Proceedings of the Seventh Southern Conference on Forest Tree Improvement, June 26-27, 1963, Gulfport, Mississippi

Historic, archived document

Do not assume content reflects current
scientific knowledge, policies, or practices.

Proceedings

of the
SEVENTH

Southern Conference
on

Forest Tree Improvement

Gulfport, Mississippi June 26-27, 1963

2

Proceedings of the

SEVENTH

Southern Conference on Forest Tree Improvement

June 26-27, 1963
Gulfport, Mississippi

Sponsored Publication No. 23 of the

Committee on Southern Forest Tree Improvement
Published by the Southern Forest Experiment Station,
Forest Service, U.S. Department of Agriculture, 1963.

Contents

Impact of the pulp and paper industry upon forestry
in the Southern United States Gunnar W.E. Nicholson

Site preparation, fertilization, other cultural practices
Donald D. Stevenson —___ sn Ne eee ee Soe Ne

Future management of the southern pines L. E. Chaiken ee

Hardwood silviculture in tomorrow’s southern forest J. S. McKnight
Better forest management through better adaptation T.E. Maki __

How much is forest genetics helping the forester by increasing
growth, form, and yield? John C. Barber _

Genetics in wood quality improvement R.L. McElwee ;

How can genetic control of diseases aid the forest manager? F.F. Jewell __

Breeding methods in tree improvement Franklin C. Cech __

How can we improve southern hardwoods through genetics?

James R. Wilcox _ sei sceest an tet atege gore OR oh eee 2 Sscaea he sor AEs 9 ee een

Physiology of trees as related to forest genetics
Wm. H. Davis McGregor me Pept, Ae sues

How far can seed be moved? Philip C. Wakeley ese Bt ie -

Management of pine seed production areas Donald E. Cole
Management of seed orchards Paul J. Otterbach
Juvenile-mature tree relationships Charles D. Webb dae Bue Rene are
Economic considerations of the genetic approach R.L. Marler
Literature cited (a consolidated list) Misi er ee 5

FRE LIStRANCS) ee eee ee eerere eee ee een Seer e rae SEM Ser eet na ee ere AE OEE EE

Previous publications and reports sponsored by the committee on

southern forest tree improvement _. 220

Papers are in the order of their presentation at the Confer-
ence. Titles and moderators of the five sessions were: How
intensive will forestry practices become? J.W. Johnson; How
much can genetics help the forest manager? B.J. Zobel; What
do we know about inheritance patterns—how can we get im-
provement through genetics? C.H. Driver; Mass production of
improved seed, R. E. Goddard; Value and speed of the genetics
approach, J.P. van Buijtenen.

“Tree Improvement in Modern Forest Management” was the
theme of the Conference, which was sponsored by the Committee
on Southern Forest Tree Improvement. Officers of the Com-
mittee then were: Chairman, J. W. Johnson, Woodlands Research
Dept., Union Bag-Camp Paper Corp.; Vice-Chairman, B. W.
Henry, Institute of Forest Genetics, Southern Forest Experiment
Station, U.S. Forest Service; Secretary, R.A. Bonninghausen,
Forest Management, Florida Forest Service. Officers now are:
Chairman, B.W. Henry; Vice-Chairman, R. A. Bonninghausen;
Secretary, Donald E. Cole, Continental Can Co., Inc.

Impact of the Pulp and Paper Industry Upon
Forestry in the Southern United States

Gunnar W.E. Nicholson
Tennessee River Pulp & Paper Co., New York

The pulp and paper industry in the 12 Southern
States has developed from its beginning about 40
years ago into a giant industry, producing more
than 60 percent of the total pulp production in the
United States. The present annual consumption of
pulpwood in the South is 25 million cords, with a
market value of over $500 million. Approximately
100,000 employees, men and women, are directly
engaged in the southern pulp and paper industry,
and the annual value of pulp, paper, and board
shipped from the 73 pulp and paper mills is over
$2 billion. The capital investment is over $4 billion.

The industry is an ideal combination of natural
resources together with people, progressive man-
agement, and technological and scientific know-
how. The industry has brought about a better bal-
ance between agriculture and industry with a higher
standard of living. The pulp and paper companies
own only 12 percent of the total 200 million acres
of commercial forest land located in the 12 South-
ern States, whereas 11% million private small land-
owners own about 74 percent, and the balance of
14 percent is owned by large landowners and the
State and Federal government.

During the early years, before modern forestry
management was practiced, the timber inventory
was being reduced year by year to such an extent
that the consensus of opinion was that timber even-
tually would have to be shipped in from other sec-
tions of the country. This trend has been changed
entirely, and during the last 25 years a surplus of
inventory has been built up, and at the same time
the annual amount of pulpwood produced has been
increased from year to year. This tremendous im-
provement in increase in productivity of the timber-
land has been brought about by a cooperative effort
by the forest industries together with the U.S.
Forest Service, the various State forestry depart-
ments, various forestry organizations such as the
Forest Farmers Association, the Southern Pulpwood
Conservation Association, the American Forestry
Association, the American Forest Products Indus-
tries, Inc., and several others, together with research
and educational facilities of the forestry schools
and universities.

One of the most important and effective actions
taken has been better fire protection, reducing the
annual acreage losses by fire from 5 percent or

higher to the present 1 percent or lower. These
losses will undoubtedly be reduced still further in
the years to come.

Other very important developments are: the
employment of more than 1,600 graduate foresters
by forest industries, the organization of school
forests and pilot forests, the establishment of train-
ing camps for thousands of 4H clubs and FFA mem-
bers. In general, a very effective public relations
campaign for publicizing the tremendous import-
ance of better forestry land management on the
whole economy of the Southern States has been
taking place over the years.

A much greater utilization of the whole tree has
also been brought about by the introduction of
modern log barkers in the sawmills, making it pos-
sible to convert slabs and edgings into chips shipped
to the pulpmills. This has added about 15 percent
of the total pulpwood required.

The usage of hardwood in pulp manufacture has
been continuously increasing, to the extent that 20
percent of the total pulpwood used is now hard-
wood. This will, of course, increase still further
in the years to come.

Other developments of improved forest manage-
ment practices have been selective cutting with seed
trees left for natural reseeding, a tremendous in-
crease in plantings of seedlings amounting to about
one million acres per year during the last few years,
seeding by airplane, cultivation of land for planting
or seeding, research on combating insects and tree
diseases.

All such activities have been essential in order
to bring about the great step-up and increase of the
timberland productivity which has been experienced
during the last 30 years or so. These efforts will
continue to an even greater extent in the years to
come.

I believe, however, that you scientists and forest-
ers who have been and who are creating new strains
of pedigreed trees will effect by far the greatest
progress in the increase of financial return for the
timberland owners as well as for the forest indus-
tries in the years to come.

The fundamental research which has been car-
ried on during the last 40 to 50 years in the Scandi-
navian countries and Western Europe, as well as

during the last 30 years or so in this country, very
clearly indicates the tremendous possibilities ahead
of you. The application will work in two ways:

1. The development of new strains resistant
to insects and tree diseases, as well as faster growth
in volume and weight in pounds of fiber per acre
will give the landowner a much greater financial
return.

2. The development of new strains will also
bring about a new specified type of fiber which will
allow the pulp and paper industry to produce new,
superior grades of paper products with new appli-
cations and usages. Encouraging results have al-
ready been achieved, and I believe we can predict
that long before the end of this century we will
produce, in planted seed orchards, all of the seed
of known and specified parentage that will be re-
quired for all seeding of forestry land, and the
requirements of the nurseries for growing of seed-
lings.

Progress in other phases of timberland manage-
ment will also take place, and it is not impossible
to think that most forests will eventually be planted
like any other agricultural crop, and harvested in
a mass-production way by very highly developed
machinery in a manner similar to our present meth-
od of cutting the wheat fields with harvester com-
bines.

