The purpose of this booklet is to facilitate the first assessment of trials established within the framework of the International Neem Network. During 1995/96 some 30 provenance trials were established in 17 countries. The proposed assessment methodology reflects the many uses of neem, however, since resources to carry out the assessment might be scarce and since not all charecters will be of relevance for a given trial, a minimum set of characters is described in the first part. The minumum set is complemented by a set of optional characters which could be considered in the future, or could be included in the present assessements, should the country wish so.
The International Neem Network was established in 1993 with the long term objective to improve the genetic quality and adaptability of Neem and to improve its utilisation. The network collaborators have undertaken activities in provenance exploration, seed collection and exchange as well as establishment of internationally coordinated trials. In addition, the network has undertaken research in seed physiology and technology, genetic diversity and reproductive biology, as well as studies on variation in chemical compounds.
National institutes in 21 countries are collaborating in the network together with a number of international organisations and projects. The Food and Agriculture Organisation of the United Nations (FAO) is entrusted the global coordination of the Network activities. The network has issued the following booklets and guidelines related to its activities:
The present booklet has been prepared by a working group1 which was established at the workshop of the International Neem Network held in Bangkok 14-15 March 1996. The guidelines were agreed upon at the workshop of the International Neem Network in Yangon, Myanmar, 28 July - 1 August 1997. The proposed assessment methodology is based on DFSC (1994 and 1996), Stewart (1989 and 1990), MacDicken et al. (1991), Stewart et al. (1992) and Kundu and Tigerstedt (1997).
Network collaborators who are uncertain about the guidelines or have practical problems in the trial assessment are welcome to contact the coordinator at the below address, where also further information on the International Neem Network and its activities can be obtained :
FAO, Forest Resources Division
Viale delle Terme di Caracalla
I-00100 Rome, Italy
Fax: 0039 6 570 55137,
The study of the field trials will examine any differences in respect to adaptability and growth between provenances of neem covering the entire distribution area over a range of sites. The overall objective is to provide guidelines for the choice of seed sources of neem for given sites and end-uses thus improving the production of goods and services in the tropical countries where neem is used. The specific objectives are:
Trial data to be provided can be divided into three major groups:
Re 1. Description of the site was included in the procedures for the design in trial establishment (International Neem Network, 1996). The form to be used in the description of the site is attached as Annex A for easy reference. Collaborators who have not completed description forms for each established trial, should do so as part of the first assessment. A copy of the form should be sent to the coordinator (FAO). The trial site description forms will be compiled in a booklet (orange).
Re 2. Description of the 25 seed sources which were exchanged between collaborators in the International Neem Network in 1995 can be found in the booklet on seed sources ("The blue booklet"). Collaborators who have included local, or other additional seed sources in their trials, should describe these sources in a similar way.
Re 3. Data collection in the trials require:
The characters to be assessed and the intensity of assessment required
depend on the age (development stage) of the trial. The present manual
covers primarily the first five years of the trials, where annual (and
for some characters, if possible, even more frequent) assessment is foreseen.
A sketch map of the trial showing the orientation and distribution of seedlots in blocks and plots may already be available (ref point 3.2 (iii) in the booklet on procedures in trial design by the International Neem Network 1996). An example is given in Annex B. It is very important for the assessment as well as for the analysis that a map showing the layout of blocks and plots is available.
Data will be recorded on individual tree basis for most characters. To identify the tree it is advisable to register the position of all trees in the plot. The identification of individual trees will enable correction of any mistake found later. The trees should be numbered in a row continuously (also dead or missing trees should have a number) as shown in Figure 1 and the field orientation (an arrow pointing towards North) should be noted. Should another data recording system be used, this should be explicitly mentioned.
Figure 1. Proposed system to number individual trees within a plot
Assessment of a large number of characters is time consuming and expensive. The potential characters to be assessed in the field trials should be carefully reviewed, and a selection of characters should be made based on the end uses of neem in the country as well as the available human and financial resources and capabilities.
In this chapter and the following chapter a distinction is made between a minimum set of characters to be assessed by all collaborators (core characters), and characters which are optional (optional characters). This distinction provides a flexible system, allowing regional and global analysis to be carried out based on the minimum set (core set) of characters, while at the same time giving individual network collaborators the possibility to design additional assessments (by including optional characters) of special importance to the country.
The characters included in the minimum set of characters should be assessed for all trees in the plot and in all plots in the trial, every year during the first five years of the trial (see Form 1 and Form 2 in Annex C). Plot border rows are also measured because at an early stage of development the plot border rows are not likely to differ significantly from the plot core area. The registration of tree position will make it possible to take into account the effect of competition in the plot border rows , if required.
