACCE, H. & LY, J. 2003. Performance traits of pigs fed sweet potato (Ipomoea batatas L.) foliage ad libitum and graded levels of protein. Livestock Re-search for Rural Development, [Online]. 15 (9). Available at: http://www.cipav.org.co/lrrd/lrrd15/9/gonz159.htm [Accessed June 2, 2009].

Dom & Ayalew


http://www.fao.org/ag/aga/agap/frg/AHPP95/95-217.pdf

http://www.cipav.org.co/lrrd/lrrd16/7/gian16050.htm




http://www.cipav.org.co/lrrd/lrrd15/9/gonz159.htm




37

HANSON, L. W., ALLEN, B. J., BOURKE, R. M. & MCCARTHY, T. J. 2001. Papua New Guinea Rural Development Handbook, The Australian National University, Brown Prior Ander-son, Melbourne, Australia, pp. 11. LAWES AGRICULTURAL TRUST 2005. GenStat® Release 4.2, Discovery Edition. VSN Inter-national Limited, Hempstead, United Kingdom. MALYNICZ, G. & NAD, H. 1973. The effect of level of feeding and supplementation with sweet potato foliage on the growth performance of pigs. Papua and New Guinea Agricultural Journal, 24:139-144. NOBLET, J., FORTUNA, H., DUPIRE, C. & DUBOIS, S. 1990. Valeur nutritionnelle de treize matières premières pour le porc en croissance. 1. Teneurs en énergie digestible métabolizable et nette. Conséquences du choix du système énergétique. Journées de Recherche Porcine en France 22:175–184. (as cited in Pérez 1997.) PAYNE, W. J. A. 1990. An Introduction to Animal Husbandry in the Tropics, Fourth Edition, ELBS, Longman Singapore. pp 658. PETERS, D., TINH, N. N., THUY, T. T. & THACH, P. N. 2001. Fermented sweet potato vines for more efficient pig raising in Vietnam. AGGRIPPA, FAO, Rome, Italy. PÉREZ, R. 1997. Feeding pigs in the tropics, Animal Production and Health Paper No. 132, FAO: Rome. ROSE, C. J. & WHITE, G. A. 1980. Apparent digestibilities of DM, OM, CP, energy and ADF of chopped raw sweet potato (Ipomoea batatas) by village pigs. Papua New Guinea Agricultural Journal, 31 (1–4):69–72. TOMITA, Y., HAYASHI, K. & HASHIZUME, T. 1985. Palatability of pigs to sweet potato-silage and digestion trial by them. Bulletin of the Faculty of Agriculture, Kagoshima University, No. 35:75–80 (as cited in Pérez 1997).

Dom & Ayalew


38

Nitrogen fixation by Casuarina oligodon L. Johnson agrofor-estry in the Papua New Guinea highlands

Johnny Minga Wemin1, Rowan Reid2 and Robert Edis2

1R&D Department, Ramu Agri Industries Ltd, P. O Box 2183, LAE, Papua New Guinea. 2Faculty of Land and Food Resources, The University of Melbourne, Victoria 3010, Australia. Corresponding author email: [email protected]

ABSTRACT Casuarina oligodon L. Johnson is a multipurpose tree species grown in the highlands of Papua New Guinea. Biological N fixation in C. oligodon under field conditions in short (5-10 years) and long (11-15 years+) fallows in the Papua New Guinea central highlands was investigated using the 15N natural abundance technique. Results from the study suggest that as much as 70 per cent of N in C. oligodon was derived from the atmosphere. The rate of N2 fixation was relatively low in short fallows of casuarina and increases as the trees mature in the long fallows. A rate of N2 fixa-tion up to 36 kg N ha -1 year -1 was estimated based on commonly practiced tree stocking rates and field conditions in the region. Although casuarina fallows tend to accumulate higher total N and C compared with equivalent period of grass fallows, the amount of N and C in the surface soils of all systems under study showed no significant difference (P>0.05). The amount of total N and C un-der longer fallows of casuarina was generally greater than shorter fallows of casuarina. A signifi-cant proportion of ‘fixed’ N was stored in above-ground biomass of trees that were more than 10 years of age. Management of the standing biomass, particularly when the fallow is converted back to the cropping phase, is therefore critical in ensuring that the farmers are able to gain maximum benefit from the ‘fixed’ N. Whilst the removal of trunk for use as fuel or building material may be an important product of agroforestry system, retaining the foliage, small branches and bark on the site is crucial in sustaining agricultural productivity.

Key words: Casuarina oligodon, agroforestry, biological nitrogen fixation, frankia, Papua New Guinea.

INTRODUCTION

Agricultural crops require nitrogen (N) in large quantities yet most crop species are not able to access N directly from the air where it is abundantly present. The process of biological nitrogen fixation (BNF) or the addi-tion of some form of N fertiliser is therefore essential for sustainable agricultural productiv-ity (Wani et al., 1995; Angus, 2001). Many subsistence farmers who depend largely on natural recycling of soil nutrients are now adopting short grass fallows, which although easier to manage, may be less effective in im-proving soil fertility and crop productivity.

Casuarina oligodon is widely used as a traditional fallow species in the highlands of Papua New Guinea (PNG). The N fixing na-tive tree species is also valued by farmers for shading coffee (Coffea arabica) and providing wood for fuel and building materials. Casua-rina species thrive under poor soil conditions because of their ability to form symbiotic rela-tionship with Frankia (bacteria), thereby transforming atmospheric dinitrogen (N2) into organic N (Gauthier et al., 1985). Upon min-eralisation of the plant residues, plant usable ammonium (NH4

+) and nitrates (NO3-) are re-

leased for enrichment of soil fertility (Smith, 1999). An accurate and reliable estimate of the



39

actual amount of N captured by individual spe-cies is useful for developing sustainable farm-ing systems, particularly in areas where N fer-tilisers are either unavailable or prohibitively expensive.

A number of workers have quantified N2 fixation rates in some Casuarina species such as C. equisetifolia and their estimates vary between 15 and 100 kg N ha-1 year-1 (e.g. Gauthier et al., 1985; Yoneyama et al., 1990; Mariotti et al., 1992; Parrotta et al., 1994, 1996). However, the extent of N2 fixation by C. oligodon or its impact on soil fertility in the tropical highlands of PNG is inadequately known (Thiagalingam, 1983; Bourke, 1989, 2007; Bino & Kanua, 1996; Talopa, 2003). The tree has potential to not only contribute to soil fertility enhancement but also provide vil-lage communities access to fuel wood and building materials that are scarce (Bourke & Allen, 2009). Inadequate research and lack of scientific information may lead to misunder-standings about the potential of C. oligodon to capture N2 and the management strategies re-quired to realize this potential through en-hanced crop production. If the amount of N captured in field grown casuarinas can be de-termined accurately and reliably, this could enable farmers to develop sustainable farming systems with reduced reliance on N fertilisers. The aim of this study was to explore the effec-tiveness of C. oligodon in capturing N2 during the fallow phase for subsequent enhancement of soil fertility.

Four land management systems were used in this study to determine N contributions by C. oligodon in the subsistence farming sys-tems. These were cropping phase, grass phase, short casuarina phase and long casuarina phase. Cultivated land in the study area is of-ten abandoned after 2 or more plantings with apparent decline in crop yield which is mainly attributed to soil fertility depletion, particularly N (Hartemink & Bourke, 2000; Bourke & Al-len, 2009). The land is rested for grass fallow up to 5 years, short casuarina fallow of 5-10 years or long casuarina fallow of 11-15+ years to recycle soil nutrients. The fallow period adopted by each landholder increases or de-creases depending on land availability within the landholdings and the urgency for re-cultivation. Fallow is shortened and cultivation

intensifies where population density is high with permanent settlement while shifting culti-vation of extended fallow is practiced in less densely populated communities (Bourke & Allen, 2009).


Site description The field study was conducted on 24

subsistence gardening and fallow sites in four village communities in the Simbu Province (latitude 6o S and longitude 145o E) of PNG (Figure 1) between December 2005 and Febru-ary 2006. The sites were situated between 1500 and 2100 m above mean sea level. The average annual rainfall is 2500 – 3000 mm. The mean temperature is 24 – 26 oC with a range of 13 to 30 oC (McAlpine & Keig, 1983; Hanson et al., 2001).

The dominant soil type was Inceptisols with a lesser composition of Mollisols and En-tisols (USDA classification) (Bleeker, 1983; Bellamy & McAlpine, 1995). These soils have characteristically well developed dark colour in A horizons, fine textured and light to dark brown clayey subsoils. Dinima, Masul and Nintul soils are mostly clay loam with darker topsoil and light brown to dark brown subsoil. Konoma soils are generally dark with patches of yellowish to greyish clayey and have sig-nificant amounts of siltstone and mudstone throughout the measured depth (0-30 cm). The soils were generally rich in organic matter.

Experimental design The experimental design comprised

four treatments (Table 1), six replicates, two soil depths and two composite samples for each soil depth. The plant sampling strategy comprised two of the treatments, six replicates and one composite sample of each plant spe-cies. The experimental sites and subjects were selected under prevailing field conditions with-out inducing external treatments. The size of each site varies between 0.08 and 1.8 ha with slopes ranging from 2- to 55 %.

Soil cores Soil cores (15 cm in diameter) were

taken at regular intervals of between 5 and 10 m depending on the size of the area following

Wemin, Reid & Edis


40

Figure 1. Map of the Simbu Province showing field research sites.

Casuarina sampling Plant sampling strategy followed that

outlined by Shearer & Kohl (1986). One (or two if the tree density was low) leafy branch was removed from the mid-canopy zone of each casuarina tree within a quadrate of 10 m x 10 m. If the mid-canopy zone was difficult to

reach, lower branches were collected. The fresh leaves and branch lets from each site were bulked together for drying (Shearer & Kohl, 1988; Danso et al., 1992; Parotta et al., 1996).

Reference plants

Reference plants are non-N2-‘fixing’ species that are presumed to have similar root architecture and explore a similar soil volume as the N2-‘fixing’ species of interest (Fried et al., 1983; Shearer & Kohl, 1986; Herridge & Danso, 1995). The reference plant should ulti-mately grow in the same place or close to the N2 ‘fixing’ plant of interest. The sampling

Wemin, Reid & Edis


a zig-zag pattern (Gunn et al., 1988). Sampling was done at two different depths (i.e. 0 - 10 cm and 10 - 30 cm). Eight to ten cores at each depth were bulked and thoroughly mixed in separate 10 litre plastic buckets. Two compos-ite samples, each weighing approximately 400-500 g was collected for drying. The samples

were transferred into paper bags and dried in an oven at 40 oC for 48 hours. The dried sam-ples were sieved (2 mm screen) before sub-sampling. A total of 96 sub-samples weighing about 200 g each were ground and analysed at The University of Melbourne, Australia.

Table 1. Land management phases observed in the study.

Level Treatment 1 End of cropping phase 2 5-10 years of casuarina phase 3 11-15 + years of casuarina phase 4 3-5 years of grass phase

41

strategy and criteria used is outlined by Domenach et al. (1989) and Peoples et al. (1989).

The three reference plant species sam-pled were; Eucalyptus pellita F. Muell, Maca-ranga pleiostemona Pam & K. Hoffm. Perry and Dodonaea angustifolia L. F. Many authors have used Eucalyptus spp. as reference plant for detecting N2 fixation in casuarina (e.g. Yo-neyama et al., 1990; Mariotti et al., 1992; Par-rotta et al., 1994; Stahl et al., 2005) but the other two species have not been used before. The reference plants were either found with casuarina or on the periphery of casuarina for-ests. The middle canopy or accessible lower branches were collected and bulked from 3-4 trees of each species.

Sample preparation Dried samples were milled into a ho-

mogeneous fine powder using an electric Ring Mill fitted with chrome steel head from Rock-Labs, New Zealand. The head was thoroughly cleaned with 70 per cent alcohol to avoid cross contamination between samples. The ground samples were weighed into tin (Sn) capsules using an Analytical Balance with 0.01 mg ac-curacy (about 10-20 mg samples for soil and 3-4 mg for plant material). The tin capsules were folded against a stainless steel micro plate into a neat 2-3 mm diameter package for isotope and elemental analysis.

Isotope and elemental analysis 15N and 13C isotope of the plant and

soil materials under casuarina fallows were determined by isotope ratio mass spectrometry (IRMS) at the Forest and Ecosystems Research Laboratory, The University of Melbourne, Creswick, Victoria. Elemental N and C were

determined by the Dumas Method (Bremner & Yeomans, 1988; Mulvaney, 1993) using an Elemental Analyser (Carlo Erba NA1500 Se-ries II) at the Soils Chemistry Laboratory, The University of Melbourne, Parkville, Victoria.

