L.D. Pryor
L.D. PRYOR, Emeritus Professor is Visiting Fellow, Department of Forestry Australian National University, Canberra
The last decade has seen tremendous expansion in the use of Eucalyptus in plantation silviculture in the warm, temperate and monsoon tropical regions of the world. The capacity of species of the genus to produce wood quickly on short rotations, in a form suited for both industrial and rural use and in climates frequently rather severe in terms of moisture relations for tree growth, has contributed to its success in many countries in the warmer latitudes of the world. Much of this good performance is due to the absence of leaf-eating insects which are so much a feature of the natural eucalypt habitat. But it is also due to silvicultural methods which have become routine practice in plantation silviculture e with the genus, especially methods of soil cultivation and fertilizing. Large and still growing demands for pulp, paper and other wood fibre products have led to some very large plantation schemes. The capacity of the eucalypt to endure rather harsh conditions and yet make stems of reasonable form and dimensions, has made it of particular importance in the rural economies of many developing countries.
It has recently been reported that there is now more Eucalyptus wood utilized outside Australia than within and this situation will become accentuated as time goes by.
Eucalypts are confined naturally to the Australasian region where there are some 450 quite distinct species and numerous variants within many of them, so that it is a formidable task to screen the genus to select a species likely to be the most suitable for a particular locality. The sheer weight of species numbers alone has presented problems of organization which often strain local resources of countries other than Australia carrying out silvicultural programmes. The preservation of the natural gene resource in Australasia for future world silvicultural development is certainly one of the pressing needs in Eucalyptus improvement. The benefits will be especially felt in the future rather than the present.
Eucalyptus in its many species provides the arboreal vegetation (mainly forest and woodland) of more than half Australia. It occurs to a very limited extent in Papua New Guinea, Mindanao, Celebes and Timor and some of the Lesser Sunda Islands of Indonesia. Like many woody genera, Eucalyptus is predominantly a randomly outbreeding genus depending very largely on insect vectors for pollination and - with one or two exceptions - without any significant vegetative multiplication taking part in reproduction under natural conditions. The numerous species which occur throughout the natural range of the genus make it unusual, if not unique amongst woody plants, especially for the great sensitivity of numerous species to habitat and the fact that individual stands of a species are often of limited extent. Those species which are extensively distributed generally show considerable intraspecific variation.
A starting point in the improvement of Eucalyptus is with the systematics of the genus, an area of research which is complex for a plant group and the object of continuous study. Undoubtedly, most species of eucalypts are well known and a good deal of current study is simply a systematic reinterpretation of already known material. But increased activity in botanical exploration within Australia, especially of sites often of limited extent and opened up relatively recently to easy access, has disclosed additional species not previously known and it must be expected that still more will be recognized and described in the future, although as little as 20 years ago it seemed there would be few still to be recognized. It is quite unlikely that any species with an extensive distribution and in large stands remains to be discovered although some redefinition of species represented by extensive populations may still be expected. Systematic studies therefore remain a significant element in improvement work and will have substantial impact in future developments in the silvicultural improvement of the genus.
Virtually without exception the species regenerate by seed with supplementary survival through vigorous vegetative rejuvenation of damaged individuals whether this be by fire, drought, grazing or man.
It is indicated from so far relatively limited studies that species are generally preferentially outcrossed but some at least include a proportion of selfing although this may be associated with a loss of vigour in selfed offspring.
PLANTATION OF Eucalyptus deglupta IN PAPUA NEW GUINEA breeding for tropical conditions
Self incompatibility occurs in varying degrees; it is known in a number of species and clearly has a bearing on their reproductive habits. In the natural stands, providing there is no substantial habitat disturbance, the species for the most part are evidently quite stable and genetic breakdown between them does not readily occur. There are exceptions in some localities but, overwhelmingly, the majority of species retain their identity unblemished even though some limited interspecific hybridizing may, and often does, take place between pairs of species when they are growing side by side and are cross compatible. If there is no human disturbance the products of this hybridizing seldom have been able to spread from the site of initiation to others. The breeding system-implies a high level of genetic diversity in each species. This is, on the one hand, often a disability for immediate silvicultural use but, on the other, it ensures an array of genotypes which is one of the bases for tree improvement work. It implies that there is an opportunity for very great improvement by well known means. For many species, generation time is often six or seven years from seedling to the production of the first seed crop, although there are a good many exceptions in which flowering may take place as early as the second year and, occasionally, one-year-old plants are known to produce flower buds and set seed. In tree improvement work this is quite rapid and there are benefits to be had from this relatively short time scale. Seed for most species takes from six to twelve months to ripen which is a further cause for delay. This, together with long generation makes any breeding programme dependent upon a sequence of generations a rather slow and tedious process. As a consequence, the principal thrust for improvement work in the next few decades will be by means of those techniques which do not depend on a turnover of generations. On the other hand, the seed is quite small and a kilogramme from most species will produce a very large number of seedlings, five hundred thousand being common enough.
