CIFOR–ICRAF publishes over 750 publications every year on agroforestry, forests and climate change, landscape restoration, rights, forest policy and much more – in multiple languages.

CIFOR–ICRAF addresses local challenges and opportunities while providing solutions to global problems for forests, landscapes, people and the planet.

We deliver actionable evidence and solutions to transform how land is used and how food is produced: conserving and restoring ecosystems, responding to the global climate, malnutrition, biodiversity and desertification crises. In short, improving people’s lives.

Tree seed source re-classification manual

Export citation

A seed source classification system is a prac tical tool in seed documentation which is an essential tool in quality assurance and marketing. Di fferent seed source classes have been used from th e early stages of tree improvement to rank the genetic history of plantation seed (Zobel and Talbert 1984, Wright 1976). Definitions of tree seed sources have been developed and applie d in many countries, and standardised (slightly differently) by OECD, EU, FAO, DFSC, GTZ and others (for convenience we call it th e OECD system). These guidelines generally rank seed sources into unclassified and selected source s and for selected sources, into tested and untested reproductive material (e.g. OECD, 2 007, Albrecht, 1993, Barner et al., 1988). The OECD system has been developed and is accordingly most suitable for plantation species in temperate indus trialised countries. However, the OECD system fa ils to include the majority of seed sources used by smallholders in the tropics. Most of these seed comes from trees scattered on farmland or from natural forests. Introduction of the class ‘farmland seed sources’ is an attempt to include this large and important category of seed sources. The ‘farmland seed sources’ may be remnants of natural vegetations or plante d trees. Only a very small part of the reproductive material utilised for agroforestry is fro m selected, let alone tested sources, and the currently used de finitions therefore disregard options for applying simple principles that can widen the genetic quality considerably and minimise the risk of inbreeding. Furtherm ore the terms ‘selected’ and ‘unclassified’ sources presuppose that central authorities ha ve the capacity to es tablish, maintain, protect and evaluate improved sources of re levance to the consumers (smallholders), which is rarely the case. We suggest that it will be more fruitful to apply common sense criteria of quality to the bulk of reproductive mate rial that is used by smallholders than to uphold strict crite ria that are difficult to apply. The simplified system used in this manual wa s first introduced in Nepal by Dhakal et al. (2005). The present document is mainly been an attempt to ‘cover’ the seed flow system with a major componen t of agroforestry species in the tropical East Africa. However, the principles applied do not have geographical restrictions but can in practice be applied anywhere. Compared to the OECD seed scheme the main change is the introduction of farmland seed sources, and the elimin ation of the terms selected and identified. These term s may still be used as quality ranking in seed documentation. Some ‘classes’ have been e liminated as distinct classes e.g. Seed Production Areas and Seed Stan ds. It has therefore been necessary to re-classify these sources in the present system . Each category of the fi ve new tree seed sources classification (Natural, farm land, plantation, Tree se ed orchard and vegetative propagules) were tested in the field.

Related publications