J.M. FIELDINGJ.M. FIELDING is officer-in-charge of silvicultural research at the Forest Research Institute of the Forestry arid Timber Bureau, Department of National Development, Canberra, Australia. Since the beginning of 1969 he has been on assignment with FAO as project manager of the United Nations Development Programme project for pilot plantations of quick-growing industrial tree species in Malaysia.
THE MOST appropriate method for mass production of improved material depends on various factors, including the nature of the species, the stage reached in the breeding programme, the aims of the afforestation programme, facilities and staff available, the extent of the demand for improved material, the method of afforestation, and the estimated relative rates of gain and relative costs of the different methods.
The transfer of provenances and the use of seed trees and seed production areas are simple and useful interim methods.
Valuable improvements can be obtained in some localities through the use of seed of a nonlocal provenance, as for example in the case of Pinus sylvestris and Picea abies (Langlet, 1964; Bergman, 1965; Stefansson and Sinko, 1967), and possibly in the case of Araucaria cunninghamii (Slee and Reilly, 1967). When desirable provenances have been located, seed should be collected from selected stands or trees.
The collection of seed from seed trees (selected, open pollinated trees growing in the forest) can give valuable improvements if selection has been careful and sufficiently intensive; this is particularly true if selection has been made to improve characteristics having high heritability (Slee and Reilly, 1967). Seed trees should not only be good phenotypes, but should produce adequate amounts of seed. High variation in seed production among the dominant trees of a stand is a characteristic of many species that must be considered when seed trees are selected.
Seed production areas are being used widely throughout the world. Pitcher (1965) stated that 35 such areas had been set up in the north central region of the United States. A seed production area is a plus stand developed by thinning to remove the poorer phenotypes and managed with the object of producing large crops of seed. Methods for establishing and managing seed production areas in the United States have been described by Cole (1959, 1963) and by Eliason (1969). In selecting a seed production area in plantations of an exotic species, consideration must be given to possible genotype-site interaction in. stand quality and seed yield.
Methods of inducing precocious flowering are needed for many species. Delayed male flowering, for example, is a major problem in mass-producing improved seed of Araucaria cunninghamii, and it is a problem in seed orchards of certain species of Pinus.
Also, improved methods are needed for increasing seed production and for reducing annual variations in the size of seed crops. The methods used at present in seed orchards include fertilizing, soil cultivation, irrigation, disease and pest control, and the positioning of seed orchards in localities and on sites favourable to the production of seed. However, when in order to increase seed production a seed orchard is located in a different region from that in which the seed is to be used, consideration should be given to possible effects from preconditioning of the seed (Rowe, 1964).
A better understanding of the factors controlling flowering and seed production in trees is needed. Physiology of flowering and seed production is a very important field for forestry research.
Better methods of estimating the size of seed crops are needed to aid the planning of seed collection programmes. For example, Silen (1967) has suggested that an estimate in the fall of the male cone production of Pseudotsuga menziesii may provide a useful, early forecast of the female cone crop of the following year.
The possibility of extending the period of cone collection has been studied for a number of species. Not only must the early collected seed develop high viability, but this viability must be satisfactorily maintained over normal storage periods (Bevege, 1965; Yanagisawa, 1965; Krugman, 1966; Pfister, 1966). Variations among the trees of a stand in the time at which cones mature and in the response of their seed to after-ripening treatments (Franklin. 1965) is an important consideration.
The high intrinsic value and present general scarcity of improved seed make it desirable to raise the greatest number of plantable seedlings practicable from the seed. This requires for many present-day forestry operations a reappraisal of seed extraction, seed storage, nursery, and planting methods in- order to realize the highest possible rate of establishment of superior seedlings in plantations.
Seed orchards are the main method used by tree breeders at present for mass-producing forest tree seed. Their object is to produce seed of known source and of good genetic quality at a reasonably low cost.
The choice between clonal and seedling seed orchards rests on various considerations including estimated rates of gain, available staff and facilities, relative precocity and abundance of the seed production of clones and seedlings, species characteristics, such as ease of vegetative propagation, and presence or absence of progeny tests suitable for conversion into seedling seed orchards. Neither clonal nor seedling seed orchards are generally accepted as superior as regards rate of gain for all heritabilities, selection intensities, and other factors (Toda, 1964; Namkoong et al., 1966).