Resources permitting, the character related to flowering and fruiting could be assessed several times during flowering and fruiting season.
The following characters should, as a minimum, be included in the yearly assessment by all collaborators (see chapter 6 for optional characters to be assessed):
The assessment results for these characters are registered using assessment form 1 and 2 in Annex C.
The health status can be assessed using the following categories, slightly modified after McDicken et al 1991 and Wellendorf 1989 (cf. DFSC 1996). All trees within each plot should be measured.
|Part of the tree affected||Severity of attack||Type of damage|
|S||Stem||Main Class||Value Sub- Class||% damage of part affected||Severity scale (from Rohrmoser)||N||No damage|
|L||Leaves||Severe||1||> 90 %||Affected plant parts dominate the appearance of the plant||I||Insect damage|
|W||Whole tree||3||60-70%||P||Animal (physical)|
|6||30-40%||Healthy plant parts dominates the appearance of the plant||F||Fire|
|Slight||7||20-30%||S||Water stress (drought)|
|8||10-20%||Symptoms scattered||L||Water logging|
When the affected part of the tree is identified for example the stem, the severity of the attack should be noted by judging the percentage damaged of the affected part for example how much of the stem is damaged. If for example around 45% of the stem was damaged, the severity of the attack was medium and subclass value was 5. Then the type of damage (or cause of damage) should be noted. Some types of damage are easy to identify e.g. "wrong species" (X) other causes can sometimes be difficult to know, however, local experiences from the trial site may assist in this task for example if the site is well known to suffer from e.g. frost or phosphor deficit.
Survival percentages (the ratio between number of living trees remaining and number of trees originally planted) for individual plots, blocks and provenances can be calculated using the health status assessments. Alternatively, all living trees in the plot can be counted separately; this is a reliable method that minimises gross errors.
Height defined as the vertical height from the ground to the highest growing point is assessed on all tress within the plot. Measurement is done with a measurement pole (in the first assessments, when trees are small) or a clinometer when trees become taller (> 10m). Measurements may be recorded to the 0.1 m accuracy.
The diameter (or the circumference/girth) should in the first five years be measured at 0.3 m. A 0.3 m stick can be used to help identifying the measuring point. Assessment is done with the use of either caliper or diameter tape. Method of measurement (diameter or circumference/girth) should be noted on the form. If using caliper two readings at right angles to each other should be taken. Both readings should be recorded. The average does not need to be calculated in the field, but can be calculated at a later stage. If there are multiple stems the diameter of the three largest stems (above 1 cm in diameter) should be recorded.
Experience has shown that the diameter at 0.3 m (or basal area at 0.3 m) provides the best parameter for estimating biomass (cf. McDicken et al 1991 and Stewart et al 1992). In addition, it is also easy to assess. Later, however, it may be more convenient to assess the diameter at 1.3 m. It should be clearly noted on the form if (or when) the assessment changes from 0.3 m to 1.3 m. This change should only be made if the diameter can be measured for all trees in all plots at 1.3 m in the same trial.
Straightness is assessed according to a slightly modified scale developed by Lauridsen et al. (1987 and 1995 cf. DFSC 1996). The trees are grouped in three main classes based on main features: straight trees, waving trees and crooked trees. Each main class is then subdivided in three. The definition of the three main classes are illustrated in Figure 2. The assessment of straightness is a judgement of the stem(s) from the ground to the crown, e.g. in the case of multiple stems the judgement should express the average of the main stems. The branching habit is thus not expressed by this character but by the following character "number of stems at 1.3 m". As straightness is based on individual judgement, it is important that the same person assess the straightness within each block and preferably also within each trial. This will make it possible to rank the provenances.
|Straight trees||Straight stem(s), free of any faults for timber||9|
|Crooked, bent or twisted trees||Severe bends||3|
|Completely crooked or twisted tree||1|
|Straight trees (value 7-9)||Straight stems have no severe bends, nor a waving appearance. The distinction between the classes can be that some stems are slightly twisted or other less severe faults for timber.|
|Waving trees (value 4-6)||Waving stems do not have crooks (severe bends) see below. The
distinction between subclasses can be more or less waving appearance
A bend (waving)
|Crooked trees (value 1-3)||A crook is a bend where a line from each end of the curve falls
outside the stem - see drawing. The distinction between subclasses
can be the number of crooks or other severe faults like twisting
Figure 2: Illustration of straightness
The total number of stems should be counted at 1.3 m height. Stems are defined as all stems and branches over 1 cm in diameter. The branches should be appearing from a point below 1.3 m (e.g. not counting branches hanging down from above).
