ÅKE GUSTAFSSON AND FRANÇOIS MERGEN
ÅKE GUSTAFSSON is Director, Institute of Forest Genetics, The Royal College of Forestry, Stockholm Sweden. FRANÇOIS MERGEN is Professor of Forest Genetics, Yale University, School of Forestry, New Haven, Connecticut, U.S.A. Other members of the drafting team were G. Siren (FAO), and KARL SAX (U.S.A.) Chairman of Section 1 of the Consultation who also prepared the summary.
Cytogenetics provides the basis for inherent variation, and it is essential to determine the chromosome constitution and nature of the genetic variation in tree species. What is the role of polyploidy and chromosome structural changes in varieties and species? What genetic differences result from changes in chromosome balance or structural changes and to what extent is variation due to "point" mutations?
The genera of gymnosperms of value in forestry have large chromosomes, polyploidy is rare and the karyotypes within genera are similar. In the Pinaceae the basic chromosome numbers are 10, 11, 12 and 13, which has led Khoshoo to conclude that the original basic number was 10, and that the higher basic numbers have been derived by chromosome fragmentation. This has been well demonstrated in Pseudolarix and Podocarpus. The basic number of 12 pairs of chromosomes is, however, characteristic of most genera of the Pinaceae and it is possible that genera with lower basic numbers could have originated by chromosome union following unequal translocation. Variation of basic chromosome numbers in certain genera of angiosperms has been shown to result from both chromosome fragmentation and by chromosome union.
Polyploidy in the Pinaceae is rare although polyploids occur in nature and can be induced. Mergen has found that both spontaneous and induced polyploids are retarded in growth and have a low survival value. The very extensive experiments of Hyun on induced polyploids in conifers provide more needed information.
The large size of chromosomes in the Pinaceae make them very sensitive to ionizing radiation. Gustafsson found many years ago that pine seeds are injured by relatively low doses of X-rays. These facts must be considered in mutation breeding and in evaluating the relative value of radiation versus chemical mutagens.
In contrast to the gymnosperms the genera of angiosperms of value in forestry vary greatly in basic chromosome number ranging from 6 to 41 pairs. The chromosomes are relatively small and polyploidy is common. In these genera induced polyploidy can be of value in forest trees as it has been in producing new types of ornamental trees and shrubs.
One of the most promising aspects of forest genetics is the production of F1 hybrids between species of diverse habitat. Species of conifers which have been geographically isolated for thousands of generations can be crossed and have produced vigorous and relatively fertile hybrids, as has been demonstrated in crosses between European and Japanese larch (Larix decidua and L. leptolepis) and between Asiatic and American species of white pine. Similar hybrid vigor has also been demonstrated in crosses between American and European species of Populus and of Platanus.
The use of ionizing radiation as a mutagenic agent has been of great value in breeding crop plants, even though desirable mutations are rare. But the long life cycle of forest trees and the space needed for testing the vigor and quality of the desired mutants makes mutation breeding a long and expensive project for forest tree improvement. Perhaps mass selection at the time of seed germination could be used in selecting for disease resistance, as has been done so successfully with cereal crops by treating the germinating seeds with the toxin of the pathogen. The recently discovered correlation between X-ray sensitivity and inherent vigor in the seeds of Zea may also be of value in selecting the superior genetic segregates in hybrid populations of forest trees.
All organisms, including those below the stage of chromosome development, characterized by two qualities, namely, replication or genetic constancy, and mutation or genetic change. Because of mutation, organisms are represented in nature by populations of genotypes. From the onset of life there is an interaction of heredity and environment, and the environmental conditions under which organisms live are subject to change. Because different populations contain individuals that are more or less well adapted to specific environments, the genetic composition of populations becomes altered as a result of natural selection. If the environmental stresses are more severe than the genetic possibilities of a population or species, the population or species will disappear and be replaced by others.
The genetic composition of a population of higher organisms changes because of mutation and recombination. In mutation, the genes or molecules of heredity are changed in various ways and often in their normal order on the chromosome. The chromosomes form linkage groups which limit the possibilities for recombination but, because of crossing-over, linkage groups are broken thereby increasing potential recombination during sexual reproduction.
During the differentiation of cells, the hereditary material becomes concentrated on the chromosomes in the nucleus, while the physiological and developmental activities become centered in the cytoplasm (Figure 1). This does not imply physiological inactivity on the part of the nucleus, nor lack of hereditary information in the cytoplasm; indeed, the genes of the chromosomes are the source of biochemical instructions to the cytoplasm, and in the cytoplasm there are "plasmagenes" or "plastogenes" that act as independent hereditary units.
FIGURE 1 - Meristematic cell: cell constituents photographed by means of electron microscopy
This chapter deals with forest trees and their cytogenetic behavior. It will be evident from the introductory discussion that the past history of a species or genus is of basic importance for its present genetic constitution, its population structure, its polymorphy and its distribution. There are also fundamental differences between species, genera and families in their cytological characters. The differences of chromosome behavior are related to
1. the basic chromosome number;
2. chromosome size and structure; and
3. degree of polyploidy.
These problems have been discussed in considerable detail by Gustafsson (1960a), with reference to the pioneer research work on tree cytology carried out by K. and H. J. Sax in the early 1930s.
Basic chromosome number
In gymnosperms the basic numbers (x) are 7 (Ephedra), 8 and 9 (several genera of Cycadaceae), 10 (Sciadopitys), 11, 12 and 13 (numerous genera); they may also reach higher values such as 19 and 20 (Podocarpus) and 22 (Pseudolarix). The last-mentioned numbers are no doubt secondary, that is, derived, and may in several cases have resulted from some sort of chromosome breakage or translocation.
The angiosperm tree genera belong to several different families of the natural system. The chromosome differences are consequently complex and varied. The basic numbers range from 6 and 7 (Cassia), through 8 (Carpinus), 9 (Hevea), 10 (Cornus), 11 (Corylus and Eucalyptus), 12 (Quercus), 13 (Acer and Ficus), 14 (Alnus and Betula) to 19 (Populus), 21 (Platanus), 23 (Fraxinus) and 41 (Tilia).
Gymnosperm and angiosperm genera also differ in average chromosome size. This does not mean that the small-chromosomed angiosperms have fewer genes than the large-chromosomed gymnosperms; it signifies only that the extragenic material of the gymnosperm chromosomes has undergone extraordinary development. Selective factors can apparently increase or decrease chromosome size, according to the type of environment or, to follow the Danish botanist Raunkiaer, to the life form of the species. See, for instance, Babcock (1947), Stebbins (1950) and Gustafsson (1951, 1960a).
Stebbins has pointed out that many woody angiosperms possess smaller chromosomes than do the related herbaceous species. This difference may depend on the fact that the angiosperm wood contains fiber cells from small-sized cambia initials or, according to Stebbins, on the true or false presumption that woody plants in general require a genetic system with a maximum amount of genie recombination, such a system being favored by many and small chromosomes. The principal gymnosperms, the Coniferales lack wood fibers and possess cambial initials about equal in size. They have, as discussed above, a fairly high but, when compared with the angiosperms, still distinctly low basic chromosome number with large-sized chromosomes. Presumably, Stebbins says, the reduction in chromosome size appeared early in the evolution of woody angiosperms.
