Overstory #130 - Seed Sources for Collection
Seed propagation is the principal mode of propagation for trees in temperate as well as in tropical regions. Seeds are unique in natural regeneration and propagation because:
- Seeds constitute unique genetic compositions, resulting from mixing parental genetic material. The result is genetic variation of the offspring, which in turn enhances ecological adaptability.
- Seeds are usually produced in large numbers and are readily available, each year or at longer intervals.
- Seeds are (usually) small concentrated packages of plants-to-be, containing nutrients for the establishment of the plant and, except for recalcitrant (short-lived) seeds, usually resistant to damage and environmental stress.
- Many seeds can be stored for long periods under cold and dry conditions.
Considering that in a regenerational context only one successful seed (or two if the species is dioecious) is necessary to replace the parent tree(s), the production of seeds during the lifetime of a tree is exorbitant. A full grown Eucalyptus camaldulensis tree may produce a million or more seeds per year and may live and produce seeds for a century. Each year's production could afforest several hundred hectares. Although seed production is smaller in most other species, it is probably never a limiting factor in natural regeneration. Each seed contains the potential for becoming a full grown tree, but in nature most of the seed production will succumb to failed dispersal, predation, infestation, natural deterioration, germination failure, etc.
The objective of seed handling is to achieve a high survival and germination rate of the seed. Seed handling encompasses a series of procedures beginning with selection of the best quality seed source, collection, processing, storage, pretreatment, and germination. Each link of this chain implies a potential risk of losing seed, and any link in the process is of equal importance (though not necessarily equally sensitive). If a seed dies due to careless handling during collection or processing, even the best storage, pretreatment or germination conditions will not bring it back to life. If a seed dies during a handling procedure, the whole preceding effort is wasted.
The whole process of seed handling begins with collection of good-quality seeds, both physiologically and genetically. The genetic quality of seed will affect a plantation for years ahead, and since the operational cost of seed handling is almost the same regardless of genetic history, expensive handling will pay better when applied to good quality seed. Referring to the genetic quality of seeds, the Australian Tree Seed Centre has adopted the slogan: 'Good seed does not cost - it pays' (Midgley 1996), meaning that the small investment in obtaining the best seed source is minor compared to the potentially better growth of the offspring.
Genetic quality of seeds
The term 'seed source' applies to the stand of trees where seed is collected. A seed source can be a number of single trees, a natural stand, a plantation, and a seed-production area or seed orchard. Seed trees are the individual trees from which the seeds are collected. Potential seed sources are identified in the planning phase; actual seed trees are often only selected during the seed collection. "Phenotype," simply put, is the tree as we observe it, and "genotype" is the genetic constitution of the tree.
A seed source should yield an appropriate quantity of seed with a high physiological and genetic quality which matches the plantation site and purpose. In general the seed trees should be of good phenotype, neither juvenile nor over-mature and good seed producers (Morandini 1962). For special planting purposes, for example conservation or provenance seed stands, special consideration on sampling for the capture of genetic diversity may be included. For plantations not intended for future seed production, genetic diversity is usually of less importance, but collection should avoid inbred seed and inferior parent trees, which may affect the performance of the plantation. If, however, the plantation is envisaged to become a seed source itself some time in the future, appropriate measures should be taken to assure reasonable genetic diversity.
The genetic constitution or inheritance carried by the seeds makes up the potential performance of the progeny: if the genetic potential is poor, the performance will remain poor regardless of environment and silvicultural efforts; if the genetic potential is good, this potential may be expressed by appropriate silvicultural measures. Genetic quality can only be proven by genetic tests (e.g. progeny tests) which are outside the scope of this article. Yet, in the selection of seed sources and seed trees of unknown genetic constitution a few measures and precautions can and should be taken in order to avoid genetic inferiority, viz.
