TWO MILLION acres of forest land were planted with tree seedlings, for which about a thousand tons of seeds from a multitude of species had been collected, in the winter of 1959.

In 1985 or thereabouts we shall know the results, but until then we cannot be sure whether the trees will be tall and straight or crooked and runty. We are not sure because the genetic quality of most of the seed was not known. In fact, the pine species probably were of a low grade because the seed collectors got much of it from low and bushy trees that they could climb more easily than taller, better trees.

It is too bad that only a few people were concerned about tree seed, that forests were left to regenerate themselves, that harvested trees often were replaced by undesirable species, that new stands were sparse or failed to come up. Many people considered it farfetched even to think about controlling the quality of the seeds.

The situation has changed quickly and radically. Forestry practices have been intensified in a hundred ways. The planting of forest seedlings has mushroomed, and millions of acres are being planted. Foresters see the need and the opportunity for improving the planting stock.

It is not easy to do so. To provide superior tree seed in the needed amounts takes time. The task is to achieve large-scale production of the best possible seed as quickly as possible.

Selection of good trees as parents, breeding stock from them, and testing the progeny will produce forests of better trees that grow much faster than they do today. In the meantime, landowners have had to settle for seed that is only a little better.

The most efficient way yet devised to produce large amounts of genetically superior seed is to make a seed orchard, a plantation of genetically improved trees that is intensively managed to produce large seed crops.

We have no production figures of seed orchards in the United States, because they are too young to produce seeds in quantity. Two American natives, loblolly pine (Pinus taeda) and slash pine (P. elliottii), produced 43 and 25 pounds of viable seeds per acre at 20 years of age in Australia. From that, we assume that an acre of seed orchard should produce enough seeds each year to plant at least 200 acres in the United States.

TO ACHIEVE genetic improvement, a number of steps must be taken before the seed orchard can be planted.

The heritable characteristics of the parent trees must be defined. Improvement of growth rate, tree form, and wood properties generally is the aim.

Sometimes special features are emphasized, such as resistance to blister rust in white pine (Pinus monticola, P. strobus), high yield of oleoresin in slash pine, or drought resistance in loblolly pine that is to be grown in some parts of Texas.

The next step is rigorous and careful selection of parent trees within the range of the species that is to be propagated. Selections are grouped by geographic races of the species if the existence of such races is known or suspected. Geographic races have been indicated in loblolly pine, Scotch pine (P. sylvestris), white ash (Fraxinus americana), green ash (F. pennsylvanica), black cottonwood (Populus trichocarpa), slash pine, shortleaf pine (P. echinata), and longleaf pine (P. palustris). Most species with a wide physiographic range may contain two or more races.

The next step is tests to determine how much genetic improvement the selected parents provide. The tests involve breeding the selections and observing the growth rate, form, and wood properties of the resulting progeny. If the young trees do well, the parents are accepted as breeding stock.

To save time, this step may be postponed until the seed orchard is established. Progeny tests can then be made while the orchard is growing up, and unsatisfactory parents are removed as the results of the tests become available.

The final step is vegetative propagation of the parents and planting in the seed orchard. Grafting is the most widely adopted propagation method for establishing a seed orchard, but air layering or rooting of cuttings can be used.

We know little about the relationships between rootstock and scion in the coniferous species. Genetic and physiological incompatibility between the two sometimes cause abnormal growth, imperfect union at the point of grafting, and death. Until the causes of incompatibility are better known, most workers use random nursery seedlings for grafting stock.

The seed orchard should be in a place where there will be no damage from wind, snow, and ice. Soil fertility and internal drainage should be excellent. The area should be easily-accessible to facilitate intensive management, which requires expensive equipment and skilled labor.

The orchard should be isolated from contaminating pollen sources. For southern pines, for example, an isolation strip at least 400 feet wide is recommended. The isolation strips could serve as pasture or cropland or they could be planted to trees of a commercial species that does not hybridize with the orchard trees.

The size of the seed orchard has a bearing on its efficiency even though the strip can be made productive.

B. J. Zobel, at North Carolina State College, calculated that a square seed orchard of 4 acres surrounded by a 500-foot isolation strip occupies 40 acres in all. He recommended that seed orchards be made as large as circumstances permit. The Georgia Forestry Commission, following this reasoning, started a 325-acre orchard for slash pine and loblolly pine.

The orchards contain a number of plants propagated from each parent tree, which constitutes a clone—a group of plants derived from a single individual by grafting, for example.