Calculation of N2 fixation Several methods have been developed

and used in the estimation of BNF in legumes and nodules producing non-legumes (Rennie & Rennie, 1983; Peoples et al., 1989). In the present study it was feasible to use the 15N natural abundance technique (Domenach et al., 1989). This technique usually reflects the 15N abundance of the major immediate N sources that are available in the soil (soil N) for non-N2 ‘fixing’ plants and from the atmosphere (N2) for N2 ‘fixing’ plants. The difference of 15N abundance values from these sources of N pro-vides the basis for estimating the relative con-tribution of atmospheric N to N2 ‘fixing’ sys-tems (Shearer & Kohl, 1988). Although Hen-sen & Pate (1987) point out that the 15N natu-ral abundance method may not be applicable to all ecosystems, many other authors agree that it is superior to other methods in quantifying N2 fixation in N2 ‘fixing’ shrubs and trees un-der established ecosystems, provided that suit-able reference plants are growing in the same site for use in the analyses (e.g. Fried et al., 1983; Rennie & Rennie, 1983; Shearer & Kohl, 1986; Ledgard & Peoples, 1988; Domenach et al., 1989; Yoneyama et al., 1990; Herridge & Danso, 1995).

The percentage of N derived from at-mosphere (% Ndfa) by C. oligodon is esti-mated using Equations (A) and (B).

Wemin, Reid & Edis


(atom % 15N sample) – (atom % 15N standard) (A) δ15N (‰) = 1000 x ___________________________________

atom % 15N standard

where δ15N (‰) is a net increase in the 15N abundance of soil N compared with atmos-pheric N2 which is expressed as parts per thou-sand (‰) because of very small difference; atom% 15N standard of the atmospheric N2 is

0.3663 which is usually constant anywhere on the earth; and atom% sample 15N is the experi-mental subject (s) (i.e. casuarina and reference plants) used to quantify N2 fixation.

42

where δ15Nref represent the level of δ15N de-tected in a reference plant growing in the same place as the casuarinas, δ15Ncas is the 15N abun-dance of C. oligodon, and B is the 15N abun-dance (‰) of fixing plants grown obtaining its entire N from the atmosphere. A B value of -0.98 ± 0.5 ‰ for Casuarina equisetifolia (Mariotti et al., 1992) was used for the calcula-tion of %Ndfa in this study.

Both the ‘fixing’ plant (Casuarina) and the non-‘fixing’ plants (reference) grow in the same soil and explore the same soil volume but casuarinas presumably obtain their N mostly from the atmosphere and reference plants from the soil N pool. The difference in δ15N abundance between N obtained from soil (reference plant) and atmosphere (fixing plant) gives the estimate of %Ndfa by the Casuarina. Boddey et al. (2000) point out that in order to obtain better estimates of N2 fixation the dif-ference should be large.

Isotopic fractionation and discrimina-tion in the samples often leads to undesirable estimates of ‘fixed’ N in ‘fixing’ systems (Ledgard & Peoples, 1988; Shearer & Kohl, 1988, 1993; Herridge & Danso, 1995; Boddey et al., 2000). In the present study, the δ15N dif-ference between reference plants and casuarina was more negative and too small in most sam-ples to sufficiently detect N2 fixation estimates of all the sites under fallow of C. oligodon. This has allowed the use of soil δ15N values (Equation C) to quantify amount of ‘fixed’ N by C. oligodon in the soil (Chalk et al., 1983; Chen & Zhu, 1986; Bergersen et al., 1989). Shearer & Kohl (1988) note that other re-searchers use δ15N of soil in some terrestrial systems but suggest that results should be in-terpreted with caution. Hence, the δ15N values of the soil under casuarina fallow were used together with δ15N of non-N2 ‘fixing’ plants and proportional contribution of ‘fixed’ N was estimated.

where % Ndfn represents the percentage of N derived from fixed N rather than other sources in the soil; δ15Nsoil and δ15Ncas are for soil and casuarinas, respectively; and δ15Nair is the at-mospheric composition which is effectively zero (Herridge & Danso, 1995; Boddey et al., 2000).

The results are expressed in both pro-portional contribution (%Ndfa) and rate of fixation (kg N ha -1). Mass of the foliage of casuarina was calculated using Equation (D) (Parrotta, 1999). Then the amount of fixed N in casuarina foliage was calculated using the mass.

where Y is dry mass in kg and Db is tree di-ameter at breast height. Statistical analysis

Statistical analysis was performed in Minitab 14. A descriptive analysis determines the means, standard errors and coefficients of variance. Two-way ANOVA was used to de-termine the difference between treatments and sites. The F and Tukey tests were done to de-termine if there was any significant difference in the treatments, between sites, N fixations, total N and C, and C:N ratios.

RESULTS AND DISCUSSIONS

Soil N and C Total N will generally contain several

forms of N including organic N, ammonium (NH4

+), nitrite (NO2-) and nitrate (NO3

-). Ni-trite is rarely available to plants as it oxidizes quickly, whereas NH4

+ and NO3- can be taken

up by plants for physiological functions and protein synthesis. Total N reflects the amount of organic N available for mineralization. Its concentrations in soils can vary widely from less than 0.02 per cent in the subsoil to more than 2.5 per cent in the peat (Bremner & Mul-

(δ15Nref – δ15Ncas) (B) % Ndfa = 100 x _________________ (δ15Nref - B)

(δ15Nsoil – δ15Ncas) (C) % Ndfn = 100 x ________________ (δ15Nsoil - δ15Nair)

Y = 0.156 + 0.019Db2 (D)

Wemin, Reid & Edis


43

vaney, 1982) and generally decreases with depth (Sierra et al., 2002). In the present study, the total N pool in the surface soil (0-10 cm) did not show statistical difference between treatments. However, total N pool in the sub-soil (10-30 cm) under long casuarina fallow did show significant difference (Table 2). The difference is twice (∆1.8 mg g -1 soil) as much as that observed in short fallow casuarina fields (∆0.8 mg g -1 soil).

The study also suggest that total N con-centrations under short and long fallow of ca-suarinas did not indicate significant dissimi-larities despite numerically higher soil N con-centrations observed in the short fallow fields (6.8 mg g -1 soil) (Table 2). These results are consistent with a previous report that showed N concentration in excess of 5 mg g -1 soil in the same region (Hartemink & Bourke, 2000). The lower N concentration under long fallows could relate to faster decomposition and min-eralization, leaching of NO3

-, N uptake by the trees and N removal in the products as timbers and fuel wood (Power et al., 2003). The oppo-site could be suggested for the sites under short fallow of casuarinas with high total N concentrations with less difference in N con-centrations between top and sub soils.

Power et al. (2003) report that the opti-mum time for maximum soil N accumulation under tree fallow was between 5 and 10 years after planting. This has been the case where higher turnover of leaf litter under short fallow of casuarinas has prospectively contributed to

high soil N levels (Bino & Kanua, 1996; Si-erra, 2002). Despite not detecting significant N2 fixation from some sites under short fallow of casuarinas, the results from two sites sug-gest that increments in soil N are a function of BNF process.

The supply of soil N ultimately de-pends on the proportion of organic matter frac-tion of the soils (Adams & Attiwill, 1986; Atti-will & Leeper, 1987; Sparling, 1998; Issac et al., 2003; Martius et al., 2004). The inorganic ions or mineral N that are easily accessible by plants are made available through the process of soil organic matter decomposition. The highest quantities of organic C were found in soils under short fallow of casuarinas (83 – 71.3 mg C g -1 soil in order of decreasing soil depth). This suggests that actively growing trees undergo rapid replacement of tissues and returned more dead tissues (litter) to the soil as compared to soils under long fallow of casua-rinas (80.8 – 53.5 mg C g -1 soil in order of decreasing soil depth). Differences in the soil organic C concentration with depth under short fallow of casuarinas (∆11.7 mg C g -1 soil) is much less than that of long fallow of casua-rinas (∆27.3 mg C g -1 soil) suggesting that, given time, leaf litter drop and fine root turn-over in the surface soils can lead to a build up of soil organic C in surface soils (Schaetzl & Anderson, 2005). A decline in organic C with depth is consistent with literature (e.g. Pandey et al., 2000).

Wemin, Reid & Edis


Table 2. Total N and C concentration and amount (t N ha -1) of N and C in the surface soils at the end of cropping phase, under Casuarina fallows and grass fallows in the PNG highlands.

Treatment Top soil (0-10 cm) Sub soil (10-30 cm)

N (mg/g soil) C (mg/g soil) N (mg/g soil) C (mg/g soil)

(1) End of cropping phase 6.42 ± 0.96a 77.50 ± 12.00a 5.50 ± 0.85a 64.83 ± 9.84b

(3.53 ± 0.52a) (43.21 ± 5.10a)

(2) Casuarina 5-10 yrs) 6.75 ± 0.70a 83.00 ± 8.50a 5.92 ± 0.62a 71.33 ± 8.35a

(3.61 ± 0.16a) (44.91 ± 2.30a) (3) Casuarina (11-15yrs+) 6.00 ± 0.63a 80.75 ± 9.54a 4.17 ± 0.64b 53.50 ± 9.09c

(4.26 ± 0.69a) (55.30 ± 9.37a) (4) Grass fallow 6.08 ± 0.75a 78.75 ± 9.60a 5.33 ± 0.75a 65.33 ± 8.35b

(3.54 ± 0.47a) (46.08 ± 5.27a)

1. Means with similar letters within column are not significantly different (P>0.05). 2. Values in parenthesis are amount of total N and C in the surface soils in t ha -1.

44

Effectiveness of reference plants in detect-ing N2 fixation

Proportional contribution of atmos-pheric N (%Ndfa) by C. oligodon was esti-mated using δ15N of reference plants and B value. Eucalyptus pellita and Macaranga pleiostemona had δ15N of -0.07 ‰ and -0.15 ‰, respectively. A B value of -0.98 ‰ from Mariotti et al. (1992) was used to satisfy Equa-tion (C ).

Two of the reference plants species used provided good estimates of N2 fixation by C. oligodon. However, the results indicate that Eucalyptus sp. was a better reference plant choice than M. pleiostemona because the δ15N values of the latter were more negative and smaller than that of casuarinas. Dodonaea an-gustifolia was used as a reference plant in this study but did not appear to be useful in esti-mating N2 fixation because δ15N was more negative and much smaller than that of casua-rinas.

The results suggest that C. oligodon has the potential to fix atmospheric N to sat-isfy as much as 70 per cent of the trees N re-quirements. As much as 70 per cent of N with an average of 38.5 ± 9.9 in C. oligodon over 10 years of age was derived from the atmos-phere through N2 fixation. In the present study, %Ndfa is comparable to previous estimates for C. equisetifolia which vary between 39 per cent (Marriotti et al., 1992) and 81 per cent (Yoneyama et al., 1990).

N2 fixation estimate using δ15N reference plants

The study sites had average tree stock-ing rates of 4590 and 2280 trees ha -1 under short and long fallows of casuarinas, respec-tively. From the 12 composite samples com-prising 6 each of short and long fallows of C. oligodon, two of the short fallow and all of the long fallow sites showed N2 fixation between 8 and 70 per cent. Very small difference of δ15N between reference plants and casuarina did not allow detection of N2 fixation in four of the sites under short fallows of casuarina (Figure 2). The rate of N2 fixation ranged from 3 to 36 kg N ha -1 year -1 (Figure 2). The %Ndfa is similar to previous estimates for C. equisetifo-lia by Mariotti et al. (1992) and Yoneyama et al. (1990) using 15N natural abundance

method. The estimate for C. equisetifolia is between 39 per cent (Mariotti et al., 1992) and 81 per cent (Yoneyama et al., 1990). Parrotta et al. (1994, 1996) also found similar rates of N2 fixation by C. equsetifolia using 15N-enrichment techniques. Their estimates ranged from 39 – 94 kg N ha -1 year -1 (%Ndfa = 50 – 70 %). Nevertheless, others had grossly over-estimated fixation rate of 400 – 850 kg N ha -1 year -1 by C. equisetifolia (e.g. Sougoufara et al., 1990).