The loading and hauling of pulpwood will be
mechanized entirely, and roads will be built for
regular tractor-trainloads. Pallet loading will be
carried out by the small landowner, leaving pallet
loads at the country roads to be picked up by haul-
ing contractors in the same manner as milk cans
are now being picked up and delivered to the
creameries,

In some instances, barking and chipping will be
carried out in the forests, and chips transported
through movable light metal or plastic pipelines to
the pulp mill.

It is quite possible that the average annual yield
of pulpwood produced per acre can be increased
to between 1 and 2 cords.

The preceding outline will, in a general way,
forecast the possibilities of improved forest manage-
ment which will bring about a very great effect
upon the financial return to the landowner. It
should also be of utmost importance from the na-
tional economy viewpoint as the increase in the
demand for wood will be very great.

We are today using 25 million cords of pulpwood
per year in the Southern States, and the forecast
is that this will be increased to between 75 and 100
million cords per year by the end of this century.
This is based upon increased usage per capita, in-
crease in population, and a very great growth in
export of pulp and'paper products to the Free
World.

The byproducts from the pulpmills, such as tall
oil, turpentine, and others, are playing an important
role today in the profitability of the industry. Such
development will undoubtedly expand very greatly
in the years to come.

There are enough indications and results from
research carried out so far over the years that
eventually a method of using the lignin in the
wood as raw material for the production of various
valuable organic chemicals will be discovered. As
the lignin amounts to nearly half of the total dry
weight of the wood, the quantities involved are
very great.

Whenever this is accomplished, the total by-
products made in the pulp mills could become of
greater value than the cellulose fiber made, and
this certainly will increase still further the demand
for additional wood, giving still greater value to
the timberland.

There is indeed a very great need for greatly
improved efficiency of forestry operation, harvest-
ing, and transportation when we realize that the
cost of pulpwood used as raw material in making
chemical pulp makes up more than half of the
total cost of the finished product.

The present low efficiency in growing, harvest-
ing, and delivering pulpwood to the pulpmill, com-
pared with the present highly developed method of
conversion of pulpwood into finished product, chem-
ical pulp in the plant, can be compared with the
driving of a Model T Ford versus the driving of
a Cadillac.

The opportunities to bring about a change in this
relationship are enormous, and this should bring
about a still brighter future for the paper industry
in the Southern States. This would also mean a
brighter future for the landowners, and greater
accomplishments to be achieved by you forest man-
agers, scientists, and forest geneticists.

The past is only a prologue to the accomplish-
ments to come.

Site Preparation, Fertilization, Other Cultural Practices

Donald D. Stevenson
Buckeye Cellulose Corp., Foley, Florida

The prediction of future developments in almost
any field of human endeavor is a chancy matter.
Such predictions are uncertain because man is en-
dowed with the intelligence not only to reach great
heights of achievement but also to make colossal
blunders. Few would have believed 40 years ago
that we could reach the type of affluent society in
which we now live or would have predicted on the
other hand a second World War, more terrible than
the one through which that generation had just
passed. Now we see the possibilities for material
and cultural progress menaced by forces which
could wipe out most of the human race.

In spite, however, of such great uncertainties
and the fears of a biologist like Dr. Ernst Mayr of
Harvard that the human race may already have
reached its peak in development of brain size and
intelligence and the trend may be in a downward
direction, I am optimistic that man can solve his
problems and will continue to produce startlingly
new methods of accomplishing constructive goals.

My optimism carries over into the field of forest
management in which here in the South we have
made notable progress. Favorable economic factors
have enabled us to introduce forest practices which
paralleled in intensity many of those developed in
agriculture. Mechanized tree farming is no longer
something to dream about but is here. In pine man-
agement we prepare sites with harrows and chop-
pers, or by chemicals. To improve drainage some
are throwing up ridges like the celery beds of south
Florida, and digging canals in deep swamps. Fertil-
izing tree crops may be a practice just around the
corner. Timber stand improvement with tractor-
mounted mist blowers or by aerial spraying with
silvicides has been carried out on extensive areas.
All this is apart from forest genetics programs from
which we anticipate sizable dividends.

The degree to which cultural practices will be
intensified is dependent on demand and supply and
a host of other economic factors of a national and
international nature. Time does not permit a de-
tailed discussion of these questions, but we cannot
forget that wood is grown by the forest industries
as a raw material for products and those products
must be sold at a profit in order to continue in
business.

The current over-abundance of available pine
timber in many areas of the South is a matter of
concern to private forest owners with wood to sell.

Nor does this situation put any pressure on the
forest industries to intensify cultural practices.

On the other hand, long-range forecasts indicate
a tightening of wood supply against demand. In
fact, the authors of the recent book, ‘‘Resources in
America’s Future’”’ (Landsberg et al. 1963), in their
interpretation of the Timber Resource Review
(U.S. Forest Service 1958) and other source ma-
terial, see some serious deficits by the year 2000.
They suggest that the median demand, as projected
in their report, is not likely to be satisfied without
serious depletion of forest stands.

The pressures on timber supply will be of two
kinds: increased demands for forest products, and
a diminishing forest land area. Timber deficits,
they believe, will have to be met by more intensive
forest management, use of substitutes, and possibly
more imports.

It is the judgment of the authors of this report
that the land resource in the United States overall
will be adequate only if yield and efficiency levels,
as projected, will be reached.

I am not prepared to evaluate critically the as-
sumptions and projections in this very comprehen-
sive study. It could be, however, that a population
of 330 million by the year 2000 will place demands
on the forest resource which will force a high
degree of intensity in forest practices joined with
careful allocation of forest land to various uses.

However we may predict the future of forestry,
intensification of forest practices by the wood-using
industries on their own land will continue to be
dependent as much on business considerations as
on economists’ projections to the year 2000. Man-
agement has to decide how much can be invested
in the growing of the tree crop in relation to rate
of return on its investment. On the more intensive-
ly managed forest lands I assume that present pro-
grams to build up growing stock to levels which
will meet anticipated requirements will be con-
tinued.

This is not to say that the industries will dis-
regard the results of research by the forest experi-
ment stations, forestry schools, and their own re-
search staffs where analysis shows that modest
investments in new cultural practices will pay off.
I simply want to emphasize the fact that the forest
industries are spending more time and effort on
cost analysis than ever before and must show a

good economic justification for additional invest-
ments in forest management.

In this connection it is instructive to take note
of a statement made by Yoho and Muench (1962)
in considering the fact that labor productivity in
the woods must be increased in the South in order
for the forest industries to remain competitive. I
quote: “the only likely way of achieving this in
forestry and logging would be through the means
by which it has been accomplished in other sectors
of our economy—namely, through increasing the
ratio of capital to labor .... In this regard, perhaps
the greatest contribution which the industrial forest
manager could make would be to cast off many of
the shackles of the classical concepts in forestry
and to concentrate upon growing a uniform product.
In this way he could increase the feasibility of sub-
stituting capital for labor all along the forest pro-
duction line.”

Will more intensive forestry cultural practices be
adopted by the thousands of small forest land-
owners and farmers who produce the bulk of the
timber in the South? One obvious answer is that
they may find it profitable to do so if they have
markets for their wood products at a profitable
price. I am afraid, however, that the answer is not
as simple as that.

W.B. Lord (1963) may have brought out the
best answer to this question. He suggests that more
important investments are open to farmers in the
farming enterprise and forestry investments must
take second place. Changes, therefore, are neces-
sary in farm organization to form larger accumula-
tions of capital before farm forestry becomes more
attractive as a business undertaking.

Realizing all the economic uncertainties in the
forestry enterprise, I shall make a few modest
predictions about forestry cultural practices in the
South with some data to back them up.

