Báo cáo khoa học nông nghiệp " Replacing fertiliser N with rhizobial inoculants for legumes in Vietnam for greater farm profitability and environmental benefits " MS4 docx

Executive Summary Increased production of high-quality inoculants and QA Much of the R&D effort during the reporting period was focussed on evaluating rhizobial strains for soybean and

Ministry of Agriculture & Rural Development

Collaboration for Agriculture & Rural Development

Project Progress Report MS4: Second Six-monthly Report

013/06VIE Replacing fertiliser N with rhizobial

inoculants for legumes in Vietnam for

greater farm profitability and environmental benefits

July 2008

6.2 Gender and Social Issues _ 24

7 Implementation & Sustainability Issues _ 23

7.1 Issues and Constraints _ 23

1 Institute Information

Project Name Replacing fertiliser N with rhizobial inoculants for

legumes in Vietnam for greater farm profitability and environmental benefits

Vietnamese Institution Oil Plants Institute (OPI)

Vietnamese Project Team Leader Ms Tran Yen Thao

Australian Organisation NSW Department of Primary Industries

Completion date (original) March 2009

Completion date (revised) As above

Contact Officer(s)

In Australia: Team Leader

Name: Dr David Herridge Telephone: 02 67631143

Position: Principal Research Scientist Fax: 02 67631222

Organisation NSW Department of

Primary Industries

Email: david.herridge@dpi.nsw.gov.au

In Australia: Administrative contact

Name: Mr Graham Denney Telephone: 02 63913219

Position: Manager External Funding Fax: 02 63913327

Organisation Oil Plants Institute (OPI) Email: yenthao@opi.org.vn

yenthao@hcm.fpt.vnyenthao9@yahoo.com

2 Project Abstract

3 Executive Summary

Increased production of high-quality inoculants and QA

Much of the R&D effort during the reporting period was focussed on evaluating rhizobial strains for soybean and groundnut in both growth room and field The Australian commercial strains CB1809 (soybean) and NC92 (groundnut) were found to be more effective than local Vietnamese strains at almost all of the 20 field sites for which data are available Thus, when the crops were inoculated with CB1809 or NC92, nodule weight, biomass yield and grain yield increased relative to the local strains at 85%, 85 and 90% field sites respectively However, the extent of the increase was different depending on sites and local strains The two Australian strains increased soybean and groundnut nodulation, biomass yield and grain yield by an average of 62%, 34% and 27%, relative to uninoculated plots and 26%, 11% and 10% relative to inoculation with local Vietnamese strains

R&D on inoculant production technology focussed on additives to both broth and peat, on survival of rhizobia in peat, and on temperature and pH effects on rhizobial growth and survival The IAS examined effects of broth (fermentor) culture dilution prior to inoculating the peat carrier as a means of extending the broth

Issues with rhizobial strain purity and maintenance, discussed at the Project Review Workshop in February 2008, prompted new recommendations for strain maintenance Other issues raised and discussed were procedures for optimising the peat carrier through adjustment of pH and moisture and for sterilisation of the peat There was no training

Farmers in Vietnam currently fertilise legumes such as soybean and groundnut with N, rather than inoculate with rhizobia Replacing fertiliser N with rhizobial inoculants would save Vietnamese farmers A$50-60 million annuallyin input costs and, at the same time, help facilitate the desired expansion in legume production There would also be positive environmental outcomes This project aims to increase production of high-quality legume inoculants in Vietnam through enhanced production capacity, implementation of a national quality assurance (QA) program at and increased inoculant R&D Participating in the project in Vietnam are the Oil Plants Institute (OPI), the Institute of Agricultural Science (IAS) and the National Institute for Soils and Fertilisers (NISF; now known as the Soils & Fertilisers Institute (SFI)) Institutions in Australia are NSW Department of Primary Industries and the University of Sydney Legume inoculant use by farmers in Vietnam will be increased through the development and implementation of an effective extension and training program for researchers, MARD extension officers and farmers The benefits of inoculants and legume nitrogen fixation will be demonstrated in the field and communicated through workshops, meetings and publications To ensure sustainability of inoculant production and use, the project will engage the private sector in marketing and ‘pilot production’ of legume inoculants, with the aim that they may scale-up production and progressively take over supply as the technology and markets are developed

conducted during the reporting period, although training scheduled for August-September