Many experienced researchers point out the importance of selecting suitable refer-ence plants and obtaining δ15N greater than N2-fixing plants in order to accurately estimate the amount of N2 derived from atmosphere (e.g. Herridge & Danso, 1995; Shearer & Kohl, 1986; Yoneyama et al., 1993). Even though three of the reference plants used met the se-lection criteria (Domenach et al., 1989; Peo-ples et al., 1989), E. pellita and Macaranga pleiostemona were useful in detecting N2 fixa-tion by C. oligodon. Theoretically, %Ndfa esti-mates depend on the δ15N values of reference plants being larger than that of N2-fixing spe-cies and close to that of the soil (Herridge, pers. comm., 2006). However, the δ15N values obtained for the reference plants were too small and more negative compared to the δ15N values of C. oligodon to satisfy Equation (C) for determining the proportional contribution of atmospheric N2 by C. oligodon in all the sites under tree fallows.

The results suggest that isotopic fractiona-

tion and discrimination cause changes in δ15N abundance of the N sources available to reference plants and C. oligodon (Ledgard & Peoples, 1988; Shearer & Kohl, 1988; Yoneyama et al., 1993; Boddey et al., 2000). According to Ledgard & Peoples (1988), isotope fractionation during trans-formations of soil N is inevitable and this alteration produces a net effect of 15N abundance in the plants that affects estimates of N2 fixation. Isotopic discrimination occurs when biological processes are in favour of 14N (lighter isotope) over 15N

Wemin, Reid & Edis


(δ15Nref – δ15Ncas) (C) % Ndfa = 100 x ______________ (δ15Nref - B)

45

Wemin, Reid & Edis


(heavier isotope) (Rennie & Rennie, 1983). Nevertheless, large discrepancies in the esti-mates (e.g. negative %Ndfa) under short fal-low of casuarinas are probably due mainly to temporal changes in 15N enrichment effect rather than fractionations of 15N abundance in the soils (Ledgard & Peoples, 1988). Because data do not come from controlled experiments, it is difficult to determine the cause precisely.

Where the δ15N value of C. oligodon is smaller than δ15N of the reference plants, N2 fixation observed is sufficient under the pre-vailing conditions. However, where it is greater than that of the reference plants, N2 fixation detected is quite small or suppressed. Although many workers obtain high positive δ15N for the reference plants (e.g. Domenach et al., 1989; Yoneyama et al., 1990, 1993), Sylla et al. (2002) report that very negative δ15N val-ues are usually found in plants grown in cooler climates. They also suggest that plants depend-ing entirely on soil N have δ15N values close to that of soil N. Discriminate uptake of the two most accessible forms of mineral N (NO3

- and NH4

+) by the plants is likely to affect δ15N val-ues of the samples as well. For example, Fal-kengren-Grerup et al. (2004) linked lower δ15N values in some plant species (e.g. Aegopodium podagraria, Geum urbanum) to higher assimi-lation of NO3

- whilst higher δ15N values in other species (e.g. Poa nemoralis, Deschamp-sia flexuosa) to higher assimilation of NH4

+. A B value specifically for C. oligodon

has not been determined. This reduced the ac-curacy of N2 fixation estimate using 15N natu-ral abundance method in this study. As B value can only be derived by growing the N2-fixing species in N-free medium, it was not possible to achieve that in the given time. As such a literature B value for C. equisetifolia was used. Boddey et al. (2000) suggest that B values for calculating BNF of field-grown plants should be preferably derived from mature plants, which would normally have δ15N at equilib-rium. However, the literature B value used in this study was most probably obtained by growing C. equisetifolia under laboratory con-ditions (Mariotti et al., 1992).

Effect of tree on N2 fixation

The general assumption is that the rate of N2 fixation quantified per hectare depends

on the number of trees per hectare (Gauthier et al., 1985). However, negative correlation (r2 = 0.31) between tree stocking rates and N2 fixa-tion by C. oligodon (Figure 3) suggests that this may not occur across a range of sites and soils. Regardless of high stocking rates of some sites, the rate of N2 fixation in all sites under short fallow and most sites under long fallow of casuarina remained below 10 kg N ha –1 year-1. On the other hand, a few sites un-der long fallow of casuarina with low tree stocking rates showed relatively high rates of N2 fixation in excess of 25 kg N ha -1 year-1.

Parrotta (1999) established a direct re-lationship between basal area and stand age of C. equisetifolia and found that as tree age in-creased from 1 to 7.5 years, basal area in-creased proportionally from 10- to 45 m2 ha -1. Since mature forests produce larger biomass and accumulate significant amounts of nutri-ents (especially N), it seems logical to assume that larger basal areas and tree volume will reflect higher rates of N2 fixation. The present study showed relatively high rates of N2 fixa-tion under long fallows with higher basal area (≥ 5 m2 ha –1) and greater tree volume (≥ 140 m3 ha-1). On the other hand, moderately low rates of N2 fixation (≤ 5 kg N ha –1 year-1) were found in trees under short fallows with basal area of ≤ 15 m2 ha -1 and tree volume ≤ 140 m3 ha-1 showed (Figures 4a and 4b).

Although trees influence soil physical and chemical properties (Young, 1997; Pa-rotta, 1999), the bulk of the ‘fixed’ N is stored in above-ground biomass. This organic N may be slowly released to the soil by the standing trees or become available when the fallow is cleared and biomass is retained on site. N2 fixation estimate using δ15N soils

Some workers used soil δ15N to deter-mine biologically ‘fixed’ N (e.g. Chalk et al., 1983; Chen & Zhu, 1986; Bergersen et al., 1989). The use of soil δ15N in this study was necessary due to more negative δ15N values of the reference plants and very small difference in δ15N values between reference plants and C. oligodon to adequately detect N2 fixation esti-mates. Estimates based on soil δ15N only re-flects retention of biologically ‘fixed’ N in the soils under C. oligodon. For this reason inter-pretation is treated with caution. However, it is

46

3

-4

74

13 3

-8

36

-4

28

-1

-10

0

10

20

30

40

N 'f

ixed

' (kg

ha-1

year

-1)

1 2 3 4 5 6Casuarina fallow sites

Long fallow Short fallow

Figure 2. Estimate of the amount of N ‘fixed’ by Casuarina oligodon during fallow period.

y = -0.0038x + 18.533R2 = 0.31

-20

-10

0

10

20

30

40

0 1 2 3 4 5 6 7 8

Thousands

Stocking rate (trees ha-1)

N 2 'f

ixed

' (kg

N h

a-1 y

ear-1

)

Figure 3. A negative correlation between N2 fixation and tree stocking rate during the fallow period in PNG central highlands

Wemin, Reid & Edis


reasonable to assume that ‘fixed’ N in the tree biomass of C. oligodon is returned to the soil and sufficiently mineralized.

On the basis of soil δ15N analysis, soils under short fallow of C. oligodon accumulated large proportions of biologically ‘fixed’ N of between 91 and 92 per cent in the top soil and

sub-soil, respectively. In the long fallow of C. oligodon, soils appear to retain slightly lower proportions of ‘fixed’ N between 78 and 86 per cent in the upper 10 cm and 10-30 cm depths, respectively. Assuming these propor-tions form the fraction of total N and if ex-pressed in mass, short fallow of casuarinas

47

y = 0.91x - 6.2281R2 = 0.27

-15

-10

-5

0

5

10

15

20

25

30

35

40

0 5 10 15 20 25 30 35

Tree Basal Area (m2 ha-1)

N2

' fixed

' (kg

N h

a-1

yea

r-1)

Figure 4a. Relationship between N2 fixation and tree basal area during the fallow period in PNG central highlands.

y = 0.0932x - 3.3334R2 = 0.59

-15-10

-505

10152025303540

-10 90 190 290 390 490

Tree Volume (m3 ha-1)

N 2 'f

ixed

' (kg

N h

a-1

yea

r-1)

Figure 4b. Relationship between N2 fixation and tree volume during the fallow period in PNG central highlands.

Wemin, Reid & Edis


would have between 6.2 – 5.3 mg N g -1 soil and long fallow would have 4.8 - 3.7 mg N g -1

soil that originate from N2 fixation. The low proportion of ‘fixed’ N in the soil could mean that bulk of the ‘fixed’ N is retained in the above-ground biomass and only a small frac-tion is returned to the soil through its dead tis-sues. It also suggests that ‘fixed’ N is lost

through such means as leaching, product re-moval and denitrification. On the other hand, high retention suggest better quality of dead tissues from much younger casuarina plants which presumably mineralized effectively and liberated ‘fixed’ N to the soils.

As noted, the proportion of N derived from ‘fixed’ N (%Ndfn) increased as soil

48

depth increases while the actual amount by mass decreased with depth. These differences were caused by differences in δ15N of soils at each measured depth. Surface soils have smaller δ15N values than subsoil. Sylla et al. (2002) report that differences in δ15N of soils are attributed to spatial variability such as soil types, soil profile characteristics and forms of available N.

CONCLUSIONS

Casuarina oligodon derived as much as 70 per cent of its N from the atmosphere. The rate of N2 fixation was relatively low in the short fallows of casuarina and increased up to 36 kg N ha -1 year -1 as the trees mature. This may reflect more N fixation occurring in re-sponse to impoverishment of soil N fertility with time. In spite of C. oligodon’s ability to contribute N in relatively N deficient soils, in-creased N2 fixation in long fallow of casua-rinas may be less beneficial to the farming communities with increased land pressure for food crop production, but seems useful for per-ennial crop combinations such as coffee/casuarina.

The study suggests that bulk of the N2 captured by fixation is stored in above-ground biomass. How the standing biomass is man-aged, particularly when the fallow is converted back to the cropping phase, is therefore critical in ensuring that the farmers are able to maxi-mize the benefit. Whilst the removal of stem wood for use as fuel or building material may be an important product of the agroforestry system, retaining the foliage, small branches and bark on the site could be critical in im-proving soil fertility and sustaining agricultural productivity.

Although casuarina fallows tended to accumulate more soil N and organic C than grass fallows, the amount of N and C in the surface soils across all systems under study showed no significant difference (P>0.05). The N fertility under long fallows (11 – 15 years+) was generally greater than that under short fal-lows of C. oligodon (5 - 10 years). Total N and C were mainly concentrated in the upper soil horizon and decreased with increasing soil depth under all systems.

Natural Frankia strain present in the

PNG highlands soils is not established. A fur-ther investigation into the natural Frankia population or introduction of more effective Frankia species, infectivity of host plant and nodulation efficiency of casuarina would im-prove N2 fixation estimates in C. oligodon for enhancement of agricultural productivity in the region. Furthermore, a value of B for C. oli-godon was not found while studying the litera-ture, suggesting that further research should include the determination of a B value to im-prove the estimate of N2 fixation. The B value can be obtained by growing C. oligodon in N-free medium where the only source of N is at-mospheric N2. Thorough investigation of BNF process of C. oligodon can help farmers deter-mine N requirements for other crops in agro-forestry systems and sustain use of available and diminishing land resources.

ACKNOWLEDGEMENTS

The Asian Development Bank – Japan provided the scholarships to the principal au-thor while undertaking postgraduate studies at The University of Melbourne. The authors wish to acknowledge The University of Mel-bourne for making it possible for the fieldwork in Papua New Guinea (PNG) with the student research support grants. The National Agricul-tural Research Institute (NARI) of PNG pro-vided logistical support for conducting the fieldwork. Drs. Stefen Arndt, Deli Chan and Mr. Ron Teo of The University of Melbourne provided invaluable technical assistance and laboratory analysis of the samples. Mr. Mark Tinah of NARI PNG produced the map of the study sites while Mr. Torry Kuria and K. De-mas of PNG Forestry Research Institute identi-fied the reference plant species. This paper is an extract from a master’s thesis which was proof-read by Prof. Zhihong Xu of The Uni-versity of Melbourne and Dr. Michael Bourke of The Australian National University. We are very grateful to them. We also thank many in-dividuals who have helped with the fieldwork in PNG. We gratefully acknowledge the con-sent of villagers whose fields were used in this study.