1. Site preparation of wild land for pine planta-
tions will be intensified. Subsoiling on some sites
may be practical. Bedding to improve drainage
shows real promise. An analysis by my company
of slash pine plantations growing on unbedded and
bedded flatwoods land shows that bedding has
increased present plantation value by $11.60 per
acre. This discounted present value is based on
the assumptions that 400 trees per acre will be
harvested at age 30 and that bedded plantations
will accumulate their current growth advantage
to that age.

A company owning a 30,000-acre property in
south-central Florida reports that it has obtained
per-acre yields of pine by bedding, as contrasted
with burning only and chopping, as follows:

Burning only
Chopping
Bedding
2. Fertilization of trees on sites where soil defi-
ciencies exist will be widely practiced. Better

5 cords at 18 years
5 cords at 15 years
8 cords at 12 years

knowledge of forest soils, including the part played
by mycorrhizae and other micro-organisms, and the
physiology of trees may well enable us to increase
yields economically with fertilizers even on average
and good sites.

On the basis of experiments with first-year appli-
cations of phosphorus to 1l-year-old slash pine plan-
tations on bedded sites, we have found that dis-
counted present value is increased through fertiliza-
tion by about $14 per acre. This, of course, assumes
that the present growth advantage will be main-
tained.

We have fertilized 630 acres of young pine plan-
tations this spring, applying 200 pounds of triple
superphosphate per acre by airplane. This is wet
“pitcher plant’? land near Carrabelle, Fla.

If this pilot project proves successful in terms of
increased growth, we shall assume a shortened rota-
tion advantage and probably fertilize additional
acreage. Fertilizer and application cost came to
$7.25 per acre. The job was contracted to Christo-
pher Dusting Service of Okeechobee, Fla. This firm
used Grumman Ag-Cat planes carrying 1,200
pounds per load.

3. Thinnings in natural stands will have to be
highly mechanized to be economical in the slash
pine belt where stagnation occurs at such an early
age. Mechanization of woods operations, however,
has still so far to go that we cannot predict what
economies there will influence stand treatments.

4. Plantations will be grown at wide spacings
without thinnings and clearcut. Uniformity of
growth will result in a more uniform product to
the industries.

5. Plantations will be cultivated as well as
fertilized.

6. Timber stand improvement, widely practiced
today, will be intensified as new, more effective
chemicals come on the market. Phil Briegleb, in
reporting to the SAF meeting at Atlanta last fall
on progress in technical forestry in the South, said
that research has shown that an investment of $5
per acre for release work can raise the value of
timber growth over the following decade by as
much as $50.

7. Insect and disease control will have to be
intensified. We are looking to the research organi-
zations to find the answers. It is a serious question
as to whether enough is being done now, or is being
planned in the immediate future, to meet the many
unsolved problems in entomology and pathology.
I believe that a larger percentage of the research
dollar should be going into basic and applied work
in these fields.

These are a few of the cultural means, along
with genetic improvement, that we shall probably
employ to grow more wood and a more uniform
product per acre. And I have little doubt that those
of you who are on hand by the year 2000 will have
seen other developments in forestry practice that
will dwarf anything yet attempted in the sixties.

Future Management of the Southern Pines

iE. Chaiken

School of Forestry, Duke University, Durham, North Carolina

The only rational basis for this discussion must
be within the framework of Don Stevenson’s obser-
vation that the intensity of our future forest man-
agement will depend only little upon our technical
prowess, and, indeed, only partially upon future
supply and demand, but will be affected largely by
the complexities of the world, national, and regional
economies, and especially those peculiar to each
forest operating organization.

But if we will presume that these relationships
will be favorable for additional investments in
forestry, which will in turn yield additional econo-
mic benefits, we might spend a few minutes in
speculating how we might intensify these invest-
ments, and hence our forest practices.

Several years ago I had the assignment of trying
to find out what the research needs were for the
management of the southern pines. It seemed to
me at the time that the best way to approach this
was to find out what the future management prac-
tices were likely to be, and that the best way to do
this was to hit the road and talk with forest man-
agers.

I finally cornered about 50 professional foresters
of many faiths and persuasions. They represented
pulp and paper companies, lumber companies, State
forestry commissions, National forests, schools of
forestry, as well as research foresters, both public
and industrial. They were a diverse group, but
they had one thing in common: each held a respon-
sible position, roughly equivalent to woodlands
manager (or assistant or staff forester) or was
recognized as qualified in some segment of forest
management or research.

To these, the following questions were put: (1)
What do you think the future forest practices will
be in the southern pine region? (2) In connection
with these practices, what are the problems needing
solution?

As might be expected, the answers to the first
question varied by objectives of management and
locality, and, of course, there was considerable dis-
cussion of what products were to be produced. But
everyone agreed that practices will become more
intensified, with greater investments in cultural
measures resulting in lower unit production costs.
Further, with the exceptions to be noted later, there
were strong opinions that future practices, particu-
larly on the larger ownerships, would be intensive
culture for the mass production of wood fiber of

high quality especially in woodlands in proximity
to the mills. This will be done by:

a. Complete ground preparation, not only for
type conversion but for many pine sites.

b. Artificial regeneration, by planting and
direct seeding.

ec. Such cultural measures as may be profit-
able: fertilization, cultivation, protection
from pests, etc.

d. The shortest possible rotations, consistent
with the production of the desired size,
quality, and uniformity of the fiber-pack-
age.

e. Efficient extraction and conversion into a
marketable and competitive product.

Divergent opinions were expressed by a small

group of foresters whose product objective is saw-
timber, and especially those whose timberlands are
located in areas of high site quality. In such areas,
natural regeneration of the pine types is relatively
easy, given a modest investment in the control of
competitors. Seed supply is consistently high, and
moisture for germination and survival is seldom
limiting. Natural reproduction is considered to be
more suited to the production of saw logs, poles,
and piling, where rate of wood increment is second-
ary to quality increment. Nor is the longer regen-
eration period as vital a factor as it is in short rota-
tions.

There are trends and countertrends, yet it is the
consensus that the relative production of sawtimber
will decline in the future, and that the management
of the forest will be geared to the production of
wood fiber on essentially an assembly line basis.
These notions could easily come from a sampling
heavily weighted with pulp company foresters.
But the fact remains that intensive forestry will be
practiced by foresters—and large holdings by lum-
ber companies in the region are declining.

This is no place to toll the bells or mourn the
passing of a vital segment of the forest products
industry. The production of lumber still absorbs
the largest share of the production of the forest.
And yet if we are to discuss future practices we
must inevitably do so with the view of future
products.

Perhaps this is an irresponsible and sweeping
generality, but we are told that the processing of
lumber would frequently be an unprofitable ven-
ture if it were not for the sale of chips from mill

residues. We are further informed that the current
market is more favorable to the lower grades of
lumber than the better grades. We see the increase
in lamination and the rapid rise in the production
of particle-board. One can reasonably question the
advisability of choosing to grow high-quality saw
logs, on a purposeful investment basis with rota-
tions in excess of 50 years.

But whether you agree or not as to whether the
lumber industry will decline or not, the most sig-
nificant point remains: the advanced practices of
the region will be directed toward the production
of wood fiber for conversion products.

It appears, then, that these advanced techniques,
whatever they may be, are likely to be applicable
mainly to the large industrial ownerships. The
problems peculiar to the small owner, and those
of the National forests too, will call for somewhat
different approaches. But let no one suggest that
the forester is not aware of the many problems
associated with these practices. He knows full well,
for example, that if the 30 million acres of large
ownerships are to be managed on a 30-year rotation,
1 million acres must be regenerated each year—
and if planted, almost 1 billion seedlings must be
planted each year. He is also aware that much is
still to be learned about planting and direct seeding,
and of the management of plantations. He knows,
too, that insects and disease will have a greater
impact upon his intensively cultured stands. But
he looks hopefully for answers, through organized
research and through his own mass empiricisms
and those of his colleagues.