2008 in Thailand and for later in 2008 in Australia was discussed at the February 2008 Project Review Workshop

Extension and training of farmers and advisors

Extension and training of farmers and advisors is a major focus of the project as a means of facilitating adoption of legume inoculation in Vietnam The extension-training program is built around simple, multi-location inoculation experiments in the legume production areas of the country The experiments involve participation of farmers and extension officers in all aspects, from the design of experiments to sowing, sampling, harvesting and interpretation of results The MARD extension service plays a large role in extension activities Data from field demonstrations will be used to produce an economic model for production and use of legume inoculants in Vietnam In addition, training courses will be organised for farmers, extension workers and researchers in methods of inoculant use, and economic as well as environmental benefits of inoculation

A total of 28 demonstration (extension) trials have now been conducted in 9 provinces At the time of writing this report, data were available for 15 demonstration trials The demonstration fields usually had two treatments: +inoculation and -inoculation Overall, inoculation of soybean and groundnut increased potential income to farmers by an average of 3.5000.000VNĐ The size of the profit varied with different sites

Questions were often asked by farmers were:

- How do the inoculants cost?

- How much do inoculants used for 1000m2 or 1 ha?

- Where can we purchase inoculants?

- Do inoculants have other benefits besides urea (N) fertilizer?

- Can we use inoculants together with plant protection products?

- Can we use legume inoculants for other crops?

- Are inoculants effected by bad weather such as heavy rain, hot weather when inoculation and during plant growth?

- Can we use inoculants together with urea (fertiliser N)?

And main requests:

- Supply inoculants for farmers to test inoculants in their fields

- Technical support for farmers to use inoculants

Involvement of the private sector in inoculant production, distribution and marketing

Careful selection of private sector partners for commercial production of legume inoculants

is critical for ensuring the sustainable supply of high quality legume inoculants to farmers in Vietnam After visiting potential inoculant producers in February 2008 and the withdrawal of Fitohoochmon from the project it was initially concluded that the prospect of the private sector producing high quality inoculants in Vietnam in the near future was low and that large scale production should be carried out by the institutes However, following this, contact was made with the Thien Sinh (Komix) company which is much better equipped to adopt legume inoculant production and a strategy has been developed for the transfer of technology from the institutes to this company A site visit in February 2008 revealed very good potential for legume inoculant production The facilities were well equipped for careful industrial microbiological application Technology transfer from the research institutes to the company was discussed and a clear strategy was determined The company also agreed to conduct extension activities through their country-wide network

4 Introduction & Background

Project Objectives and Expected Outputs

The Vietnamese government (MOI, MARD) is committed to increase the area sown to legumes from the current 780,000 ha to >1,000,000 ha by 2010, with particular focus on soybean and groundnut in the Mekong Delta, the Central Coastal region and upland (highland) areas of the North, Central and North The legumes are used for production of food, oil and protein meal, and are grown as rotation crops with rice (Mekong Delta), as intercrops in the upland areas with cassava, sugar cane, rubber, fruit and maize and as cover crops in the sandy coastal soils ACIAR Small Project LWR2/98/27 (Increasing yield and

nitrogen fixation of soybeans, groundnuts and mungbean in Vietnam through Rhizobium

inoculation) identified that legume production in Vietnam currently relies on expensive imported fertiliser N, rather than cost-effective inoculants containing rhizobia Replacing fertiliser N with rhizobial inoculants would save Vietnamese farmers A$50–60 million annuallyin input costs and, at the same time, help facilitate the desired expansion in legume production There would also be positive environmental outcomes

Details of the economic benefits of replacing fertiliser N with rhizobial inoculation were outlined in the proceedings of the technical workshop to terminate LWR2/98/27 However, for this to happen, high-quality inoculants need to be readily available in the market The current capacity of inoculant production in Vietnam is about 40,000 packets annually, and would need to be increased to about 500,000 packets annually to meet potential demand Inoculant quality is also poor (LWR2/98/27 project) and would need to be improved Shelf life and distribution and marketing are issues that would also need to be addressed Moreover, there is limited awareness of the benefits of inoculants and methods of application among Vietnamese farmers and extension workers