Wemin, Reid & Edis


49

REFERENCES

ADAMS, M. A. & ATTIWILL, P. M. 1986. Nutrient cycling and nitrogen mineralization in euca-lyptus forests of Southeastern Australia. Plant and Soil, 92:319-339. ANGUS, J. F. 2001. Nitrogen supply and demand in Australian agriculture. Australian Journal of Experimental Agriculture, 41:277-288. ATTIWILL, P. M. & LEEPER, G. W. 1987. Forest Soils and Nutrient Cycling. Melbourne Uni-versity Press, Melbourne, 202 pp. BELLAMY, J. A. & McALPINE, J. R. 1995. Papua New Guinea Inventory of Natural Resources, Population Distribution and Land Use Handbook. 2nd edition. PNGRIS Publication No. 6, Can-berra, Australia, 155 pp. BERGERSEN, F. J., BROCKWELL, J., GAULT, R. R., MORTHORPE, L., PEOPLES, M. B. & TURNER, G. L. 1989. Effects of available soil nitrogen and rates of inoculation on nitrogen fixa-tion by irrigated soybean and evaluation of δ15N methods for assessment. Australian Journal of Agricultural Research, 40:763-780. BINO, B. & KANUA, M. 1996. Growth Performance, Litter Yield and Nutrient Turnover of Casuarina oligodon in the Highlands of Papua New Guinea. In: PINYOPUSARERK, K., TURNBULL, J. W. & MIDGLEY, S. J. Recent Casuarina Research and Development. CSIRO, Canberra, pp. 167-170. BLEEKER, P. 1983. Soils of Papua New Guinea. CSIRO, Australia, 352 pp. BODDEY, R. M., PEOPLES, M. B., PALMER, B. & DART, P. J. 2000. Use of the 15N natural abundance technique to quantify biological nitrogen fixation by woody perennials. Nutrient Cy-cling in Agroecosystems, 57:235-270. BOURKE, R. M. 1989. Food, coffee and casuarina: an Agroforestry system from the Papua New Guinea highlands. In: NAIR, P. K. R. (Ed.), Agroforestry Systems in the Tropics. Kluwer Aca-demic Publishers, Dordrecht, pp. 269-275. BOURKE, R. M. 2007. Managing the species composition of fallows in Papua New Guinea. By plant-ing trees. In: CAIRNS, M. F. (Ed.), Voices from the Forest: Integrating Indigenous Knowledge into Sustainable Upland Farming. Resources for the Future Press, Washington, DC, pp. 379-388. BOURKE, R. M. & ALLEN, B. 2009. Village food production systems. In: BOURKE, R. M. & HAR-WOOD, T. (Eds.), Food and Agriculture in Papua New Guinea. ANU E Press, The Australian National University, Canberra, pp. 193-269. BREMNER, J. M. & MULVANEY, C. S. 1982. Nitrogen – Total. In: PAGE, A. L., MILLER, R. H. & KEENEY, D. R. (Eds.), Methods of Soil Analysis: Part II Chemical and Microbiological Properties. 2nd edition. American Society of Agronomy, Maddison, pp. 595-624. BREMNER, J. M. & YEOMANS, J. C. 1988. Advances in nitrogen methodology: Laboratory tech-niques for determination of different forms of nitrogen. In: WILSON, J.R. (Ed.), Advances in Nitrogen Cycling in Agricultural Ecosystems. C.A.B. International, Wallingford, UK, pp. 339-414. CHALK, P. M., DOUGLAS, L. A. & BUCHANAN, S. A. 1983. Use of 15N enrichment of soil miner-alisable N as a reference for isotope dilution measurements of biologically fixed nitrogen. Canadian

Wemin, Reid & Edis


50

Journal of Microbiology, 29:1046-1052. CHEN, D. L. & ZHU, Z. L. 1986. Nitrogen supply of subsoil in flooded rice fields. Soils (China), 18:36- 37. DANSO, S. K. A., BOWEN, G. D. & SANGINA, N. 1992. Biological nitrogen fixation in trees in agroecosystems. Plant and Soil, 141:177-196. DOMENACH, A. M., KURDALI, F. & BARDIN, R. 1989. Estimation of symbiotic dinitrogen fixation in alder forests by the method based on natural 15N abundance. Plant and Soil, 118:51-59. FALKENGREN-GRERUP, U., MICHELSEN, A., OLSSON, M. O., QUARMBY, C. & SLEEP, D. 2004. Plant nitrate use in deciduous woodland: the relationship between leaf N, 15N natural abundance of forbs and soil mineralisation. Soil Biology and Biochemistry, 36:1855-1891. FRIED, M., DNASO, S. K. A. & ZAPATA, F. 1983. The methodology of measurement of N2 fixation by nonlegumes as inferred from field experiments with legumes. Canadian Journal of Microbiology, 29:1053-1062. GAUTHIER, D., DIEM, H. G. & DOMMERGUES, Y. R. 1985. Assessment of N2 – fixation by Casua-rina equisetifolia inoculated with Frankia ORS021001 using 15N methods. Soil Biology and Biochemis-try, 17:375-379. GUNN, R. H., BEATTIE, J. A., REID, R. E. & van de GRAAFF, R. H. M. (Eds.) 1988. Australia Soil and Land Survey Hand book: Guidelines for Conducting Surveys. Inkata Press, Melbourne, 300 pp. HANSON, L. W., ALLEN, B. J., BOURKE, R. M. & McCARTHY, T. J. 2001. Papua New Guinea Rural Development Handbook. The Australian National University, Canberra, 326 pp. HARTEMINK, A. E. & BOURKE, R. M. 2000. Nutrient Deficiencies of agricultural crops in Papua New Guinea. Outlook on Agriculture, 29 (2):97-108. HENSON, A. P. & PATE, J. S. 1987. Evaluation of the 15N natural abundance method and xylem sap analysis for assessing N2 fixation of understory legumes in Jarrah (Eucalyptus marginata Donn ex Sm.) forest in S.A. Australia. Journal of Experimental Botany, 38:1466-1457. HERRIDGE, D. F & DANSO, S. K. A. 1995. Enhancing crop legume N2 fixation through selec-tion and breeding. Plant and Soil, 174:51-82. ISSAC, L., WOOD, C. W. & SHANNON, D. A. 2003. Hedgerow species and environmental con-ditions effect on soil total C and N and mineralization patterns of soils amended with their prun-ing. Nutrient Cycling in Agroecosystems, 65:73-87. LEDGARD, S. F. & PEOPLES, M. B. 1988. Measurement of nitrogen fixation in the field. In: WILSON, J. R. (Ed..), Advances in nitrogen cycling in agricultural ecosystems. C.A.B. Interna-tional, Wallingford, UK, pp. 351-367. MARIOTTI, A., SOUGOUFARA, B. & DOMMERGUES, Y. R. 1992. Estimation of nitrogen-fixation using the natural abundance method in a plantation of Casuarina equisetifolia (Forst). Soil Biology and Biochemistry, 24:647-653. MARTIUS, C., HOFER, H., GRACIA, M. V. B., ROMBKE, J. & HANGARTH, W. 2004. Litter fall, litter stocks and decomposition rates in rainforests and Agroforestry sites in central Amazo-nia. Nutrient Cycling in Agroecosystems, 68:137-154.

Wemin, Reid & Edis


51

McALPINE, J. R., KEIG, G. & FALLS, R. 1983. Climate of Papua New Guinea. CSIRO, Austra-lia, 200 pp. MULVANEY, R. L. 1993. Mass Spectrometry. In: KNOWLES, R. & BLACKBURN, T.H. (Eds.), Nitrogen Isotope Techniques. Academic Press, Inc. San Diego, CA, pp. 11-57. PANDEY, C. B., SINGH, A. K. & SHARMA, D. K. 2000. Soil properties under Acacia nilotica trees in a traditional agroforestry system in central India. Agroforestry Systems, 49:53-61. PARROTTA, J. A. 1999. Productivity, nutrient cycling, and succession in single- and mixed-species plantations of Casuarina equisetifolia, Eucalyptus robusta, and Leucaena leucocephala in Puerto Rico. Forestry Ecology and Management, 124:45-77. PARROTTA, J. A., BAKER, D. D. & FRIED, M. 1994. Application of 15N-enrichment method-ologies to estimate nitrogen fixation in Casuarina equisetifolia. Canadian Journal of Forestry Re-search, 24: 201- 207. PARROTTA, J. A., BAKER, D. D. & FRIED, M. 1996. Changes in dinitrogen fixation in matur-ing stands of Casuarina equisetifolia and Leucaena leucocephala. Canadian Journal of Forestry Research, 26:1684-1691. PEOPLES, M. B., FAIZAH, A. W., RERKASEM, B. & HERRIDGE, D. F. (Eds.) 1989. Methods for Evaluating Nitrogen Fixation by Nodulated Legumes in the Field. Australian Centre for Inter-national Agricultural Research, Canberra, 76 pp. POWER, I. L., THORROLD, B. S. & BALKS, M. R. 2003. Soil properties and nitrogen availabil-ity in silvopastoral plantings of Acacia melanoxylon in North Island, New Zealand. Agroforestry Systems, 57:255-237. RENNIE, R. J. & RENNIE, D. A. 1983. Techniques for quantifying N2 fixation in association with nonlegumes under field and greenhouse conditions. Canadian Journal of Microbiology, 29:1022- 1035. SCHAETZL, R. J. & ANDERSON, S. 2005. Soils: Genesis and Geomorphology. Cambridge Uni- versity Press, NY, 840 pp. SHEARER, G. & KOHL, D. H. 1986. N2-fixation in field settings: Estimations based on natural 15N abundance. Australian Journal of Plant Physiology, 13:699-756. SHEARER, G. & KOHL, D. H. 1988. Natural 15N abundance as a method of estimating the contribution of biologically fixed nitrogen to N2-fixing systems: potential for non-legumes. Plant and Soil, 110:317- 327. SHEARER, G. & KOHL, D. H. 1993. Natural abundance of 15N: Fractional contribution of two sources to a common sink and use to a isotope discrimination. In: KNOWLES, R. & BLACK-BURN, T. H. (Eds.), Nitrogen Isotope Techniques. Academic Press Inc. NY, pp. 89-125. SIERRA, J., DULORMNE, M. & DESFONTAINES, L. 2002. Soil nitrogen as affected by Glir-cidia sepium in a silopastoral system in Guadeloupe, French Antilles. Agroforestry Systems, 54:87- 97. SMITH, A. P. 1999. Nitrogen cycling and grain legume nitrogen fixation in ameliorated sodic

Wemin, Reid & Edis


52

soils. M.Agri.Sc. The University of Melbourne. SOUGOUFARA, B., DANSO, S. K. A., DIEM, H. G. & DOMMERGUES, Y. R. 1990. Estimat-ing N2 fixation and N derived from soil by Casuarina equisetifolia using labeled 15N fertilizer: some problems and solutions. Soil Biology and Biochemistry, 22:695–701. SPARLING, G. 1998. Forest into pasture and pasture into forest: Impacts of soil organic matter and nutrient contents. In: NOBLE, A. D. & RANDALL, P. J. How Trees Affect Soil. Rural Indus-tries Research and Development Corporation. Barton, ACT, 124 pp. STAHL, L., HOGBERG, P., SELLSTEDT, A. & BURESH, R. J. 2005. Measuring nitrogen fixa-tion by Sesbania sesban planted fallows using 15N tracer technique in Kenya. Agroforestry Sys-tems, 65:67-79. SYLLA, S. N., NDOYE, I., GUEYE, M., BA, A. T. & DREYFUS, B. 2002. Estimates of biologi-cal nitrogen fixation by Pterocarpus lucens in a semi arid natural forests park in Senegal using 15N natural abundance method. African Journal of Biotechnology, 1:50-56. TALOPA, P. 2003. Role of Casuarina in the Agroforestry systems in Papua New Guinea: our knowledge about Casuarina oligodon L.Johnson. Coffee Industry Newsletter. No. 2, pp. 11-14. THIAGALINGAM, K. 1983. Role of Casuarina in Agroforestry. In: MIDGLEY, S. J., TURNBULL, J. W. & JOHNSTON, R. D. (Eds.), Casuarina Ecology, Management and Utiliza-tion. CSIRO, Melbourne, pp. 175-179. WANI, S. P., RUPELA, O. P. & LEE, K. K. 1995. Sustainable Agriculture in the semi-arid tropics through biological nitrogen fixation in grain legumes. Plant and Soil, 174:29-49. YONEYAMA, T., MURAKAMI, T., BOONKERD, N., WADISIRISUK, P., SIRIPIN, S. & KOUNO, K. 1990. Natural 15N abundance in shrub and tree legumes, Casuarina, and non-N2 fix-ing plants in Thailand. Plant and Soil, 128:287-292. YONEYAMA, T., MURAOKA, T., MURAKAMI, T. & BOONKERD, N. 1993. Natural abun-dance of 15N in tropical plants with emphasis on tree legumes. Plant and Soil, 153:295-304. YOUNG, A. 1997. Agroforestry for soil management. 2nd edition. CAB International, Wallingford, UK, 320 pp.