For some species, seed production is earlier and more prolific in a clonal than in a comparable seedling stand. This is advantageous, particularly because of the present urgent need for improved seed. However, sound comparative data are not available for many species. In the case of species such as Pinus radiata, more abundant seed production is to be expected in the early years of a clonal seed orchard, but in the case of some other pines, as for example Pinus pinaster (Illy, 1966), clonal and seedling seed orchards differ little in this regard.
Selfing for many important forest plantation trees results in distinct inbreeding depression and in a low production of viable seeds. Fowler (1965) reported selfing in Pinus resinosa to be greater in the lower than in the upper part of the crown. Selfing within a particular ramet is likely to be greater in the early years of a seed orchard when the crowns are wide apart and when precocity of flowering is most obvious. Very little pollen is produced during the early years of some pine seed orchards, and the cones contain only few viable seeds. For this reason and to allow for the later thinning of inferior genotypes, close initial spacing has been used in certain orchards. Minimum number of clones needed to avoid serious inbreeding effects in clonal seed orchards has been estimated by Stern (1959). Self-incompatibility also can be exploited to reduce the frequency of spontaneous selfing in orchards (Cram, 1969).
Methods for establishing seed orchards are well developed for many important conifer and hardwood species (Tyystjarvi and Kärki, 1969; Kellison, 1969). Grafting is commonly used to propagate selected phenotypes. Techniques are described in a later section.
Trimming and training the branches with the object of reducing height growth or of developing crowns like those of fruit orchard trees (from which seed may be easily collected) have usually not been advantageous. It is costly and may substantially reduce the yield of seed (Kellison, 1969; van der Sijde, 1969).
Harvesting the seed is generally a major cost in seed orchard management, particularly for species which produce small seed crops or which grow rapidly in height. For species, such as Pinus radiata, with serotinous, retained cones, the cost of seed collection in a region where all summers are cool and moist can be reduced by collecting the cones at intervals of several years instead of annually. However, in other regions (including probably all localities in which it is grown in Australia) the cones need to be collected annually and shortly after the maturation of the seed.
Seed harvesting in the southeastern United States has been speeded up through the use of new equipment. Mechanical tree shakers have been developed for species whose cones can be readily shaken from the branches, and vacuum sweepers have been used for collecting disseminated seed lying on the ground (Kellison, 1969).
The standard horticultural practices of fertilization, irrigation, cultivation, and cover cropping have been used in various seed orchards with the object of producing larger crops of seed. In some seed orchards, a mown turf is maintained to aid access and to help in the collection of seed by vacuum sweepers (Kellison, 1969).
An ideally located seed orchard is easy of access, within close proximity to an existing installation such as a forestry headquarters, convenient for easy fire protection, on a well-drained, reasonably fertile soil, and on gentle topography which can be readily covered by motorized equipment.
The area of orchard needed for a given yield of seed varies widely depending on the rate of seed production of the species or provenance concerned. In the absence of seed-yield data of existing orchards, estimates of the area of seed orchard required can be only very approximate.
Giertych (1965) summarized arguments for and against systematic seed orchard layouts and concluded that systematic layouts have the more advantages.
Squillace (1967) queried the narrow isolation zones sometimes used around seed orchards. He removed all the male strobili from a 2-hectare Pinus elliottii seed orchard with a 120-metre surround, and yet the female strobili produced normal amounts of germinable seed. Obviously, enough is not yet known about distances over which effective pollination can occur to specify minimum widths of isolation strip for most species.
Kellison (1969) advocates a minimum of 25 and preferably 50 to 100 clones for a clonal seed orchard. The planting of a large number of clones gives scope for a later reduction in the number due to graft incompatibility, thinning of inferior genotypes, and other factors.
Cheap methods for making mass-controlled pollinations would have great practical value in forestry. Tree breeders could make large-scale use of high specific combining ability, or they could produce seed of valuable species-hybrids or provenance-hybrids on a commercial scale.
Present methods for mass-controlled pollinations require the bagging of flowers to isolate them against stray pollen. This involves the use of a large amount of labour over a brief season. This makes the mass production of such seed difficult and usually impracticable. However, controlled pollination may not be uneconomic, since the high cost of pedigreed seed may be only a small item in the total cost of establishing a plantation.