Capacity gaps are evident at the national and institutional level The major gap at the national level is the lack of a coordinated, focussed national legume inoculant program At the institutional level, the gaps are capacity for medium-scale inoculant production and associated quality assurance (QA) as well as R&D and training capacity The proposed project would address these issues of production, quality, distribution and marketing and farmer education Involvement of the private sector in both production and marketing will ensure the long-term viability of the concept The project objectives are to:

1 Increase production of high-quality inoculants for soybean, groundnut and other legumes in Vietnam through enhancement of production capacity (personnel and equipment) at participating institutions, implementation of QA, and increased inoculant R&D;

2 Increase farmer interest and use of inoculants in Vietnam through development and implementation of an effective extension and training program on inoculants and legume nitrogen fixation for researchers, MARD extension officers and farmers through demonstration trials, workshops and meetings, and publications;

3 Ensure the long-term viability of the project through involvement of the private sector

in this ‘pilot production’ of legume inoculants, with the aim that the private sector would progressively take over production as the technology and markets are developed

The Project is aligned with the CARD Program Strategic Objective 2 ‘Improved productivity and links to markets for the rural poor in the Mekong Delta and Central Coast regions’,

through Objective 2.1 ‘Increase rural productivity’, using principally Strategy 1 ‘Increase productivity and competitiveness of the agricultural system’

Project Approach and Methodology

The project strategy is to enhance inoculant production, quality, distribution and marketing and farmer education through the collaborating institutions It will involve both Government institutions – Oil Plants Institute (OPI), the Institute of Agricultural Science (IAS) and the Institute for Soils and Fertilisers (ISF) – as well as private sector companies (Fitohoocmon Fertiliser JSC, Cu Chi Bio-Chemical Fertiliser JSC and Humix) The latter would be involved initially in marketing and distribution of inoculants and would be provided with advice and technical expertise to improve and expand their inoculant production capabilities

In time, it is envisaged that the private sector would take over inoculant production, leaving

QA to the public institutions Involvement of the private sector in both production and marketing will ensure the long-term viability of the concept

Increased production of high-quality inoculants and QA

The focus will be on rhizobial strains and their maintenance, inoculant production (fermentation) technologies, quality assurance of the production process and products and training in the production technologies and QA

Rhizobial strains - Existing strains from the collections in Vietnam and/or other institutions,

eg ALIRU (Australia), NifTAL (University of Hawaii), Suranaree University (Thailand) will

be utilised where appropriate Decisions will be made about the most appropriate strains for

inoculant production through a program of research and development

Strain maintenance - Protocols and operation manuals for maintaining strain effectiveness

and recognition to ensure stability of inoculant quality during long-term storage will be developed and implemented During the course of the project a decision will be made about which institute/s will take responsibility for the maintenance of a culture collection and verification of strains for inoculant production The institute/s will be equipped with the necessary materials to continue processing germplasm for inoculant production

Production technology – Draw on experiences from Thailand and Australia to develop

production technology of inoculants at medium-scale in Vietnamese institutes through:

• Modifications to broth formulations and experiments in procedures for maintaining sterility

and dispensing broths into the inoculant carrier

• Testing appropriate forms of inoculant (peat, granular, liquid) that allow compliance with quality control standards and are easy to use, supply and transport Economic benefits will

be determined by assessing their effectiveness in laboratory and field trials

• Strain selection: Different strains will be tested for survival in inoculant products and

during delivery of products to the field

Quality assurance - Australian QA protocols will be used initially as a model From that,

QA protocols, training and working manuals specific to inoculant production in Vietnam will

be developed jointly between Australian and Vietnamese project scientists Decisions will be made about which institute will be equipped to carry out on-going routine quality control of inoculants in Vietnam

Training on inoculant production and QA - Vietnamese researchers from institutes will be

trained in Vietnam by Australian collaborators and at Suranaree University of Technology (Thailand) in inoculant production, QA and laboratory management, as well as R&D in rhizobiology

Extension and training of farmers and advisers

The extension-training program for farmers and extension officers will be built around simple, multi-location inoculation experiments in the legume production areas (Mekong Delta, the Central Coastal region and upland (highland) areas of the North and Central and the South East) They will involve participation of farmers and extension officers in all aspects, from the design of experiments to sowing, sampling, harvesting and interpretation of results It is expected that the MARD extension service will play a large role in extension activities Data from field demonstrations will be used to produce an economic model for production and use of legume inoculants in Vietnam In addition, training courses will be organised for farmers, extension workers and researchers in methods of inoculant use, and economic as well as environmental benefits of inoculation This extension-training program will be conducted by Vietnamese researchers in collaboration with Australian counterparts, who will assist in collating and preparing extension materials for translation and transfer to Vietnam