Wemin, Reid & Edis


53

Relative impacts of some insecticides on infestation and dam-age by leaf-eating caterpillars and on non-target arthropod diversity in head cabbage (Brassica oleracea var. capitata) crops in Samoa

Adama A. Ebenebe1 , Kelemeni Navucu1 and David J. Hunter1

1C/o The University of the South Pacific, School of Agriculture & Food Technology, Alafua Campus, Samoa. Corresponding author email: [email protected]

ABSTRACT

Field experiments were conducted at Alafua, Samoa, to compare the relative efficacies of eight commercial insecticides for controlling leaf-eating caterpillars, as well as for relative impacts on non-target arthropod diversity in head cabbage crops. The insecticides evaluated were: ‘Attack’ (pirimiphos methyl + permethrin), ‘Bifenthrin 8SC’ (bifenthrin), ‘Steward 150SC’ (indoxacarb), ‘Suncis 25EC’ (deltamethrin), ‘Delfin WG’ (Bacillus thuringiensis), ‘Orthene’ (acephate), ‘Suncloprid 20SL’ (imidacloprid), and ‘Sunthene 75SP’ (acephate). Each insecticide was applied once a week at manufacturer’s recommended dilution. Weekly observa-tions were made for incidence of damage and infestation by leaf–eating caterpillars, and for types of non-target arthropods on cabbage plants and in pitfall traps. Severity of damage and yield of cabbage were measured at harvest. Results obtained indicated that each of the insecticides evalu-ated, except ‘Suncloprid 20SL’, reduced damage by leaf-eating caterpillars to significantly lower levels compared to the untreated control (P<0.001). However, results also showed that ‘Bifenthrin 8SC’, ‘Steward 150SC’, and ‘Suncis 25EC’ gave the best results under high pest pressure. None of the insecticides appeared to exert serious negative impacts on non-target arthropods in the cabbage crops. We recommend that ‘Bifenthrin 8SC’ and ‘Suncis 25EC’, which are not sold in Samoa at present, should be made available to cabbage farmers for use in IPM programmes.

Key words: insecticides, cabbage, leaf-eating caterpillars, non-target arthropods, Samoa.

INTRODUCTION

The range of locally produced leafy vegetables sold in the local markets in Samoa is quite limited, consisting mainly of cabbages, watercress, and taro leaves. In view of the im-portance of vegetables in the human diet, as well as their contribution to household income generation, the Samoa Ministry of Agriculture and Fisheries (MAF) has been supporting ef-forts to increase production of vegetables in the country (Falevaai, 2010; Memea, 2010). Although these efforts cover a wide variety of vegetables, cabbages (Brassica oleracea var.

capitata and B. chinensis) are still the most widely produced and marketed leafy vegeta-bles in the country.

Production of cabbages in Samoa is often constrained by various factors including climatic factors, cost of production inputs, dis-eases and, especially, invertebrate pests. In-sects (mainly leaf-eating caterpillars), the giant African land snail (Achatina fulica Bowdich), and slugs are the major invertebrate pests of cabbages in Samoa. Together, these constraints ensure that the supply of locally produced cab-bages is often erratic and prices remain rela-tively high for much of the year.



54

Leaf-eating caterpillars which have been reported on cabbages in Samoa are Cro-cidolomia pavonana (Fabricius) (Amosa, 2005; Ebenebe et al., 2006), Hellula sp., Plutella xylostella L., Spodoptera litura (Fabricius) (Taefu, 1977; Hollingsworth et al., 1984; Hollingsworth and Aloali’i, 1985; Amosa, 2005; Ebenebe et al., 2006), and occa-sionally Trichoplusia sp. (Ebenebe et al., 2006). There is very little published informa-tion on the extent of damage by these pests in Samoa. However, Ebenebe et al. (2006) have reported peak incidences of infestation of up to 72.8 % by C. pavonana, 27.1 % by H. undalis, 13.3 % by S. litura, 2.2 % each by P. xylostella and Trichoplusia sp., and up to 100 % yield loss in untreated head cabbage crops. In an earlier study, Hollingsworth et al. (1984) had recorded a 22 % peak incidence of Hellula sp. These previous reports demonstrate that out-breaks of leaf-eating caterpillars in cabbage crops in Samoa can reach levels that warrant direct pest control intervention.

Cabbage growers in Samoa, especially those who grow for sale, often resort to insecti-cide applications to minimise crop loss due to leaf-eating caterpillars. The usual farmer prac-tice is weekly spraying of insecticides (Robinson, 2010), often referred to as pro-gramme-spraying.

In an effort to improve cabbage pro-duction, Samoa MAF is currently focusing re-search efforts into the development of inte-grated pest management (IPM) strategies for managing cabbage caterpillars. The IPM ap-proach is focusing on pest monitoring and in-secticide application when pest numbers reach a threshold level of 10 % infestation. The pro-gramme aims to discourage the current farmer practice of weekly spraying of insecticides.

Although insecticides are relatively easy to acquire in Samoa, the range of prod-ucts is limited and individual products are of-ten out of stock. This situation encourages re-peated applications of the same products. Both programme spraying and repeated use of the same product are not consistent with the prin-ciples of IPM and pesticide resistance manage-ment which advocate the rotational use of pes-ticides with different chemical properties, ap-plied only when needed. There is a need to

broaden the range of products available for farmers to practice rotational application.

In this paper, we report the findings of a study conducted to evaluate the relative effi-cacies of some insecticides against leaf-eating caterpillars in cabbages in Samoa. In addition, potential impacts of the insecticides on non-target arthropods in cabbage crops treated with the insecticides were also assessed. The overall aim of the study was to identify insecticides that could be introduced to broaden the selec-tion of products available for rotational appli-cation against leaf-eating caterpillars of cab-bages in Samoa.


Two field trials were conducted at the School of Agriculture and Food Technology, The University of the South Pacific, Alafua Campus, Alafua (15 – 17.S, 171 – 173.W), Sa-moa. The first and second trials were done dur-ing May to July 2008, and August to October 2008, respectively.

For each trial, seedlings of head cab-bage (variety ‘KK cross’) were raised in boxes in a nursery and later transplanted into a field plot at four weeks after germination. The field plot consisted of 36 sub-plots, 3.25 m x 3.25 m each, arranged in four blocks (replications) of nine (to correspond with nine experimental treatments). Spacing between plants was 50 cm x 50 cm. Plots were maintained by regular practices (hand-weeding, watering, fertiliser application) as needed, but in equal measure for all treatments.

At two weeks after transplanting, eight insecticides (treatments) were applied into separate sub-plots in a randomized complete block design. The insecticides evaluated were:

‘Attack’ (475 g/l pirimiphos methyl + 25 g/l permethrin in the form of emulsifiable concen-trate), ‘Bifenthrin 8SC’ (bifenthrin 80 g/l), ‘Steward 150SC’ (150 g/l indoxacarb in the form of suspension concentrate), ‘Suncis 25EC’ (2.5 % deltamethrin), ‘Delfin WG’ (Bacillus thuringiensis subspecies kurstaki), ‘Orthene’ (970 g/kg acephate), ‘Suncloprid 20SL’ (20 % w/v imidacloprid), and ‘Sunthene 75SP’ (750 g/kg acephate). Of these eight insecticides, only ‘Attack’, ‘Steward’ and ‘Orthene’ are currently sold in

Ebenebe, Navucu & Hunter


55

Samoa. Each insecticide was diluted according

to manufacturer’s recommendation (‘Attack’ at 40 ml/20 L knapsack sprayer, ‘Bifenthrin’ at 20 ml/20 L knapsack sprayer, ‘Steward’ at 0.5 ml/L of water, ‘Suncis’ at 8 ml/20 L knap-sack sprayer, ‘Delfin’ at 5 g/15 L of water, ‘Orthene’ at 15 g/20 L of water, ‘Suncloprid’ at 50 ml/L of water, and ‘Sunthene’ at 15 g/20 L knapsack sprayer). Each insecticide was ap-plied to target plants with a 3-L handheld sprayer. Spray mixture was directed to individ-ual plants until thorough wetting of plant sur-face was achieved. All applications were re-peated on the same day each week, up till nine weeks after transplanting (one week before harvest). An untreated control was included as the ninth treatment in the experimental design.

Observations made in both Trial 1 and Trial 2 were: (1) incidence of damage by leaf-eating caterpillars, (2) incidence of infestation by leaf-eating caterpillars, (3) occurrence of non-target arthropods in cabbage plant canopy, (4) types and numbers of arthropods caught in pitfall traps, and (5) yield of cabbage (percentage of cabbage plants with marketable heads and average weight per head). In addi-tion, data was recorded for severity of damage by leaf-eating caterpillars in Trial 2 (at har-vest), using a severity rating scale as follows:

0 = no damage 1 = slight damage on outer leaves 2 = slight damage on wrapper leaves 3 = moderate damage, 2 wrapper leaves must be removed 4 = heavy damage, more than 3 wrapper leaves must be removed 5 = very severe damage, no harvestable head

Data on incidence of damage, inci-dence of infestation (Trial 1 only), and yield were analysed by ANOVA and LSD using GenStat Discovery Edition 3 statistical soft-ware. Incidence of infestation in Trial 2 was analysed by Friedman’s test, and data on di-versity of non-target arthropods were analysed using Simpson’s reciprocal index (1/D). The level of significance for statistical analyses was P≤0.05.


Incidence of damage by leaf-eating caterpil-lars

Results obtained (Table 1) show that incidences of damage in all insecticide treat-ments, except ‘Suncloprid’, were significantly lower than in the untreated control in Trial 1. This suggested that insecticide treatments suc-cessfully reduced the occurrence of caterpillar damage in treated plots. In Trial 2, only ‘Bifenthrin’ and ‘Steward’ treatments had sig-nificantly lower incidences of damage than the untreated control by harvest (Table 1). ‘Steward’ gave the best protection against cat-erpillar damage in this trial, while ‘Suncloprid’ had the same incidence of damage (100 %) as the untreated control.

Figure 1 and Figure 2 show the comparative progression of incidences of damage by leaf-eating caterpillars during Trial 1 and Trial 2, respectively. The figures show that in both seasons, leaf-eating caterpillars began attack-ing the cabbages soon after transplanting and incidences of damage increased steadily as the seasons progressed. However, whereas inci-dences of damage remained relatively low in all insecticide treatments except ‘Suncloprid’ during Trial 1 (Figure 1), only ‘Steward’ main-tained a relatively low incidence during Trial 2 (Figure 2).



Table 1. Incidence (%) of damage by leaf-eating cater-pillars at harvest.

Treatment Trial 1 Trial 2 Untreated (Control) 77.2a 100.0a Attack 2.8c 87.5ab Bifenthrin 2.8c 80.6b Delfin 35.8b 94.5ab Orthene 23.9bc 97.2ab Steward 6.2c 33.3c Suncis 0.0c 86.1ab Sunthene 16.3bc 94.5ab Suncloprid 64.6a 100.0a P-value <0.001 <0.001

Means in the same column with the same letter are not significantly different (P>0.05).

56

0

10

20

30

40

50

60

70

80

90

2 3 4 5 6 7 8 9 10

Weeks after transplanting

% dam

aged

plants

Control

Steward

Orthene

Bifenthrin

Attack

Sunthene

Suncis

Delfin

Suncloprid

Figure 1. Progress of incidence of damage (%) by leaf-eating caterpillars - Trial 1.

0

10

20

30

40

50

60

70

80

90

100

2 3 4 5 6 7 8 9 10

Weeks after transplanting

% dam

aged

plants

Control

Steward

Orthene

Bifenthrin

Attack

Sunthene

Suncis

Delfin

Sunchloprid

Figure 2. Progress of incidence of damage (%) by leaf-eating caterpillars - Trial 2.



57

Variance component analysis showed that incidences of damage were significantly

higher during Trial 2 than Trial 1 (Figure 3).

0

10

20

30

40

50

60

70

80

90

100

Control

Steward

Orthene

Bifenthrin

Attack

Sunthene

Suncis

Delfin

Suncloprid

Treatments

% dam

aged

plants

Trial 1

Trial 2

Figure 3. Comparison of incidence of damage in Trial 1 and Trial 2.

χ2 (p<0.001)

The reason for the higher incidence of damage observed during Trial 2, relative to Trial 1, was not investigated. However, it is well documented that climatic conditions can influence pest occurrence. For example, it has been reported from Malaysia and Indonesia that populations of diamondback moth (Plutella xylostella) are higher during drier periods (Ooi, 1986; Sastrodihardjo, 1986). Therefore, the observed lower levels of dam-age during Trial 1, relative to Trial 2, might be partly attributable to the generally higher pre-cipitation experienced during Trial 1 (total monthly: 132.1 mm, 536.4 mm, and 77.9 mm during May, June and July, respectively) than during Trial 2 (total monthly: 23.3 mm, 92.1 mm, and 155.6 mm during August, September and October, respectively). However, it is also noteworthy that no brassica crop was planted in the vicinity of the experimental area for about five months prior to the commencement of Trial 1. This probably meant that popula-tions of leaf-eating caterpillars were generally

low in this area at the start of Trial 1, thereby resulting in slower rate of population build up in the experiment. In contrast, there were other brassica crops growing in the location when Trial 2 was planted, which might have contrib-uted to the generally higher incidences of dam-age observed in this trial relative to Trial 1. Whatever the reason for the observed differ-ences, one conclusion that can be drawn from this study is that the rate of development of leaf-eating caterpillar damage may vary from crop to crop. Consequently, monitoring should be an integral practice in the management of these pests so that specific management ac-tions are taken only when necessary. A thresh-old of 10 % infestation is currently being adopted by the Crops Division of MAF, Sa-moa, under an Australian Centre for Interna-tional Agricultural Research (ACIAR)-funded ‘Integrated Pest Management (IPM) of Bras-sica crops’ project (Robinson, 2010).