What are the specific practices and what are some
of the main problems associated with them? Brief-
ly, they break down into:

1. Seed and seedling production.—lIf artificial
regeneration becomes common practice, it is evi-
dent that large quantities of seed and seedlings will
be required. As for quantity of either, there seems
to be no great concern, although some adjustments
will certainly be needed. With rotations being
shorter, seed production areas will become even
more necessary, with the collection and storage of
seed in good seed years to meet anticipated de-
mands.

As for quality, everyone seems confident that
the programs in tree improvement will produce
seed and seedlings with superior qualities. This,
they point out, is one of the most compelling reasons
for shifting over to artificial regeneration.

Those who propose planting programs believe
that more efficient nursery practice will provide
seedlings of better vigor and improved field survi-
val and growth. They are counting on improved
culture in the nursery and better methods of lifting,
handling, packaging, transportation, and storage of
seedlings.

2. Site preparation.—This has been thoroughly
covered by Don Stevenson, but I should mention
one matter of concern to the foresters in the north-
ern part of the region. They wonder whether the

techniques, mainly mechanical, but also fire, will
cause deterioration of the heavier soils, as in the
Piedmont and parts of the upper Coastal Plain.
Although bulldozing or any method of topsoil move-
ment is no longer advocated, they are alert to the
possibility of soil compaction through the use of
heavy equipment, and of erosion through disturb-
ance.

3. Techniques of artificial regeneration.—The
regeneration period—that most vulnerable point
in the rotation—will come more frequently. It is
here that losses are most likely to occur, and costs
to soar. We need to know how to insure successful
regeneration, with our blooded stock, and how to
do this with the least cost. It may be surprising
that so many organizations have proceeded with
large-scale programs of artificial regeneration with
so little actual experience and background in woods
planting. It is presumed that the planting tech-
niques are the same as those of the extensive old-
field plantings of the past 20 to 30 years. Yet those
who have observed the misshapen roots of planted
trees, as reported by Trousdell and others, are
wondering now if trees are currently being planted
less carefully than back in the days of the Civilian
Conservation Corps and whether it makes much
difference if the roots are ‘‘balled-up” anyway,

Since there should be little difference in root
development between plantings in old fields as
opposed to well-prepared woods sites, they feel that
if root arrangement proves significantly to affect
survival and growth, then more care can be exer-
cised in planting, and if this cannot be done effi-
ciently, then there is always the alternative of
direct seeding.

As a matter of fact, the success of direct seeding
during the past few years has given the forester
considerable confidence. Its advantages are well
known, as are the disadvantages. The forester feels
that broadcast seeding has adequately served the
purpose of quick regeneration of large areas, but
future seeding will be mainly by spots or rows.

So it appears that the forester has several alter-
native techniques for artificial regeneration. One
bothersome point, for which he has no ready an-
swer, is the invasion of volunteer seedlings into the
neatly spaced plantation; if heavy, this could be
quite annoying.

4. Growing stock levels.—The forester present-
ly has only scant information on which to base his
decisions as to spacing in his plantations. He Knows,
for example, that close spacing generally yields
more total volume of fiber for the first 15 or 20
years or so than the wider spacings—but the wider
spacings produce more usable wood more efficient-
ly extracted. And he also knows that it is cheaper
to plant 500 trees per acre than 1,000. So the trend
is toward 8 by 8, 8 by 10, 10 by 10, and wider.
But how should this vary by site quality? Or by
intensity of ground preparation? How does spacing
affect wood quality and fiber yield? Or its effect
upon competing species? What are the growth re-
sponses to thinning? How might spacing be varied

with the use of intensive cultural practices, such as
fertilization, cultivation, and pest control?

At present, and for some time to come, the for-
ester will be managing three distinct populations:
the natural stands, old-field plantations, and forest-
origin plantations. How different are these popula-
tions in their growth responses to different levels
of growing stock?

It is well to note that few foresters have precisely
similar objectives and resources for timber manage-
ment. Some are more concerned with wood quality
than others; some have limited and consolidated
holdings, while others may have scattered lands
distant from the mills. Financial structures and
objectives may differ; one may choose to produce
greater yields per acre at lower returns on invested
capital, while another may be vitally concerned
with generating the highest rate of return. So it
seems futile to search for regional ‘“‘optimum stock-
ing.” These foresters are quite capable of choosing
their own optima, given the biological responses
of any array of stocking levels and treatments.

5. Site improvements.—As one forester in Vir-
ginia put it: ‘“‘We can’t buy reasonably good forest
land any more; in fact, because of population pres-
sures and all sorts of progress, we’re being pushed
off our better lands now. We are going to have to
learn how to grow crops of trees on the dry deep
sands, the organic wetlands, and the heavily eroded
soils.’ So we see sizable programs of scrub-oak
clearing in the sandhills and ditching for water con-
trol in the wetlands. How will the pines perform?
And in between the drylands and the wetlands
there are a lot of shallow soils, sites 50 to 70, where
pine trees are growing slowly.

What can be expected in site improvement? The
addition of nutrients through fertilization, the con-
trol of moisture both excess and deficient, modifica-
tion of organic hardpans—are all possible. Another
approach would be some modification of the trees
instead of the soil: perhaps, the development of
drought-resistant and flood-resistant strains, or the
adaptation of species to site, such as the current
trials with sand pine on the dry sites.

6. Control of insects and diseases.—As recently
as 15 years ago the average forester in the region
dismissed as minor nuisances the few forest insects
and diseases he encountered. He had been aware,
of course, of the threat of the southern pine beetle
and others, but as the years went by without major
depredations, this too was dismissed. He took com-
fort in the knowledge, for example, that littleleaf
disease was restricted to stands on depleted soils.
He learned that the frequently extensive damage
caused by tip moth in young old-field plantations
became less important as the plantations became
older. He dutifully salvaged his Ips-killed trees,
or absorbed the loss, with the view that this was
one of the hazards to be expected in growing tim-
ber, especially in the dry years.

As long as the forester was busily engaged with
the many household chores needed before the truly
professional job of management could begin, and

as long as he was dealing with (and frequently
liquidating) naturally grown, second-growth tim-
ber, such things as insects and diseases were of
minor concern. More pressing jobs needed to be
done. Land acquisition, boundary surveys, fire
protection, road construction, inventories, etc., took
all available time. Following these came the job
of large-scale rehabilitation of understocked, de-
pleted, and idle lands.

But during the course of all this, the forester
began to take a closer look at his trees because he
is determined to let nothing upset his carefully
planned program of management, and certainly
not his ‘“forester-grown” timber stands. He dis-
covered Pales weevil, the cone insects and diseases,
Fomes annosus, and tip moths on the tops of 80-
foot trees where they were not supposed to be.
When he plants 600 trees to the acre he wants 600
to survive, and if any thinning needs to be done he
wants to control it.

With all his enthusiasm for the intensive produc-
tion of trees, the forester seems a bit uneasy, for
he knows that his forest will be vulnerable to the
common insects and diseases. He is aware that even
if no new pest arises to plague him, there will be
a greater impact upon his intensively cultured
stands than ever before. Even so, he is quite optim-
istic about his ability to cope with this problem. He
hopes that research will provide him with effective
control techniques, particularly with the use of
systemics. He is counting on the development of
genetic strains resistant to all pests. But most of
all he feels that he is flexible enough to change his
practices if and when the pests become dominant.

7. Methods of harvesting.—Another point of
agreement among almost all foresters in the region
is that along with our intensive culture of the
forest we must streamline our methods of extraction
of products, and this will strongly influence our
management practices. The need for more efficient
extraction techniques is evident. Decreasing supply
of woods labor, the drudgery of it and inefficiency,
increasing minimum wage standards, increasing
taxation, and the tremendous rise in paperwork
connected with it all surely indicate that improve-
ments must be made.