There is an interesting connection between chromosome size and radiosensitivity demonstrated in recent years by the research of Sparrow and his coworkers. Sparrow and Miksche (1981) showed a good positive correlation between growth inhibition by ionizing radiation and the size of the interphase nuclei in shoot meristems. The effect of chromosome size on radiosensitivity in forest tree species was verified by Gustafsson and Simak (1968), and Wettstein et al. (1969) who compared radiation effects on species of Populus, Pinus, and Picea. Members of the Pinaceae have large nuclei when compared to angiosperms and the nuclear volume of the pines is approximately five times that of the oaks (Quercus spp.) although both genera have a somatic number of 2n = 24. This might explain the greater radiosensitivity in the Pinaceae.
Since Gustafsson and Simak (1968) reported the high sensitivity of pine and spruce to radiation, considerable information has accumulated. Pedigo (1980, 1982) and Platt (1983) have described the effects on Pinus taeda, Sparrow et al. (1963) on Pinus strobus, Mergen and Stairs (1982) on Pinus rigida, and Brandenburg et al. (1962) on Pinus monophylla.
Of considerable interest are the effects of low-level chronic radiation on pine trees. Pinus rigida trees have been killed after an exposure of five years to g -rays at the rate of not more than 8 r per day during eight months of each year. Quercus trees exposed to the same conditions survived this chronic gamma radiation, probably because of their smaller nuclear volume. However, the R1 progeny from both genera was much more variable than the control, reflecting genetic effects that had occurred in the chronically irradiated trees. This suggests that an appreciable number of genetic and physiological changes can occur in trees as a result of cumulative low-level chronic radiation.
Polyploidy
Only few genera of gymnosperms contain polyploid species. Examples are Ephedra, Gnetum, and Welwitschia. Juniperus chinensis tetraploid and Sequoia sempervirens is a naturally occurring hexaploid species. Although polyploidy does not play a significant part in the variation pattern of the Pinaceae, polyploid plants are occasionally found in nurseries and in nature. Aneuploid and mixoploid conditions were found in dwarf Picea abies seedlings (Kiellander, 1950; Illies, 1959) and a twin seedling of Abies firma was found to be tetraploid (Kanezawa, 1949). There are isolated reports of natural polyploidy in Larix. Christiansen (1950) located a mature tetraploid Larix decidua, and Chiba and Watanabe (1962) found among 2-year-old transplants of Larix leptolepis 8 polyploid seedlings - 2 had diploid roots, while the remaining 8 were all tetraploid. A single tree that arose from a cross between Larix decidua and L. occidentalis was triploid (Larsen and Westergaard, 1938). Spontaneous polyploid or mixoploid trees have been reported in four species of pine. Zinnai (1953) located 5 tetraploid Pinus densiflora seedlings; in Pinus elliottii mixoploid seedlings with 2n, 3n, and 4n chromosome complements were described by Mergen (1960); the presence of polyploidy in Pinus sylvestris in Sweden was reported by Johnsson (1959); and Nishimura (1980) described a tetraploid seedling of Pinus thunbergii that arose from a all-embryonic seed.
Gymnosperm species have been made polyploid by colchicine treatment (Merger, 1959). Polyploidy was successfully induced in all species attempted, and the overall changes in the seedlings or trees were similar. In general, the needles were shortened and thickened, the number of cells was reduced, branching was coarser, flowering was suppressed, and dwarfing of the trees was common. With most species tetraploids are not considered desirable, but they are of use as an intermediate step in the production of triploids.
Efforts to bypass the tetraploid sporophyte stage have been made by treating microsporangiate strobili of Larix leptolepis (Illies, 1956) and Pinus nigra and Pinus mugo (Mergen, 1959) during microsporogenesis. Diploid pollen grains were produced but no results from the progeny of this type of pollen are available. This method is promising, however, and will undoubtedly receive further attention. Further information about polyploidy in gymnosperms is given by Gustafsson (1960a), Mehra (1960) and Mergen (1963).
Natural polyploidy often arises after hybridization between different species or populations with the taxonomic status of species, and the subsequent doubling of the chromosome number, owing to the formation of unreduced gametes (a condition known as amphiploidy or allopolyploidy). Such a type of polyploidy is common among angiosperm crop plants (Nicotiana, Gossypium, Triticum, Brassica) and also among wild species (Galeopsis Rubus, Poa) and new combinations can be produced artificially, an example being Triticale which is a new "genus" combining Triticum and Secale. In numerous other instances, polyploidy is intraspecific in origin (the term used being autopolyploidy). Limits between allo- and autopolyploidy are fluid (Müntzing, 1938). The genus Dactylis is interesting in this respect since the naturally occurring tetraploids are often considered to be interspecific polyploids, although the corresponding diploids are no doubt closely related (Müntzing, 1956).
FIGURE 2 - Karyotype evolution: chromosome changes during evolution illustrated by the karyotypes (idiograms) of Pseudotsuga, Larix and Pseudolarix.
An interesting case was reported by Wright (1969a) in white ash, Fraxinus americana, which is divided into three "ecotypes," one northern (2n = 46), one intermediate (containing polyploidy) and one southern (with 2n = 46, 92, 138). The pumpkin ash, Fraxinus tomentosa is a rare hexaploid species (2n = 138), probably derived from a cross between a diploid green ash and a tetraploid white ash (Wright, 1959b). In Fraxinus, according to these analyses, both autopolyploidy and amphiploidy occur.
The most outstanding feature of intraspecific polyploid races consists of changes in developmental rhythm and ecological behavior, as first illustrated by Müntzing, (1938) and further elaborated by Stebbins (1950, 1956) and Müntzing, (1956, 1959). The general "gigas" appearance, the increase in vegetative growth, the alteration of incompatibility reactions and the changes in ecological requirements enable autopolyploids to extend the cultivation area of the diploids. Such polyploids are also useful in forest tree breeding, an idea put forward by Nilsson-Ehle in his early studies on the reaction and potentialities of the autotriploid giant aspen, Populus tremula. In gymnosperms, and in numerous hardwoods, allopolyploids may be even more suitable than autopolyploids for direct use in practice.
Disploidy and secondary polyploidy
The term disploidy (Chiarugi, 1932) signifies that different basic numbers, phylogenetically connected, occur in a genus. The term is not common in the literature but is quite useful. In tree genera disploidy is not infrequent. In the genera Cycas and Microcycas, for example, there occur the basic numbers 11, 12 and 13, which are probably derived from each other. In Podocarpus, with basic numbers of 11, 12, 19 and 20, some sort of chromosome breakage, leading to disploidy, probably occurs. According to Barlow (1959), the Australian genus Casuarina shows x = 8, 9, 10, 11, 12, 13, and so on. Even more interesting from an evolutionist's point of view is the genus Pseudolarix, related to Larix, where it seems possible to define the probable phylogenetic changes by means of chromosome analysis (Figure 2). Pseudolarix amabilis with x = 22 has two pairs of chromosomes with submedian or median centromeres and 20 pairs with almost terminal centromeres. In Larix all chromosomes have median, submedian or subterminal centromeres. It is feasible that Pseudolarix originated from Larix by means of chromosome breakage at the centromere regions of 10 chromosomes of the haploid set, whereas two chromosomes have remained unchanged (cf. Mergen, 1961).