Involvement of the private sector in production, distribution and marketing

Two (and possibly three) Vietnamese companies will be involved in the project, Fitohoocmon Fertiliser JSC and Cu Chi Bio-Chemical Fertiliser JSC (and Humix) It is envisaged that the market for inoculants will grow during the course of the project from the current low level and that the private sector will progressively become involved as the production technology is developed and the market for the inoculants expands The companies will initially be involved in marketing and inoculant distribution Training workshops will also be open to researchers from the collaborating private companies Project scientists will provide technical expertise to the companies throughout the project to troubleshoot production problems and increase production capacity and product quality

5 Progress to Date

5.1 Implementation Highlights

5.1.1 Increased production of high-quality inoculants and QA

The focus is on rhizobial strains and their maintenance, inoculant production (fermentation) technologies, QA of the production process and products and training in the production technologies and QA

Table 1: Nodulation of CB1809 and local strains on soybean

Nodule number/plant

No Strains

Total nodule number

Main roots Secondary roots

Graphs 1, 2 and 3 summarise responses to inoculation with strains CB1809 (soybean) or NC92 (groundnut) of plant nodulation, biomass and grain yield, respectively Responses ranged from small to large to inoculation depending on field sites

Graph 1 Range of nodulation responses

to inoculation

-20 0 20 40 60 80 100 120 140 160 180

There were large effects of inoculation on nodulation at 70% of the field sites (Graph 1) At those sites, nodule weight increased by 43–166% Nodulation responded moderately at one site (34%) and, at the rest (25%) of the sites, response were small with an average increase of 10% The overall average increase in nodulation using the superior Australian strains was 62%

With crop biomass, there were large responses (44–60%) to inoculation at 50% of the field sites, moderate responses (22–37%) at 30% of the sites and small responses (5–19%) at the remaining 20% of sites (Graph 2) Increases in grain yield from inoculation were smaller than the increases in nodulation and biomass yield (Graph 3) There were large responses (40–68%) at 20% of sites Moderate responses (20–37%) were recorded at 55% of the sites and small responses (4–19%) at the remaining 25% of sites The overall average increases in biomass yield and grain yield using the superior Australian strains were 34% and 27%, respectively

There were large differences in nodulation, biomass yield and grain yield responses amongst the rhizobial strains Australian commercial strains CB1809 (soybean) and NC92 (groundnut) were more effective than local Vietnamese strains at almost field sites (Appendix 2) Data analysis shows that when the crops were inoculated with CB1809 or NC92, nodule weight, biomass yield and grain yield increased relative to the local strains at 85%, 85 and 90% field sites respectively However, the extent of the increase was different depending on sites and local strains Graphs 4, 5 and 6 show the increase in nodulation, biomass and grain yield of soybean and groundnut when inoculated with CB1809 and NC92, respectively, compared with inoculation using local strains

Graph 2 Range of crop biomass responses to inoculation

0 10 20 30 40 50 60 70

The two Australian strains, CB1809 and NC92, increased soybean and groundnut nodulation

by an overall average of 26%, relative to the local Vietnamese strains (Graph 4) Biomass yields were increased by an average of 11% (Graph 5) and grains yields increased by an average of 10% (Graph 6), relative to the local strains

For each of the measures, there were large variations according to the particular site For nodulation, the range was 0–70% For biomass yield, the range was 0–30% and for grain yield, the range was 0–32%

Graph 4 Nodulation increases with CB1809 and NC92 compared to local strains

0 10 20 30 40 50 60 70 80

Graph 5 Biomass increases with CB1809 and NC92 compared to local strains

05101520253035

Graph 6 Grain yield increases with CB1809 or NC92 compared to local strains

0 5 10 15 20 25 30 35

5.1.1.2 Rhizobial strain maintenance

There is a need to ensure that the rhizobial cultures used in inoculant production are maintained as authentic, pure strains that originate from the same source and are effective in

fixing nitrogen

On reviewing the QA of inoculant strains within each institute (i.e OPI, IAS, SFI) during the project visit to Vietnam in February 2008, it was determined that there was a need for more training in the identification and maintenance of pure rhizobial cultures The lack of in-depth knowledge of the morphological identification of the rhizobial strains may have resulted in the production of poor quality of legume inoculants for the field trials