In speaking of pulpwood logging, one of our
foresters commented, ‘‘The tree itself, free of limbs,
forms a nice package, but we proceed to sail into it,
cutting it up into confetti at the stump—and then
struggle to re-assemble these pieces on a truck or
railroad car.”

But whether we leave the tree as confetti or
whether we go into long-length logging, it is certain
that mechanization will necessitate large-scale op-
erations. For one thing, we will have to change
some of our notions about thinnings—although I
am not so sure that we will be thinning in our
short rotations. But if we do thin our stands, spac-
ings must be such as to permit access of the mech-
anized equipment, and volumes must be high e-
nough to justify its use.

And at the same time we must accept systematic
selection of trees to be removed as thinnings: whole
rows or some such. But no longer will we spend
time making judicious professional decisions as to
which tree to cut and which to leave. Because of
the anticipated advances in tree improvement, all
our trees will be alike as peas in a pod, and the only
decision to make is how much to cut.

Trend Toward a Monotype

What will the intensively cultured forest be like?
If the forester has his way, there will be sizable
areas of pure, even-aged stands with a minimum of
competing species. No one conceives of vast areas,
for the land-use pattern in the region is such, as
are the sites, that it is unlikely that there will be
many stands in excess of a thousand acres homo-
geneous as to species and age.

The forester doesn’t seem to be impressed by
the several classic examples of failure of tree
monotypes. He points out that pure even-aged
stands of the southern pines have existed in the
region for many years, without untoward losses.
He also knows that any forest management will
increase the risk of economic loss, and that although
the unmanaged wild and mixed-species forest may
be the “safest,” it is certainly not the most produc-
tive.

He further argues that even though risks can be
scattered, and perhaps minimized, by scattering his
stands and age classes, these will be more complex
and costlier to manage. And yet, paradoxically, if
he achieves success in his search for improved
strains of trees, he may incur even greater risks,
for how vulnerable will be the monotype, especially
if founded upon an increasingly narrow genetic
base?

Just a few words in conclusion. I don’t suppose
that what I have said is really new, for many of
these practices are currently being applied. But I
would like to insert a word of caution. The stream-
lining of our forest operations is inevitable as we
strive toward greater efficiency, for to remain com-
petitive in this free enterprise system we must
increase production with decreased unit cost. And
here may be the trap; may we not be led into what
might be called forestry by the average? To stream-
line, we must smooth out the bumps and depres-
sions. Yet these bumps and depressions represent
the biological variations inherent in forestry, many
of which may be profitably ignored because they
would be too costly to recognize. To ignore others,
however, could lead to substantial losses of produc-
tion opportunity.

No one advocates the tree-by-tree forestry typical

of the Europeans. But let us beware of the sausage
grinder.

Hardwood Silviculture in Tomorrow's Southern Forest
J.8. McKnight

Southern Hardwoods Laboratory, Southern Forest Experiment Station
Forest Service, U.S. Department of Agriculture, Stoneville, Mississippi

Some 101 million acres, comprising half the
southern forest, are currently dominated by hard-
woods. Seventy million of these acres are capable
of growing high-quality timber rapidly, and are
likely to continue in forest indefinitely (Briegleb
and McKnight 1960). The Forest Service’s Timber
Resource Review (1958) estimated that the Nation’s
total wood requirements will double in the next
decade. For the past half-century, these 70 million
acres have been supplying more than half the Na-
tion’s needs for factory lumber and veneer logs,
and they will have to meet at least an equal portion
of the increased demand for these products in the
future. In addition, they will have to grow vast
amounts of pulpwood. Southern pulp mills sex-
tupled their consumption of hardwoods in the last
decade, and the end of this expansion is nowhere
in sight.

These facts make it plain that we must intensify
our forestry in good hardwood stands as well as
in our wonderful pine forests. To what extent silvi-
culture may be practiced in southern hardwoods is
the question to be dealt with in this paper.

Some recent results with cottonwood show off
the possibilities. In a plantation established by the
Missouri Conservation Commission on a good Mis-
sissippi River bottom site near Hannibal, dominant
trees averaged 6 feet per year in height growth
during the first 14 years. Diameter growth over
the same period averaged about one-half inch annu-
ally. This was achieved with little or no weed con-
trol (Wylie 1961).

By carefully cultivating to control weeds during
the first year, Chapman and Dewey Lumber Com-
pany of Memphis, Tenn., established a cottonwood
plantation on a favorable old-field site in the bot-
toms of the Coldwater River. The trees, which are
now 9 years old, have annually grown 7 feet in
height and 1 inch in diameter (McKnight 1963).

Intensive cultivation for the first 2 years re-
warded another plantation owner, near Vicksburg,
Miss., with an average annual growth of 15 feet
in height and almost one and three-fourths inches
in diameter the first 3 years, according to Virginia
McKnight (1962). Considering the values inherent
in species such as walnut and cherry and the tech-
nical qualities of oak and hickory coupled with
their reasonable growth rates, there is no reason

to doubt that comparable opportunities exist in
species other than the poplars.

Achievement of intensive silviculture will depend
a great deal on recognition of appropriate sites for
the culture of hardwoods. Good hardwoods can
be grown on sites of four physiographic classes:
bottom lands of major rivers, uplands with rich
soils, bottom lands of creeks and small streams, and
some swamps. Within these classes, the growth rate
and quality of major timber species are greatly
influenced by soils and local drainage.

In several recent reports (Broadfoot 1960, 1961,
1963; Broadfoot and Krinard 1959) the growth
capabilities of particular hardwood species are
related to the series and phase of soils on which
they occur. Growth relationships of white oak to
soils of northeast Mississippi have been reported
(McClurkin 1963). These are but examples of the
classification work that is sure to become more
exact and to form the basis for intensified silvicul-
ture. Already the soils on some important forest
areas have been fully classified and mapped by
experts. With accurate soil maps and corresponding
capability ratings for hardwood species, the silvi-
culturist can reliably choose the single or several
species best adapted to each of his sites.

The intensity of silviculture that can be justified
will be measured, partially at least, by soil produc-
tivity. On a site that will without amendment pro-
duce 400 cubic feet per acre annually, very inten-
sive stand culture is likely to be well repaid. On
a site capable of only 80 cubic feet, less intensive
measures may be justified.

Hardwood silviculture of the future will recog-
nize the demands for other use of land on which
the forests grow. Our hardwood forests are the
ideal habitat for game. Some bottom land sites
may be inundated by new dams and by waterways.
Others may always be in demand for agricultural
purposes. The intensity of silviculture must be
guided by the interrelationship of timber produc-
tion with other uses of the forest or the site. Un-
balanced multiple use may deter intensive silvi-
culture. For instance, this past spring deer browsing
severely damaged hardwoods in a plantation-spac-
ing study of the Southern Hardwoods Laboratory.
Some is both expectable and tolerable, but an over-
population of deer (one deer per 7 acres) seeking
food in the forepart of the growing season has

1 Maintained by the Southern Forest Experiment Station in cooperation with the Mississippi Agricultural Experiment Station

and the Southern Hardwood Forest Research Group.

kept new growth nipped back almost to the ground
and very likely ruined a 220-acre study.

Thus, the application of silviculture to both
natural and seeded or planted stands will be guided
by site-species relationships and take into consider-
ation the multiple uses of the forest or possible
alternative use of the land.