According to Barner and Christiansen (1962), Pseudotsuga taxifolia has 13 pairs of chromosomes, the 2 shortest of which are telocentric, that is, they possess terminal or almost terminal centromeres. If these 2 chromosomes are regarded as being derived from a single long chromosome with a median or submedian centromere, the idiogram is quite similar to that of Larix. There are, in addition, other generic peculiarities which point to a fairly close phylogenetic relationship of these 2 genera.
When the idiograms of gymnosperm species and genera have been worked out in more detail, a series of conclusions relating to the connection of chromosome alterations and phylogeny will be possible.
Nucleus and cytoplasm
In the introduction it was pointed out that in the cytoplasm there are constituents which can also be considered self-replicating and, in a sense, independent of the nuclear genes. This fact lies behind the terms genome (the sum of nuclear genes, Winkler, 1920) and plasmon (the sum of the hereditary factors of the cytoplasm, Wettstein, 1926) as well as the less clear term plastom (the hereditary factors of the plastics, Renner, 1934, Michaelis, 1957/58, and Gustafsson and Wettstein, 1957/58). Of general interest is the plasmatic basis of male sterility in hermaphroditic species, such as maize, sugar cane, sugar beet, onion and Dactylis, which makes this plasmatic factor an important tool in breeding for hybrid vigor or heterosis. The situation is similar where there is differentiation into different sexes and cytoplasmic factors influence the formation of female or male sex organs (examples being Aquilegia, Godetia, Bryonia Satureja, Cirsium and Streptocarpus). In this connection, the classic studies of Correns (1908, 1916) are of great importance. The conifers are generally monoecious but in many species, such as Pinus sylvestris, individuals are found which are predominantly male or female. A systematic selection for dioecy may be quite useful in breeding for heterosis or hybrid vigor and the use of male-sterile individuals in crossing work. By contrast, the change from monoecy or dioecy (for instance in Salix or Populus) to hermaphroditism may be advantageous for the production of inbred lines. Such changes have been effected in hemp (Sisov, 1937; Sengbusch, 1952), with changes from dioecy to monoecy or hermaphroditism.
In the last few years there have been indications that plasmatic inheritance can be introduced into a species via induced mutations in nuclear genes (Figure 3). With regard to some chlorophyll lethals, for instance, nuclear mutations induce irreversible changes in the plastics or plastogenes. If the nuclear genes are then removed, the plastic breakdown persists and is transmitted plasmatically through the mother plant to the offspring (Wettstein, 1961, and unpubl., Gustafsson 1960b). It would be highly desirable if plasmatically and genically conditioned male sterility or dioecy could also be induced or isolated in tree species.
With regard to disease resistance, some data obtained by Langner (1952) indicate the occurrence of a cytoplasmic background of susceptibility to needle-cast in crossings between Larix decidua and L. leptolepis.
Species and population structure
The species is in one sense the fundamental unit of evolution. However, the definition of a species depends on the viewpoints of the observer and experimenter. In 1922 Turesson tried to make a sharp division within and between species or groups of species founded not on morphological but on biological characters. He distinguished between:
1. ecotypes, which are interfertile subunits;2. ecospecies, which can form hybrids although with a decreased hybrid fertility and a reduced viability of the hybrid offspring;
3. coenospecies, consisting of one or more ecospecies which may exchange genes among themselves but cannot recombine with other groups of species.
Turesson's concepts were later especially advocated by Clausen, Keck and Hiesey in their outstanding studies on the Californian flora (see Clausen, 1951). However, the concepts have not met with general acceptance. For the purposes of this report it is sufficient to consider that in general a species is heterogenous and polymorphic, that it is divided into populations adapted or in the process of adaptation to various climates, sites and niches. Whether or not the populations within a species are morphologically different from one another is then of secondary importance; likewise, whether the populations form more or less sharply delimited ecotypes or constitute clines with gradual changes in physiological and morphological characters (Huxley, 1938; Langlet, 1959a, b) is also of secondary importance for this discussion.
FIGURE 3. - Gene dependent plastom mutation: cytoplasmatic changes, in this case plastid aberrations, conditioned by gene mutations.
The degree of heterogeneity and the intensity of gene recombination depend largely on the type of fertilization, that is, self- or cross-fertilization, or, even, on the disappearance of fertilization in parthenogenetic and vegetatively propagating species. The occurrence of natural selfing is fully established in a number of tree species including conifers. Such is the case in the uniform Picea omorika, according to Langner (1959). However, even then crosses occur, often with pronounced segregation in later generations. In small populations or isolates, selfing may set in with consequent increase in homozygosity and general decrease of viability of the homozygous variants, due to inbreeding effects. In hermaphrodite or monoecious tree species the degree of natural selfing may vary from year to year, depending on the amount of flowering and spread of the pollen within the stand (for details of the distance of pollen flight, see Andersson, 1955). This fact must be seriously considered when natural regeneration is applied in silviculture. It has been pointed out that in species which are usually cross-fertilizing like Picea abies, Pinus sylvestris, and Pinus monticola, self-fertilization may occur quite readily and some fully self-compatible variants are encountered (Sylvén, 1910; Plym Forshell, 1953; Barnes et al., 1962; Eklundt Ehrenberg, 1963). Self-fertile biotypes arise spontaneously or in experiments under the action of mutagenic agents, by loss-mutation or destruction of the incompatibility genes and alleles.
However, in numerous plant species, including conifers and angiosperm tree species, there exist a series of transitions between full self-incompatibility to pronounced selfing ability. Nevertheless, the species are for ecological, historical or migrational reasons divided into populations which intercross in nature and are more or less adapted to their special habitats and as constituents of natural plant communities, with competition and co-operation of the individual biotypes both between and within species.
A most interesting analysis concerns the effects of self-fertilization contra cross-fertilization in 4 trees of Pinus monticola (Barnes et al., 1962), 2 of which were self-fertile, and 2 partially self-sterile. Pollen of the individual trees was mixed and the mixture used for fertilization. The proportion of outcrossed and selfed seed varied, depending on the genotypes of the mother tree and of the pollen parent. In some instances, self-pollen was as effective as foreign pollen; in most cases, however, it was less effective.
In apomictic species, strongly heterozygous hybrids may breed true and spread widely in nature, owing to their ability to undergo asexual seed formation or vegetative modes of propagation. Such species are rather rare among the economically valuable forest trees, although apomixis has been found to occur in Alnus, Euonymus, Sorbus and other genera, and many species, like the aspens, poplars and willows, do propagate vegetatively.
Tree populations and their adaptation
It is often considered that natural populations are in general well-adapted to environmental conditions. In a recent paper by Duffield (1962, page 9) for instance, the following sentence is found: "As a breeding procedure, therefore, induction of mutations is simply gambling against extremely high odds, for in organisms as well-adapted as most forest trees, virtually any change is likely to be for the worse" (the italics are the present writer's). On the contrary, a tree population often fails to acquire a complete adaptation to its habitat. This holds true for flowering ability as in Picea abies which flowers irregularly over large parts of its area of distribution, or for seed setting which either does not occur at all or is hampered in various ways, as in the case of Pinus sylvestris and Picea abies at high altitudes and latitudes (Simak and Gustafsson, 1954). However, the survival and growth of the vegetative phase may often be imperfect as can be seen in the inadequate frost and cold hardiness of local populations under severe conditions. For Scandinavian populations of Pinus sylvestris (Figure 4) this was fully realized by Wibeck (1933), and then worked out in more detail by Eiche (1962 and unpubl.); see also Gustafsson (1962). It must not be forgotten that a natural environment is always changing and that local tree populations have to undergo successive processes of adaptation which also lead to genetic changes of the potential variability. In addition, historical contingencies such as the original heterozygosity, the mode of migration from less to more severe conditions, irregular attacks or catastrophes due to insects, fungi, rodents, fires, and so on, may greatly influence the constitution of the local population.