We have no firm experimental evidence on the minimum number of clones necessary for adequate cross-pollination. Most European workers give a range of 9 to 50 clones. Dr. Zobel reported that most seed orchards in the Southeastern States contain 15 to 25 different clones on each acre.

Clones should be arranged in a random pattern to permit statistical analysis of clonal differences. Two plants from the same clone should have at least two other plants in between to reduce pollination within the same clone.

Spacing should be wide enough to allow good development of tree crowns and rate of growth. Final spacings of 20 x 20 feet to 30 x 30 feet are recommended. The wider spacing is for the large-crowned species. For slow-growing conifers with narrow crowns, initial spacing may be 15 x 15 feet, to make sure that the pollen production, which is low at the start, will be used more efficiently.

Spacing can be increased by removing every other tree when the trees become crowded.

A permanent sod cover in the seed orchard prevents erosion, but it competes with the trees for moisture and nutrients. It probably is best to clean-cultivate the whole area or around the base of the trees during the years that they grow rapidly in order to conserve moisture.

Irrigation may be feasible when moisture is critical.

Nutrient levels in the soil may be improved by applications of fertilizer and the use of leguminous cover crops.

The best system of orchard management must be worked out for each species and soil region.

Seed collection would be simplified if the orchard trees could be kept at a maximum height of 25 to 30 feet. Pruning to achieve that and to develop rounded, bushy crowns would be simple in hardwoods but does not offer much promise for conifers. Severe pruning to limit the height may reduce seed production of coniferous species because most of the cones are found in their upper parts.

TREE SEED ORCHARDS in the United States produced a small amount of seed in 1961. It will be at least 1970 before the orchards will add much to the supplies of seed of some species. Fewer than 500 acres of tree seed orchard had been established in 1961, but the area was being expanded rapidly, notably in the Southeast.

T. O. Perry and Wang Chi Wu, at the University of Florida, estimated that 6 thousand acres of orchard will supply all of the slash and loblolly pine seeds needed. We have no estimates of this kind for other species.

Seed orchards in North America in 1961 included such species as slash pine, loblolly pine, shortleaf pine, eastern white pine, western white pine, ponderosa pine (P. ponderosa), red pine (P. resinosa), Douglas-fir (Pseudotsuga menziessi), noble fir (Abies nobilis), and hemlock (Tsuga heterophylla).

The parents in only two of the seed orchards had been progeny tested—a small experimental slash pine seed orchard near Lake City, Fla., for high yield of oleoresin, and a shortleaf seed orchard near Union, S.C., for resistance to the littleleaf disease.

Seed orchards will produce large quantities of seed efficiently, although genetic improvement may be slight at the start. They have practical advantages over random collection of seed from natural stands. Large amounts of seed can be collected with less effort and at the proper time. Seed yields per bushel of cones will be greater because of better cross-pollination. Seed orchards, above all, afford good opportunities for stimulating and protecting abundant cone crops through intensive management practices.

SEED-PRODUCTION areas can serve until seed orchards come into production. They are good natural or planted stands of seed-bearing age, which have been thinned to provide the best possible growing space for the remaining trees.

As a rule, not more than 30 to 40 of the best formed trees are left after one or more cuts. Genetic improvement by this practice is uncertain, but since it insures that seed comes from the best trees, it is an improvement over random seed collection without any control over the source of seed. This type of control is important because poor form can be inherited. One southern pine tree, whose poor form can be passed on to the next generation, may produce enough seeds to plant 30 to 40 acres.

Several thousand acres have been converted to seed-production areas for slash pine, loblolly pine, red pine, eastern white pine, western white pine, ponderosa pine, sugar pine (P. lambertiana), and Douglas-fir. The total area in seed-production tracts is being extended rapidly over the whole Nation. The most rapid expansion is taking place in the pine belt of the Southeast.

MANY FACTORS influence the production of forest tree seed. In most species, 3 to 6 months elapse between flowering and seed ripening. In the pines, however, fertilization takes place a year after pollination, and the seed ripens in the fall of the second year.

Each species exhibits a rather well-defined pattern of seed production throughout the life of a tree. The age at which flowering starts and the span of productive years vary sharply among the species. Vigorous trees with large crowns, however, produce the largest crops for a given age and species.

Seed crops from the same tree fluctuate from year to year, partly because of weather. Heavy rainfall during pollination may wash a great deal of pollen to the ground and so reduce pollination to much below the normal. Late frosts and strong winds can cause a considerable loss of flowers. Available moisture when flower primordia—the first flower cells—are being formed is important.

