P. F. WAREING
P. F. WAREING is Professor of Botany, University College of Wales, Aberystwyth (United Kingdom). Other members of the drafting team were H. A. Fowells (U.S.A.), T. Ingestad (Sweden) and G. Sirén (FAO).
Some of the factors determining tree productivity are discussed in this chapter. The photosynthetic capacity of the tree depends upon the photosynthetic rate and the total leaf area engaged in photosynthesis. Although differences in photosynthetic rates between tree species have been reported, improvements in tree productivity by breeding for increased rates of leaf production would appear to hold greater immediate prospects of success than breeding for increased photosynthetic efficiency. Some of the factors determining the rate of leaf area increase are examined and it is concluded that in the early stages of growth species with large leaves are at an advantage but, as the complexity of the shoot-system increases, the high degree of branching of conifers probably compensates for the small area of the individual needles. The "leaf area duration," which is the integral of the leaf area over the growing period, is an important factor in determining the annual production of dry matter by the free. The evergreen hate* of most conifers gives an advantage over deciduous broadleaved species.
The seasonal pattern of extension and radial growth also profoundly affects the annual production of the tree; rapidly-growing species, such as poplar and Eucalyptus, generally have a long annual period of extension growth, or else they make several flushes of growth each year, an example being Pinus radiate. It is suggested that breeding for repeated flushes of growth in other Pinus species may be advantageous.
Some of the special problems associated with the growth of trees in close stands are examined. Whereas the growth rate of individual trees appears to be higher for certain broadleaved species than for conifers during the early years, when the canopy is closed the productivity of coniferous forest is frequently higher than for broadleaved forest due to the evergreen habit of conifers and to the fact that they intercept a higher proportion of incident light.
Some of the factors determining the ability to grow under limiting soil nutrient conditions are examined. It appears that species able to succeed on poor soils are not necessarily less demanding than other species, but are more efficient in taking up the available soil nutrients.
The present state of knowledge regarding the hormonal control of cambial activity is reviewed and the importance of both auxins and gibberellins is stressed.
Experiments on the transition from the juvenile to the adult flowering condition in seedling trees suggest that the attainment of the adult condition occurs when the tree attains a certain size, and thus the number of cycles of growth and dormancy through which the tree passes is not important. The practical implications of this conclusion for tree breeding are indicated.
The need for more study of the physiology of root growth is stressed, especially in relation to the reestablishment of transplanted seedlings. Nursery transplanting practice should be related to endogenous periodicity in root growth, which appears to depend on the state of dormancy of the shoot.
Experiments on the rooting of dormant "winter" cuttings and of leafy "summer" cuttings are described. Rooting of winter cuttings is greatly stimulated by the presence of nondormant buds, but root initiation in summer cuttings is promoted by the presence of expanded leaves.
Interest in tree physiology has greatly increased in recent years and there have been several general articles showing the bearing of tree physiology on forestry (Kramer, 1956; Richardson, 1960; Kozlowski, 1961).
The aim of this chapter is not so much to present a compilation of the present state of knowledge of tree physiology in relation to breeding but rather to analyze, from a physiological standpoint, some of the fundamental problems facing the tree breeder, especially in his attempts to improve tree productivity, and to indicate the direction which future research should take. In the past plant breeding, whether of agricultural crops or of forest trees, has been largely empirical; for example, the crop breeder has often not known the nature of the physiological factors determining yield and has largely had to work in the dark because the plant physiologist has so far paid relatively little attention to this problem. There is now, however, a fairly widespread appreciation that breeding must be based on a clear understanding of the physiological and biochemical processes underlying and controlling the desired characters. It is clear, therefore, that if progress is to be made there must be far more close co-operation between tree breeders and physiologists in the future.
Analysis of tree growth
The production of wood involves primarily the synthesis of cell wall material, which is derived from the carbon dioxide of the air fixed in photosynthesis. Now the total photosynthetic capacity of a tree depends on:
1. the photosynthetic rate, that is, the rate at which carbon dioxide is assimilated into organic material;2. the total leaf area engaged in photosynthesis.
In order to examine the factors affecting the productive capacity of a tree it is necessary, therefore, to measure these two parameters separately. This problem has received considerable attention from agricultural crop physiologists who have developed the techniques of "growth analysis." Such an approach has been very little used in relation to forest trees although it has been applied to Pinus sylvestris (Rutter, 1957), to the oilpalm (Rees, 1962, 1963), and to the tropical woody species Trema guineensis (Coomb, 1960) and Musanga crecopioide (Coomb and Hadfield, 1962). The basic parameters used in growth analysis are the net assimilation rate E, and the leaf area index L. The net assimilation rate can be defined as the mean rate of dry matter production per unit leaf area. It should be noted that the productivity of the tree depends upon the net increase in dry weight resulting from the balance between photosynthesis and respiration. The net assimilation rate E can be estimated from successive determinations of dry weight and leaf area at short intervals, usually of one or two weeks, during the growing season. The leaf area index L is defined as the total area of leaves present per unit area of land. The value of L may range from much less than 1 for young plantations to as much as 8 or 9 for dense plantations of conifers (Büsgen and Münch, 1929). The product (E x L) gives the crop growth rate, which is the rate of dry matter production per unit area of land.