Calculated per time unit and perhaps also per generation, forest trees show slow adaptation to changing environmental conditions. This is even more delayed by poor flowering conditions or failure to set seed. (In certain districts of Sweden, for example, a good seed set of Pinus sylvestris occurs once in 30 or 40 years.) Generally, in northern populations of pines and spruces, the juvenile stage, up to 15 or 20 years, is most sensitive to an extreme type of climate. Having passed this stage, the stand as well as the individual biotypes can be considered vegetatively adapted. Depending therefore on the climatic conditions following field germination or planting, the local populations are more or less adapted to the conditions of a given locality, involving long-lasting consequences for the generations to come.
FIGURE 4. - Adaptation of pine populations: indigenous populations of Pinus sylvestris in Sweden are often less adapted to the local climate than introduced (in this case more northern) populations.
|
Prov. |
Survival % |
Undamaged shoot % |
Height cm |
|
6 |
80 |
77 |
58 |
|
23 |
57 |
55 |
67 |
|
36 |
19 |
12 |
42 |
|
107 |
45 |
34 |
58 |
Tree populations are, in general, imperfectly adapted to many forms of human intervention, such as methods of thinning and they are - as wild species - not adequately selected to serve man's needs in timber quality, pulp production or chemical composition. This state of affairs makes forest tree breeding even more urgent, with the object of eventually bringing about an intensified domestication of tree species.
Genotypic composition of cross-fertilizing populations
It is of fundamental importance in plant breeding to analyze the genotypic composition of a population: how the genes are distributed on the chromosomes, their effects in the heterozygous and homozygous condition, their free or limited recombination. Studies on Drosophila by Mather (1943, 1960), Dobzhansky (1951) and their co-workers, have provided a fairly good picture of population behavior and constitution, under natural as well as experimental conditions, and under severe and mild selection pressures. The extreme heterozygosity of natural populations, involving breakage and rearrangements of chromosome segments, is an established fact; likewise, the abundance of mutations in the heterozygous state, which act as lethals and semilethals when homozygous and decrease viability, is well proved. Many such mutations increase viability above "normal" when heterozygous, an example being the chlorophyll lethals Gustafsson (1954). Eiche (1955) has shown that such chlorophyll lethals are common in natural populations of forest trees. In any case, there is a complex balance between all types of genes and mutations, heterozygous as well as homozygous. Owing to linkage phenomena the potential variability is restricted in the early stages of selection, but is released when crossing-over and recombination of genes in nearby chromosome segments have taken place.
Unfortunately, not much is known about the occurrence of spontaneous inversions or translocations in tree species. Sugihara (1940) has shown that translocations occur in local populations of Cephalotaxus drupacea, since quadrivalents and hexavalents are said to occur during meiosis of biotypes of this species. Small inversions may be present in every species without having been noticed in the few studies on meiosis so far made. A careful analysis of the idiograms of related species, for instance in Larix (Figure 6) or Pinus among conifers, would reveal whether species differentiation takes place not only by gene mutation but also by minute and gross chromosome rearrangements. It is evident from hybridization work that far-reaching crosses can be made in numerous species and genera of conifers. This fact can be utilized for the production of hybrid vigor (Righter, 1946, 1960; Hyun, 1960). In nature, species hybridization is often associated with a good deal of species introgression (Anderson, 1949; Stebbins, 1960); that is, genes are transferred from one species to another by hybridization and later back-crossing of the hybrid to one or the other of the parent species. Such introgression has been reported also for species of Alnus, Quercus and Pinus.
FIGURE 5. - Idiograms of larch species: related species often have different karyotypes (idiograms) as is shown by these idiograms of Larix decidua and L. sibirica.
The harmonious development of many tree populations is intimately associated with symbiotic phenomena, for instance with the occurrence of mycorrhiza. This leads to the mutual interaction and continuous adaptation of entirely different groups of organisms. Moreover, as shown by numerous workers, a stand of trees often or regularly forms an enormous "convivium" resulting from a far-reaching root symbiosis with transport of nutrients, hormones, exudates, water and other materials from one individual to another. The biological consequences of such mutual biotic influences have so far not been considered from a genetic or biometric point of view.
Quantitative inheritance and its genetic background
In the preceding paragraphs the complex constitution of cross-fertilizing plant species has been emphasized. The populations react to natural and artificial selection in the manner of polygenic systems consisting of numerous genes, most of which have slight individual effects, at least in the heterozygous state, and form the genetic basis of quantitative characters. The theory of quantitative inheritance has been developed by Mather in numerous papers (Mather, 1960). As early as 1915 Nilsson-Ehle pointed to the existence of such complex hereditary systems and wrote: "It is evident from the rather comprehensive analyses made in different countries that the characters important in practice, provided they are quantitative, must in general be conceived as being made up of several and sometimes numerous genetic factors, which obey the Mendelian laws... All hereditary properties, not only external morphological characters of little importance in the breeding program, but also physiological or biological characters, such as winter hardiness, genetic resistance to disease, earliness, lodging resistance, germination capacity, and so on, clearly segregate after crossings and form new combinations. In general, quantitative properties act in this way, examples being size properties and protein content in wheat. In no investigation has the author found characters of practical importance with another behavior... The segregation may be more or less complex - which according to the theory presented here depends on the number and action of the different genetic factors - but there is no doubt that different properties act basically alike. The composition of characters of practical importance as depending on a great number of genetic factors is of the greatest importance for the principles and methods of plant breeding" (op. cit., page 57, translated).
However, when Nilsson-Ehle, East, Fisher, Mather, and others stress the importance of genes with small effects, it must not be forgotten that cultivated plants also deviate from their wild ancestors in several drastically and abruptly changed characters (Schwanitz, 1957). Mutations with major effects have continually contributed to the process of domestication. Around these major changes recombination and further mutation is at work, fitting large and slight changes together into a balanced whole, where numerous genes influence the same quantitative character. In cross-fertilizing species, deleterious genes and mutations also exert slight modifying effects in the heterozygous state. The reason why single gene changes are so difficult to analyze in studies on the quantitative inheritance of forest trees is simple enough: the design of experiments and methods of measurements are often imprecise and inadequate, and environmental effects conceal slight genetic effects.
Disease resistance and major gene effects
The genetics of disease resistance in forest trees has become important in recent years. In a review published in 1962, Heimburger stated the problem in the following way: "The genetic background of resistance in the host can be polygenic or governed by a smaller number of major genes, although in most cases resistance to disease in plants has been found to be based on a combination of polygenes and major genes. Major genes governing resistance to disease constitute the basis for most of the spectacular advances in breeding for disease in agriculture and horticulture." On the other hand Heimburger subsequently stated: "It must be kept in mind, however, that most cultivated plants have been produced by selection and breeding mostly on the basis of polygenes and that many new and useful characters, such as disease resistance, can be materially enhanced in this manner if no other closely related material with superior resistance are available" (op. cit., page 358). The data on disease resistance in forest trees are vague and contradictory. This is mainly because the inheritance of resistance has largely been studied in species crossings, where the hybrid condition complicated the physiological action of the genes for resistance derived from quite different sources. That species hybrids between resistant biotypes will often turn out to be susceptible is perhaps to be expected. Heimburger himself cites hybrids between Pinus griffithii and Pinus strobus, where the "inhibiting mechanism broke down, the infection spread rapidly from the needles to the stem and heavy mortality of the seedlings was the result" (op. cit., page 360). On the other hand, cases are known where one species makes the species hybrid almost entirely resistant, an example being the cross between the canker-resistant Larix leptolepis and Larix decidua (Figure 6).