The crown diameter should be measured using a measuring tape. The first measurement should be taken parallel to the row of trees in the plot from one edge of the crown to the other edge (using the projection of the crown on the ground). The next measurement should be taken along the axis perpendicular to the first one. Both measurements should be recorded.
Flowering and fruiting should be assessed quantitatively and qualitatively.
Stage and quantity are assessed from the ground. In the case of tall trees, binoculars can be used. Flowers and fruits are investigated more closely on a number of sample trees by collecting a few samples from the crown. The total number of flowers/fruits can be estimated by counting the number on an average flower/fruit bearing branch and multiplying by the number of such branches.
For quantitative evaluation of flowering and fruiting of individual trees a logarithmic score should be used:
|Average No. Flowers or Fruits||Score||No. of Flowers or Fruits|
The score was developed by Wellendorf (Sirikul et al.1991).
The following score should be used for the qualitative evaluation of development stages of flowering and fruiting of individual trees:
|D||Dry empty fruits|
In practise a quantitative score (1-9) is given for each development stage (B-D) present (cf. assessment form 2)
The maturity of the fruit can be judged by the colour and the firmness of the fruit. The colour of mature neem seed is green-yellow and the seed is firm (not hard, not soft). Figure 3 shows the colour of immature fruit, mature fruit and over mature fruit.
Figure 3: From left side of the picture is shown the colour of immature neem seed, mature neem seed, and over mature neem seed.
The seed production is not of equal importance to all participating countries. As a minimum, one seed collection of three or more provenances should be made when the trial is 5 years old. For each of the selected provenances the total production (in kilos) of each of the plots in the trial should be recorded. The provenances to be collected can be chosen according to the most important uses of neem in the country e.g. in a country where seed production is of significant importance, the three provenances producing most seed could be collected and assessed; and in a country where poles and other wood products are the main products of neem, the provenances to be collected could be the three tallest provenances.
A canvas sheet or alike should be placed under the trees in the plot when collecting seed. All seed should be collected.
When the seed is collected, a sample of 100 fruits from each collected
plot should be cut with a sharp knife. Out of the 100 fruits, the number
of full and normal seeds (with no damaged or empty kernel) should be counted
by looking at the appearance of the kernel (cutting test in the field).
When the trials get older, many of the optional characters will increase in relevance. Therefore, the International Neem Network might decide to elaborate common guidelines for a sub-set of these characters at a later stage.
The International Neem Network at its meeting in Yangon, Myanmar in 1997 decided to continue the existing working group on trial assessment with the objective to develop and improve the guidelines on optional characters that relates to adaptability and end uses of neem. Characters under consideration include: drought resistance, estimation of above ground biomass, calorific value and fodder value.
In addition, another working group was established at the Yangon meeting to look more into the possible assessment of chemical compounds in neem seed. In the initial phase this group will look into the following questions: (i) identification of the most important chemical compounds to assess; (ii) techniques for quantification; and (iii) procedures for assessment, including practical arrangements and cost considerations. This working group will report back to future network meetings and liase with the working group on optional characters.
This chapter will only give a general overview of a number of optional characters at present under consideration. The methods to assess some of these characters are well described in other manuals; for referenceplease refer to the list of literature in chapter 10. If time and resources permit, some or all of the following optional characters of relevance to the country could be included in field assessments.
A sample taken from the seed collection of provenances (see 9. in chapter 5) from each of the plots could form the basis for assessing seed size (weight and length), germination % and for assessing the azadirachtin and oil content in the seeds. This will of course only provide information for the provenances that are collected.
It is important, in general, to keep the seed samples at separate the plot level, in order to be able to determine any effect from the environment (differences between plots) for the characters assessed.
A sample of the fruits should be taken from seed collected at plot level. The fruits should be weighed and the length assessed.
A sample of the processed seed (collected at plot level) should be taken, and the seed should be germinated to estimate the germination %.
Samples should be analysed by a biochemist to assess the azadirachtin content. The azadirachtin content per kg seed can then be compared between provenances. Additional information will be provided later by the working group on chemical compounds.
Samples of seed should be sent to a biochemist for analysis of oil content. The oil content per kg seed can then be compared between provenances. Additional information will be provided later by the working group on chemical compounds.
In countries where neem is used for fodder, leaf samples can be collected and send for chemical analysis by animal nutrition experts.
The assessment of total above ground biomass requires a destructive method, and the assessment could be done at the time when the trial needs thinning or when the trial is not supposed to be continued any longer.