A Silviculture For Natural Stands

Within the next decade there will be a marked
increase in the effort to eliminate weed-species and
culls from stands of hardwoods on qualified sites.
But stand improvement is only preliminary silvi-
culture. For mixed hardwoods on good sites, some
form of partial cutting will always be necessary to
avoid premature harvesting of trees capable of
making premium products. The hardwood silvi-
culture of the future will result in a mosaic of
even-aged stands and will amount to group selec-
tion guided by the growth capability and potential
value of the individual trees within the group (Put-
nam et al. 1960). Thinnings will not only promote
good growth but will aid in quality control for
specific products. Obviously they will be made
practical by the expanding markets for pulpwood
from southern hardwood forests.

Intermediate cuts will be made with an eye to
favoring species and trees that have the least likeli-
hood of producing epicormic branches and that have
little sign of diseases or insects. Some crop trees of
potentially superlative value may be protected from
boring insects through the periodic spraying of the
tree boles with long-lasting insecticides or repel-
lents. In preliminary tests by R.C. Morris of the
Southern Hardwoods Laboratory, annual spraying
of boles in June with a 0.25 percent water emul-
sion of BHC has prevented attack by borers for
the past 4 years.

In many cases, some site treatment will be re-
quired to obtain adequate regeneration of the de-
sired species. Possibilities of such treatment are
suggested by recent pilot-scale tests along the banks
of the Mississippi River, where considerable acre-
ages of high-quality site were taken over by the
tolerant but undesirable boxelder after most of the
pioneer cottonwood was logged. In a cooperative
effort the Southern Hardwoods Laboratory and the
Anderson-Tully Company logged or deadened all
overstory trees except cottonwood and then plowed
trenches to a depth of 8 inches and a width of 7
feet—the purpose being to prepare a bare, moist
seedbed for cottonwood. So far, this site treatment
has been very successful when applied in the spring
on silty and sandy loam soils (Johnson 1962). Re-
sults on clay soils or when plowing was done in
the fall have been erratic.

How about nutrition? Will fertilizers be used
in hardwood silviculture? In natural stands near
Tallulah, La., applications of nitrogen have signifi-
cantly increased both diameter and height growth
of 20-year-old sweetgum, willow oak, and water
oak. It is too early to predict the economic signifi-
cance of fertilizers, but these early results indicate
a likely place for their use.

10

Of several soil amendments tested, supplemental
watering has proved most helpful. Broadfoot (1958)
found that where clay soils were kept covered with
water during the winter, trees grew twice as fast
the next growing season as they did on sites where
no water had been impounded. These temporary
lakes also attract waterfowl into hardwood forests.
Particularly on broad slack-water sites, this mul-
tiple use management to aid both game and timber
is being employed by a number of landowners.
To avoid killing of timber, water is trapped only
during the winter months and drained off early in
spring.

Plantation Possibilities

Plantability of hardwoods has until recently been
low, but certain species are now emerging as good
candidates for particular sites—cottonwood, yellow-
poplar, sweetgum, sycamore, green ash, and some
of the oaks. Also, the possibilities of direct seeding
are unlimited for open sites or situations where the
forest needs to be converted to more desirable
species. Discovery of appropriate techniques for
repelling rodents from acorns will advance the
reforestation of good hardwood sites. Physical
properties of the soil as well as the nutrient condi-
tion may need to be amended to assure survival
and successful growth of future hardwood planta-
tions.

Of course, site will dictate to a certain extent
what can be planted or which species should be
favored, but Charles A. Heavrin of the Anderson-
Tully Company has calculated a Species Value
Index to compare the most important species in
growth, form, and market value in west Tennessee.
The Index is in terms of the dollar value of the
annual growth per tree. Using an arbitrary 24-inch
diameter for maturity and taking into consideration
the average height and form of each species, he
divided the volume of the average tree of each
species by age to get the average annual growth per
tree.

Then he converted the average annual growth
per tree into dollars, using the Hardwood Market

Report. The resulting Index appears below:
Species Value Index
Black walnut $1.61
Yellow-poplar typ
Cottonwood .70
Willow .06
Cherrybark oak 42
Green ash .40
White ash .36
Nuttall oak 130
Sycamore 34
Sweetgum Pod
Red maple .30
Pecan 26
Cypress 45)
American elm B22,
Hackberry .20
Overcup oak 19
Hickory mill
Honeylocust 10

It can be seen that a yellow-poplar can increase
in value at the rate of $0.72 per year, while hickory
is only increasing $0.11 annually—a species differ-
ential of 6 to 1. But cottonwood, with a lower mar-
ket value than cherrybark oak, has a greater index
because of its fast growth.

Of course relative values may change over the
years. For instance, pecan has very recently become
an extremely popular furniture wood, and veneer
and lumber companies are competing for logs.
Pecan is a fast grower on suitable sites and in addi-
tion provides food for deer, squirrels, and turkey.
These considerations might well raise its position
in the desirability index. Change is inevitable,
but such an index, if carefully conceived and
coupled with site-species evaluations, can provide
an admirable guide to choice of species to plant.

Equally important will be site preparation and
cultivation appropriate to the value index assigned
and to the species to be planted. Illustrating the
effects of differences in preparation is a cottonwood
plantation on riverfront soils near Greenville, Miss.
When it was established 5 years ago, it was given
minimum site preparation, cultivation, and weed
control. Another planting was made 2 years later
on a well-prepared adjacent site where weed con-
trol was absolute and cultivation was intensive.
Now the younger stand averages 42 feet tall and
6.5 inches d.b.h., while the older is the same height
and has an average diameter of 6.7 inches.

In another place sycamore planted in trenches
made by a fireplow was killed by excessive compe-
tition, whereas in cultivated plantations it per-
formed well.

Rundown sites on alluvial soils may be rejuvena-
ted by extremely deep plowing. Trials with a pan
plow indicate that old pastures can be turned over
to a depth of 16 inches for a cost of $15 per acre.
Full benefits have yet to be determined, and so
does the value of deep ditching to aid moisture
retention next to planted trees on droughty sites.
However, site preparation and maintenance are
sure to be part of hardwood plantation manage-
ment of the future.

Hal

When the development of superior planting stock
assures us of uniformly good growth and survival,
most of the guesswork will disappear from planta-
tion spacing. We will be able to space pretty strictly
on the growth requirements of the trees and the
need for cultivation. With present-day variation
in growth and survival, the conservative course
is to space close within the rows, so as to provide
against possible heavy mortality and also to allow
some choice between survivors of differing form
and vigor.

As a hedge against high risks of planting one
species, plantations may be of two or more species.
For instance, sycamore, sweetgum, and ash are
likely to be planted with cottonwood on riverfront
sites. Nuttall oak and green ash may be planted
in alternating rows on slackwater sites. The degree
of intermingling will be determined by factors such
as shade tolerance, requirements for growing space,
and machine plantability.

Until effective and economic methods of repelling
rabbits and deer are developed, special food plants
for such game may have to be established to entice
the animals away from seedlings or seed and allow
for maximum development of the trees.

Eventually these plantations will be started with
stock developed or selected for good tree form,
fast growth, resistance to insects and disease, and
heavy yield of high-value wood products. In the
South such planting stock may first be achieved
with cottonwood, which is exceptionally suitable
for genetics and tree improvement research. Breed-
ing techniques are relatively simple with Populus,
and improvement once expressed in a single geno-
type can be maintained indefinitely on a commer-
cial basis by way of vegetative propagation.

It should be clear from the foregoing that genetic
improvement of the important plantable species
would greatly enhance the returns from money
spent on cultural techniques. It is almost a pass-
word phrase for the geneticist but it is worth em-
phasizing that we shall more and more realize that
genetic improvement, once established, results in
long-term profits without further costs.