There are abundant data available from work on agricultural and horticultural plants to show that major genes are involved in resistance. Knight (1946) lists 33 crop plants in which major gene resistance to 84 pests and diseases has been demonstrated. Recent work by Briggs, Flor, Favret and others has shown that there are numerous genes producing resistance within a species, and that these genes are often not distributed at random over the genome of the host plant but tend to be grouped in genetical segments, concentrated on a few chromosomes. Briggs was the first to suggest such a phenomenon in bunt disease in wheat. The genetics of resistance to mildew in barley (Figure 7) is possibly the most striking example (Favret, 1960a, b). Eighteen different factors for resistance to mildew are distributed over 14 loci; 17 lie on chromosome 5 and one on chromosome 4. Thirteen loci form a large "isophenic segment" of 45 to 50 cross-over units. This isophenic segment can be divided into four sections according to the nature of the alleles for resistance. Some loci contain only one allele for resistance, others are built up on a series of 2 to 5 alleles. Three loci are closely linked forming a genie complex in a short segment with a length of about one cross-over unit. Most genes for resistance are dominant or semidominant, although recessive factors also occur.
FIGURE 6. - Disease resistance in species hybrids, in this case of conifers, may be "recessive" (upper figure) or "dominant" (lower figure).
It is certainly suggestive that this complex system of resistance has been found in several carefully studied species. This may argue for a similar condition also in conifers and broadleaved trees. Although many genes are involved each of them has in general a "major" effect. Therefore, inheritance is not "polygenic" in the sense of Heimburger, following Mather, in showing small additive effects, but is "multi-factorial" in the sense of Nilsson-Ehle, perhaps based on a series of small duplications, as suggested by Favret in his mutation work, many factors for resistance being dominant or semidominant. In addition, it must be emphasized that disease resistance and field tolerance of a disease may involve different phenomena.
The concept of heritability
In studies on quantitative characters in trees, such as volume production, height or diameter increment, stem form and wood properties, the experiments should be designed to differentiate between genetic and environmental influences and to estimate the extent to which a certain phenotype is determined by heredity and by the environment. In animal breeding the long generation time and the high cost of progeny testing make these problems as important as in tree breeding and the concept of heritability (h²) (cf. Lush, 1948) has been devised for that part of the total phenotypic variance (VP) due to genetic factors with additive effects (VA.). Originally the term heritability referred only to the correlation between parents and offspring but in tree breeding heritability is often used in two senses following the ideas presented by Lush (1948). In the broad sense, heritability refers to the functioning of the whole genotype and is used in contrast to environmental variance. The narrow definition of heritability includes only the average effects of the genes, carried from parents to offspring in meiosis (chromosome segregation) and subsequent fertilization (chromosome recombination). "This narrow meaning of heritability is used when the main emphasis is on expressing what fraction of the phenotypic difference between parents may reasonably be expected to be recovered in the offspring" (Lush op. cit.). See also Toda, 1957; Zobel, 1961; and Eklundh Ehrenberg, 1963.
FIGURE 7. - Resistance to a fungus disease, in this case barley mildew, is often conditioned by many genes, dominant or recessive, which lie in specific segments of chromosomes, called isophenic segments.
Heritability in the broad sense is a restatement of Johannsen's contrast of genotype and phenotype. The concept is beautifully illustrated by the experiments on flower color and temperature in Primula sinensis described by Erwin Baur (1919) in his classical textbook on genetics (Figure 8). In those forest trees in which vegetative propagation is a common procedure, clonal analysis easily reveals the genotypic component of variation. This is an important principle in the selection of plus trees because numerous factors, such as soil properties, water balance, stand density, irregular thinning, and so on, may result in a plus development of genetically inferior trees or a minus development of genetically superior trees. However, stem form, branching habit, and especially branch angle, are often highly fixed genetically. The seed properties of Pinus sylvestris demonstrate in an elegant manner the relative influence of heredity and environment (Simak and Gustafsson, 1954). Characters such as seed color and seed size depend for their expression on the ripeness of the seed and are easily influenced by environmental changes. By contrast the characters affecting seed shape, that is, the ratio of length to breadth, surface structure and wing shape, are less changed by environment. The morphological details of seed which are specific to the individual biotypes, for example, a curved or tapering tip or a strongly marked asymmetrical hilum, are highly fixed and independent of changes in climate or soil. These highly fixed characters permit precise control of pine material used in grafting and seed orchards.
FIGURE 8. - Genotype response in different environments. A. Shows the phenotypic response of a Primula genotype to slight changes in temperature. B. Illustrates how the seed shape of Pinus genotype is largely independent of climatic conditions, whereas seed color is greatly changed by environment. C. Shows that the branch angle of Pinus sylvestris has high "heritability", in both the broad and the narrow sense.
FIGURE 9. - Genetic principles of plant breeding.
Wood properties have in the last few years attracted great interest (Zobel, 1961; Ericson, 1960, 1961; Well wood and Smith, 1962). In some species, the correlation of specific gravity between selected plus trees and their clones in clonal tests is rather pronounced. The higher the basic density of a clone of Pinus sylvestris the higher is its pulp yield and the tearing strength of its pulp. Well-wood and Smith (op. cit.) have shown that in Pseudotsuga taxifolia and Tsuga heterophylla, there is no connection between the external features of a plus tree and its internal wood properties. In breeding for increased pulp yield in these species, the wood properties must be analyzed separately.
Heritability in the narrow sense is, with certain restrictions, a useful concept. If the experiments are correctly designed, mathematical analysis will indicate the size of the genetic component of variance at least in the offspring of crosses within populations. The important problem is to remove the nonspecific environmental influences. If the plus trees used in breeding are carefully selected, biotypes with minus hereditary characters will automatically be excluded from seed production. This is a negative but important procedure. However, the positive aspect of selection will be emphasized if the plus trees are carefully selected, tested in clone tests and crossed with suitable partners. It is encouraging to note that provenance tests, using mixtures of seed, have revealed a distinct connection between the phenotypic status of the parent stands and their offspring; Petrini (1959) described such a situation in a 50-year-old provenance test of Pinus sylvestris. In her studies on plus trees versus minus trees, Eklundh Ehrenberg (1963) has shown that the progeny of selected plus trees are superior in growth and development to the progeny of minus trees.
However, the object of tree breeding is to exploit not only general but also specific effects of combination. The formation of seed orchards containing 30, 40 or more clones derived from plus trees in a given region is an act of safety that can be recommended for forest areas with severe and varied site and climatic conditions. On the other hand, in more productive regions with favorable climatic conditions the number of clones might be reduced and, as experience accumulates, more emphasis can be put on specific combining ability. In seed orchards formed to exploit hybrid vigor by species and provenance crosses, the number of different clones may after careful progeny tests be reduced still further. This is especially the case when natural regeneration is not practiced and future generations of forests will be planted. Mathematical calculations of the relation of parent trees to offspring (that is, heritability in the narrow sense) will often fail if specific combining effects are not considered.