Collect from a diverse genetic base
A narrow genetic base implies a risk of inbreeding. In a population of few flowering individuals the risk of self-pollination is high, and unless the species has a strong inbreeding barrier, many seeds of a small breeding population may be inbred. Isolated trees or trees flowering out of phase with the majority of the population are more likely to self-pollinate and consequently produce inbred seed. Therefore, such trees should be rejected as seed trees,
Neighbouring trees in natural stands are often half sibs or full sibs (Griffin 1990). Species with short-range pollination and dispersal are more likely to create groups of related individuals in the stand than species with long distance pollination and dispersal. This is especially to be considered in natural stands of a single dominant species, e.g. Tectona grandis, Acacia senegal, Brachystegia spp. and many pines and eucalypts. A distance of 100 meters between seed trees is usually considered a minimum in natural stands, but it depends on collection purpose (Gray 1990, Palmberg 1985). Genetic diversity is also assured by collecting from a large number of seed trees. Special sampling techniques are applied for special collections like trials or ex situ conservation (Eldridge et al. 1992, Palmberg 1985).
The genetic history (e.g., the number of mother trees) of seed sources of planted material is important. Plantations raised from a narrow genetic base (i.e. few mother trees) should be rejected as seed sources. Obviously this is even worse in clonal plantations, unless specifically designed for seed production.
Many exotic plantations are known to have originated from few mother trees during the first introduction. For example, mahoganies (Swietenia spp.) cultivated in many parts of Asia are believed to originate from a small number of seed trees in Honduras and Belize. Unless new material from a broader genetic base has been introduced later, plantations raised from seeds of such trees are likely to suffer from inbreeding depression. Other examples of exotic plantations based on a few mother trees are Cupressus in Kenya and Gliricidia in Sri Lanka.
Compared to natural stands, neighbouring trees in plantations are less likely to be related, provided the total genetic base is broad. This is because both seeds and plants are usually mixed during the establishment. Consequently, distance requirement for seed trees is less strict in plantations than in natural stands. However, as neighbouring trees are likely to be pollinated by the same pollen cloud, seeds collected from two adjacent trees may have a higher probability of being related on the paternal side than two distant trees.
Plantations raised from a broad genetic base and superior phenotypes are good seed sources. Since both seeds and plants have been mixed during establishment, the risk of neighbouring trees being related is not higher than for distant trees. Therefore distance requirement during sampling is less important.
Avoid collecting from trees that appear inferior
The phenotype (the tree as we see it) is a product of both genotype and environment. A poor phenotype can be caused by detrimental environment and the progeny may perform excellently when grown under favourable conditions in plantations. For example, Lake Albucutya provenance of Eucalyptus camaldulensis grows bent and crooked in its natural environment where it is heavily exposed to wind and sand flow. Grown in plantations (e.g., in Israel) the progeny grows fast and straight.
Yet, phenotypic selection does have a justification: if the phenotypic quality is good, then we know that the tree has the genetic potential for good performance; if the phenotype is poor, then we do not know the cause. Hence, in environments with moderate environmental stress a certain selection of seed trees is appropriate. Trees with exceptionally poor phenotypes (multiple stems, forking, attack by diseases etc.) should be avoided.
To avoid detrimental genetic effects in seed collection the following practical measures are recommended (For. Com. 1994):
- Avoid seed collection from sites where seed crops are sparse or heavy crops restricted to isolated trees, i.e., give preference to stands with heavily fruiting trees in close proximity to each other.
- Within the preferred stand, spread each collection over the largest possibly number of widely dispersed trees; collect from at least 15 trees which are preferably at least 100 m apart.
- Collect from vigorous trees of good form; some defects due to physical damage (e.g. from fire or falling trees) can be ignored.
Species and provenance
Within a species (or other taxonomic sub-unit) a large variation exists in terms of ecological adaptation and growth forms. In botanical ecology the term ecotype designates a special growth site (habitat), e.g. dry zone, humid zone or high altitude. In forestry the term provenance has come into common use as the place of origin of the planting material because it designates both the ecotype and the growth habit (e.g. fast growth, straightness of stem or other desired traits). For example, Eucalyptus camaldulensis grows over most of the Australian continent.
Despite its morphological similarity, various ecotypes occur according to different ecological conditions. Variations of growth habits have been revealed through provenance trials, i.e., trials of comparative performance of different seed sources grown under similar conditions. Provenance names such as Petford and Lake Albucutya are known sources of seeds of the species whose progeny has proven superior growth habit in many areas of the world with climate and soil similar to the original site.