Because of the very substantial genetic variation in species, it must be assumed that the process of natural selection which has been going on for a very long time in the Australian habitat has led in many cases to very precise site adaptation so that separate populations within species of Eucalyptus generally differ one from the other throughout the geographic and climatic range of the species. This capacity for precise site adaptation is quite general in eucalypts and is a condition of basic value in any improvement programme. The amplitude of this variation is readily assessed only for the climatic and edaphic range of the natural habitat. Because of Australian topographic and edaphic characteristics and latitudinal location it does not offer always the opportunity for selection for particular climatic conditions by a homoclime comparison. This is especially so with regard to the ability to endure low temperatures and - in a lesser way - to adaptation to calcareous soils. Many species are able to thrive as a silvicultural crop in site conditions distinctly different from those of any part of their natural habitat. Evidence of such capacity depends on the known results of trials with any given species planted as exotic and therefore accurate records of performance of eucalypt species planted outside Australasia are vital to the assessment of the potential ecological amplitude of each; a good example is seen in E. camaldulensis which has been very successful in northern latitudes of more than 42°, whereas it is limited to about 38° in Australia and E. globulus thrives in low latitudes and high altitudes in India (Octacamunde) and also on the Alto Plano of Bolivia although it does not occur naturally above 500 m altitude or below 36° of latitude.
The designation of species depends upon systematic treatment. In plant groups this is not uniform, partly because different authors have had somewhat different standards, but also because, in some taxa, groups that may reasonably be recognized as species are not always the same in their characteristics from genus to genus. Broadly, eucalypt species occur in two ways in relation to their total populations. Firstly, there are those species of limited geographic extent and unusual habitats represented by small populations confined to particular sites either of climatic or edaphic conditions. The size of the natural populations is not necessarily an index to the utility of the species in silviculture. Some successful or potentially very successful ones have come naturally from extremely limited sites indeed, as for example E. dunnii. On the other hand, there are species with quite extensive geographic ranges covering distances of hundreds, even thousands of kilometres and in altitude hundreds of metres, within which there is considerable population diversity. This is usually reflected through different genetic make up in the different stands. There are of course intermediate circumstances. One of the general features in species of Eucalyptus which are widespread, apart from variation in accordance with different sites, is the graded variation which is associated with graded habit conditions. This is clinal variation, which is usually recognized on the basis of morphology but appears to be almost always accompanied by gradation in physiological features too. Clinal variation presents problems in terms of systematics but provides a genetic condition of high value in terms of tree improvement. Some of the most important species display variation of this type throughout their range and therefore a knowledge of such patterns is of weighty significance when searching for the best provenance for any particular silvicultural use.
The first step in any tree improvement programme is the designation of species and their introduction for trial. A principal shortcoming or source of error, often is the precise identity of the material involved, since it is easy and at several points possible to have misidentifications which can be quite misleading. However, once a species is known as likely to be successful - if it is one which is at all widespread in its natural stands - then a very considerable genetic diversity can be expected in the natural populations, and precise attention to provenance of seed together with trial plots to determine which provenance is most suited to the project becomes essential. Many of the physiological characters which depend upon provenance are those related to climate, a capacity to withstand lower than usual temperature being one of the so far most examined. There are also species, provenances of which may occur on special sites such as calcareous soils, where in most of the range the species is on non-calcareous substrata. This may provide an opportunity for the introduction of provenances with the physiological capacity to endure calcareous sites which the species for the most part cannot tolerate well, as for example, in E. camaldulensis. The same applies to ability to resist saline conditions and to other special physiological capacities.