Hyun's commercial production of Pinus rigida x taeda seed in the Republic of Korea is the classic forestry example of mass-controlled pollinations. For this work pollination bags have been used on a large scale. This is an interim method for the production of hybrid seed until hybrid seed orchards come into production.
Research is needed into methods of controlling pollination on a large scale. The aim should be to increase yield of seed per pollination and to reduce costs. Hyun (1969) reported that, due to genetic incompatibilities, controlled pollinations in intraspecies and interspecies crosses of Pinus rigida x taeda generally yield fewer seed per cone than either wind pollinations, or controlled crosses within populations. Controlled backcrosses to either parent yield much more seed than the cross pollinations between parents.
Snyder and Squillace (1966) report disappointingly low yields of seed from controlled crosses of southern pines compared with open pollinations. Also general experience in controlled pollination of Pinus radiata in the Australian Capital Territory shows that the yield of seed per bagged flower is low. On the other hand, Callaham (1967) obtained high seed yields from controlled pollinations. Anyone who has attempted it knows that controlled pollination depends upon genetic relationships between the entities being crossed, climate during the breeding season, experience of the breeder, and care with which the work is done.
Breeders of species, such as Araucaria cunninghamii, which produce female flowers for a considerable number of years before producing any pollen, have the opportunity of making large-scale controlled pollinations. Pollen can be applied by a blower or duster to the unbagged flowers where they receive little or no stray pollen.
The use of diluted pollen makes mass-controlled pollinations more efficient and cheaper (Callaham, 1967; Hyun, 1969).
The development of other methods for mass-controlled pollinations lies in research into the control of the production of male and female flowers. Possibilities include the development or selection of clones which are malesterile or which are wholly or mainly self-incompatible (Cram, 1969), and the development of selective chemicals to control the process of sexual reproduction.
Certain hardwood species, such as Populus spp. (Einspahr and Benson, 1964; Stettler and Howe, 1965), offer the possibility of mass-producing seed cheaply by the controlled pollination of flowers on detached shoots in the greenhouse. Using the same method, Chiba (1952) produced a small amount of viable seed of Cryptomeria japonica, and Linhart and Libby (l 967) a few seeds of Sequoia sempervirens. Sweet (1969) describes a similar method that would be suitable for most conifers; it involves grafting shoots bearing female flower buds onto branches where mass pollination could be done conveniently.
The high cost of controlling pollination by bagging, can be eliminated by producing hybrid seed in orchards. Hyun (1969) and Cram (1969) report experience in production of F1 and backcross hybrids in orchards. Pro auction of hybrid seed in orchards deserves more investigation, for it has the potential of' overcoming a major obstacle to the use of hybrids.
The most famous large-scale grafting of forest trees is probably the grafting of Pinus nigra near Paris during the first half of the nineteenth century. More than 100000 successful grafts were made (Bouvarel, 1960).
Grafted stock is generally agreed to be too expensive for routine forestry planting. But since the second world war the development of clonal seed orchards has necessitated large-scale grafting. For example, the Foundation for Forest Tree Breeding in Finland has a grafting programme which aims to produce approximately 200000 successful grafts annually for the establishment of seed orchards (Tyystjarvi and Kärki, 1969).
The grafting of forest trees follows simple horticultural practices, although the methods used vary in their details with the species, localities and technicians concerned.
Conifers are generally easy to graft, the top-cleft or side grafts being most commonly used. However, to ensure good success, it is essential that the stock be in a healthy, vigorously growing condition, that the scions be healthy and capable of vigorous growth, that the grafts be guarded against desiccation and not, exposed to frost damage, and that conditions conducive to fungal infection of the graft surfaces be avoided.
The grafting of broadleaved species such as Eucalyptus requires closer control of environmental conditions and of the actual grafting process than is necessary for conifers. Grafting of deciduous trees generally follows the well-known techniques used for multiplication of fruit trees.
More research is needed into graft incompatibility, which results in lack of grafting success or in the later death of the graft and which also (at least, in the case of Pinus radiata) appears to cause the unhealthy growth of some clones.