A sample of trees from each provenance at plot level is felled. Then
the total biomass is separated into size components. Each of the size
components are weighed. Dry weight and specific gravity of the wood in
each size component are estimated based on fresh weight, dry weight and
volume of the wood samples from each size class. Dry weight of foliage
can be assessed in a similar way (cf. e.g. DFSC 1994).
Basic information should be collected in the office before going to the field in order to be able to verify and complement this information in the field visit. It might be useful to organise the field assessment by dividing the assessment group in two teams:
Partly completed forms are exchanged between the two teams at appropriate intervals. This provides a check for gross errors (e.g. measuring the wrong row of trees) as each team has the measurements of the other team at hand after the first exchange of forms. It also ensures that the person recording data in the forms has sufficient time to monitor the assessment.
The phenology can hereafter be assessed by the two teams:
It would also be possible to organise the assessment for one team taking
all measurements at the same time. This way the notetaker ("rapporteur")
will have a difficult task (busy!) which might slow down the assessment.
If this method is used special care should be taken to verify the plot
and tree position.
The following list is a proposal for basic equipment to take to the field:
All assessed data should be properly recorded and stored. The result of the assessment should be reported at national level. A copy of the trial assessment report, including all completed assessment forms, should be sent to the coordinator of the International Neem Network (FAO - see address in the introduction).
Danida Forest Seed Centre 1996: Note on field assessment of ex situ conservation stands. Danida Forest Seed Centre, Humlebaek, Denmark.
Danida Forest Seed Centre 1994a: Evaluation of field trials established within the framework of the FAO Project on Genetic Resources of Arid and Semi Arid Zone Arboreal Species. Preliminary note on assessment Methodology. Danida Forest Seed Centre, Humlebaek, Denmark.
Danida Forest Seed Centre 1994b: Preliminary Assessment Reports. Trial No. 1-2. Evaluation of field trials established within the framework of the FAO project on Genetic Resources of Arid and Semi Arid Zone Arboreal Species. Danida Forest Seed Centre, Humlebaek, Denmark.
Kundu, S.K. and P.M.A. Tigerstedt 1997: Geographical variation in seed and seedling traits of Neem (Azadirachta indica A. Juss.) among ten populations studied in growth chamber (under publication, personal communication).
McDicken, K.G., G.V. Wolf and Briscoe (editors) 1991: Standard Research Methods for Multipurpose Trees and Shrubs. 92 pp. Windrock International Institute for Agricultural Development.
Pinyopusarerk, K., B.V. Gunn, E.R. Williams and L.D. Pryor 1993: Comparative, geographical variation in seedling morphology of three closely related Red Mahoganies, Eucalyptus urophylla and E. pellita and E. scias., Aust. J. Bot. 41: 23-34.
Sirikul, W., H. Wellendorf and J. Granhof 1991: Provenance x site interaction in cone setting of Pinus caribaea var. hondurensis in Thailand. Forest Tree Improvement 24. Arboretet, Horsholm, Denmark.
Stewart, J.L. 1989: Aspects of biomass estimation in multipurpose trees, pp. 31-324 in: Breeding Tropical Trees: Population Structure and Genetic Improvement Strategies in Clonal and Seedling Forestry. Proceeding IUFRO conference, Pattaya, Thailand, November 1988 (Eds. Gibson, G.I., A.R. Griffin and A.C. Matheson). Winrock International and Oxford Forestry Institute.
Stewart, J.L. 1990: Evaluation manual. International Trials of Central America Dry Zone Hardwood Species. 51 pp. + Appendices 13 pp. Oxford Forestry Institute.
Stewart, J.L., A.J. Dunsdon, J.J. Hellin and C.E. Hughes 1992: Wood Biomass Estimation of Central American Dry Zone Species. 83 pp. Tropical Forestry Papers 26, Oxford Forestry Institute.
Thomsen, A. and O. Souvannavong 1994: The International Neem
Network. Forest Genetic Resources Information. 22: pp. 49-51. FAO.
1 - Mr. C.J.S.K. Emmanuel, AFRI, P.O. Krishi
Mandi, Bhagar Ki Kothi, New Pali Road, Jodphur, Rajasthan, 342 005, India.
Mr. A. Gaye, Institut Senegalais de Recherches Agricoles, Parc Forestier du Hann, B.P. 2312, Dakar, Senegal.
Mr. B. Boontawee, Silvicultural Research Division, Royal Forest Department, Chatuchak, Phaholyothin Road, Bangkok 10900, Thailand.
Ms. A. Thomsen & Mr. L. Graudal, Danida Forest Seed Center, Krogerupvej 21, 3050 Humlebaek, Denmark