Differences between species in photosynthetic efficiency have been the focus of considerable interest in recent years. At one time it was thought that the photosynthetic efficiency was very similar in a wide variety of agricultural crops, but it now appears that there are significant differences between species in this respect. Similar differences have been reported for several tree species; thus, photosynthetic rates were reported to be higher in Douglas fir (Pseudotsuga taxifolia) than in Pinus strobus or Picea abies (Polster, 1955), and higher in Quercus than in Pinus seedlings at low light intensity (Kozlowski, 1949). Differences in photosynthetic rates have also been reported among the clones of species and hybrids of Populus (Huber and Polster, 1955; garner, 1955). The existence of these genetically determined differences in photosynthetic rates clearly offer opportunities to the tree breeder to improve the productivity of the tree, and several authors have stressed the importance of using photosynthetic rates as a basis for selection. However, the methods at present available for measuring photosynthetic rates are somewhat laborious and time consuming and appear to be most suitable for screening clonally propagated material. Moreover, the maximum rates of photosynthesis measured over short periods are not maintained over longer periods, and there are considerable differences between species in this respect (see Kozlowski, 1963). By contrast, Hellmers and Bonner (1959) have argued that the maximum photosynthetic efficiency is approximately the same not only for agricultural crops (including both temperate and tropical species) but also for forest trees, and lies between 2.0 and 2.5 percent under field conditions. This conclusion is not necessarily incompatible with the observed differences in photosynthetic rates referred to above, but clearly it is important to determine how much scope there is, within this possible limiting figure, for further improvement in the photosynthetic efficiency of various species of forest trees.
In considering the prospects of improving yield in agricultural crops, Watson (1956, 1958) has pointed out that differences in productivity are due more to differences in leaf area index than to differences in net assimilation rate, and he has concluded that improvements in productivity by breeding are likely to result mainly from increased leaf growth. It is necessary, therefore, to discuss this subject from the standpoint of tree breeding.
The importance of leaf area for diameter growth in stands of forest trees has been studied by a number of authors, and the subject has recently been reviewed by Matthews (1963). Thus several authors have found a correlation between total needle weight and stem diameter in various conifers (Schmidt, 1953; Ovington, 1957; Schöpfer, 1961); for example, the highly productive Abies alba is reported to carry almost twice as large a total needle surface as Picea abies of the same size of stem (Büsgen and Münch, 1929). Nevertheless, although the importance of leaf area has been appreciated by the silviculturist and tree breeder, it is desirable to examine in greater detail the possibilities for further improvement of productivity through affecting the leaf area index.
The growth rate of the tree depends the rate at which the leaf area increases, since the latter determines the overall photosynthetic capacity. Hence a species which rapidly develops a large leaf area has a high growth rate. Now the rate of increase in total leaf area of a plant depends on:
1. the rate of production of leaf primordia by the apical meristems;
2. the area attained by the individual leaves;
3. the number of apical meristems, as determined by the branching habit of the species.
In general there is, of course, an inverse relation between leaf number and leaf size so that species producing large numbers of leaves, such as conifers and willows (Salix spp.) have small leaves, while those producing large leaves, such as Acer, Platanus and Populus have correspondingly fewer. In the early stages of growth, when the number of growing apices is small, it appears that the production of large leaves is an advantage in the rapid development of leaf area. Thus, a comparison of the growth of first-year seedlings of a species such as Acer pseudoplatanus with that of first-year Pinus seedlings indicates how much more rapidly the leaf area of the former species increases. Although the pine seedling produces new leaves (needles) at a much higher rate than the broadleaved species, the greater area of the latter more than compensates in leaf area production. On the other hand, as the trees increase in size and complexity of branching, the number of shoot apices producing leaves increases and hence the total rate of increase in leaf area by a conifer is probably no longer significantly smaller than that of a broadleaved tree, so that after a few years the differences in growth rate between the two types of tree are reduced (see below). In general, it would seem desirable to select and breed forms with large leaves or needles, but it is evident that there is an urgent need for more precise information on the factors determining the rates of leaf area increase in trees.
Another important factor affecting the annual production of dry matter by the tree is the leaf area duration which is the integral of the leaf area index over the growth period and takes account of the magnitude and persistence of the leaf area index L (Watson, 1958). It is clear that evergreen conifers are able to carry on photosynthesis whenever light, temperature and water conditions are favorable. Although there is some difference of opinion as to whether photosynthesis by conifers in the winter is of any importance, there seems no doubt that in regions experiencing mild winters there is a significant increase in dry matter at this period (Hagem, 1962). Probably more important, however, is the ability of evergreen conifers to conduct photosynthesis in the early spring and late autumn, when conditions of light and temperature are favorable. One of the major factors limiting the productivity of arable crops in North Temperate regions is the fact that they present low values of the leaf area index L even in May, when light intensities may be very favorable for active photosynthesis; greatly increased yields could be obtained if the leaf area index of a crop were to move quickly to the optimum and remain there for a large part of the growing season (Nichiporovich, 1960; Watson, 1956, 1958). Although deciduous forest trees in temperate regions do not normally come into leaf until, say, late April or May, nevertheless a deciduous forest attains a high leaf area index very rapidly, once unfolding of the buds occurs. This is because the presence of preformed leaf initials within each bud allows very rapid increase in leaf area, and also because the presence of many layers of leaves on the branches at various heights allows a high leaf area index to be rapidly attained. Thus, beech trees may rapidly reach a leaf area index of 5 or 6 when they come into leaf in the spring. In this way, deciduous trees are able to take advantage of the favorable light conditions in the spring much more rapidly than arable crops and the advantages of the evergreen habit are reduced.