It was therefore recommended that for future research (laboratory, growth room and fieldwork), only commercial strains from Australia would be used These were the strains that were most effective in field trials conducted in 2007–08

As the purity of the cultures of NC92 and CB1809 used previously in research was doubtful,

it was decided that all existing cultures of these strains be discarded OPI was selected as the curator of the strains to be used in the commercial production of legume inoculants in Vietnam It was recommended that ALIRU re-supply OPI with fresh agar cultures and freeze-dried cultures of NC92 and CB1809

It was decided that:

• on receipt of the cultures issued from ALIRU, scientists of OPI would check the purity of cultures by streaking onto YMA and CRYMA plates Once the purity of the strains had been established, sub cultures of the strains would be made, and copies issued to the other institutes

• when OPI issued cultures to the other institutes, they would be accompanied by photographs of cultural growth showing colony morphology on agar in Petri plates

In conjunction with this strain identification, each institute would streak out the culture that they were using in the starter broths, photograph colony morphology and send it to the other institutes including ALIRU This cross checking between laboratories would ensure that cultures of the rhizobial strains used in the production of inoculants were identical

5.1.1.3 Inoculant production technology - experimental

Experiments on inoculant production were conducted at SFI, IAS and OPI Each institute prepared inoculants using different sources of peat and measured survival over time The IAS measured survival of rhizobia in peat amended with worm casts and coconut coir dust thus changing the water holding capacity of peat Both OPI and IAS investigated the effect of different additives on rhizobia in inoculants formulated as liquids, a technology that would reduce the high cost of sterilisation of peat carrier, while the ability to culture rhizobia using cost effective ingredients was investigated at SFI

The SFI produced inoculants using sterilised peat with three local strains of soybean rhizobia and two local strains of groundnut rhizobia The Australian commercial inoculant strains CB1809 (for soybean) and NC92 (for groundnut) were also included as reference strains The aim of the experiment was to compare the survival of local strains and Australian strains

in peat culture

Broth cultures were grown in erlynmeyer flasks and transferred to a fermentor after growth The broth was checked for contamination after growth in both the erlynmeyer flask and fermentor Peat was collected, milled, mixed with lime and sterilised by autoclave in the packet The pH of the peat was 3.0 to 4.0 after collection and increased to 6.8 to 7.0 after the addition of lime After sterilisation, peat was suspended in sterile water, diluted to 10-6 and plated on glucose-peptone media to check for contamination The packet was then sealed and injected with broth Broth was added to peat at a ratio of one part broth to four parts peat The initial moisture content was less than 10% and increased to 40% after injection The inoculant was incubated for one week and then stored at room temperature

The number of rhizobia in peat was measured at time intervals up to 6 months, immediately after production, after 2 weeks, 1, 2, 3 and 6 months (Table 2) The initial counts for soybean rhizobia, CB1809 and local strains SL1, SL2, SL3 were 5 x 109, 2.3 x 109, 1.8 x 109 and 2.1 x

109 respectively and decreased to 2.7 x 108, 2.5 x 108, 1.2 x 108 and 1.3 x 108 after 6 months

At 3 months the numbers of viable cells were still over 109 and ranged from 1.1 to 2.2 x 109 However, all numbers were determined from viable plate counting and no plant infection tests were used to confirm presence of rhizobia Numbers can be overestimated if only determined using plate culturing methods because of poor distinction between rhizobial and contaminant colonies Although contamination of peat was low after sterilising, contaminants may grow quickly upon addition of moisture and out-compete rhizobia

Table 2 Number of rhizobia in peat during storage time

Number of rhizobia (CFU/g) Rhizobial strains

Innitial 2 weeks 1 month 2 month 3 month 6 month CB1809

Local strain SL1

Local strain SL2

Local strain SL3

5.0 x 109 2.3 x 1091.8 x 1092.1 x 109

6.2 x 109 4.8 x 1093.4 x 1093.2 x 109

5.6 x 1092.9 x 109 2.2 x 1093.0 x 109

3.5 x 1093.0 x 1092.0 x 1091.8 x 109

2.2 x 1091.3 x 1091.3 x 1091.1 x 109

2.7 x 108 2.5 x 1081.2 x 1081.3 x 108

Ingredients for growth of rhizobia in large scale fermentors can be expensive The SFI investigated the replacement of laboratory grade yeast extract with industrial yeast extract and soybean extract All rhizobial strains grew to 109 cfu/mL indicating that fermentation media can be prepared using more cost effective and accessible ingredients (Table 3)