58

Incidence of infestation by leaf-eating cater-pillars

Three species of leaf-eating caterpillars were observed to be causing damage in cab-bage during Trial 1. They were C. pavonana, P. xylostella and H. undalis, with C. pavonana being the most prevalent species (based on data from untreated control plots) (Table 2). Analysis of incidences of infestation recorded

during this trial showed that the insecticides were mostly able to keep infestations by each caterpillar species at significantly lower levels than the untreated control (Table 2). The ex-ceptions were ‘Orthene’ and ‘Suncloprid’ against P. xylostella, and ‘Suncloprid’ against H. undalis. Overall, ‘Suncloprid’ was the least effective in controlling infestations of the cat-erpillars.

The observation that C. pavonana was the most frequently encountered species during Trial 1 agrees with earlier findings by Ebenebe et al. (2006). However, S. litura, which is an-other leaf-eating caterpillar reported to attack cabbages in Samoa (Taefu, 1977; Hollingsworth et al., 1984; Hollingsworth & Aloali’i, 1985; Amosa, 2005; Ebenebe et al., 2006), was not encountered in Trial 1. This indicates that this species does not always con-tribute to damage in cabbage crops in Samoa.

In Trial 2, C. pavonana, P. xylostella, H. undalis, as well as S. litura (not present in Trial 1) were encountered, with S. litura being the least frequently observed. Friedman’s Test showed that there were no significant differ-ences among treatments in terms of incidence of infestation by C. pavonana and P. xylos-tella, but infestation by H. undalis and S. litura were significantly more noticeable with ‘Delfin’ than with the other treatments (Table 3).


Table 2. Incidence (%) of infestation by each caterpillar species – Trial 1.

Treatment C. pavonana P. xylostella H. undalis Control (Untreated) 42.24c 4.54b 7.44b Attack 0.62a 0.31a 1.23a Bifenthrin 0.92a 0.31a 0.31a Delfin 12.64b 0.31a 1.54a Orthene 3.28a 1.97ab 0.00a Steward 3.47a 0.00a 0.00a Suncis 0.00a 0.31a 0.00a Sunthene 1.85a 2.12a 0.31a Suncloprid 17.66b 3.08b 5.86b P-value <0.001 0.012 <0.001

Means with the same letter within a column are not significantly different (P>0.05).

Table 3. Sum of ranks on incidence of infestation by caterpillar species – Trial 2.

Treatment C. pavonana P. xylostella H. undalis S. litura Control (Untreated) 5.09 0.92 0.00 0.00 Attack 0.46 6.01 1.39 0.00 Bifenthrin 0.00 1.39 0.00 0.00 Delfin 5.09 2.78 4.63 0.92 Orthene 1.39 2.31 0.46 0.00 Steward 0.46 0.00 0.00 0.00 Suncis 0.00 1.39 0.92 0.00 Sunthene 0.00 4.16 0.92 0.00 Suncloprid 5.09 0.92 0.00 0.00 P-value (χ2 approx.) 0.067NS 0.499NS 0.042* 0.042*

‘NS’ means that differences are not statistically significant (P>0.05).


59



Severity of damage by leaf-eating caterpil-lars

Due to a generally high incidence of damage in all treatments in Trial 2, severity of damage by caterpillars was assessed at harvest in order to compare the direct impact of cater-pillar damage on head yield.

Results (Table 4) show that ‘Steward’ had significantly lower severity rating than all treatments except ‘Bifenthrin’. ‘Orthene’ and ‘Suncis’ had moderate damage ratings, while ‘Suncloprid’ had the highest severity rating possible, which was the same as the untreated control.

Although it was observed in Trial 2 that only ‘Steward’ and ‘Bifenthrin’ had sig-nificantly lower final incidences of damage by caterpillars compared to the untreated control (Table 1), severity of damage assessment showed that four of the eight insecticides evaluated (‘Steward’, ‘Bifenthrin’, ‘Orthene’, and ‘Suncis’) had significantly lower damage severity ratings than the untreated control (Table 4). This strongly indicates that similari-ties in incidences of damage among treatments do not necessarily mean that the extent of damage suffered by plants in the treatments would be similar. The observation suggests that the evaluation of efficacy of insecticides against leaf-eating caterpillars in head cab-bages should take into account both incidence of damage and severity of damage.

The observation that most (five of eight) insecticide treatments in Trial 2 suffered moderate to heavy damage (Table 4), despite

the generally low incidences of infestation re-corded (peak of 5.1 %, 6.0 %, 4.6 %, and 0.9 % for C. pavonana, P. Xylostella, H. undalis, and S. litura, respectively), suggests that the 10 % incidence of infestation currently being adopted by Samoa MAF may be too high dur-ing periods of high pest activity. Yield of cabbage

All insecticide treatments, except ‘Suncloprid’, produced significantly more marketable heads than the untreated control in Trial 1; numbers of heads produced in ‘Suncloprid’ and the untreated control were very low (Table 5). The trend for average weight of heads produced was the same.

Similar to Trial 1, results from Trial 2 showed that all insecticide treatments, except ‘Suncloprid’, produced significantly more marketable heads than the untreated control (Table 5). However, unlike Trial 1 in which most insecticides produced statistically similar numbers of marketable heads, ‘Suncis’ and ‘Steward’ had significantly more marketable heads than ‘Attack’, ‘Sunthene’ and ‘Suncloprid’ in Trial 2. In terms of mean weight of heads produced, ‘Suncis’, ‘Steward’ and ‘Bifenthrin’, produced the heaviest heads, which were significantly superior to those pro-duced with ‘Delfin’, ‘Sunthene’, ‘Suncloprid’ and the untreated control. Mean weight of heads produced with ‘Delfin’ and ‘Suncloprid’ were not significantly higher than those from the untreated control (Table 5).

Overall, both marketable heads and weight of heads were lower in Trial 2 than in Trial 1 for all treatments (Table 5), which may be attributable, at least partly, to the generally higher incidence of damage observed in Trial 2 compared to Trial 1 (Table 1). Also, less fa-vourable environmental factors during Trial 2, especially lower rainfall, may have contributed to the observed lower yields in this trial. But regardless of the observed yield differences between the two trials, the fact is that yield is usually the main concern of the farmer, no matter when the crop is being grown. There-fore, based on yield obtained within each sea-son, it may be concluded from the comparative performances of the eight insecticides evalu-ated in this study that any of the insecticides, except ‘Suncloprid’, may be included in rota-

Treatment Severity (scale 0-5)+ Untreated (Control) 5a Attack 4ab Bifenthrin 2cd Delfin 4ab Orthene 3bc Steward 1d Suncis 3bc Sunthene 4ab Suncloprid 5a P-value <0.001*

Means with the same letter are not significantly differ-ent (P>0.05). + See Materials and Methods section.

Table 4. Severity of damage by leaf-eating caterpillars at harvest – Trial 2.

60

tions for the integrated management of leaf-eating caterpillars in Samoa. However, the re-sults strongly showed that ‘Steward’, ‘Bifenthrin’, and ‘Suncis’ should be preferred during growing periods, or in areas, with high pest pressure.

In general, the biological pesticide ‘Delfin’ (Bacillus thuringiensis) did not per-form as well as the other insecticides (Table 5). This observation probably lends some cre-dence to a previous claim by Kessing and Mau (1991) that tests of Bacillus thuringiensis in Hawaii and elsewhere suggested that it was only partially effective against H. undalis (one of the leaf-eating caterpillars in this present study).

Diversity of non-target arthropods Effectiveness in controlling pest spe-

cies should not be the only biological criterion for selecting insecticides for use. The ecologi-cal sustainability of management practices should be among the factors that should be considered in decision-making with regards to pest management inputs. In this trial, an at-tempt was made to assess the relative diversity of arthropods other than leaf-eating caterpillars in cabbage plots being treated with the eight insecticides. The following are the findings.

(a) In crop canopy Apart from leaf-eating caterpillars, spi-

ders, small ants, whiteflies, aphids, earwigs, ladybird beetles and other small beetles, grass-hoppers, and lacewings were the other arthro-pods encountered during weekly observations of cabbage plant canopy in both Trial 1 and

Trial 2. Whiteflies were generally present on many plants during each observation in both trials, but always in very low numbers (no more than 1 to 3 individuals on a plant). Spi-ders and small ants were also seen occasion-ally. The other arthropods mentioned were en-countered even less frequently. The larva of an unidentified ground beetle was also encoun-tered on two occasions; one of the larvae was captured alive and placed in a petri-dish con-taining a large caterpillar of C. pavonana and it consumed the caterpillar. Due to the very low levels of occurrence of the arthropods en-countered in the cabbage canopy, no numerical data was recorded for further analysis.

(b) In pitfall traps A total of 687 and 3501 arthropods

comprising of insects (Coleoptera, Orthoptera, Hemiptera, Hymenoptera, Dermaptera and Dictyoptera), spiders, and myriapods were col-lected using pitfall traps during Trial 1 and Trial 2, respectively (Table 6).

The specific activities of the different non-target arthropods recorded in this study were not investigated. However, Odontoma-chus sp., Chlaenius samoensis, spiders and centipedes are known to be carnivorous, there-fore they may have been preying on other ar-thropods, including leaf-eating caterpillars, in the cabbage plots. Millipedes occurred in rela-tively high numbers, but although some milli-pedes are known to cause plant damage (Radford, 1976), most species are feeders on decaying plant matter. The millipedes caught in pitfall traps in this study (mostly Trigoniu-lus sp.) were most probably organic matter



Table 5. Proportion of plants with marketable heads and mean weight of heads produced in each treatment – Trial 1 and Trial 2.

Plants with marketable heads (%) Mean weight of heads (g) Treatment Trial 1 Trial 2 Trial 1 Trial 2 Untreated (Control) 16.1b 5.9c 60c 22d Attack 100.0a 56.9b 441ab 322ab Bifenthrin 96.9a 83.3ab 537ab 418a Delfin 78.1a 63.9ab 383b 196bcd Orthene 95.8a 72.2ab 554ab 330ab Steward 100.0a 88.9a 560ab 453a Suncis 100.0a 91.7a 595a 459a Sunthene 95.0a 55.6b 463ab 202bc Suncloprid 33.7b 11.1c 152c 40cd P-value <0.001* <0.001* <0.001* <0.001*

Means with the same letter within a column are not significantly different (P> 0.05).

61



Table 7. Simpson’s reciprocal index for non-target arthropods in different treatments based on pitfall trap catches.

Control Steward Orthene Bifenthrin Attack Sunthene Suncis Delfin Suncloprid Trial 1 5.4 6.4 1.9 7 5.1 5.5 4 3.4 2.3 Trial 2 4.9 5.4 4.1 3.6 7.1 6.6 5.6 4.7 5.6

Table 6. Pitfall trap catches of non-target arthropods in all treatments – Trial 1 and Trial 2.