The formation of viable hybrids between pairs of species is common enough in many plant genera. On the other hand, in some groups of plants this is absent. In eucalypts the occurrence of interspecific hybrids is common but species breakdown seems to have been avoided in the natural circumstances largely by an economical barrier. Interspecific hybrids occur at stand junctions but their progeny is limited and they do not invade the parent stands. If this isolating situation is circumvented or eliminated, either by human interference with the natural habitat as has occurred very extensively in Australia in the last two hundred years, or the elimination of it when the species are planted together as exotics, interspecific hybridizing may become extensive. There are recognized in Eucalyptus some seven or eight taxa which might be ranked as subgenera and between which it appears that hybridizing does not occur but within which it is usually found from rarely to frequently between some pairs of species. It is relatively easy to find individual trees in the field which are hybrid between many pairs of species, capable of interbreeding when they occur naturally side by side. It is less usual in the natural habitat to find extensive breakdown between species except perhaps in excessively disturbed areas where conditions have favoured the spread of consequent hybrid swarms. Because of the size of the genus and the features by which the species are recognized and classified it is not always easy to recognize naturally occurring interspecific hybrids. To be reasonably sure that any suspect individual is a hybrid, rather rigorous examination is necessary by proceeding beyond morphological and field examination to look at least at the progeny from an open pollinated seed set. In spite of the fact that interspecific hybridizing is generally not associated with a genetic breakdown between species there are a few circumstances where in the natural environment there has been interspecific mixing on a fairly extensive scale and areas where such genetic intergrades occur. This poses both particular problems and opportunities for tree improvement. Such breakdowns in the natural stands evidently depend on unusual natural conditioning circumstances.
The position with regard to interspecific hybrids is of special importance for an introduction programme. There is no doubt that in the past seed has left Australia already interspecifically hybridized because in seed collection this condition has been overlooked (for example E. "algeriensis" and E. rudis); At times also interspecific hybrids have occurred in exotic collections from which seed has been produced and has further complicated the matter of species choice. This is not only because in subsequent generations after initial trials there is genetic degrade of the material, but also because the F1 interspecific crosses in a number of cases give stands of superior value. The bringing together in arboreta of collections of species not ordinarily associated in the natural environment is a potent source of interspecific hybridizing and one which has had a major impact in the situation in some countries. Where this has occurred successive generations of Eucalyptus derived from an initial hybrid source show a pattern of segregation which leads to very considerable degrade in the quality of subsequently produced stands as a whole as for example in Brazil with E. urophylla.
This situation leads to a condition in which to arrest declining productivity there must be a return to pure species and a programme to ensure the elimination of such genetic contamination.
There is a long way to go and tremendous opportunities in this field. Some of the ultimate potential is dependent on technical developments which will have an impact on tree species improvement generally. These are techniques such as clonal culture either by meristem or single cell culture or by some method of raising trees not at present susceptible to the method, by stem cuttings. A breakthrough in any of these methods to permit clonal culture will open the way to considerable improvement, with at the same time the introduction of risks and pitfalls against which adequate safeguards should be built into any programmes so implemented. Until such radical developments occur some of the steps which can be taken have been explored in eucalypts so that programmes which apply to them can be taken up with confidence where there is a substantial silvicultural operation foreshadowed or in existence.
The first step in eucalypt improvement is to obtain in a particular area the provenance of a species most suited for the purpose. Once a species has been designated as that desired, then if there is variation from natural provenances in performance it should be examined fully. In most eucalypt species there is such variation and a programme to enable designation of the best source can lead to very substantial gains. In a great many cases - probably in most cases - seed in the quantities needed for a significant programme must be produced in cultivated stands since supply from the natural provenances is now generally limited. This, however, will become increasingly so as time goes on, not only because there are variations from year to year in the seed crop, but the natural stands are diminishing rather than increasing and are more and more being placed under long term protection so that procurement can be ensured in continuity for experimental plots but is unlikely to be maintained in lots suitable for large scale commercial activity.
Thus the development of seed producing plantations becomes particularly important. This can follow two general patterns: firstly, seedling seed orchards or seed producing plantations which do no more than maintain the genetic integrity of the species and the particular provenance involved by ensuring that it is suitably isolated from potentially contaminating genetic sources. The maximum distance from another stand with which a given stand can easily cross is dependent for the most part upon the distance that insect vectors travel to effect pollination. It appears that two kilometres is ample distance in most cases but the number of assessments of this is still small and attention should be given to accumulating information on this aspect. A very effective way of isolating seed producing areas is to plant the seed producing plot in a plantation of another species with which there is a complete barrier to interbreeding. Because more than one species often is used in silvicultural programmes this is quite often feasible. For example, isolation can be provided by E. citriodora if the seed producing plot is to be E. grandis or E. saligna, and for given species such arrangements can be planned. The seed producing plots themselves should be planted at reasonably close spacing to allow early and heavy thinning so that trees of the best form and vigour are retained for culling at rather wide spacings so that full crowns will develop and good seed crops result. Enough experience has been gained to show that systems of this kind work well although the precise details of spacing, time of thinning, methods of seed collection from often finally large trees have to be worked out in accordance with the facilities in the locality in which the programme is implemented.