Another important factor affecting the leaf area and growth rate of trees which has not received sufficient attention in discussions on productivity is the seasonal pattern of extension and radial growth. As is well known, some tree species continue active extension growth and leaf production throughout the growing season, whereas in other species, extension growth is limited to the expansion of leaf primordia laid down in the resting bud and when these have been expanded no further new leaves are developed. Intermediate between these two extreme types are many species in which there is continued apical activity and expansion of new leaves after expansion of the bud, but in which extension growth ceases in midsummer, that is, at the end of June or in July. Now it is clear that the pattern of shoot extension of the tree will markedly affect its rate of leaf production and growth rate. In species in which extension-growth is maintained throughout the year, the leaf area, and hence the photosynthetic capacity of the tree, is continually increasing throughout the season. By contrast, in trees in which extension growth is restricted to one brief flush of growth in spring, the leaf area remains constant for much of the season and the new material formed in photosynthesis is not used for the development of new leaves but in the development of buds, radial growth and the development of reserves. Assimilates used in radial growth are not wasted, of course, but it is clear that for maximum efficiency in wood production a balance must be struck between radial growth, which does not increase the photosynthetic capacity of the tree, and assimilates "plowed back" into leaf material which does further increase its productive capacity.
Trees of the genus Pinus present an interesting case in that extension growth proceeds by very rapid flushes so that a species such as Pinus sylvestris may have completed its growth by early June; nevertheless, the leaves continue to grow for several more months by means of a basal intercalary meristem so that the tree steadily increases its leaf area throughout the growing season. This and other species of Pinus, and certain broadleaved trees such as Quercus robur and Fagus sylvatica, may produce a second or third flush of growth (the so-called "lemmas shoots ") in midsummer when the daylength and other conditions are favorable.
Accurate data on the relation between yield and the duration of extension growth appear to be lacking, but probably it is not mere coincidence that high yielding trees of Populus and Eucalyptus have long growth periods. It is also of interest to note that the very rapidly growing Pinus radiata produces several flushes of growth each season and in the North Island of New Zealand it may never completely cease growth even in the winter. In the opinion of the author, this question of the duration of extension growth is of paramount importance to the tree breeder and merits very careful study. Moreover, considerable ecotypic variation in growth period exists, so that this character is very susceptible of modification by breeding and holds much more immediate prospect for improvement than increased photosynthetic efficiency. For example, Pinus radiata is not adapted to cooler temperate climates but it may well be possible to select other species of Pinus for repeated flushes throughout the growing season and thereby greatly improve the annual growth rate. It is clear however that, if it is planned to breed for extended growth periods, this will probably be most feasible for areas having a relatively mild winter. It is known that forest trees are rather delicately adjusted to the length of the frost free period and that both latitudinal and attitudinal races may be distinguished. Daylength and temperature responses play an important role in this adaptation, as is demonstrated in the paper by Holzer (1963).
Growth of trees in plantations
The productivity of forests has been the focus of much interest in recent years (Ovington, 1958, 1961; Hellmers and Bonner, 1959). This work has been concerned primarily with the productivity of the forest as an ecosystem. The following discussion is more concerned with the analysis of some of the physiological factors involved in the productivity of trees grown in closed stands. In the early stages of growth in plantations there is little or no competition between the trees for light, soil nutrients or water. Under these conditions the growth curves are generally exponential in shape, indicating that the annual gain in dry weight is increasing steadily as the leaf area, and hence the capacity for photosynthesis, increases. Thus, the crop is growing at an ever-increasing rate during this "exponential" phase. In due course, however, the canopy closes and competition for light becomes severe so that, as new foliage is developed in the upper parts of the tree, the heavily shaded lower branches die off. In the early stages of growth of the plantation the leaf area index L will be less than 1, but as growth proceeds it will steadily increase until, when the canopy is closed, it reaches a steady value when the rate of production of new leaves is balanced by the rate of death of foliage in the basal region of the tree.
During the early stages, before the canopy is closed, species having a high rate of leaf production will make the most rapid growth, and during this phase the growth of certain broadleaved species, such as Betula verrucosa, Alnus glutinosa, Acer pseudoplatanus and poplars, is more rapid than that of conifers, such as Pinus Picea and Abies (Büsgen and Münch, 1929). When the canopy is closed, however, the requirements for rapid growth of the crop as a whole change, and the capacity for rapid leaf production would seem to be of less advantage if increased leaf area is simply matched by a corresponding loss of foliage in the basal region; moreover, under certain conditions increased area of canopy may result in increased transpiration losses, which could more than offset the gain from increased photosynthetic area. On the other hand, differences in photosynthetic efficiency probably now become all important. The most efficient crop will be one which completely intercepts the whole of the incident light, and utilizes it efficiently in photosynthesis. The lowermost leaves will exist at very reduced light intensity and they can survive only if the production of assimilates by photosynthesis can balance their loss by respiration (the light intensity at which this occurs is, of course, the "compensation point "). Shade tolerance, involving a low compensation point and the ability to retain leaves under low light intensities, is, therefore, an advantage under a dense canopy. As Matthews (1963) has pointed out, it may be of significance that the more productive species of Abies, Picea and Pseudotsuga taxifolia are tolerant of shade and carry their needles for at least 3.5 to 4 years and often for 6 or 7 years, whereas the less productive Pinus and Larix species are intolerant of shade and carry their needles for 1 to 3 years only. For any given spacing, the former species will carry a denser crown than the latter species and presumably thereby intercept and utilize the incident light more efficiently. It is generally assumed that leaf senescence and abscission do not occur until the light intensity falls to the "compensation point" but it is known that hormonal relationships also play an important part in leaf abscission and it is possible that in some species leaf fall occurs at light intensities above the "compensation point" because of a disturbance of the hormone levels in the leaves. If this is so, then it may be possible to select for greater leaf retention at low light intensities and thereby to increase the area of the crown for a given species. More information is obviously needed about the factors determining leaf abscission, especially in evergreen conifers. The techniques of growth analysis have been applied by Donald (1961) to problems of competition for light in arable crops and pastures, and it would seem likely that the application of this approach to forest crops would be valuable.