The use of cuttings for the establishment of forest, plantations has very attractive possibilities because of' the likelihood of high genetic gain in selection among clones. In addition, at least for species such as Pinus radiata, additional improvement may be possible through effects of cyclophysis. However, because cheap, effective methods for the mass production of rooted cuttings are not yet available for most forest trees, the large-scale use of cuttings in forestry is restricted at present to the willows,. certain poplar species, and to a lesser extent, Cryptomeria japonica, Chamaecyparis obtusa, Cunninghamia lanceolata and Pinus radiata.
Pinus radiata is the only pine for which methods for large-scale propagation by cuttings have been developed. Plantations of Pinus radiata cuttings are being established in Tunisia. In Australia experimental plantations of Pinus radiata cuttings have been established almost annually since 1938, and the practicability of raising the species on a large scale from cuttings is being studied in New Zealand (Thulin, 1969) and from rooted needle fascicles in Chile.
The rooting ability of poplars varies widely among the different species: it is generally good in the balsam poplars, variable in P. alba, P. nigra and P. deltoides, and poor in aspen and the sections Turanga and Leucoides (Sekawin, 1969).
A large number of experiments (mostly empirical) have been carried out on the rooting of cuttings of forest trees. Information resulting from this work is briefly as follows:
1. Very large differences occur among species (even among the species of a particular genus) in the rooting ability of their cuttings, and substantial provenance differences within species have also been reported.2. Rooting ability of the cuttings of most, if not all, trees and shrubs declines as the plant ages. High rooting potential of the shoots is generally regarded as a juvenile characteristic of a plant. There appears to be a correlation (at least, in some species) between the occurrence of juvenile characteristics such as juvenile leaf form and the rooting ability of the shoots.
3. Rooting ability and nursery growth of cuttings may be strongly influenced by the physiological condition of the parent plant (ortet), by the nutrition of the ortet (Mahlstede and Haber, 1957; Ooyama, 1958; Enright, 1959; Hartmann and Kester, 1959; Fielding, in press), and by the trimming of the crown of the ortet (Garner and Hatcher, 1955; FAO, 1956; Franclet, 1969).
4. Rooting ability and nursery growth of cuttings taken from young rooted cuttings may be considerably greater than those of cuttings taken from the older parent tree from which the rooted cuttings originated (Ooyama and Toyoshima, 1965; Fielding, in press).
5. The length of the juvenile period of high rooting ability in some species may be extended or possibly maintained indefinitely, and adult trees of certain species, the cuttings of which are difficult to root, can be rejuvenated, thereby developing a clone which can be rooted relatively easily from cuttings.
6. Rooting ability of the shoots on a plant tends to vary from season to season throughout the year.
7. Rooting ability and nursery growth of cuttings are strongly influenced by variations in the nursery environment (atmospheric and edaphic).
8. Rooting ability of cuttings of many species has been increased by use of growth-regulating substances. The action of growth inhibitors appears to influence the rooting of the cuttings of some species.
9. Rooting ability and nursery growth are influenced by such characteristics of the shoots as their position in the crown of the tree, their thickness, and length, and the presence or absence of male strobili.
An intensification of physiological and biochemical research into the rooting of cuttings is needed. Such research may lead to the development of methods for the economic, large-scale production of rooted cuttings of important forestry species for which such methods are now unavailable.
Foresters and tree breeders must control and record the genetic identity of all reproductive material (Barber, 1969). The tree breeder has responsibility to maintain exact identification of all material with which he works and to identify properly all materials exchanged or released for planting. Precise identification is mandatory to ensure that desirable combinations can be repeated and that pedigrees can be traced to locate parents with particular gene combinations.
The forester must control genetic identity to ensure that the best trees are used for each planting operation. His responsibility is great and the exercise of control of genetic identity is difficult in many situations. If he receives seedlings from a nursery, he must be able to place his confidence in the records and care of all operations from seed collection through nursery shipping. Seed collection is the most difficult step to control. The collector must be honest in collecting seed and in identifying each container of fruits, cones or seed. Subsequent operations must be planned to ensure the maintenance of identity. Carelessness and dishonesty are the most frequent sources of error.
Workable schemes have been developed and are functioning that will provide adequate control of genetic identity. The Scheme for Control of Forest Reproductive Material Moving in International Trade, adopted in 1967 by the Organisation for Economic Co-operation and Development (OECD), has been accepted by several countries and probably will set the pattern for future development. This scheme can well serve to control genetic identity of forestry reproductive materials throughout the world. Each nation should develop mechanisms to provide for control within its borders. Each forester and administrator should make it his moral responsibility to ensure that the best material is used, and that its identity is accurately documented through all phases from collection to plantation establishment.
