Individuality in Forest Trees

ASSOCIATE PROFESSOR OF BOTANY

EVER since man became a sedentary agriculturalist he has been faced with the problem of understanding the nature of plants, how they grow, and how they affect each other. Patient trial and error through centuries of time produced many important rules for the successful cultivation of plants, but contributed relatively little to a knowledge of their fundamental nature. Only in the last several decades have we come to understand green plants as individuals able to manufacture their own food and carry on their own growth processes so long as environmental conditions, including adequate supplies of water, minerals, and light, are favorable. From here it was but a step to the development of the concept that the principal relationship between plants is one of competition for required components of the environment. Thus individuals growing close together are thought to alter each other's environment by utilizing available supplies of water, nutrients, and light; stresses induced by overutilization result in a general reduction of growth and often in the death of some of the competing organisms. This theory is admirably supported by facts gathered not only through centuries of empirical experience in growing plants, but also through more recent experiments designed specifically to test the effects of competition.

Although the competition theory was developed primarily through experience gained with herbaceous agricultural plants, it has been widely accepted by forest botanists and used to explain relationships between forest trees. Thus, under forest conditions, the success of some trees and the failure of others is generally ascribed to differences in the competitive ability of the individuals. This idea is the touchstone from which most botanists proceed in their analysis of forest tree behavior. It also underlies the universally accepted idea in forestry that under given environmental conditions each tree has, in theory, an ideal growing space. The regular spacing of trees in a forest tree plantation or in artificially thinned forests is a reflection of this theory.

It now appears that the interpretation of the tree as a strictly individual entity is, in many instances, open to serious question. Evidence amassed, particularly in the last decade, indicates that root grafts or fusions between the roots of two or more trees are not uncommon. The existence of continuity of conducting tissues between trees opens the possibility that biological activities in one may be directly influenced by another through biophysical and biochemical forces exerted through natural root grafts. These forces exerted directly by one tree on another may play an important role, along with competition, in shaping the nature of the forest community.

Three categories of root grafts occur in nature. Although the most common type are grafts between roots of an individual tree, these are germane to the present discussion only when they enter into complexes of grafts between trees. Interspecific grafts between roots of trees of different species, although of considerable biological interest, are relatively rare so far as we know now. Intraspecific grafts between roots of trees of the same species are common and it is with these that most of the following paragraphs will deal.

For a long time natural root grafts have been known as a botanical curiosity and reports of their occurrence have appeared sporadically throughout the literature of botany and forestry. These came primarily from direct sightings of grafts or from observations that certain seemingly dead stumps continued to five and grow in diameter. It was discovered that these living stumps were invariably connected to living trees through root grafts (a conclusion we have confirmed on numerous occasions by injecting radioisotopes into stumps and detecting them in adjacent trees). Additional reports are found in the literature of plant pathology where root grafts were found to be responsible for the transmission of disease organisms from tree to tree.

Prior to 1950, although dozens of articles contained reports of root grafts, there was little recognition of their widespread occurrence or their potential significance in forest ecology. Among a few notable exceptions was the work of Professor F. S. Page '13 of Dartmouth College who, in 1927, after an intensive survey of several old stands, suggested that root grafting might play a major role in the nutrition of white pine trees.

One of the great deterrents to extensive studies of root grafting was the amount of labor required to unearth roots in the search for grafts. This problem was greatly simplified with the arrival of the nuclear age and a general availability of radioisotopes. Appropriate techniques were soon developed for the injection of radioisotopes into trees, and shortly thereafter it was learned that these were readily transported through root grafts to adjacent trees. There, with the aid of portable radiation counters, it was possible to detect the radioisotope in minute quantities and, if gamma radiation was utilized, to detect it, deep within the sapwood. Not only can grafted trees be detected by the radiation method but it is possible to trace "hot roots" from the grafted tree back to the injected tree and thus with a minimum of excavation expose the graft.