Genetic principles of plant breeding
It is necessary now to summarize the genetic factors involved in plant breeding, all of which are also acting in nature (Figure 9). These are:
1. selection;2. hybridization;
(a) F1 heterosis or hybrid vigor in crosses between species, populations and individuals,(b) F2 to Fn recombination and transgression,
© back-crossing of F1 to Fn individuals to one or other parent;
3. spontaneous or induced mutation due to molecular instability, ionizing radiations or chemical mutagens;
4. natural or artificial polyploidy, resulting in allo- and autopolyploidy.
Numerous devices are available for developing the long-term work of breeding. These include vegetative propagation, the use of male sterility, production of inbred or homozygous lines and their subsequent mass propagation by grafting. X-ray analysis depicts embryo and endosperm development. Mass infection to reveal genetic resistance to disease can be done in glasshouses. Climatic control of various kinds can be used to study genotype-phenotype relationships and the physiological and ecological conditions leading to early and profuse flowering. Correlations of characters in parent and offspring at an early stage simplify the correct selection of parent trees for crossing work. The necessity of mathematical analysis must be stressed. Numerous other scientific and technical methods, many not yet available, can be expected to accelerate research and speed the practical application of results.
A fascinating aspect of modern biology is the general applicability and validity of genetic principles in many individual species and genera. This makes possible the transfer of results obtained from bacteria, fungi, or flies to populations of trees or to man. The application of these results may be limited by the size of the organism or the length of its reproductive cycle. There is something fatalistic in the geneticist's view of life, that hereditary material consisting of fixed units with a high degree of constancy is transferred unchanged from generation to generation, although with the possibility of mutation. However, genetic material does not exist in a vacuum and through mutation and recombination, genotypes have been selected and molded for thousands or millions of years in varying environments. In other words, a genotype without an environment is a meaningless concept.
This interaction of genotype and environment is seen in the practice of silviculture. The economic value of a stand may be increased through thinning, pruning, and other forms of management, but no forester can produce a plus stand from one which is failing because of unsuitable provenance. It is impossible to apply breeding and genetics to forest trees without due consideration of many other disciplines of forestry. In this respect there may be a difference between forestry and agriculture.
In agricultural and horticultural crops, distinct changes have occurred during their development from the original wild state. These differences involve morphological changes, physiological changes leading to increased production, and variations in the underlying karyotype. Karyotype changes involve the number and structure of chromosomes and the arrangement of genes in the chromosomes. A typical example is provided by comparing the wild diploid species of Triticum and Aegilops with the hexaploid commercial species Triticum aestivum. Even within diploid crop plants the modernization has been remarkable. Contrast, for example, the hybrid maize of the corn belt of the United States of America with the primitive maize discovered in the caves of New Mexico and South America (Mangsladorf, 1958). In addition, some species such as the lupine have quite recently become used as crop plants. Alkaloid-free cultivars were developed with great skill by Sengbusch in the late 1920s, and the lupine have since been steadily improved by further mutation and recombination.
A similar process of domestication has begun in some forest trees, initiated by Nilsson Ehle's discovery of the fast-growing triploid aspens (Nilsson-Ehle, 1936). Using crosses between tetraploids and diploids, it is now possible to produce great quantities of triploids. Further improvement has been obtained by Johnsson (1953 and unpubl.) who has shown (cf. Gustafsson, 1960b) that hybrids between Populus tremula and Populus tremuloides are more productive in Sweden than the native Populus tremula (Figure 10). By introducing tetraploid Populus tremula into the crossing program Johnsson was able further to increase yield and, at the same time, transfer resistance, or at least tolerance, to Valsa nivea from the tetraploid to the hybrid. On the other hand, numerous cultivars of Populus are highly bred clones at the diploid level, also involving various species crosses. However, it appears that poplar breeding has often been casual, and in the future careful planning of the crosses will lead to even better results.
FIGURE 10. - Height growth of species hybrids and triploids in Populus - illustrating the combination of two breeding methods.
In conifers, as in broadleaved trees and agricultural crops, domestication will cause far-reaching changes in population structure. Hybrids are easy to produce in some tree genera, particularly Pinus and Larix. Many combine hybrid vigor with disease resistance or tolerance. Nothing is known about triploid hybrids in Pinus but possibly some will be of value in the future. At the intraspecific level, population crosses are likely to be of more immediate practical importance. For example, the central European, Polish or west Russian provenances of Picea abies transferred to Scandinavia are often more productive than the native Scandinavian populations. Seed and seedlings of these provenances have been imported in large quantities and selected plus trees of native and foreign origin are already established in seed orchards in Sweden (Andersson and Andersson, 1962). Wide out-crosses have been made in experimental clonal seed orchards and these will fundamentally change the population structure when used as F1 seed. The stands may be highly heterogenous but will be highly productive. Further plus tree selection may be made in these stands.
The rate of domestication in forest trees will naturally be slower than that of annual or biennial crops and will vary according to the species and characters under selection. The end product must vary with the needs of utilization and the ecological conditions of the growing site. Populations delivered by plant breeders into practice have to be completely adapted to the climates and sites for which they are planned. This implies the full conformity of the sum of the genotypes in the population and their environment, and this fundamental principle thus provides the beginning and end of this chapter.
ANDERSSON, E. 1949. Introgressive hybridization. New York. 109 p.
ANDERSSON, E. 1955. Pollenspridning och avståndsisolering av skogsfröplantager [Pollen distribution and isolation by distance of seed orchards]. Norrlands Skogsv. Förb. Tidskr., p. 35-100.
ANDERSSON, E. 1955. Pollenverbreitung und Abstandsisolierung von Forstsamenplantage. Z. Forstgenet., 4: 150-152.
ANDERSSON, E. & ANDERSSON, S. 1962. Förteckning över Domänverkets och Skogevårdsstyrelsernas skogsfröplantager [List of seed orchards owned by the Swedish Forest Service and the Local Forest Conservation Boards]. Stockholm, Samarbetsnämnden för skoglig växtförädling och genetik. Limited distribution. 158 p.
BABCOCK, E. B. 1947. The genus Crepis. Part One. The taxonomy, phylogeny, distribution and evolution of Crepis. Berkeley, University of California Press. 197 p.
BARLOW, B. A. 1959. Chromosome numbers in the Casuarinaceae. Aust. J. Bot., 7 (3): 230-237.
BARNER, H. & CHRISTIANSEN, E. 1962. The formation of pollen, the pollination mechanism, and the determination of the most favourable time for controlled pollination in Pseudotsuga menziesii. Silvae Genet., 11: 89-102.
BARNES, B. V., BINGHAM, R. T., & SQUILLACE, A. E. 1962. Selective fertilization in Pinus monticola Dougl. II. Results of additional tests. Silvae Genet., 11: 103-111.
BAUR, E. 1919.Einführung in die experimentelle Vererbungslehre. Berlin. 410 p.
BRANDENBURG, M. K., MILLS, E. L., RICKARD, W. K. & SHIELDS, L. M. 1962. Effects of acute gamma radiation on growth and morphology in Pinus monophylla Torr. and Frem. (Pinyon pine). Radiation Botany, 2; 251-264.
CHIARUGI, A. 1962. La cariologia nelle sue applicazioni a problemi di botanica. Atti della Società Italiana Progresso delle Science, XXI Riunione Roma, 1932, III: 1-38.