The provenance name normally designates the distinct location of origin of the seed source, for example named from the nearest town, lake, river or hill. An ideal provenance is characterized by (after Barner 1975):
- It is composed of a community of potentially interbreeding trees of similar genetic constitution (and significantly different from the genetic constitution of other provenances).
- It is sufficiently large for the collection of reproductive material in quantities significant for forest practice.
- It can be defined by means of boundaries that can be identified in the field.
Although boundaries of gene flow (interbreeding) may be difficult to define in areas with more or less continuous population, the provenance concept is practical in forestry and should be included in seed documentation.
Planning of activities based on biological systems must necessarily be flexible and adjustable since these systems are often unpredictable or things can change rapidly. A crop failure of one species or seed source may be compensated for by a larger collection of another species or seed source. A sudden abundant fruiting of a rarely fruiting species should be taken advantage of by a large collection of that species. In some species, processing can be delayed without detrimental consequences, e.g., for dry orthodox seeds; in other species lack of preparation or shortage of capacity of the processing unit may ruin an otherwise successful seed collection. In some cases seed availability may influence the planting programme, especially for those species where a storage buffer is lacking, e.g., for recalcitrant seeds. Hence, planning and management of seed collection involves the whole seed-handling process.
Barner, H. 1975. Identification of sources for procurement of reproductive materials. In: Report on FAG/DAN IDA Training Course on Forest Seed Collection and Handling. Vol. 2: 42-64, FAO Rome.
Eldridge, K., Davidson, J., Harwood, C. and Wyk, G. van. 1992. Eucalypt domestication and breeding. Oxford Science Publication.
For. Com. 1994. Seed manual. Procedures for seed collection, handling, storage, purchase and accounting. Forestry Commission, Hobart, Tasmania.
Gray, R. 1990. Professional seed collection. In: Conference Proceedings: Sowing the seeds, direct seeding and natural regeneration conference. Greening Australia.
Griffin, A.R. 1990. Effects of inbreeding on growth of forest trees and implications for management of seed supplies for plantation programmes. In: Reproductive ecology of tropical forest plants. (Bawa. KS. and Hadley, M., eds.). Man and the Biosphere Series, Vol. 7. pp 355-374.
Midgley, S. 1996. Seed collection strategies in a changing world. In: IUFRO: Innovations in tropical tree seed technology. Proceedings of the IUFRO Symposium of the Project Group P.2.04.00, 'Seed Problems', Arusha, Tanzania. 1995.
Morandini, R. 1962. Forest seed handling, equipment and procedures, I: Seed production, collection and extraction. Unasylva 15:4, 1-15.
Palmberg, C. 1985. Sampling in seed collection. In: Forest tree improvement. FAO Forestry Paper No. 20: 41-45. FAO Rome.
This article was adapted with the kind permission of the author and publisher from:
Schmidt, L. 2000. Guide to Handling of Tropical and Subtropical Forest Seed. Danida Forest Seed Centre. Humlebaek, Denmark.
This exceptional guide covers forest tree seed handling from scientific, practical and administrative perspectives. For further information about the book and a wide range of other publications contact:
About the author
Lars Schmidt is Chief Technical Adviser of the Indonesia Forest Seed Project, a Danish-Indonesian support project to the Indonesian forest seed sector. Lars is a biologist specialising in tropical forest ecosystems and tropical forest seed. He has been adviser to international and bi-lateral forestry projects in Malawi, the Philippines and Indochina. In Indochina he was Technical Adviser for Vietnam, regional training adviser and regional coordinator on conservation of Forest Genetic Resources. He is presently on leave from Danida Forest Seed Centre, Denmark. His publications include mainly technical guidelines and articles. Address: Indonesia Forest Seed Project, Taman Hutan Raya Ir. H. Juanda No. 120, Dago Pagar, Bandung 40198, Jawa Barat, PO Box. 6919 Bandung 40135, Indonesia. Tel/fax: 62-22-2515895. E-mail: firstname.lastname@example.org
Related editions to The Overstory