By means of this and other techniques, the occurrence and function of root grafts have received increasing study from botanists around the world, particularly in the United States, Finland, and in the Soviet Union.

Here at Dartmouth, for the past three years, we have studied root grafting in white pine forests growing on a wide range of sites in New Hampshire, Maine, and Vermont. Our results indicate that grafting is a much more common phenomenon than previously suspected. Thirty to eighty per cent of the trees in all stands studied by us were found to be grafted to one or more adjacent trees. Grafting occurred in stands of all ages, and in stands growing in coarse sand, in loams, in stony soils, in soils free of stones, in well-drained soils, in poorly drained soils, and in extremely wet peat soils. We have concluded that given proper spacing conditions, grafting will occur among white pine trees growing under the whole range of conditions tolerated by the species population.

Evidence from other investigations suggests that our results may be applicable to a wide range of forest tree species. Notable work by Yli-Vakkuri in Finland indicates that up to 28 per cent of the trees in older Scots pine forests are grafted. In the U.S.S.R., Junovidov has estimated 26, 30, and 45 per cent, respectively, of the trees in young forests of Siberian pine, larch, and fir are grafted. Among 35-year-old trees in a plantation of Crimean pine, Beskaravainya found 50 per cent to be grafted. Studying a German spruce stand, Wichmann found 462 living stumps and 3334 living trees. Since the root system of each stump was grafted to at least one living tree, and many of the trees were undoubtedly grafted to each other, this stand must have been a maze of root grafts. The extreme of root grafting has been reported by Kuntz and Riker for dense stands of jack oak growing in Wisconsin. Using radioisotope tracers they found practically all trees to be united with their neighbors.

In summation, we have compiled a bibliography of 171 references containing reports of intraspecific root grafting in 105 species of forest trees. Based on the degree of grafting mentioned in each report and the number of times grafting has been reported for each species, we have concluded that grafting is a common phenomenon in 63 of these species including such important trees as white, red, and Scots pine, European and American beech, Douglas fir, eastern hemlock, red oak, and American elm. Thus, there seems little doubt that intraspecific root grafting is a widespread phenomenon of considerable potential importance.

The role of intraspecific grafts in the physiology and ecology of grafted trees is slowly unfolding. The movement of water, minerals, food, dyes, and poisons, up to distances of 43 feet between trees, indicates that grafts function in the transport of substances. In a study of the rate of water intake by freshly cut stumps, we have found a correlation between the rate of intake and weather conditions conducive to high water losses through the leaves of trees grafted to the stumps. Apparently stresses on the water conducting system, developed as a result of water loss from the intact trees, were felt throughout the grafted root system and were, in part, responsible for regulating the rate of intake by the stumps.

Another indication of how one tree may influence another is provided by the observation that radioisotopes moving in the root system of one tree are frequently shunted through grafts to the root system of another tree. The observation that radioisotopes move predominantly from tall, vigorous oaks to small oaks, has prompted the suggestion that the former may aid in the survival of the latter.

The best understanding of the ecological importance of root grafting would come from the observation of groups (unions) of grafted trees over long periods of time. To shorten the time requirement, we have separately studied the kinds of unions that occur in white pine stands of various ages and then fitted the pieces together in a time sequence. The following pattern has emerged. Early in the life of a stand, some trees become grafted together in unions, while other trees remain as individuals. As the stand matures and the trees become larger, root systems increase in size, and a union may coalesce with an adjacent union or individual. Throughout this period, the tendency to add new members to the union is counteracted by the death of some of the members. However, death proceeds from the top down and the root system of a tree often remains alive and functional within the graft complex for years after the top has decayed and toppled over. In the older stands, the union is often represented by one to several robust trees surrounded by one to several living stumps. Thus as a particular union matures the more vigorous trees come to dominate the group, while less vigorous companions succumb leaving only their roots to function in the service of the union.