As has already been seen, in the early c sages of growth certain broadleaved species show a higher growth rate than conifers, but this position is reversed later when the canopy is closed, and there seems good evidence that the productivity of coniferous forest is considerably higher than broadleaved forest (Ovington, 1958, 1960). The greater productivity of conifers in close stands, despite the greater potential growth rate of certain hardwoods as individual trees, is probably partly due to the fact that the conifer canopy intercepts a higher proportion of the incident light than does the canopy of many hardwoods (Ovington, 1960). Ovington (op. cit.) has also pointed out that the conical shape of the individual conifer crown insures that a large area of leaf is effectively illuminated. Thus the greater productivity of coniferous forest has been attributed to evergreen habit, dense canopy, and conical crown shape. There is, however, little direct experimental evidence on this subject and quantitative studies of these various factors in relation to productivity are badly needed. It is possible that the productivity of broadleaved species in close stands could be improved by breeding, if the factors limiting production at present were more clearly understood.
A very important factor in estimating the final yield of a tree crop is, of course, the ultimate height reached by the trees. Since there is a correlation between height growth and yield, a species such as Betula verrucosa or Alnus glutinosa, which has a relatively small maximum height, will clearly give a lower ultimate yield than species such as Fagus sylvatica, which attain much greater heights. There is not necessarily any direct relation between the rate of growth in youth and the ultimate height attained, and both quick-growing species of Larix, Populus and Eucalyptus, and slow-growing ones such as Abies species may reach a large size. In the case of quick-growing species, the maximum rate of height growth is reached earlier than with slow-growing ones. Büsgen and Munch (1929) stated: "Thus poplar, alder, ash and birch attain their maximum rate of growth even in the second to fifth year, while beech, spruce and silver fir do so only in their third or fourth decade."
The nature of the factors determining maximum height in any tree species is not understood. As the shoot system of the tree increases in size and complexity there is a gradual reduction in the annual growth increment, both in the branches and in the main stem. This process has been called "aging" (Moorby and Wareing, 1963). When the main axis shows aging, so that there is no longer a clearly developed "leader," the tree develops a rounded crown and further height growth effectively ceases. Weber (1891) suggested that the height of the tree is ultimately limited by the height to which water and "sap" can be lifted. Another possibility is that the maximum size of the tree is determined by the distance over which phloem translocation between leaves and roots is effective. Since the maximum height is such an important character of the tree for the forester, more study of the physiology of aging is clearly necessary as a basis for any improvement by breeding.
Mineral nutrition and water relations
In the preceding discussion special emphasis has been placed upon the role of light in tree productivity. It is clear, however, that the growth rate will also be markedly affected by a number of other environmental factors, including temperature, rainfall and soil conditions. This third factor of soil conditions is now to be considered briefly.
It hardly needs to be stated that the growth rate is very sensitive to the mineral nutrient conditions and Watson (1958) has shown that, for agricultural crops, the effect of mineral nutrition is not so much on the photosynthetic rate but upon the rate of leaf production. It will be assumed for the present discussion that this also applies to forest trees. For arable crops it is usually feasible to rectify mineral nutrient difficulties by manuring. In forestry, however, the conditions under which it is economic to apply fertilizers are somewhat limited at present, and in many areas forest trees are grown on soils in which mineral nutrient levels are low. A high premium will therefore be placed upon species able to make active growth under limiting mineral nutrient conditions. The forester is well aware that some species are better able to grow under low nutrient conditions than others, but the physiological basis of such differences has hardly been studied at all so far. It is necessary to know how it is that certain trees, such as Fagus sylvatica, Castanea sativa, Pinus sylvestris and Picea sitchensis can grow actively on poor soils. There seems good evidence that the mycorrhizal habit is important under certain conditions but it is also necessary to know whether there are differences in the tree species themselves which render it possible for certain species to grow under low nutrient conditions. The following questions may be posed:
1. Do such species have low demands for mineral nutrients? Can they produce large amounts of cell wall material in relation to small amounts of protoplasm ?2. Are these species more efficient in taking up the available soil nutrients, either because of an efficient uptake mechanism or because of an extensive root system?
3. Are these species efficient in re-utilizing the mineral nutrients taken up, for example, by transport from senescent to young leaves?
These are questions on which there is very little information at present. It is generally believed that conifers are less demanding than broadleaved species, but if the ratio of nutrients absorbed to dry weight production is taken as the measure of the efficiency of utilization by different species, then the difference between conifers and broadleaved species is less striking and there are numerous exceptions to the rule that broadleaved trees are more demanding (Leyton, 1958 a, b; Ingestad, 1962). As might be expected, fast-growing species generally have high demands but, when their greater dry matter production is taken into account, their requirements are not necessarily higher than those of slow-growing species. Thus, the information available suggests that species which are able to succeed on poor soils are not necessarily less demanding than other species, but rather that they are more efficient in taking up the available soil nutrients. A somewhat different approach has been adopted by Ingestad (op. cit.), who has argued that the most satisfactory basis for comparison of species which differ widely in their anatomical and morphological characters, for example, broadleaved species and conifers, is the width of the growth curve about the optimum in response to varying mineral nutrient levels. Tolerant species will have a broad, flat-topped curve, indicating wide tolerance to varying nutrient levels, whereas species of low tolerance will show a narrow response curve. Using this method of approach he concluded that Pinus sylvestris is comparatively tolerant whereas Betula verrucosa is of low tolerance. On the other hand, the growth response to increased nutrient supply was often small in Pinus sylvestris but great in Picea abies and Betula verrucosa. The requirements of mineral nutrients per unit dry matter produced at maximum growth was comparatively high in Betula verrucosa and low in Picea abies. Further details of Ingestad's work are given in his paper published in the Proceedings of the World Consultation on Forest Genetics and Tree Improvement (Ingested, 1963).