Most forest trees bear seed in cycles. A good crop may occur at intervals of 2 to 10 years, depending on the species. Between good years, crops are much lighter and sometimes fail completely. One reason may be that food reserves stored in the stem are exhausted after a bumper seed crop and a number of years are needed to replenish these reserves.

Trees vary also in their inherent fruitfulness. Some trees do not produce a good seed crop even under the most favorable circumstances, while others consistently rank high.

Some of the factors can be manipulated to cause premature flowering, stimulate heavy seed production, and lessen the year-to-year fluctuation.

The time at which flower primordia are first differentiated in the bud determines when stimulation treatments can be applied most effectively. Female flowers start to form during August or during September in Pinus, Taxus, and Pseudotsuga. Male flowers in these genera generally are initiated several weeks before their female counterparts. Environmental and inherent factors, however, cause considerable variation within a species.

Heavy thinning—release—generally increases seed production in well-stocked stands. Longleaf, loblolly, and slash pine bear two to three times as much seed several years after release increases the growing space, moisture, and nutrients for each tree. Single trees with no competition from neighboring trees or other vegetation produce the largest seed crops.

The highest seed production per acre, however, requires a closer spacing than for maximum production per tree. Maximum seed production per acre for slash pine, for example, can be obtained with spacing about 25 x 25 feet.

Release cuttings to develop seed-production areas can best be applied before the stand reaches pole size. The remaining trees can then maintain large crowns and high vigor. If release is delayed too long, the trees will not respond.

Fertilization generally raises seed production of forest trees and causes slash pine (and perhaps other trees) to start bearing seed at an earlier age. One to two pounds of commercial fertilizer per inch of diameter (at breast height) applied around each tree will promote health and vigor and thus benefit seed production.

White oak (Quercus alba), sugar maple (Acer saccharinum), beech (Fagus grandifolia), loblolly pine, slash pine, sugar pine, and Douglas-fir have responded to single applications of commercial fertilizers, and their seed crops have increased twofold to fivefold. No firm recommendations can be made, however, as to amount and type of fertilizer until nutrient requirements for each species have been determined. Fertilizer ratios can then be calculated on the basis of deficiencies shown by soil analysis.

Stem injury, which retards the normal phloem transport of organic food substances, stimulates production of seed of a number of forest trees. Stems may be girdled or banded. A girdle is made by removing a strip of bark and cambium in two semicircles or a spiral around the tree. Wire or metal strips can be used to band the tree; this treatment, however, may be less effective than girdling. Longleaf pine, slash pine, loblolly pine, red pine, and Scotch pine have yielded two to three times a normal cone crop after they were girdled.

Seed stimulation by means of stem injury should be used with caution, because this type of treatment may become harmful in the long run. Repeated wounding may kill the trees by weakening them too much or by making them more susceptible to damage from insects and diseases.

Root pruning, a less injurious treatment, stimulates production of seed. Careful pruning may be repeated without much damage. Root pruning combined with fertilization is most likely to increase yields in seed orchards with a minimum risk to the trees.

SEVERAL DISEASES attack flowers, fruits, and cones of forest trees. Seed losses from disease can become serious on conifers. As far as we know, damage to hardwoods usually is slight.

Among the cone rusts that occur on various conifers, the one that attacks slash and longleaf pines is the most important. It is caused by the fungus Cronartium strobilinum and destroys annually nearly 20 percent of first-year slash pine cones. The fungus infects the female flowers. As the infected flowers grow into a cone, they swell to several times natural size and in late spring turn to a bright yellowish-orange color. Diseased cones as a rule are heavily attacked by cone moth (Dioryctria) larvae, and in turn serve as a source of further insect infestation of neighboring healthy cones. Practically all diseased cones are shed by late summer.

Evergreen oaks, particularly live oak (Quercus virginiana), and the runner oaks (Q. pumila and Q. minima) are alternate hosts to the fungus. It is advisable therefore to establish seed orchards of slash pine north of the range of these alternate hosts.

Weather conditions greatly influence the severity of cone rust. Some years are more favorable than others for the buildup of the disease on the evergreen oaks. There also appears to be a direct relationship between heavy infection and long periods of high relative humidity and rainfall.

Ferbam, a fungicide, has given good protection against cone rust. It can be applied with an orchard sprayer in January and February, when the flowers of slash pine are susceptible to infection. The most effective time to spray seems to be just before or immediately after each 18-hour (or longer) period when the relative humidity exceeds 85 percent. Delay of only a day or two may mean a complete loss of protection.