Number of individuals caught in all traps Organisms Trial 1 Trial 2 Odontomachus sp. (Hymenoptera: Formicidae) 64 (9.3) * 306 (8.7) Unidentified small ants (Hymenoptera: Formicidae) 237 (34.5) 1117 (31.9) Achora sp. (Coleoptera: Tenebrionidae) 141 (20.5) 722 (20.6) Chlaenius samoensis (Coleoptera: Carabidae) 4 (0.6) 7 (0.2) Copris sp. (Coleoptera: Scarabaeidae) –** 4 (0.1) Unidentified click beetle (Coleoptera: Elateridae) 3 (0.4) 31 (0.9) Other unidentified beetles (Coleoptera) – 33 (0.9) Unidentified grasshopper (Orthoptera: Acrididae) 2 (0.3) 7 (0.2) Teleogryllus oceanicus (Orthoptera: Gryllidae) 63 (9.2) 173 (4.9) Earwig (Dermaptera: Forficulidae) 6 (0.9) 48 (1.4) Pycnoscelus surinamensis (Blattodea) 2 (0.3) 3 (0.1) Unidentified bug (Heteroptera) – 78 (2.2) Spiders (Araneae) 16 (2.3) 209 (6.0) Millipedes (Diplopoda) 93 (13.5) 319 (9.1) Centipedes (Chilopoda) – 8 (0.2) Miscellaneous 56 (8.2) 436 (12.5) Total 687 (100) 3501 (100)

*Figures in parenthesis represent percentage of all individuals caught in all traps. **Not a category in Trial 1.

feeders. One of the millipede species, Trigo-niulus lumbricinus, has been reported in com-post, rotten sawdust, and old logs in Samoa (Hinckley, 1967). Teleogryllus oceanicus (Pacific field cricket) also occurred in rela-tively high numbers in both trials, but its role in the cabbage crop is uncertain. In general, crickets are omnivorous and will feed on or-ganic matter, fungi, small insects and some seedling plants. Various unidentified small ants were also common in the pitfall traps, but their role in the cabbage plots was not certain. In general, ants are omnivorous and some spe-cies are known to be predatory on other arthro-pods. The single most abundant species of ar-thropods caught in the pitfall traps was Achora sp. (Tenebrionidae), which accounted for 20.5 % and 20.6 % of total catches in Trial 1 and Trial 2, respectively. The role of this species in the cabbage crop was also not investigated, but some members of Tenebrionidae are known to feed on dung, dead insects and carrion. The

miscellaneous group in Table 6 is comprised of several unidentified arthropods each of which occurred only once or a couple of times during the collection period. They included true bugs, bees, spotted millipede, long-horned grasshopper, wasps, and other unidentified ar-thropods.

Assessment of pitfall trap catches using Simpson’s reciprocal index (1/D) showed that non-target arthropod diversity was generally not higher in the untreated control plots com-pared to the insecticide-treated plots, neither was there a clear trend among the insecticide treatments (Table 7). Furthermore, diversity index for each treatment was generally low compared to the maximum attainable values (12 and 16 in Trial 1 and Trial 2, respectively), which is not unexpected since the cabbage crops were grown in monoculture. In general, these observations suggest that the insecticide treatments did not have significant negative impacts on non-target arthropods.

62

CONCLUSION

Good nutrition is vital to good health (Anonymous, 2005), and consumption of fruit and vegetables plays a vital role in achieving it [(World Health Organisation (WHO), 2003)]. Consumption of fruit and vegetables in devel-oping countries is generally low (WHO, 2003), and the negative consequence on global human health status is high. Lock et al. (2005) esti-mated that worldwide mortality attributable to inadequate consumption of fruit and vegeta-bles is 2.635 million deaths per year. Accord-ing to this source, increasing consumption to 600 g per person per day could reduce the total worldwide burden of disease by 1.8 %, with reductions of up to 31 % for ischaemic heart disease and 19 % for ischaemic stroke. In the Western Pacific region where Samoa is situ-ated, the estimated mean daily intake of fruit and vegetables is well below 600 g per person (Lock et al., 2005); one of the reasons for this is affordability. Cabbage, being the most widely sold locally produced leafy vegetable

in Samoa, can suffer very high yield loss due to leaf-eating caterpillars. This contributes to shortages and high prices. From this present study, it is concluded that any of the insecti-cides evaluated, except ‘Suncloprid’, may pro-vide satisfactory protection against leaf-eating caterpillar damage in cabbages. However, ‘Bifenthrin 8SC’, ‘Steward 150SC’ and ‘Suncis 25EC’ should be preferred, especially for periods of very high pest pressure.

ACKNOWLEDGEMENTS

We wish to thank The Secretariat of the Pacific Community for providing some of the insecticides used in this research. We also thank Mr. Ian Faleono and Mr. Lea Mataia (The University of the South Pacific, School of Agriculture & Food Technology, Alafua Cam-pus, Samoa) for assistance with field work. Much appreciation is also extended to staff of Samoa Meteorological Services, Mulinuu, for providing the precipitation and temperature data for Alafua during the study periods.

REFERENCES AMOSA, E. 2005. Managing crucifer pests against brassica crops. South Pacific Agricultural News, 22 (3):3. ANONYMOUS, 2005. Dietary Guidelines for Americans, 2005. 6th Edition, 71 p. Washington, DC: Government Printing Office, January 2005. U.S. Department of Health and Human Services and U.S. Department of Agriculture. U.S. [Online]. Available at: http://www.health.gov/dietaryguidelines/dga2005/document/pdf/DGA2005.pdf [Accessed 13 Aug. 2010]. EBENEBE, A. A., SULIFOA, J. B. & BOE, G. P. 2006. Species complex and importance of leaf-eating caterpillars of head cabbage in Samoa. Journal of South Pacific Agriculture, 13 (1 & 2):6-10. FALEVAAI, S. 2010. Distribution of vegetable seedlings to farmers. Crops Division Newsletter, page 3, Issue 5, April-June 2010. Ministry of Agriculture and Fisheries, Samoa. GENSTAT DISCOVERY EDITION 3. http://www.vsni.co.uk/ HINCKLEY, A. D. 1967. Associates of the coconut rhinoceros beetle in Western Samoa. Pacific Insects, 9 (3):505-511. HOFFMAN, M. P. & FRODSHAM, A. C. 1993. Natural Enemies of Vegetable Insect pests. Co-operative Extension, Cornell University, Ithaca, New York, 63 pp. HOLLINGSWORTH, R. & ALOALI’I, I. 1985. Insect control on head cabbage. Crop Protection



http://www.health.gov/dietaryguidelines/dga2005/document/pdf/DGA2005.pdf




63

Leaflet No. 7. Department of Agriculture and Forests, Apia, and Samoan German Crop Protection Project, Nuu. 4 pp. HOLLINGSWORTH, R. G., MATALAVEA, S. & HAMMANS, H. 1984. Insecticide trial against pests of head cabbage in Western Samoa. Alafua Agriculture Bulletin, 9 (3):17-19. KESSING, J. L. M. & MAU, R. F. L. 1991. Hellula undalis (fabricius). [Online]. Available at: http://www.extento.hawaii.edu/kbase/crop/type/hellula.htm [Accessed 15 July 2008]. LOCK, K., POMERLEAU, J., CAUSER, L., ALTMANN, D. R. & MCKEE, M. 2005. The global burden of diseases attributable to low consumption of fruit and vegetables: implications for the global strategy on diet. Bulletin of the World Health Organisation, 83 (2):100-108. [Online]. Available at: http://www.who.int/bulletin/volumes/83/2/en/100.pdf [Accessed 14 Sept. 2009]. MEMEA, T. 2010. Training vegetable farmers. Crops Division Newsletter , Issue 5, April-June 2010, page 4. Ministry of Agriculture and Fisheries, Samoa. OOI, A. C. P. 1986. Diamondback moth in Malaysia. pp. 25 – 34. In: Talekar N. S. & Griggs, T. D. (Eds) Diamondback moth management. Proceedings of the First International workshop, 11 to 15 March 1985, Tainan, Taiwan. The Asian Vegetable Research and Development Center. RADFORD, A. 1976. Giant millipede burns in Papua New Guinea. Papua New Guinea Medical Journal, 18:138–41. ROBINSON, A. 2010. Pest management in cabbages. Crops Division Newsletter, Issue 5, April-June 2010, page 5. Ministry of Agriculture and Fisheries, Samoa. SASTRODIHARDJO, S. 1986. Diamondback moth in Indonesia. pp. 35 – 42. In: Talekar N. S. & Griggs, T. D. (Eds) Diamondback moth management. Proceedings of the First International work-shop, 11 to 15 March 1985, Tainan, Taiwan. The Asian Vegetable Research and Development Center. WORLD HEALTH ORGANISATION (WHO), 2003. Diet, nutrition and the prevention of chronic diseases. Report of a joint WHO/FAO expert consultation. WHO Technical Report Series 916. 149 pp. [Online]. Available at: http://whqlibdoc.who.int/trs/WHO_TRS_916.pdf [Accessed 13 Aug. 2010].



http://www.extento.hawaii.edu/kbase/crop/type/hellula.htm

http://www.who.int/bulletin/volumes/83/2/en/100.pdf

http://whqlibdoc.who.int/trs/WHO_TRS_916.

64

GUIDE FOR AUTHORS The Journal of South Pacific Agriculture (JOSPA) welcomes papers that can contribute to the development of agriculture and allied industries in the Pacific Islands. TYPES OF CONTRIBUTIONS The following types of contributions will be considered, on priority basis, for publication in JOSPA: i. Original high-quality research papers of relevance to agriculture in the South Pacific region. ii. Review papers on topics of relevance to agriculture in the South Pacific region. iii. General papers based on agricultural research carried out in the South Pacific region and aimed at extension workers and agricul-

tural teachers/trainers in the region. iv. Short communications of relevance to agriculture in the South Pacific region. v. Research papers on the processing of agricultural produce, floriculture, and other agro-allied topics that can contribute to sustain-

able livelihoods for people in the Pacific Islands. Reviews This section is mainly for reviews of new books relevant to the areas covered by JOSPA. Reviews can also cover other areas, such as relevant CD-ROMS, computer packages and, where appropriate, conference/workshops. The format for reviews should follow those in past issues of JOSPA. REPEAT OF EXPERIMENTS AND ANALYSIS Papers that report original research should only be submitted if the experiments have been repeated at least twice, with a minimum of three replications per experiment. Appropriate statistical analysis is required. Exceptions may be made to these requirements if they are justifiable. The final decision rests with JOSPA. PREPARATION OF MANUSCRIPT Electronic format: Manuscripts should be prepared electronically in MS Word. Language: • Manuscripts must be written in English (U.K.). Authors who are not very proficient in English are strongly advised to have their

manuscript proof-read by someone who is proficient in English prior to submission. • Use of local names of things should be avoided. But if any must be used, it should be typed in italics and enclosed in inverted

commas. A brief explanation of what the term stands for should be given in parentheses at first mention. Paper size, margin, length of manuscript: Manuscripts should be double-spaced on one side of A4 paper with a 2.5 cm margin on all sides. The length of the manuscript should not exceed 5000 words, excluding references, figures and tables. Organisation of manuscript: The arrangement of the various components of the manuscript should follow this order: • Title (should be clear, informative, not too long, capitalize first word and proper names only, no full stop, Times New Roman size

14, bold type) • Name(s) of author(s) (first name, initials, surname, Times New Roman size 12, not bold) • Full address(es) of author(s), including organisational affiliation(s), telephone, email and fax number (if available) of the corre-

sponding author (provide all as footnotes) • Abstract (should be clear, descriptive, no formulae, no common names of biota, no trade names of pesticides, no acronyms, not

exceeding 250 words, single paragraph) • Key words (3 – 6 items) • Introduction • Materials and Methods • Results • Discussion • Conclusion • Acknowledgements • References • Appendices (if necessary) Note: Illustrations (photographs, line drawings, etc.) should be included only if they clarify the text. Length of paper, font, line spacing, margins, paragraph style: • Full length research papers should not exceed 5000 words (excluding abstract, list of references, figures and tables), Times New

Roman size 12 only, double line spacing within paragraph (except in tables), single column, justify. • Bold and capitalize main titles (i.e. Abstract, Key Words, Introduction, Materials and Methods, Results, Discussion, Conclusion,

Acknowledgments, References), do not number them. • Space before and after a main title in the body of the manuscript should be 12 points. • Indent first line to indicate a new paragraph (do not use space to separate paragraphs). • Author(s) may use subtitles for sub-components under a main title within the main body of the manuscript (use sentence case, bold

type, no underlining). • Author(s) may use second-level subtitles under first-level subtitles (use sentence case, italics, do not bold or underline). • Author(s) may use third-level or fourth-level subtitles (sentence case, may use letters or numerals to number subtitles, use italics,

do not bold or underline). • No space after subtitles (indent the first line of each paragraph). • Do not indent main titles or subtitles. • Avoid using long sentences and paragraphs. Avoid beginning a paragraph with words like: And, But.