Table 3 Number of rhizobia grown on different growth media

Media Strains

4.2 x 1091.6 x 1092.7 x 1082.6 x 108

2.8 x 1093.4 x 1095.6 x 1095.2 x 109

A comparison of strain tolerance to temperature and pH revealed that there was little difference in growth of strains in media at different temperatures and growth of all strains was best at 30oC Numbers were low at 25oC and lowest at 37oC No cells grew at >45oC (Table 4) Differences were observed in growth of strains at pH 5.5 with Vietnamese strains achieving 10 fold higher numbers than Australian strains indicating acid tolerance that may

be an advantage in acid soil conditions (Table 5)

Table 4 Effect of temperature on growth of strains

Growth of strains Temperature (0C)

++

-

+ +++ ++

-

- no growth; + weak growth, around 104 – 105 CFU/ml; ++ moderate growth, 106 – 107 CFU/ml

+++ good growth, 108 – 109 CFU/ml

Table 5 Effect of pH on growth of rhizobial strains

- no growth; + weak growth, around 104 – 105 CFU/ml; ++ moderate growth, 106 – 107 CFU/ml

+++ good growth, 108 – 109 CFU/ml

The aims of laboratory experiments at the IAS were to determine the factors influencing the number of viable rhizobia in inoculants Higher numbers of rhizobia in the inoculant have greater potential to outcompete resident soil microflora after inoculation of legumes and increase N2 fixation The cost of inoculation would also decrease because of the need for smaller volumes

The IAS prepared inoculants using the Australian inoculant strains NC92 (groundnut) and CB1809 (soybean) Experiments were carried out to determine the effect of dilution of the broth culture on final numbers of rhizobia in peat mixed with worm casts and coconut coir dust after incubation (Table 6) Growth in the peat mixture of broth cultures diluted using yeast mannitol broth (YMB, to give final broth concentrations of 0.1%, 1%, 10% and 30%) was compared with growth of undiluted cells containing 109cfu/mL The peat mixture (70 g) was injected with 38 mL of broth After four weeks of growth, the results indicated that injection with undiluted broth cultures was necessary to achieve maximum numbers of 6.18 x

107 for NC92 and 5.85 x 108 for CB1809 in the peat mixture The ratio of viable cells of CB1809 injected at 0.1%, 1 % broth dilutions with undiluted broth indicated that no growth had occurred in the peat mixture possibly due to the presence of contaminant organisms (in the order of 106 /g peat mixture) Contaminants may have been encouraged by the addition of YMB to dilute broths before injection All counting was done using viable plate counts on yeast mannitol agar with and without Congo red (YMA and CRYMA respectively) and cell morphology of colonies was observed after gram staining under the light microscope

Table 6 Rhizobial cell number at different culture dilutions

No Rhizoial strains Culture dilution (%)

Cell number (CFU/g)

In another experiments at the IAS, survival of rhizobia was measured after injection of broth into mixtures of peat with worm casts and coconut coir dust (Table 7) The viable numbers of rhizobia in peat, peat (70%) mixed with worm casts (30%) and peat (40%) mixed with worm casts (30%) and coconut coir dust (30%) were counted after growth in carriers for one month

It was perceived that both worm casts and coconut coir dust increased the water holding capacity of peat Only 15 mL broth could be added to 70 g peat while 20 mL could be added

to peat mixed with worm casts and 35 mL to peat, worm cast and coconut coir dust mixture (by feel only moisture content not measured) It was also observed that coconut coir dust absorbs and loses moisture very easily Survival of rhizobia improved when peat had been amended with worm casts and both casts and coconut coir dust Although the carriers were injected with different quantities of broth, the numbers after one month did not reflect the different inoculum ratios at injection There was, however, higher numbers of contaminants

in peat mixtures than in peat alone and as counting was done by viable plate counts it is not clear if rhizobial numbers were overestimated due to incorrect identification of colonies It was recommended that experiments on carrier mixtures be carried out after adjusting carriers

to different moisture contents using protocols outlined later in this report It was also

recommended that all experiments include confirmation of the presence of rhizobia using plant infection tests also outlined later in this report