CHIBA, S. & WATANABE, M. 1952. Tetraploids of Larix kaempferi appeared in the nurseries. J. Jap. For. Soc., 34: 276-279.
CHRISTIANSEN, H. 1950. A tetraploid Larix decidua Miller. Det Kgl. Danske Videnskap. Selsk., Biol. Meddel., 18 (9). 9 p.
CLAUSEN, J. 1951. Stages in the evolution of plant species. New York. 206 p.
CORRENS, C. 1908. Die Rolle der männlichen Keimzellen bei der Geschlechtsbestimmung der gynodioecischen Pflanzen. Ber. disch. bot. Ges., 26: 686-701.
CORRENS, C. 1916. Untersuchungen über Geschlechtsbestimmung bei Distelarten. Sitz. Ber. Kön. Preuss. Akad. Wiss., 20: 448-477.
DOBZHANSKI, TH. 1951. Genetics and the origin of species. 3rd ed. New York, Columbia University Press. 364 p.
DUFFIELD, J. W. 1962. Forest tree improvement: old techniques and the new science of genetics. H. R. MacMillan Lectureship, University of British Columbia, Columbia, 12. 16 p.
EICHE, V. 1955. Spontaneous chlorophyll mutations in Scots pine (Pinus silvestris L.). Medd. Statens skogsforshn. Inst. Stockh., 45: 1-69.
EICHE, V. 1962. Nya aspekter på plantavgång och bristande vinterhärdighet i norrländska tallkulturer [New aspects on plant mortality and deficient winter hardiness in North-Swedish pine plantations]. Skogen: 423-426, 432.
EKLUNDH EHRENBERG, C. 1963. Genetic variation in progeny tests of Scots pine (Pinus silvestris L.). Studia Forestalia Suecica, 10: 1.-135.
ERICSON, B. 1960. Studies of the genetical wood density variation in Scots pine and Norway spruce. Statens skogsforskn. Inst. Stockh., Avdel. skogsproduktion, Rappt. 4, 1-52.
ERICSON, B. 1961. Skogsträdsförädling med sikte på ökat massautbyte. Tek. Vetenskaplig Forskning, 32: 194-203. (Statens skogsforskn. Inst. Ser. Uppsatser 81).
FAVRET, E. A. 1960a. Spontaneous and induced mutations of barley for the reaction to mildew. Hereditas, 46: 20-28.
FAVRET, E. A. 1960b. Induced mutations for resistance to diseases. Genet. Agr., 13: 1-26.
GUSTAFSSON, Å. 1951. Some aspects on variation and evolution in plants. Evolution, 5: 181-184.
GUSTAFSSON, Å. 1954. Mutations, viability and population structure. Acta Agr. Scand., 4: 601-632.
GUSTAFSSON, Å. 1960a. Polyploidy and mutagenesis in forest tree breeding. Proc. 5th World For. Congr., Seattle, U.S.A., p. 793-805.
GUSTAFSSON, Å. 1960b. Chemical mutagenesis in higher plants. Chemische Mutagenese Erwin-Baur-Gedächtnisvorlesungen I (1959). Abhandl. Deut. Akad. Wiss. Berlin, Kl. Med 1, 14-29.
GUSTAFSSON, Å. 1962. Genetick och växtförädling i skogsbrukets tjänst [Genetics and plant breeding as applied to Swedish forestry]. Svenska Skogsv. Fören Tidskr., 111-150.
GUSTAFSSON, Å. & SIMAK, M. 1958. Effect of X- and g -rays on conifer seed. Medd. Statens. skogsforskn. Inst. Stockh., 48 (5): 120.
GUSTAFSSON, Å. & WETTSTEIN, D. von. 1957-58. Mutationen und Mutationszüchtung. Handb. Pflanzenzücht., 1: 612-699.
HEIMBURGER, C. 1962. Breeding for disease resistance in forest trees. For. Chron., 38: 356-362.
HUXLEY, J. 1938. Clines: an auxiliary taxonomic principle. Nature, Lond., 142: 219-220.
HYUN, S. K. 1960. Mass production of control-pollinated seed of conifers. Proc. 5th World For. Congr., Seattle, U.S.A., p. 787-792.
ILLIES, Z. M. 1952. Colchicineversuche on Larix decidua Miller und Picea abies (L). Karst. Z. Forstgenet., 1: 36-39.
ILLIES, Z. M. 1956. Veränderungen der Pollengrösse bei Lärche nach Blütenbehandlung mit Colchicin. E. Forstgenet., 5: 112-115
ILLIES, Z. M. 1959. Polysomatie im Meristem von Einzelbaumabsaaten bei Picea abies. Silvae Genet., 7: 94-97.
JOHNSSON, H. 1953. Hybridaspens ungdomsutveckling och ett fõrsök till framtidsprognos [The early development of hybrid aspen and an attempt to predict its future production]. Svenska Skogsv. Fören. Tidskr., p. 73-96.
JOHNSSON, H. 1959. Föreningen för Växtförädling av Skogsträd 1936-1958 [The Association for Forest Tree Breeding 1936-1958]. Norrtelje Tidn. Boktr. AB. 68 p.
KANEZAWA, R. 1949. Polyploids appeared in twin seedlings of forest trees. J. Jap. For. Soc., 31: 22-24.
KIELLANDER, C. L. 1950. Polyploidy in Picea abies. Hereditas, 36: 613-516.
KNIGHT, R. L. 1946. The role of major genes in the evolution of economic characters. J. Genet., 48: 370-387.
LANGLET, O. 1959a. Norrlandstallens praktiska och systematiska avgränsning. Svenska Skogsv. Fören. Tidskr., 1959: 425-436.
LANGLET, O. 1959b. A cline or not a cline - a question of Scots pine. Silvae Genet., 8: 13-22.
LANGNER, W. 1952. Kreuzungaversuche mit Larix europaea D.C. und Larix leptolepis Gord. Z. Forstgenet., 1: 40-56.
LANGNER, W. 1959. Selbstfertilität und Insucht bei Picea omorika (Pancic) Purkyne. Silvae Genet., 8: 84-93.
LARSEN, C. S. & WESTERGAARD, M. 1938. Contributions to the cytogenetics of forest trees. I. A triploid hybrid between Larix decidua Miller and Larix occidentalis. Nutt. J. Genet., 36: 523-530.
LUSH, J. L. 1948. Heritability of quantitative characters in farm animals. Proc. 8th Intern. Congr. Genet., p. 356-375.
MANGELSDORF, P. C. 1958. Reconstructing the ancestor of corn. Proc. Am. Phil. Soc., 102: 454-463.
MATHER, K. 1943. Polygenic inheritance and natural selection. Biol. Rev., 18: 32-64.
MATHER, K. 1960. Evolution in polygenic system. Accad. Naz. dei Lincei. Evoluzione e Genetica 1960 (Colloquio Internazionale, Roma, 1959), p. 131-152.
MEHRA, P. N. 1960. Cytology and breeding of conifers. Proc. 5th World For. Congr., Seattle, U.S.A., II: 728-733.
MERGEN, F. 1958. Natural polyploidy in slash pine. For. Sci., 4: 283-295.
MERGEN, F. 1959. Colchicine-induced polyploidy in pines. J. For., 57: 180-190.