65

Short communications: • This applies to research that can be covered in short research note format. • Should not exceed 2000 words (excluding abstract, list of references, figures and tables). • A short abstract summarising the main findings is required (not exceeding 100 words, single paragraph). • Key words (3-6) are required. • The body of the text may be in continuous form, or organised under a selection of main titles (see ‘organisation of manuscript’). • A list of references is required. Should adhere to other general formatting as outlined in this guide. International System of Units (SI): • Use SI units (i.e. modern metric system). • Use standard abbreviations for units that follow numerical values (e.g. 25 cm, instead of 25 centimetres), unless otherwise consid-

ered more appropriate. • With the exception of angles and hyphenated use, always leave a space between a numerical value and the unit symbol following

it. Examples of correct use: (100 g, not 100g); (37 °C, not 37°C); (‘a 10-m tall tree’, not ‘a 10m tall tree or a 10 m tall tree’); (an angle of 12°, not an angle

of 12 °) Tables: • Should be numbered according to sequence of appearance in the paper, have a clear descriptive caption directly above the table, do

not embed in text, place one table per page. • Indicate approximate positions in the text. • Use brief but informative column headings. Standard abbreviations of units of measurements should be included in parentheses. • Do not use bold type for titles or other contents. • Line spacing: single. • Avoid large tables (split them into few smaller tables, if possible). • Use horizontal borders to demarcate heading rows and to underline the last row of data entry. • Use tabs to align contents into columns, leave sufficient space between columns. Do not draw tables with grid lines (whether

vertical or horizontal, visible or suppressed). • If needed, short explanatory notes may be included directly below a table. Graphs: • Should have a clear descriptive caption placed directly below the graph. • Do not embed in the text, save on separate page (see instruction for figures under ‘organisation of manuscript). • When prepared in applications such as Excel, they should be exported and saved as MS Word document. • No grid lines (but will be allowed in exceptional cases). Photographs, maps, scanned items: • The print version of JOSPA is published in black and white only. Authors are strongly advised to use either high resolution black

and white illustrations, or very high resolution colour illustrations which will retain clarity when converted to black and white. Poor quality illustrations will be rejected.

• Save pictures in JPEG format (do not embed in main body of manuscript). • Photographs, scanned pictures, maps, and similar items should be submitted as a separate file (but do indicate their approximate

locations in the manuscript). Lettering in/on illustrations Lettering on diagrams, photographs, graphs, etc. should be sufficiently large and bold to permit legible reproduction when the illustra-tion is reduced to a size suitable for inclusion in the journal. (Test the legibility of your illustration by reducing the size by 50%.) Captions: • Should be clear, descriptive, and not too long. • Capitalise first word and proper names only, write scientific names in full. • Should be formatted as follows: Table 1. Relative abundance of flowers in two varieties of tomato following treatment with hormones. Figure 1. Size of banana fruits at eight weeks after de-suckering. • All illustrations (diagrams, graphs, photographs, line drawings, etc.) should be referred to as figures. They should be numbered

consecutively (according to order of appearance in the paper). • Capitalise the first letter when referring to a table or figure within the text, regardless of its position in the sentence. Examples: Table 1 shows that ......; This trend is clearly shown in Figure 10. ** All figures and tables must be referred to in the text. Names of organisms: With the exception of common domestic animals and crops, the preferred scientific name of organisms (according to the International Codes of Nomenclature) should be given in full at the first mention of the English common name. Thereafter, the common name may be used, provided there is no ambiguity. Common names should be avoided in titles and abstracts (but English common names of crops may be used, provided there is no ambiguity). Names of pesticides • The internationally recognised common names (names of active ingredients) of pesticides and other chemicals should be used.

Commercial (trade, brand) names may be used in inverted commas (but the internationally accepted common name should still b e provided at first mention). Example:

Glyphosate is the common name (or active ingredient) in Roundup, and Roundup is the commercial (or trade) name of a herbi-cide.


66

• Common names should be presented in lower case letters, except at the beginning of a sentence. The trade name, if used, should always start with a capital letter, and should be enclosed in inverted commas. Examples: The use of glyphosate is widespread in .....; Glyphosate is widely used ..... ‘Roundup’ was applied.... ; The application of ‘Roundup’ resulted in ......

**Chemical names may also be used where appropriate. Formulae: • Wherever possible each formula should be typewritten (not scanned), with adequate space around the formula. (Consider placing

the formula inside a textbox to avoid shifting of items during subsequent formatting of the paper.) • If several equations are used in the manuscript, number them consecutively on the right-hand side in parentheses. • Superscripts/subscripts should be very clear; ensure that numerals and letters are distinguishable (e.g. between zero and letter O). • At first use, provide the meaning of all symbols in the equation immediately after the equation. • In general, use internationally accepted symbols and avoid ambiguity. Reference to numbers in text: In general, do not write the numbers zero to nine in numerals within sentences, except in special cases (e.g. when followed by units of measurement, in ranges). Numbers from ten onwards may be written in numerals within sentences, but numerals should not be used to begin a sentence. Referencing within text: • Use only surnames of authors (in initials) when citing references. • References should be cited as follows:

...the data (Greenland & Craswell, 1989 ) showed... ; or ….Greenland & Craswell (1989) showed….. • If a reference quoted in the text has more than two authors, it should be quoted as follows:

….. the data (Greenland et al., 1980) …… ; or ….. Greenland et al. (1989) showed….. • Referencing from a book should be as follows:

McKeen (2002, pp. 18-20) stated that….. ; McKeen (2002, p. 20) stated that….. • Where it is certain that the author cannot be identified, use ‘Anonymous’ or ‘Anon.’ in place of author name.

Example: Anon. (2003) reported that …; or Anonymous (2003) reported that …… **Choose and use one format only.

• References cited together in the text should be arranged chronologically. Example: ...requires two inches of rainfall annually (May, 1989; Glen, 2000; Duke, 2001).

• Where an author has more than one publication in the same year, letters of the alphabet (a, b, etc.) should be assigned to each pub-lication in the order of first appearance within the text. Examples: Lindsay (1972a) found that...., but it tends to decrease with age (Lindsay, 1972b). Lindsay (1972a, b) found that.... **These publications must carry these assigned letters in the list of references.

• When the date of publication is not indicated, cite as follows: Jones (n.d.) found that ……..

• If the author of a publication is a recognised corporate body (e.g. an international organisation, company, government department), the name of the body should be written in full at first mention, with the standard abbreviation in parentheses. Thereafter, the abbre-viation may be used in the rest of the text. Example: The Food and Agriculture Organisation (FAO) (2000) has declared .... (at first citing). The FAO (2000) noted that ...... (subsequent citing).

** It is preferable that the full corporate name is used in the list of references. List of references: • Only publications cited in the paper should be listed. • All publications cited in the body of the text should be included in the list of references. • Surnames and initials of authors should be provided (do not use et al.) • It is the responsibility of the author(s) to ensure that author name(s) and year of publication are correct, both within the body of the

text and in the list of references. • All references should be listed at the end of the paper in ascending alphabetical order (A-Z), according to the last name of the au-

thor. • Publications for the same author(s) should be arranged chronologically.

** You can use the ‘sort’ function of your word processor to arrange the references. • Use italics for titles of books, journals, newspapers, newsletters, and theses (dissertations). • References from periodicals should include the last name(s) of the author(s) followed by the initials (all uppercase letters), year of

publication, title of paper, full name of the periodical (in italics), volume number (in bold type) and page numbers. Example: LINDSAY, W. L. 1972. Zinc in soils and plant nutrition. Advances in Soil Management, 24:147-186.

• When present, the issue/part number should be written in parenthesis after the journal volume. Example: LINDSAY, W. L. 1972. Zinc in soils and plant nutrition. Advances in Soil Management, 24 (2):147-186.

• References to books should include the last name(s) of author(s) followed by initials (all uppercase letters), year of publication, title of book, edition, publisher and location, number of pages. Example:

MAJOR, R. & GREEN, V. S. 1980. Growing Rice in the Savannas. 2nd edition. Marquee Publishers, Brenton, 250 pp. • Where a reference involves only a chapter in a book whose various chapters are authored by different persons, the reference

should be listed as follows: LAL, R., KANG, B., MOORMAN, F. R., ANTHONY JOU, S. R. & MOONMAW, J. C. 1975. Soil Management Problems and Possible Solutions in W. Nigeria. In: Bornemisza, E. & Alvarado, A. (Eds.), Soil Management in Tropical America. North Carolina State Univer- sity, USA, pp. 372-408.


67

• References to online publications (articles, ejournals, ebooks): BROWN, K.C. 2008. Growing button mushroom. Urban Agriculture Newsletter, [Online]. 42 (6). Available at: http://www.veg.ug/articles [Accessed 20 Jan. 2010]. MVUNGI, J. B. & MATHI, C. J. 2000. A new incubation technique for small-scale poultry farms. Poultry Digest, [Online] 2 Feb., 13 (2):1 -6. Available at: http://get.fraf.cw.twn/ejournals/ [Accessed 12 Feb. 2009]. PASCAL, W. 2002. The third degree. [e-book] Herman: Zorrow Press. Available at: Municipal Library/Digital Library/e-books http://get.fraf.cw.twn /E-books [Accessed 6 March 2009].

**Note: Underline URLs. Do not use URLs in the body of the text (unless for a special reason). • Newspaper articles should be listed as follows: TUNAKI, M. M. 2003. Pollution scandal: top officers implicated. The Sentry, 2 Aug., p.1. • List a conference paper as follows: Name(s) of author(s) and year. Title of paper. In: (name of the editor or organisation). Full title

of conference (italics). Location, date. Publisher, place of publication. Example: SUZUKI, S. H. 2008. The impact of deforestation on rural water supply in Naiger. In: 10th International Conference on Cli-

mate Change. Hankok, Neverland, 1-5 Nov. 2008. Global Concern, New Ville. • Annual reports: Provide name of corporate author, year of publication. Title of annual report (italics), place of publication: Pub-

lisher. Example: EXTENSION DIVISION 2008. Strategic developments - annual report 2007-2008. Sebania Ministry of Agriculture & Fish-eries, 63 pp.

• Work that has already been accepted for publication in a periodical should be listed as ‘in press’ (provide the name of the periodi-cal and volume number).

• Unpublished work (including research data, internal reports) and personal communications should not be listed under references, but may be mentioned in the text; give as much details as appropriate.

• Master’s and doctoral theses that have been approved by the relevant educational institution may be included in the list of refer-ences. Example:

SULIFOA, J. B. 2007. Evaluation of Some Management Strategies against Lepidopterous Pests of Head Cabbage in Samoa. M.Sc. Alafua: University of the South Pacific.

Spelling & grammar checker: Authors are strongly advised to use their word processor’s spelling and grammar checking function (set to U.K. English) to check their manuscript prior to submission. General: • Be consistent in style (for example, use either organisation or organization throughout your paper). • The responsibility of proof-reading a paper rests with the author(s). SUBMISSION OF MANUSCRIPTS JOSPA accepts a contribution with the understanding that the submission is based on the author(s) original, unpublished work. i. Manuscript must be submitted electronically, either by e-mail, or on a CD mailed to the address provided below. Please do not

submit hard copies. ii. Submission of a manuscript implies that it has the approval of the author’s organisation under whose authority the work was con-

ducted, it has not been published elsewhere, and it is not being considered for publication elsewhere. iii. Manuscript should be accompanied by a cover page with the following information: • Title of paper • Name(s) of author(s) • A short paragraph summarizing how the paper will contribute to the development of agriculture, agro-allied industries and sustain-

able livelihoods in the South Pacific region. • A declaration stating that the paper has not been submitted in full or in part for publication elsewhere, and will not be submitted in

full or in part for publication in any language elsewhere if accepted for publication in JOSPA. (Exceptions to this requirement are submission in the form of an abstract only, or as part of a public lecture, or an academic thesis.)

• In the case of multiple authors, a statement declaring that all authors consent to the submission. • Name and contact details of two persons who may be contacted as reviewers for the paper. iv. Send your manuscript to:

JOSPA Managing Editor, USP-SAFT, Alafua Campus, Private Mail Bag, Apia, Samoa. Email: [email protected] or [email protected] .

**You may also send other correspondences regarding your submission to these addresses. REVIEW AND REVISION • All papers will be subjected to review by referees, and editorial revision, to ensure that contributions are relevant and of satisfac-

tory quality. • Where necessary, the manuscript will be returned to the author(s) for revision before the paper is finally accepted for publication. • Paper sent to author(s) for revision must be returned to the JOSPA Managing Editor within the time frame advised to avoid delays

in the publication of your paper. • The final submission may be subject to minor editorial revision to suit JOSPA style. REPRINTS One hard copy of the journal and two reprints of your contribution (if produced) will be sent to the corresponding author at no cost. Additional copies will be available at a cost. **Manuscript may be submitted without strict adherence to JOSPA’s house style, except for the general arrangement of components (organisation of manuscript). If the paper is acceptable for publication, the final version must be formatted according to the style out-lined in this guide.


http://www.veg.ug/articles




http://get.fraf.cw.twn/ejournals/



Date post:	18-Jul-2020
Category:	Documents
Upload:	others
View:	1 times
Download:	0 times

Address for Correspondence - University of Hawaii...(Consultant) UNITED KINGDOM Dr. Lalith...

Documents