Table 7 Number of rhizobial cells in different carriers after one month

Number of rhizobial cells (CFU/g)

5 NC 92 Peat + worm casts + coconut coir dust 2.5 x 109

6 CB 1809 Peat + worm casts + coconut coir dust 6.8 x 108

Liquid inoculants at the IAS were prepared by adding PVA (0.5%), gum Arabic (0.17%) and sodium alginate (0.5%) to broth as adhesives for better survival of rhizobial cells on seed After one month rhizobial numbers were highest in PVA (4.15 x 108 /mL for NC92 and 6.88

x 107 /mL for CB1809) and Na alginate (1.04 x 108 /mL for NC92 and 5.38 x 107 /mL for CB1809) and lower in gum Arabic (4.42 x 106 /mL for NC92 and 8.73 x 107 /mL for CB1809) (Table 8)

recommendations for revised methods for conducting experiments

The following section outlines those recommendations related to experimental design and protocols for testing of peat-based legume inoculants Some suggestions are made to improve the designs of experiments presented by institutes at the project meeting at OPI on 18th and

19th February 2008 New strains of rhizobia CB1809 and NC92 will be supplied from ALIRU to OPI These will then be distributed for use in all research in 2008 Sterile Australian peat will also be supplied to institutes as a reference to compare the quality of Vietnamese peat All demonstration trials will be set up using Australian peat cultures of CB1809 and NC92

General information about peat quality

The quality of inoculants in Australia was improved, following widespread nodulation failures, by the amelioration of five main factors affecting survival in peat (Roughley and Vincent, 1967) Firstly, the origin of the peat was shown to be important Survival of clover, lucerne and cowpea rhizobia varied according to the location and depth of the peat source The peats tested varied according to their colour and texture but no explanation was given by the authors as to the cause of variation in survival Secondly, pH was shown to be critical and acid peats could be amended with calcium or magnesium carbonate Thirdly, peat sterilisation, preferably by gamma irradiation, was considered essential particularly for the growth and survival of slow-growing rhizobia presumably allowing them to out-compete faster-growing contaminants Fourthly, when rhizobia were added to peat previously dried at

100oC, they survived poorly due to both the heat of wetting generated upon inoculation and the production of inhibitory substances originating from the heat treatment Finally, moisture contents of 40 to 50% proved optimal for growth and survival of a range of rhizobial strains prepared as peat cultures Later, accumulation of salt in the peat deposit, caused by several dry seasons, was found to adversely affect rhizobial survival (Steinborn and Roughley, 1975) There may be many other factors affecting the quality of peat used for legume inoculants and the ultimate measure is the growth and survival of rhizobia

Adjusting the pH of peat

Adjusting pH of peat should be done carefully allowing time for equilibration The reaction between limestone and H+ in peat will depend on particle size of both limestone and peat The more finely milled the ingredients the faster the reaction Moisture content of peat is also important to allow the reaction to occur The amount of limestone required to change the

pH will depend on organic matter and clay content as well as buffering capacity of the peat After mixing peat and limestone should ideally be allowed to react for several weeks before

pH is tested It may also be necessary to measure pH over a longer period of time Finely milled agricultural lime (Aglime, calcium carbonate with some impurities passing through a

150 µm mesh) is the best limestone to use to adjust pH Builders lime is too caustic and other lime may be too weak

Water holding capacity of peat and optimizing moisture content

Particle size, organic matter and clay content of peat will play a role in water holding capacity and water potential It is desirable to increase the water holding capacity of the peat

so that larger amounts of broth culture (and hence more cells) can be introduced to peat before incubation Experimental design to test survival at different moisture contents is detailed as follows:

Optimum moisture content of peat or mixtures of peat with other ingredients should be determined at both IAS and SFI using only one strain of rhizobia to minimise the number of samples and the treatments listed in Table 9 Before injecting peat with broth, peat must be sterilised with a moisture content of 20% The efficacy of sterilisation should be measured by injecting broth without bacteria and measuring growth of contaminants over time for one