MERGEN, F. 1961. The chromosomes of Pseudolarix amabilis. Cytologia, 23: 213-216.
MERGEN, F. 1963. Evaluation of spontaneous, chemical, and radiation-induced mutations in the Pinaceae. FAO/ FORGEN 63-1/1.
MERGEN, F. & STAIRS, G. R. 1962. Low level chronic gamma irradiation of Pitch-Pine-Oak forest - its physiological and genetical effects on sexual reproduction. Radiation Botany, 2: 206-216.
MERICLE, L. W., MERICLE, R. P. & SPARROW, A. H. 1962. Cumulative radiation-damage in oak trees. Radiation Botany, 2: 265-272.
MICHAELIS, P. 1957-58. Plasma-Vererbung. Hand. Pflanzenzücht., I: 140-175.
MÜNTZING, A. 1936. The evolutionary significance of auto polyploidy. Hereditas, 21: 263-378.
MÜNTZING, A. 1956. Chromosomes in relation to species differentiation and plant breeding. Wageningen, Conference on Chromosomes, p. 1 -37.
MÜNTZING, A. 1959. Darwin's views on variation under domestication in the light of present-day knowledge. Proc. Am. Phil. Soc., 103 (2): 190-220.
NILSSON-EHLE, H. 1915. Den modärna ärftlighetsläran och dess betydelse för växtodlingen [Modern genetics and its importance for plant husbandry]. Stockholm, Nordiska Bokhandeln. 83 p.
NILSSON-EHLE, H. 1936. Über eine in der Natur gefundene Gigantsform van Populus tremula. Hereditas, 21: 379-382.
NISHIMURA, S. 1960. Chromosome numbers of poly-embryony seedlings of Pinus thunbergii Parl. J. Jap. For. Soc., 42: 263-264.
PEDIGO, R. A. 1960. The effects of ionizing radiation on the ecology of Pinus taeda. Bull. Ecol. Soc. Am., 41: 94.
PEDIGO, R. A. 1962. The effects of ionizing radiation on the ecology of Pinus taeda L. Unpublished Ph. D. dissertation, Emory Univ.
PETRINI, S. 1959. De två äldsta svenska tallproveniensförsöken [The two oldest pine provenance trials in Sweden]. Medd. Statens skogsforsk. Inst. Stock., 48 (11): 1-49.
PLATT, R. B. 1963. Ecological effects of ionizing radiation on organisms, communities, and ecosystems. Proc. 1st Nat. Symp. Radioecol. AIBS and Reinhold Publ. (In press)
PLYM FORSHELL, C. 1953. Kottens och fröets utbilding efter själv- och krosbefruktning hos tall (Pinus silvestris) [The development of cones and seeds in the case of self- and cross-pollination in Pinus silvestris L.]. Medd. Statens skogsforskn. Inst. Stockh., 43 (10): 1-42.
RENNER, O. 1934. Die pflanzlichen Plastiden als selbstständige Elements der genetischen Konstitution. Ber. Math. - Phys. Kl. Sächs. Akad. Wiss. (Leipzig), 86: 241-266.
RIGHTER, F. I. 1946. New perspectives in forest tree breeding. Science, 104: 1-3.
RIGHTER, F. I. 1960. Forest tree improvement through inbreeding and interspecific hybridization. Proc. 5th World For. Congr., Seattle, U.S.A., p. 783-787.
SCHWANITZ, F. 1957. Die Entstehung der Kulturpflanzen. Berlin. 151 p.
SENGBUSCH, R. VON. 1952. Ein weiterer Beitrag zur Vererbung des Geschlechts bei Hanf als Grundlage für die Züchtung eines monözischen Hanfes. Z. Pflanzenzücht, 31: 319-338.
SIMAK, M. & GUSTAFSSON, A. 1954. Fröbeskaffenheten hos moderträd och ympar av tall [Seed properties in mother trees and grafts of Scots pine]. Medd. Statens skogsforskn. Inst. Stockh., 44: 2, 1-73.
SISOV, L. A. 1937. Breeding of hemp. In Vavilov, N. I. Handbook of plant breeding. Leningrad, 1937. (Cited from R. von Sengbusch, 1952)
SPARROW, A. H. & MIKSCHE, J. P. 1961. Correlation of nuclear volume and DNA content with higher plant tolerance to chronic radiation. Science, 134: 282-283.
SPARROW, A. H. et al. 1963. The radiosensitivity of gymnosperms. I. The effect of dormancy on the response of Pinus strobus seedlings to acute gamma irradiation. Radiation Botany. (In press)
STEBBINS, G. L. 1950. Variation and evolution in plants. New York, Columbia University Press. 643 p.
STEBBINS, G. L. 1956. Artificial polyploidy as a tool in plant breeding. In Genetics in plant breeding, Brookhaven Symp. Biol. No. 9, p. 37-52.
SUGIHARA, Y. 1940. On the multipartite chromosome-ring in Cephalotaxus drupacea Sieb. et Zucc. Sci. Rep. Of Tohoku Imp. Univ. Sendai, 4: 13-18.
SILVÉN, N. 1910. Om pollineringsförsök med tall och gran [Experiments in self-pollination with pine and spruce]. Medd. Statens skagsförsöksanstalt, 7: 219-228.
TODA, R. 1957. Variation and heritability of some quantitative characters in Cryptomeria. Silvae Genet., 7: 87 -93.
TURRESON, G. 1922. The genotypical response of the plant species to the habitat. Hereditas, 3: 211-350.
WELLWOOD, R. W. & SMITH, J. G. H. 1962. Variation in some important qualities of wood from young Douglas fir and hemlock trees. Res. Publ. Faculty of Forestry, Univ. Brit. Columbia, No. 50, p. 1-15.
WETTSTEIN, D. VON. 1961. Nuclear and cytoplasmic factors in development of chloroplast structure and function. Can. J. Bot., 39: 1537-1545.
WETTSTEIN, D. VON, GUSTAFSSON, A. & EHRENBERG, L. 1959. Mutationsforschung und Züchtung. Arbeitsgemeinschaft für Forschung Landes Nordrhein-Westfalen, 73: 7-48, 59:60.
WETTSTEIN, F. VON. 1926. Über plasmatische Vererbung, sowie Plasma- und Genwirkung. Nachr. Ges. Wiss. Göttingen. Math.-Phys. Kl., p. 250-281.
WIBECK, E. 1933. Till skogens rasfrågå på lokala kallorter [Race problems of Scots pine in frigid localities]. Norrlands Skogvs. Förb. Tidskr., p. 227-250.
WINKLER, H. 1920. Vererbung und Ursache der Parthenogenese im Pflanzen- und Tierreich. Jena. 166 p.
WRIGHT, J. W. 1959a. Silvical characteristics of White Ash (Fraxinus americana). Northeast. For. Expt. Sta., Upper Darby, Pa. Sta. Paper No. 123, p. 1-19.
WRIGHT, J. W. 1959b. Silvical characteristics of Green Ash (Fraxinus pennsylvanica). Northeast For. Expt. Sta., Upper Darby, Pa. Sta. Paper No. 126, p. 1-18.
ZINNAI, I. 1953. The morphological characters and the fertility of the pollen of a tetraploid Japanese red pine induced by the colchicine method. J. Jap. For. Soc., 35: 245-248.
ZOBEL, B. 1961. Inheritance of wood properties in conifers. Silvae Genet., 10: 65-70.