Since water deficiency is a major factor limiting tree growth in many areas of the world, it is clearly a subject of very great importance. There are signs of increased interest in the physiological and ecological aspects of plant water relations, and many aspects are already understood, such as the factors affecting rates of transpiration, the mechanism of water uptake by the roots and its movement in the tree. A useful summary of this subject has been given by Kramer and Kozlowski (1960). On the other hand, although the subject of drought resistance has been studied for many years, any understanding of the physiological basis of drought resistance is very incomplete. For example, although it is well known from field observations that tree species differ very widely in their ability to survive under conditions of very low soil moisture, very little is known regarding the physiological basis of these differences. In addition to information concerning these striking differences in ability to withstand severe drought conditions, more knowledge is required as to the effects of less extreme variations in water supply on the growth of economically important tree species. This problem raises the question of how such differences in growth responses of tree species can be measured and a valuable contribution to this subject has recently been made by Jarvis and Jarvis (1963).
It is clear that if the breeder is to improve the growth and survival of trees under water stress, detailed physiological studies of the kind described by Ferrell (1963) will be required along with the breeding work.
The tree breeder is very much concerned with the quality of the timber which a tree produces, as well as with the volume of wood. The properties of a timber affecting its utilization, such as strength, working properties, appearance, and so on, can be related to the anatomical features of its constituent cells and to the relative abundance of different cell types. Since the cells of the wood are formed by the division of the cambium and by the subsequent differentiation of the cambial derivatives, it is clear that an understanding of the physiology of cambial activity is of fundamental importance for those concerned with breeding for wood properties. It has long been known that auxin plays an essential role in cambial activity in woody plants, but recent work indicates quite clearly that normal cambial activity involves also the recently discovered class of hormones known as gibberellins. Indeed, there is a remarkable synergism between auxins and gibberellins in cambial activity (Wareing, 1958). Auxin alone canoes comparatively little cambial division, but the cambial derivatives formed undergo fairly normal differentiation and wall thickening. With gibberellin alone the cambium undergoes some division, but there is no differentiation of the cells so formed. With auxin and gibberellin together there is a great increase in the rate of cambial division and the derivative cells undergo differentiation to give apparently normal wood.
The cambium appears to be unable to supply its own auxin and gibberellin requirements, and it appears that young expanding leaves are the chief site of auxin synthesis. It is not surprising, therefore, that there is a close correlation between cambial activity and extension growth of the shoot; the initiation of cambial activity in the spring is dependent upon the presence of expanding buds, and there is good evidence that the developing leaves export auxin to the stem below. When extension growth ceases there is a marked fall in auxin production and in "diffuse-porous" broadleaved trees wood formation usually ceases at the same time. In "ring-porous" species the cessation of extension growth appears to mark the time of transition from spring to summerwood formation. The auxin necessary for cambial division during summerwood formation appears to come from the mature leaves. In certain conifers, such as Pinus species, cambial division continues long after extension growth has ceased, and here also it is possible that the needles produce the necessary auxin. Larson (1962) has made a detailed study of the earlywood-latewood transition in conifers and has concluded that variations in tracheid characters can be related to changes in the levels of auxin production by the needles, in response to photoperiod and other environmental factors.
Although these studies on the role of hormonal factors in cambial activity and xylem differentiation constitute a valuable addition to our understanding of these processes, there are still many problems on which we need much more information, especially in relation to breeding for wood quality. For example, we need to know what are the important "internal" and "external" factors controlling cell wall thickness in the wood. A start has been made in this direction for various conifers (Wodzicki, 1961; Richardson, 1963), but much more work is required. Our knowledge of the biochemistry of wan formation increases steadily, but our knowledge of the biosynthesis of cellulose is still very incomplete. More knowledge is also required as to what determines fiber and tracheid length, and especially what determines the changes in fiber and tracheid length which occur with age. If it were possible to predict, from a study of the fiber or tracheid characters in the seedling stages, the wood qualities of the mature tree, breeding and selection for quality would be greatly accelerated and assisted. This problem is discussed in further detail in Chapter 9. There is, as Richardson (1960) has pointed out, a need for far more work in the general field of "physiological wood anatomy."
In the present discussion, attention has been focused upon hormonal factors in cambial activity, but as Kozlowski points out, (Kozlowski, 1963), other factors, such as moisture availability, are important and indeed any factor which becomes limiting may be said to "control" cambial activity.
It is generally assumed that radial growth is determined by the conditions of the environment, especially by light intensity, temperature, mineral nutrition and soil water. There is little doubt that this is frequently the case, but it may also be the case that, just as the rate of height growth may be limited by the pattern of extension growth of the species, so radial growth may sometimes be limited by internal factors such as hormone levels. This possibility merits further study.
The present state of knowledge regarding the physiology of flowering is dealt with extensively by Kozlowski (1963) and this subject win not, therefore, be further discussed here, except in relation to the problem of juvenility in woody plants. The existence of a juvenile period during which flowering will not occur is clearly one of the major obstacles facing the tree breeder. This problem has been the subject of a number of investigations in recent years, but a real understanding of the physiological basis is still lacking. Nevertheless, a number of facts have been established which can form a basis for further work. The question arises as to whether the attainment of the flowering condition depends upon the attainment of a certain size, or whether it is necessary for the tree to undergo a number of cycles of growth and dormancy. This question has now been answered for several species (Longman and Wareing, 1959; Robinson, 1962). Three series of Betula verrucosa seedlings were grown under one of the three following conditions:
1. continuous growth under long days;
2. periodic growth, with alternate long days, short days and chiding (to remove dormancy);
3. normal daylength and temperature conditions.