The discovery of the mechanisms by which a few trees are able to dominate a union is a major objective. Perhaps because of some initial advantage of size, vigor, or heredity these trees are able to establish gradients that cause water and minerals absorbed by the root complex to move primarily to them at the expense of their less vigorous companions. Another mechanism by which dominant trees may influence smaller trees grafted to them was suggested by the results of this past summer's field work.

The initiation, in the spring, of cambial growth (growth in diameter) within the twigs, branches, and trunk of a tree is thought to be triggered by a hormone moving down the stem from the site of its production in the expanding buds. A study of the stumps of detopped trees known to be free of grafts indicated that, although many stay alive for the first growing season after cutting, none exhibit diameter growth. Examination of the cambial regions of these stumps indicated that the cells were alive and healthy and that stored foods were present. The failure of the cambium to resume activity was apparently due to the lack of the triggering factor. Conversely, stumps grafted to living trees resumed normal cambial activity probably following the transfer of the triggering factor through grafts from the intact tree. The interesting point here is that one tree has been able to initiate activities in the other. This raises the possibility that large trees whose buds break first would be able to initiate cambial activities in smaller trees grafted to them before their own buds break. The implications of this are not wholly clear, but premature initiation may introduce a serious element of disharmony into the normal processes of development and mobilization of stored foods. Perhaps this is one of the ways that larger trees gradually bring about the demise of the tops of smaller trees while their roots are retained as part of the root complex.

Now to evaluate the ecological importance of root grafting. Often abandoned or disturbed land is occupied by dense stands composed of a single species; for example, white pine and red spruce in New England, red, white, and jack pine in the Lake States, and loblolly, shortleaf, and longleaf pine in the Southern Coastal Region. These single species stands, and others like them throughout the world, are supposed to exhibit competition par excellence since the trees are crowded and each tree has almost identical requirements. Attrition over a period of a few years is great, and often up to 98 per cent of the trees may be eliminated in five to ten decades. The usual explanation for this ecological paring process is severe competition.

Our experience with white pine, however, indicates that the importance of competition as a factor in stand development may be severely overrated. It seems likely that competition plays an important role during the first five or ten years of the existence of a stand. Thereafter unions begin to form and the stand must be visualized as being composed of unions of grafted trees interspersed with non-grafted individuals. The number of trees in any stand occurring in unions or as individuals is primarily a function of the age of the stand and the space between trees. In the extreme, it is conceivable that an entire stand may be united in a single anastomosing root network. Competition may eliminate the less fit individuals, but, within the unions, which often contain a majority of trees in the stand, forces operating through root grafts will determine which trees will remain as dominant entities.

Whether or not our results apply to other species is a matter for future research to decide, but it seems likely that some of the numerous species where root grafting is common will behave in a similar way. In other species, it is probable that root grafting may have different effects; for example, in oak stands root grafts may aid in the survival of smaller trees while in white pine they hasten the demise of smaller trees.

At any rate, these findings, that many forest trees do not always occur as individual entities and that biological forces other than competition may play a major role in shaping a developing stand, open a wholly new aspect of basic research in interplant relationships. They also bring into question widespread practices in applied forestry based either on the concept of the forest tree as an individual or on the concept of competition as the dominant biotic force in stand development.

Suggested Reading

Bormann, F. H. and B. F. Graham. 1959. The occurrence of natural root grafting in eastern white pine, Pinus strobusL., and its ecological implications. Ecology, Vol. 40: 677-691.

Kuntz, J. E. and A. J. Riker. 1956. The use of radioisotopes to ascertain the role of root grafting in the translocation of water, nutrients, and disease-inducing organisms among forest trees. Proceedings, International Conference on Peaceful Uses of Atomic Energy (Geneva, Switzerland), Vol. 12: 144-148.

An extreme case of root grafting discovered between roots ofwhite pine trees growing on a deep peat soil. The black circlesindicate tree trunks. Grafts, both self and intraspecific, may beidentified wherever lines outlining roots merge rather than cross.A number of grafts have been removed in the areas marked C.