BARBER, J. C. 1969. Control of genetic identity of forest reproductive materials. Second World Consult. Forest Tree Breeding. FAO-FO-FTB-69-11/3.
BERGMAN, F. 1965. [Moving spruce provenance in central Norrland.] Arsb. Fören. Skogsträdsförädl, 1964: 85-103.(In Swedish) (Abstract, For. Abs., 27, N° 3738, 1966)
BEVEGE, D. I. 1965. An investigation of cone and seed maturity of slash pine in southern Queensland. Austr. For., 29(3): 135-148.
BOUVAREL, P. 1960. Les vieux pins laricio greffés de la forêt de Fontainebleau. Silvae Genet., 9: 41-44.
CALLAHAM, R. Z. 1967. Hybridizing pines with diluted pollen. Silvae Genet., 16(4): 121-125.
CHIBA, S. 1952. Preliminary report on a method of controlled pollination of Cryptomeria japonica D. Don by rooted cuttings having female flowers. J. Jap. For. Soc., 34: 278-281.
COLE, D. E. 1959. The establishment and management of seed production areas. Proc. 5th South. Conf. Forest Tree Improvement, p. 63-67.
COLE, D. E. Management of pine seed production areas. 1963 Proc. 7th South. Conf. Forest Tree Improvement p. 44-49.
CRAM, W. H. 1969. Use of self-incompatibility factors for hybrid seed production. Second World Consult. Forest Tree Breeding. FAO-FORGEN-69-11/17.
EINSPAHR, D. W. & BENSON, M.K. 1964. Production and evaluation of aspen hybrids. J. For., 62: 806-809.
ELIASON, E. J. 1969. Development seed production areas an interim solution. Second World Consult. Forest Tree Breeding. FAO-FO-FTB-6911/8.
ENRIGHT, L. J. 1959. Effects of stock plant fertilization upon rooting of cuttings of Picea abies, Pinus resinosa and Pinus strobus. J. For., 57: 336-338.
FAO. 1956. Tree planting practices in temperate Asia: Japan. Rome. FAO Forestry Development Paper N° 10.
FIELDING, J. M. Factors affecting the rooting and growth of Pinus radiata cuttings in the open nursery. Austr. Forestry and Timber Bur. Bull. 45. (In press).
FOWLER, D. P. 1965. Effects of inbreeding in red pine. Pinus resinosa Ait. III. Factors affecting natural selfing. Silvae Genet., 14(2): 37-46.
FRANCLET, A. 1969. Bouturage des résineux et autres espèces difficiles. Second World Consult. Forest Tree Breeding. FAO-FO-FTB-69-11/4.
FRANKLIN, J. F. 1965. An exploratory study of cone maturity in noble fir. U.S. Forest Serv., Pacific Nthwest Forest Range Exp. Sta. Res. Note PNW 21.
GARNER, R. J. & HATCHER, E. S. J. 1955. The interplay of Factors influencing rooting behaviour of shoot cuttings. Rept 14th int. hort. Congr., Netherlands, 1955. Wageningen, Veenman and Zonen.
GIERTYCH, M. M. 1965. Systematic layouts for seed orchards. Silvae Genet., 14(3): 91-94.
HARTMANN, H. T. & KESTER, D. E. 1965. Plant propagation. New Jersey, Prentice Hall. 559 p.
HYUN, S. K. 1969. Mass-controlled pollination. Second World Consult. Forest Tree Breeding. FAO-FO-FTB-69-11/6.
ILLY, G. 1966. Recherches sur l'amélioration génétique du pin maritime. Ann. Sci. for Nancy, 1966.
KELLISON, R. C. 1969. Establishment and management of clonal seed orchards of pine. Second World Consult. Forest Tree Breeding. FAO-FO-FTB-69-11/7.
KRUGMAN, S. L. 1966. Artificial ripening of sugar pine seeds. U.S. Forest Serv., Pacific Sthwest. Forest Range Exp. Sta. Res. Pap. PSW-32.
LANGLET, O. 1964. Proveniensvalets betydelse för produktion och skogstradsforadling av gran. Svenska Skogsv-Foren. Tidskr., 62(2): 145-155.