The first series (grown continuously) reached a height of about 3 meters in just over 12 months from sowing, and these then commenced flowering. The other series had failed to flower when the experiment was discontinued after 27 months; at this time the second series had received 7 cycles of growth and rest. Thus it appears that periodic cycles of growth and dormancy are not necessary for birch seedlings to attain the adult flowering condition, and that flowering win occur when a certain size has been reached, even if growth is continuous. Similar results have been obtained with Larix and blackcurrant (Ribes rubrum).
It is important to distinguish between the attainment of the ripe-to-flower condition and the initiation of flower primordia. It is well known that flower initiation may not occur in certain species every year (an example being Fagus sylvatica) even though the adult condition has been reached. It happens that in Betula long days promote flowering, and hence the same daylength conditions favor both vegetative growth and flowering. This is not so in all species, however. In adult blackcurrant bushes short days promote flowering.
It is necessary for seedling blackcurrants to be grown first under long days until they attain the maximum size for flowering, and then to transfer the plants to short days, when flower induction occurs. In Larix leptolepis, on the other hand, flower initiation is favored by training the shoots horizontally or bending them downward (Longman and Wareing, 1958). In this species flowering was attained in 3 to 4 years from sowing by first growing the seedlings as rapidly as possible in a greenhouse under long days, until they attained a height of about 3 meters and then placing the whole plants horizontally. Such horizontal plants flowered whereas the vertical "controls" showed no flowers. From the practical point of view, therefore, the most effective treatment to induce early flowering of the seedlings seems to be to grow them to a certain size as rapidly as possible, and then to apply various flower-inducing conditions which win vary from species to species. Some of the environmental and other conditions favorable to flowering, once the adult condition has been attained, are described in Chapter 10 of this report.
To understand further the "phase-change" problem, it is important that we should now study the physiological and biochemical differences between the juvenile and adult phases, and the factors controlling the phase change; in other words, the physiological basis of the size factor. The interesting discovery that treatment of adult ivy (Hedera helix) with gibberellic acid causes reversion to the juvenile condition (Robbing, 1957), whereas treatment of juvenile Eucalyptus causes transition to the adult condition (Scurfield and Moore, 1958), seems to indicate that hormonal factors are important in phase change.
The importance of hormonal factors in the flowering of trees is also shown by the extremely interesting studies of Japanese workers on the effects of gibber epic on the promotion of flowering in certain conifers. This work is summarized in the paper by Sato (1963).
Limitations of space prevent a detailed treatment of the many interesting physiological problems presented by the seed of forest trees. Although the problem of seed dormancy, especially in relation to winter chilling, is still incompletely understood, knowledge of the role of endogenous germination promoters and inhibitors is steadily increasing. The present state of knowledge of the physiology of dormancy in forest tree seed has recently been summarized by Villiers (1961).
Some of the most important practical problems arise in connection with the storage and longevity of tree seed,, and the existing literature on this subject has been exhaustively reviewed by Holmes and Buszewicz (1958). Although the most favorable conditions for the storage of tree seed, such as low moisture content and cool temperatures, have been determined in some detail, very little is known regarding the biochemical changes occurring during storage. A better understanding of such changes might suggest means of prolonging the viability period of seeds by chemical treatments.
When transplanting seedlings in the nursery and later to their final position in the forest, it is clear that success is dependent upon the seedlings being able to re-establish roots in their new position before excessive water stress occurs, and that the majority of losses at this stage are due to drying of the seedlings before new roots are established. In view of the great practical importance of this subject, it is surprising that so little attention has been paid to the physiology of root growth in trees. It has been found empirically that certain times of the year are more favorable for transplanting than others, but there seem to have been very few physiological studies on this problem.
In order to obtain establishment of a transplanted seedling it is clearly necessary that there should be rapid resumption of root activity, involving both the growth of existing roots and the initiation of new lateral roots. The initiation of secondary roots appears to involve both auxin and some other unknown factor (Torrey, 1956). It is not known whether the auxin involved in this root initiation is supplied from the shoots or by the root system itself. Richardson (1958), however, observed that secondary root formation in Acer saccharinum in spring was stimulated by the presence of buds on the shoot and that in disbudded seedlings the effect of the buds could be replaced by applied auxin. In this species, therefore, the growth of roots and shoots appears to be interrelated through hormonal factors.
The evidence regarding seasonal periodicity of root-growth in trees is conflicting, some authors claiming that root growth can occur at any time of the year when temperature and soil moisture conditions are suitable, while others have claimed that there is evidence of endogenous periodicity (see Ladefoged, 1946) in root growth. As Richardson (op. cit.) has pointed out, this conflict is probably partly due to failure to distinguish between "imposed" and "physiological" dormancy. Richardson found in Acer saccharinum that root growth ceased with leaf fall in the autumn but resumed in the spring before there was any detectable bud growth. Removal of bud dormancy by chilling was necessary before root growth could occur, however, and disbudding prevented root growth. Thus, it appears that the presence of nondormant buds, is necessary for the initiation of root growth in the spring in this species.
Although the evidence regarding periodicity of root growth in conifers is confusing it does appear that in certain species of Pinus there are peaks of growth in the spring followed by a period of quiescence during shoot elongation and then a second burst of root-activity during July-September (Wight, 1933; Nelson, 1963). The absence of root activity during shoot elongation in the spring clearly constitutes, a hazard for freshly transplanted seedlings and indicates that transplanting must be carried out in sufficient time before bud break to allow some root growth to take place in the spring, or else delayed until later in the season, to coincide with the usual period of root activity. Very little is known regarding periodicity of root growth in the conifers, but the difficulty in establishing certain species in the spring may well be due to their having similar patterns of root growth to those of Pinus. In view of the very great economic implications of this subject, it is clear that more fundamental knowledge is urgently required.