Other cone rusts are similar in many ways to the one we described. Cronartium conigenum, for instance, infects cones of Chihuahua pine (Pinus leiophylla) in Arizona and New Mexico. The fungus kills up to 50 percent of the cones on groups of trees and up to 90 percent on single trees. Alternate hosts are Quercus emoryi and Q. hypoleucoides. Melampsora Farlowii infects cones of eastern hemlock (Tsuga canadensis). Infected cones turn yellow and die.

Chrysomyxa pyrolae infects cones of black spruce (Picea mariana), blue spruce (P. pungens), Engelmann spruce (P. engelmannii), Norway spruce (P. abies), red spruce (P. rubens), and white spruce (P. glauca). Infected cones turn yellow and produce no seed. Alternate hosts are species of Pyrola and Moneses.

Among diseases that affect the seed crops of deciduous trees are a blight (Taphrina) and powdery mildew (Erisiphe aggregata). Both damage female catkins of alders (Alnus). The blight Taphrina pruni causes plum pockets on wild plum.

MANY INSECT pests prey on the flowers, fruits, and cones of most commercial forest trees in North America.

Insect depredations are nearly always completed by the time seed is extracted; rarely is seed damaged during storage. Dynamics of insect populations coupled with seed crop fluctuations result in considerable variation of seed losses in successive years, among stands of the same tree species, and even between neighboring trees.

The average extent of insect damage, however, is large enough to cause concern to managers of seed orchards and seed-production areas.

The seedworm Laspeyresia youngana can infest as much as 79 percent of the cones on white spruce and Sitka spruce (P. sitchensis) in Alaska.

Up to 64 percent of the seed in cones of white spruce (P. glauca) can be destroyed by the fly larva, Pegohylemyia anthracina, in Saskatchewan and Ontario. This insect appears to do more damage when the white spruce cone crop is small.

Surveys in California have shown cone moths (Dioryctria) to damage 17 to 73 percent of Douglas-fir cones.

Seed chalcids (Megastigmus) in the same State have destroyed as much as 21 percent of Douglas-fir seed.

A beetle, Conophthorus lambertiana, can destroy 25 to 75 percent of the cone crop of sugar pines over large areas in the West.

The white pine cone beetle (Conophthorus coniperda) in 10 years destroyed more than 95 percent of the white pine cone crops on the Massabesic Experimental Forest in Maine.

Cone moth larvae damage 10 to 60 percent of maturing slash and longleaf pine cones in many seed-production areas in the South. Two species of pine seedworms have infested up to 90 percent of the slash pine cones in some localities of north Florida and destroyed as much as half of the seed in damaged cones. A small sucking insect, Gnophothrips piniphilus, has killed up to 20 percent of the slash pine flower crop in Florida.

The major insect pests that affect seed of forest trees belong to four large orders—the beetles (Coleoptera), the moths (Lepidoptera), the flies (Diptera), and the wasps (Hymenoptera). Nearly always are the insects in the larval stage when they destroy seeds.

Seed chalcids, seedworms, and acorn weevils feed almost exclusively within seeds. The Conophthorus beetles, Barbara moths and Dioryctria moths feed throughout the cone. The Conophthorus beetle, attacking coniferous trees, bores an egg gallery in the cone axis about the time cones start their second year of development. Developing larvae then proceed to destroy internal scale and seed tissues. Cone moths produce more than one generation in the South. Their larvae feed on pine flowers, cones, vegetative buds, shoots, and tree trunks at different times throughout the year.

To protect seed crops from insect damage in seed orchards and seed-production areas, we must have detailed knowledge of their life histories and habits. Much of this essential information is still lacking. Results from several tests with insecticide sprays indicate, however, that chemical control can give at least partial protection to seed crops.

A measure of protection of sugar pine cones from the cone beetle has been obtained by application of 2 pounds of DDT in 2 gallons of diesel oil sprayed from a helicopter. First- and second-year cones of slash and longleaf pine in the South have been protected from cone moth larvae with a 0.5-percent water emulsion of benzene hexachloride, applied by hydraulic sprayer or mist blower.

During the period of establishment, seed orchards may need protection from insects other than those attacking seeds or cones. Grafts of southern pine, for instance, need protection from the beetle, Pityopthorus pulicarius, which bores in at the graft union and kills the scion. Monthly sprays with benzene hexachloride or DDT, applied until the graft union heals over, have controlled the insect. Pine tip and shoot moths hinder the development of young planted pine in many parts of the country. These insects can be controlled in newly established seed orchards with emulsions of DDT or benzene hexachloride.