LINHART, Y. B. & LIBBY, W. J. 1967. Successful controlled pollination on detached cuttings of coast redwood. Silvae Genet., 16(5-6): 168-172.
MAHLSTEDE, J. P. & HABER, E. S. 1957. Plant propagation. New York, Wiley. 413 p.
NAMKOONG, G., SNYDER, E. B. & STONECYPHER, R. W. 1966. Heritability and gain concepts for evaluating breeding systems such as seedling orchards. Silvae Genet., 15(3): 76-84.
OOYAMA, N. 1958. The effects of leaf sprays of urea on rooting of Cryptomeria japonica cuttings. Studies on rooting difficult races in the Kitayama district of Kyoto. Gov. For. Exp. Sta. Meguro, Tokyo, Bull., 105: 163-185. (In Japanese, with English summary)
OOYAMA, N. & TOYOSHIMA, A. 1965. Rooting ability of pine cuttings and its promotion. Gov. For. Exp. Sta. Meguro, Tokyo, Bull., 179: 99-125. (In Japanese, with English summary)
ORGANISATION FOR ECONOMIC CO-OPERATION AND DEVELOPMENT. Decision of the Council establishing an OECD Scheme for the control of forest reproductive material moving in international trade. 142nd Meeting, May. Paris.
PFISTER, R. D. 1966. Artificial ripening of grand fir (Abies grandis) cones. Nthwest. Sci., 40(3): 103-112.
PITCHER, J. A. 1965. Cone and seed yields in white spruce seed production areas. Second Genet. Workshop Soc. Amer. For. and Seventh Lake States For. Tree Imp. Conf., p. 76-77.
ROWE, J. S. 1964. Environmental preconditioning, with special reference to forestry. Ecology, 45(2): 399-403.
SEKAWIN, M. 1969. Poplar propagation. Second World Consult. Forest Tree Breeding. FAO-FO-FTB-69-11/5.
SILEN, R. R. 1967. Earlier forecasting of Douglas fir cone crop using male buds. J. For., 65(12): 888-892.
SLEE, M. V. & REILLY, J. J. 1967. The production of improved tree seed in Queensland. Proc. FAO World Symp. Man-made Forests, p. 1183- 1203.
SNYDER, E. B. & SQUILLACE, A. E. 1966. Cone and seed yields from controlled breeding of southern pines. U.S. Forest Serv., Southern Forest Exp. Sta. Res. Pap. SO/22.
SNYDER, A. E. 1967. Effectiveness of 400 foot isolation around a slash pine seed orchard. J. For., 65(11): 823-824.
STEFANSSON, E. & SINKO, M. 1967. Försök med tallprovenienser med särskild hänsyn till nörrländska höjdlägen. Stud. For. Suec. No 47.
STERN, K. 1959. Der Inzuchtgrad in Nachkommenschaften von Samenplantagen. Silvae Genet., 8(2): 37-42.
STETTLER, R. F. & HOWE, G. E. 1965. The production of homozygous tree material. Second Genet. Workshop Soc. Amer. Foresters and Seventh Lake States For. Tree Imp. Conf., p. 67-69.
SWEET, G. B. 1969. A method for the rapid production of controlled pollinated seed in Douglas fir. Second World Consult. Forest Tree Breeding. FAO-FO-FTB-69-11/11.
THULIN, I. J. 1969. Breeding of Pinus radiata through seed improvement and clonal afforestation. Second World Consult. Forest Tree Breeding. FAO-FO-FTB-69-9/5.
TODA, R. 1964. A brief review and conclusions of the discussion on seed orchards. Silvae Genet., 13(1/2): 1-4.
TYYSTJARVI, P. & KÄRKI, L. 1969. Mass production of grafted conifers. Second World Consult. Forest Tree Breeding. FAO-FO-FTB-69-11/2.
VAN DER SIJDE, H. A. 1969. Bending of trees as a standard practice in pine seed orchard management in South Africa. Second World Consult. Forest Tree Breeding. FAO-FO-FTB-69-11/9.
YANAGISAWA, T. 1965. The effect of cone maturity on the viability and longevity of coniferous seed. Gov. For. Exp. Sta. Meguro, Tokyo, Bull., 172:45-94.