The formation of adventitious roots on the stems of cuttings is clearly a closely related subject. Although there have been many investigations in horticulture on the rooting of cuttings, knowledge of its physiological basis is surprisingly meager. Considering, first, the rooting of cuttings of broadleaved species, it is convenient to distinguish between cuttings taken in the dormant winter condition (" winter cuttings ") and those taken in the leafy condition in the summer (" summer cuttings "). It has long been known that the rooting of a winter cutting is, greatly stimulated by the presence of swelling buds in the spring, and it has generally been assumed that the buds supply endogenous hormones which accumulate at the base of the cutting and promote root formation. In temperate species only nondormant (that is, chilled) buds stimulate rooting; if winter cuttings are taken in the autumn and maintained under warm conditions, the rooting response is very poor; if, however, the cuttings are first maintained at chilling temperature of 0° to 5°C (32-41°F) for several weeks, to overcome the dormancy of the buds, then such chilled cuttings root readily when transferred to warm conditions. In temperate climates the chilling requirements will usually be met by the normal winter temperature conditions, but in regions with very mild winters the dormancy of the buds may not be fully removed and the possibility of artificial chilling of cuttings before planting may be advisable. In the United Kingdom, dormant or "winter" cuttings of Populus x canescens do not root very readily, but it was found that exposing the cuttings to 16 weeks of chilling in a refrigerator greatly improved the subsequent rooting of the cuttings (Wareing and Smith, 1962).
In view of the stimulating effect of swelling, buds on the rooting of "winter" cuttings, and the fact that young expanding leaves appear to be rich sources of auxin production, it might be expected that actively growing, leafy "summer" cuttings would root more readily than shoots which have ceased extension growth, but this does not appear to be the case. Provided that mature leaves are present, it appears to make little difference whether there are actively growing or dormant buds present (Wareing and Smith, 1963); this would seem to indicate that mature leaves can supply the endogenous auxin necessary for rooting. Various environmental factors, such as light intensity and temperature, are well known to affect the rooting of cuttings. Recently it has been shown that daylength conditions also exert a marked effect on the rooting of leafy cuttings (Nitsch, 1957) and, where they are maintained under controlled conditions (as, for example, in "mist culture"), it may be advantageous to extend the natural daylength in the spring and autumn by artificial illumination.
In forest practice, vegetative propagation by cuttings is most frequently used for species of Populus, in which the rooting of certain species is greatly facilitated by the root primordia which are already present in the cortex of the parent shoots. However, with the advent of new techniques such as the use of hormones and "mist culture," it is possible to root cuttings of a wide range of species, including many conifers, and further physiological studies may well lead to yet more advances in technique. For example, basic research might profitably be directed to seeking an explanation for the aging associated with loss of capacity to strike in stem cuttings as compared with the ease with which this is achieved in cuttings taken from very young seedlings. This capacity in juvenile material is very widespread in woody plants and the explanation, too, may be a general one. Although it is not yet economic to raise clonal material for general stock, there may well be advantages in raising clonal rootstooks for seed orchards and other special purposes.
BARNER, J. 1955. Der tagiszyklishe Verlauf von Assimilation und Atmung im Lichte stoffproduktions analytischer Vergleichsuntersuchungen. Ber. Dtsch. Bot. Ges., 68: 271-274.
BÜSGEN, M. & MÜNCH, E. 1929. The structure and life of forest trees. London, Chapman and Hall.
COOMB, D. E. 1960. An analysis of the growth of Treema guineensis. J. Ecol., 48: 219-231.
COOMB, D. E. & HADFIELD, W. 1962. An analysis of the growth of Musanga crecopioides. J. Ecol., 50: 221-234.
DONALD, C. M. 1961. Competition for light in crops and pastures. Soc. Exp. Biol. Symp. 15. London, Cambridge Univ. Press.
FERRELL, W. K. 1963. Tree physiology-genetics studies at Oregon State University. FAO/FORGEN, 63-5/4.
HAGEM, O. 1962. Additional observations on the dry matter increase of coniferous seedlings in winter. Medd. fra Vestlandets Forstlige Forsoksstasjon, 37 (11): 5.
HELLMERS, H. & BONNER, J. 1959. Photosynthetic limits of forest tree yield. Proc. Soc. Amer. For., 1959: 32-35.
HOLMES, G. D. & BUSZEWICZ, G. 1958. The storage of seed of temperate forest tree species. For. Abstracts, Leading Article Series No. 26.
HOLZER, K. 1963. Physiological investigations on Picea abies (L.) Karst. in view of genetics. FAO/FORGEN, 63-5/3.
HUBER, B. & POLSTER, H. 1955. Zur Frage der Physiologischen Ursachen der unterschiedlichen Stofferzeugung von Pappelklonen. Biol. Zbl., 74: 370-420.
INGESTAD, T. 1962. Macro element nutrition of pine, spruce and birch seedlings in nutrient solutions. Medd. fran Statens Skogsforsknings Institut, 51: 7.
INGESTAD, T. 1963. Comparison of nutritional properties in forest tress. FAO/FORGEN, 63-5/9.
JARVIS, P. G. & JARVIS, M. S. 1963. Water relations of tree seedlings. I. Growth and water use in relation to soil water potential. Physiol. Plant, Copenhagen, 16: 215-235.
KOZLOWSKI, T. T. 1949. Light and water in relation to growth and competition of Piedmont forest tree species. Ecol. Monogr., 19: 207-231.
KOZLOWSKI, T. T. 1961. Challenges in forest production. Challenges of Forestry. Syracuse, New York, State Univ. Coll. of For.
KOZLOWSKI, T. T. 1963. Physiological implications in tree improvement. FAO/FORGEN, 63-5/1.