Better Forest Management Through Better Adaptation

T. E. Maki
School of Forestry, North Carolina State College, Raleigh

Adaptation, in the forest genetics context, is a
process of evolutionary adjustments that fit indi-
vidual trees or stands to better survive, grow, and
reproduce in their environments. The idea and the
principle are scarcely new. In history and litera-
ture we find evidence that primitive forest dwellers
already recognized and understood the importance
of adaptation, of putting the right tree in the right
place. Perhaps the earliest record of adaptation
of forest trees is found in the beginning stanzas of
the Finnish epic poem Kalevala, dating back to
900 B.C. In this poem the bard recounts a project
undertaken by the hero Pellervoinen (possibly a
precursor of Paul Bunyan) who set out to perform
a direct seeding operation with a keener awareness
of species adaptation than perhaps is exhibited by
many highly trained foresters of our times. To
quote in part from Kalevala:

oc

... On the firm soil he plants acorns,
Spreads the spruce seeds on the mountains,
And the pine seeds on the hill-tops,

In the lowlands he sows birches,

On the quaking marshes alders,

And the basswood in the valleys,

In the moist earth sows the willows,
Mountain ash in virgin places, .. .
Junipers on knolls and highlands,

Thus his work did Pellervoinen.”’

In the intervening centuries since Pellervoinen’s
heroic direct seeding operation, many silviculturists
have, by trial and error, acquired a considerable
degree of sophistication and awareness concerning
adaptation and its significance in forest manage-
ment, at least as regards a few major species. For
example, in the South today, in fewer and fewer
instances are loblolly pines being recommended or
planted on deep sand sites, and possibly only a
very few million slash pines annually are being
set out much above latitude 34° N. These signs of
restraint indicate the beginning of real progress
in an area of primary importance to sound forest
management.

Adaptation in forest trees involves adjustment
to two major environmental factors, climate and
site. Climate includes such variables as range,
level, and duration of temperatures, length of
frost-free growing season, amount, intensity, distri-
bution, and character of precipitation, and hours of
sunlight. Site is construed to include such vari-

12

ables as soil moisture, depth and texture of surface
soil, total soil depth, chemical and physical charac-
teristics of surface and subsoil, topography and
hydro-geology, depth to water table, and drainage.
Progress toward fuller understanding of adapta-
tion, and especially of its significance and practical
employment in forest management, has not been
spectacular or rapid. Research in this area repre-
sents the efforts of many investigators in many
lands over a long period of time. No attempt will
be made here to review the very large number of
studies that have been published on this subject,
but it may be of interest to mention a few; also
it may be relevant to note that a rather large
number of past investigations have dealt mainly
with provenance in relation to climatic factors.

Some Lessons from Study of Provenance

Species occupying a wide geographic range or
wide variety of habitats have been the main objects
of provenance studies in the past, and, in general,
have been the ones exhibiting the most pronounced
racial diversity. The Frenchman Vilmorin is gener-
ally credited as the first to prove the heritability
of racial characteristics of pines. During the years
from 1820 to 1830 he conducted extensive experi-
ments in the Loiret Department of France, using
seeds from French, Scottish, German, and Russian
(Riga) sources of Scotch pine. Interestingly enough,
the seed source from the Riga region produced the
finest trees, which he described as having straight,
beautiful cylindrical stem and ‘‘slight boughs”
(Engler 1905). Moreover, he noted that the second
generation of the Scotch pines from the Riga source
grown from seed harvested from the Loiret plant-
ings possessed the same fine qualities as the first
generation.

In the United States one of the earliest studies
of adaptation was a ponderosa pine provenance
experiment begun in the fall of 1911 in Idaho
(Weidman 1939). In the 22 progenies under in-
vestigation, marked differences in both height and
diameter growth were observed, along with dif-
ferences in several foliage characteristics. One of
the more important findings from this study was
the observation that a source making the best
growth at first was later overtaken by a source
showing a steadily sustained growth rate and great-
er resistance to extremes of climate at the planting
location.

In the South, the pioneering study of loblolly
pine provenances established in 1926 in Bogalusa,
La., has served to dramatize the enormous influence
that geographic source of seed may exert on the
success of planting this species (Wakeley 1944).
Subsequently, striking differences among proven-
ances of loblolly pine in survival, growth, or
drought hardiness have been demonstrated by such
studies as that of Zobel and Goddard (1955), and
similar results have been reported for shortleaf
pine, for example, by Auganbaugh (1950).

In a recent study of 188 origins of Scotch pine,
Wright and Bull (1963) recognized 14 ecotypes,
basing differentiation on such characteristics as
seed size, height growth, summer foliage color,
autumnal coloration, first-year bud formation, sec-
ond-year growth initiation, second-year leaf length,
and type of root system. However, these differences
were observed only on 1- to 3-year-old seedlings
in uniform nursery beds, hence judgment of their
importance needs to be deferred until outplantings
have been observed over a sufficiently long time.

At this point I want to emphasize and punctuate
the dangers inherent in early decisions based on
short-run provenance tests. Many seed origins may
appear fully adapted to a new environment at the
start, and continue to show promise even for 2 or 3
decades, and then succumb to, for example, a single
sudden drop in temperature, particularly if the
change is unseasonal. A good example of this was
observed in some slash pine plantings in North
Carolina during the past year. A one-night freeze
when temperatures dropped to approximately 20°
F. in October 1962 killed over ninety percent of
the slash pine seedlings on an organic loam site
situated approximately at latitude 34°30’ N. and
longitude 77°30’ W. In an older planting of slash
pine on a deep sand site at about the same latitude,
the same freeze killed or damaged most of the
saplings (already 8 to 10 feet tall) in depressions
and basins throughout the plantation. These in-
stances are minor, but they point to the real possi-
bility of similar occurrences on a large enough
scale to knock management plans into “a cocked
hat.” No introduction of exotics or extension of
existing range of native species should be made
without careful study of probable adaptability.
It is not the average values of temperature, rain-
fall, frost-free season, etc., that are necessarily
important, but the extremes of high and low and
the probable duration of such extremes.

Other Ecotypes

Most of the provenance studies of the type men-
tioned above have been concerned wholly or chiefly
with discovery and evaluation of climatic ecotypes.
But other adaptations are also important. One of
the most vital considerations is that a species or
strain must be adapted to the soil as successfully
as it must be to the climate of that site. Unfortun-
ately past investigations that have taken account
of the interrelationships with quality of site and
soil are meager indeed, and the price for this lack

13

or oversight or failure has been high. As one ex-
ample near home to most of us, we may readily
recall that in the middle thirties tremendous quan-
tities of black locust seedlings were planted, hope-
fully for soil stabilization purposes, on eroding
Piedmont upland fields. Today the ragged, riddled
remnants, still persisting in places, dismally attest
to the failure of foresters to adequately assess the
edaphic requirements of this species. Even the
heroic Southwide study of pine seed sources (Wake-
ley 1952) was limited in its major emphasis to
latitude and longitude of origin, with soil-site
aspects included only in a secondary and incidental
way. Considering the size of that study it is under-
standable enough why the decision to limit vari-
ables was made, but it is unfortunate that an assess-
ment of edaphic ecotypes could not be made at the
same time on the same heroic scale. This lack in
this instance is certainly not unique. More often
than not the worth of an ecotype or strain is as-
sessed with only one environmental factor as a
criterion. This limitation is dangerous; frequently
the interaction of two or more factors may be the
decisive one, a point not to be overlooked in the
progeny testing programs that are burgeoning in
the South right now.