A further step is in the development of clonal seed orchards where the theoretical background which has been used, particularly with regard to Pinus, appears to have full relevance in Eucalyptus although evaluation of the genetic gain in eucalyptus still awaits precise assessment for such orchards. As with pine, it has been found that grafted clones tend to display a rather high proportion of graft incompatibility which can be circumvented only by having many replacements so that there will be enough compatible grafts to make up ultimately the clonal population in the orchard. Unfortunately, as with pines, it is not always possible to foresee in young grafted stock even up to a year or two old which will be compatible and there are good reasons for avoiding this problem where possible. In some species of economic importance it is known this can be done effectively by growing clones from cuttings even though there are considerable constraints at present in the way cuttings may be taken. This varies from E. deglupta in which striking from cuttings is easy even from quite old material so that a clonal seed orchard can be built up this way or even by using air layers from quite large trees, which is also very successful. There are indications that some other species may be fairly easily handled this way too but the large majority of eucalypts will not strike from cuttings or form roots on air layers once the plant has proceeded significantly beyond the juvenile stage. With fast growing species such as E. grandis, selections can be made from trees already of substantial stature at four or five years of age by felling outer lines in plantations and selecting the best trees in these lines. Coppice shoots from the stump will strike quite readily from most trees giving at least 50% average success, provided they are taken under Australian conditions during three or four months of spring or early summer. Then they strike readily under standard mist spray. Research and development has proceeded sufficiently far to indicate that this is a highly successful method and the development of seed orchards on this basis has been accomplished in E. grandis and E. saligna. It may be found when fully examined to apply to many other eucalypt species. The possibility of rejuvenation of cutting material by subculturing from some initially established cuttings, even where the first established ones may have been difficult to strike, is a developmental aspect which merits thorough exploration. This is evidently a phenomenon exploited traditionally in horticultural plants and it appears the same occurs in at least one species of eucalypt, E. Camaldulensis.
It seems possible that even if the initial clonal development depends on grafting, rejuvenation of the scion may follow with progressive subculturing by cuttings.
There is strong evidence to show that interspecific F1 hybrids frequently will outperform pure species in terms of desirable produce. The production of interspecific F1 hybrids spontaneously and unintentionally, and especially the incorporation of their offspring in plantations, can be a source of degrade as outlined earlier. On the other hand, if this could be exploited in the way that the use of hybrids has been in some agricultural crops, there is no doubt that a very considerable improvement in production would follow.
One of the most effective systems can be found in an advanced experimental stage in Zambia. It involves production of hybrids between E. tereticornis and E. grandis simply by planting the species together in seed producing areas and roguing the progeny to select the F1 hybrids. This can be done visually with ease. While this is a programme that has a degree of chance and unpredictability as to the quantities produced, it does work but it hinges upon a peculiarity which is almost confined to E. tereticornis in that this species is in part at least wind pollinated and there is no need for particular manipulation to secure seed set which occurs in part at least when E. grandis is planted nearby. Various refinements to this system are possible such as the clonal production of E. grandis individuals known to be fully self incompatible. These are known to exist. In addition, manipulated pollination to increase seed set may be appropriate even in this wind pollinated species. With the large majority of eucalypts, however, it is necessary to carry out fully manipulated crossings and there are certain parts of this operation which have not yet been developed to a production scale. If self incompatible clones of the female parent are built up - and these could be done by grafting where graft incompatibility is of less significance than otherwise - emasculation will not be necessary and stored pollen can then be placed by hand on the stigma of the female parent to be followed by seed set. The evidence from spontaneously occurring hybrids such as the well known E.X. trabutii (an interspecific hybrid between E. botryoides and E. camaldulensis) which has performed so very well in the Mediterranean area, or the numerous cases where hybrids between E. tereticornis and E. saligna or E. tereticornis and E. grandis have been seen to perform well, together with more precise experimental trials of other interspecific hybrid combinations, indicate that this is a method of very considerable potential.
It has been shown in many combinations that F1 interspecific hybrids have a high level of uniformity - at times more so than that of the parent populations themselves.
The use of interspecific hybrids opens the way for the production of combinations which will sometimes bring together the most economically desirable characters from a pair of species. At present, the nature of the F1 in this way and therefore its economic value is not readily predicted and evaluation depends on actual trials.