KRAMER, P. J. 1956. The role of physiology in forestry. For. Chron., 32: 297-308.
KRAMER, P. J. & KOZLOWSKI, T. T. 1960. Physiology of forest tress. New York, McGraw Hill.
LADEFOGED, K. 1946. Untersuchungen über die Periodizität im Ausbruch und Langenwachstum der Wurzeln. Forstl. Forsøgsv. Danm., 16: 1-256.
LARSON, P. R. 1962. Auxin gradients and the regulation of cambial activity. In Tree physiology. New York, Ronald Press.
LEYTON, L. 1958a. The relationship between the growth and mineral nutrition of conifers. In The physiology of forest tress. New York, Ronald Press.
LAYTON, L. 1958b. The mineral nutrient requirements of forest plants. Encycl. Plant Physiol., 4: 1026-1039.
LONGMAN, E. A. & WAREING, P. F. 1958. The effect of gravity on growth, apical dominance and flowering in fruit trees. Nature, Lond., 182: 379-381.
LONGMAN, K. A. & WAREING, P. F. Early induction of 1959 flowering in birch. Nature, Lond., 184: 2037.
MATTHEWS, J. D. 1963. Some applications of genetics and physiology in thinning. Forestry, 36: 2.
MOORBY, J. & WAREING, P. F. 1963. Ageing in woody plants. Ann. Bot., Lond., 27.
NELSON, C. D. 1963. The production and translocation of Photosynthate - C14 in conifers. In The formation of wood in forest trees. New York, Academic Press.
NITSCH, J. P. 1957. Photoperiodism in woody plants. Proc. Amer. Soc. Hort. Sci., 70: 526-544.
NICHIPOROVITCH, A. A. 1960. Photosynthesis and the theory of obtaining high crop yields. Field Crop Abst., 13: 169-175.
OVINGTON, J. D. 1957. The volatile matter, organic carbon and nitrogen contents of tree species grown in close strands. New Phytol., 56: 289-388.
OVINGTON, J. D. 1958. Some biological considerations of forest production. In The biological productivity of Britain. London, Institute of Biology, p. 83-90.
OVINGTON, J. D. 1960. The accumulation of energy in forest plantations in Britain. J. Ecol., 48: 639-646.
OVINGTON, J. D. 1961. The productivity of some British wood lands. In Recent advances in botany. Toronto, Univ. of Toronto Press.
POLSTER, H. 1955. Vergleichende Untersuchengen über die Kohlendioxydassimilation und Atmung der Douglasie, Fichte, und Weymouthskiefer. Arch. Forstw., 4: 689-714.
REES, A. R. 1962. Dry-matter production by evergreen perennials. Nature, Lond., 195: 1118-1119.
REES, A. R. 1963. Relationship between crop growth rate and leaf area index in the oil palm. Nature, Lond., 197: 63-64.
RICHARDSON, S. D. 1958. Bud dormancy and root development in Acer saccharinum. In The physiology of forest trees. New York, Ronald Press. p. 409-425.
RICHARDSON, S. D. 1960. The role of physiology in forest-tree improvement. N.Z. For. Serv. Reprint No. 23.
RICHARDSON, S. D. 1963. The external environment and tracheid size in conifer seedlings. In The formation of wood in forest trees. New York, Academic Press.
ROBINSON, L. W. 1962. Juvenility in woody plants. Ph. D. thesis. Univ. of Wales.
ROBBINS, W. J. 1957. Gibberellic acid and the reversal of adult Hedera to a juvenile state. Amer. J. Bot., 44: 743-746.
RUTTER, A. J. 1957. Studies in the growth of young plants of Pinus sylvestris L. I. The annual cycle of assimilation and growth. Ann. Bot., 21: 399-426.
SATO, K. 1963. Some physiological actions of gibberellins on forest trees. FAO/FORGEN, 63-5/2.
SCHMIDT, H. 1953. Investigations on crown and increment in Norway spruce of the Bavarian foothills. Forstwiss. Cbl., 72: 276-286.
SCHÖPFER, W. 1961. Quantitative determination of the assimilating organs of Norway spruce. Schrift der Landesforstverwaltung, 6: 828-847.
SCURFIELD, G. & MOORE, G. W. E. 1958. Effects of gibberellic acid on species of Eucalyptus. Nature, Lond., 181: 1276-1277.
TORREY, J. G. 1956. Chemical factors limiting lateral root formation in isolated pea roots. Physiol. Plant. Copenhagen, 9: 370-388.
VILLIERS, T. A. 1961. Dormancy in tree seeds. Proc. Internat. Seed Testing Ass., 26: 516-536.
WAREING, P. F. 1958. Interaction between indole-acetic acid and gibberellic acid in cambial activity. Nature, Lond., 181: 1744-1745.
WAREING, P. F. & SMITH, N. G. 1962. Physiological studies on the rooting of cuttings. Report on Forest Research, 1961/2. London, Forestry Commission.
WAREING, P. F. & SMITH, N. G. 1963. Physiological studies on the rooting of cuttings. Report on forest research, 1962/63. London, Forestry Commission.
WATSON, D. J. 1956. Leaf growth in relation to crop yield. In The growth of leaves. London, Butterworth, p. 178-194.
WATSON, D. J. 1958. Factors limiting production. In The biological productivity of Britain. London, Institute of Biology p. 25-32.
WEBER, R. 1929. Cited by Büsgen and Münch, p. 10. 1891.
WIGHT, W. 1933. Radial growth of the xylem and starch re serves in Pinus sylvestris. New Phytol., 32: 77-96.
WODZICKI, T. 1961. Investigation on the kind of Larix polonica Roc. wood formed under various photoperiodic conditions. Acta Soc. Bot. Polon., 30: 111.