Perhaps the first study in North America that
made any attempt to include site quality along with
altitudinal and latitudinal influence was under-
taken in 1912 on Douglas-fir in the Pacific North-
west (Munger and Morris 1936). This study was
not designed to evaluate soil-site effects in a critical
way, and at age 17 no differences, indeed, were
observed. Other studies have, however, demon-
strated that there are such things as edaphic eco-
types, meaning that some species do maintain,
among other characteristics, a definite root form
or habit, and that such ecotypes may not adapt
readily to other environments. For example, in
Wisconsin the failure of several red pine plantings
on sandy soil was thought attributable to the origin
of planting stock raised from seed collected from
trees on rich lacustrine clays (Wilde 1954). Exten-
sive deterioration of red pine plantations in Penn-
sylvania also attests to the hazards of ignoring site
quality in assigning species to planting chances,

In longleaf pine rather marked differences in
root systems have been observed that seem to be
related to origin of seed, the more fibrous root sys-
tems being associated with moister habitats. In
jack and red pine Youngberg (1952) found that
seedlings from sand dune sites grew at a much
slower rate than those originating from parentage
on granitic outwash and other fertile soils. The
cited studies and observations, and many others,
clearly indicate the necessity of paying heed to
soil and site factors wherever and whenever the
management situation involves species assignment,
racial diversity, or ecotypic variation of any kind.

Beyond Provenance and Site

The problems of adaptation in forestry are not
restricted to reactions and susceptibilities that may

be encountered in new environments associated
with large changes in geography and climate, or
in site and soil. In a very real sense, adaptation
is also involved when it comes to taking advantage
of individual tree variation within restricted local
environments.

Fecundity and precocity are two characteristics
that are of vital concern in tree improvement, and
they are inherent at least to some degree within
otherwise apparently uniform populations. But
under a different photoperiod or a different therm-
operiod, a given source may turn out to be neither
fecund nor precocious, or vice versa.

Differences among species in the capacity to
tolerate intra-stand competition are so well known
as to be taken for granted. But this capacity ex-
tends also to strains within species, and to indi-
viduals within strains. The rigidity of control in
this characteristic has not been adequately deter-
mined but its importance in forest management can
be appreciated easily enough; it concerns spacing
directly in relation to maximum production of fiber,
volume, or both per unit area of land.

Variation also occurs among trees, at least in
loblolly pine, with respect to their capacity to
respond to fertilizer applications affecting wood
properties; that is, some trees may show a marked
growth response without a concomitant drop in
specific gravity or some other properties (Zobel
et al. 1961). Some individual trees apparently have
greater tolerance to changes in fertility levels of
the site, insofar as effects on anatomical character-
istics are concerned.

Phenological variations, relating to active periods
of shoot elongation and diameter increment, are
also observable not only among species and differ-
ent environments, but within species in the same
environment; they may, in fact, account for a very
significant portion of dry matter production differ-
ences among trees. Variation of this type can also
determine how successfully a species can become
established in a new environment, particularly one
involving higher altitudes or greater latitudes
where killing frosts may decapitate, decimate, or
destroy the early starters or the late season grow-
ers. By the same token, these variations may also
be closely associated with apparent resistance to
pests; a strain that appears resistant to an infesta-
tion, such as tip moth, on average or better sites,
may prove to be essentially nonresistant on sites
of poor quality. These variations, and similar ones,
provide a basis for selection among species and
among strains within them for improved adapta-
tion. But to capture the full potential and to make
intelligent use of such superior specimens or strains
will require rigorous assessment of their adaptation
to the environments in which they are to be grown
to specified sizes.

What About Hardwoods?

From foregoing examples and observations an
impression may have been created that adaptation

14

phenomena are confined to softwood or coniferous
forestry. Nothing could be farther from a true
picture. Hardwoods in general are even more likely
than conifers to exhibit greater sensitivity to large
changes in geography and soils, climate and site.
(For example, a freezing temperature on the night
of May 1, 1963, killed back all new shoot growth on
most of the native strains of yellow-poplars in
sapling-size plantations on the Hill Forest in Dur-
ham County, N.C., but in local sources of loblolly,
Virginia, and shortleaf pine plantings of the same
age, it appeared to cause no damage.) Local cli-
mate, indeed, is likely to have a greater bearing on
the outcome of various silvicultural practices in
hardwoods than in pines, as is suggested in the
studies by Hough (1945) and others. Because many
hardwoods regenerate from old root stocks and
stump sprouts more readily than from seed, the
opportunity for perpetuating given traits or charac-
teristics through hardwood silviculture is greater
than in pines, at least in our existing populations.
Conversely, the problem of getting rid of undesir-
able strains may prove more difficult,

Most hardwood species lack the capacity for suc-
cessful invasion of situations involving primary
succession; hence, where hardwood establishment
is attempted on open land, extra help in the form
of cultivation and fertilization may be almost in-
variably necessary to assure a_ successful start.
Even on existing forest soils on cutover lands some
form of relatively drastic site preparation may be
necessary. For example, our studies on yellow-
poplar planting have shown that this species when
planted into spots where logging slash has been
burned will attain an average total height of 18
feet in 6 years, but outside these severely burned
spots on otherwise identical soil it will attain a
height of only about 9 feet in the same length of
time.

Although knowledge about hardwood silviculture
and hardwood adaptation problems is meager, par-
ticularly for species in the South, we are in a better
position with this group of species than we were
with pines to draw on the considerable backlog of
research and experience accumulated over the cen-
turies in the field of horticulture.

What About Breeding and Hybridization?

In striving for better adaptation, we are not
limited to selections from superior phenotypes in
existing populations. Much improvement may also
be achieved from breeding and crossing for specific
purposes, as indeed has already been done and is
being done, for example, in projects of the Industry-
North Carolina State Cooperative Tree Improve-
ment Program. Through such work we may ulti-
mately achieve crosses or strains with greater photo-
synthetic efficiency, drought hardiness, or other
characteristics that make for better utilization of
existing resources of soil, water, and air. It is a
foregone conclusion, however, that no miraculous
outcome is in the offing, just solid, and perhaps
sometimes annoyingly gradual, improvement. As an

illustration, our study of loblolly pine has indicated
a high degree of correlation between certain foliage
constituents and usable volume in individual trees.
One might loosely assume that some trees are much
more efficient than others in extracting the avail-
able nitrates, phosphates, and other molecules or
elements from the soil. What is more likely is that
certain apparently rapid-growing trees have had
access to greater amounts of the constituents that
appear in greater concentration and amount in the
foliage of the larger specimens. It seems very
likely also that new strains or new hybrids selected
or developed for rapid growth will not attain their
full hereditary potential, if at all, unless given an
environment of high native fertility, or one made
high through intensive culture.

In Peroration

In a general, rambling discourse of this sort,
indulging in peroration seems pardonable, justified,
and maybe even necessary. I have attempted to
show how adaptation in forestry from a genetic
viewpoint is vital to sound forest management.
Many aspects involving other facets of interactions
and susceptibilities of genotypes to their environ-
ments have been passed over or mentioned only in
passing because others on this program will treat
them in detail. But lessons from the past stand out
clearly enough even in a general exposition to point
out where we have operated stupidly in the past
and to suggest how and where we might operate
more cleverly in the future. A few specific areas
relevant to this southern forest region bear reitera-
tion now.

We have replaced longleaf pine with slash and
loblolly on deep sands and other droughty sites.

We have introduced slash pine on loblolly sites
on much too large a scale.

We have planted loblolly on ‘“poop-out’’ spots
all over the Piedmont, spots so poor that a demand-
ing species like loblolly will die before reaching
marketable size, whereas other species without

15

cultural measures or loblolly with intensive cultural
measures would have made the grade.

We have tended largely to ignore or to malign
such species as Virginia pine and pond pine, over-
looking their splendid qualities such as tolerance,
respectively, of infertile sites or wet pocosins; their
vegetative vigor, sure-fire regeneration capacities,
and the tremendous capacity of pond pine to sur-
vive or to recover from severe damage by fire.

We have attempted planting hardwood species
on eroding old fields without even the minimum
of cultural assistance.

We have brought slash pine too far up from its
native botanical range.

We have assumed that shortleaf pine is more
drought resistant than loblolly, confusing its capaci-
ty to sprout at early ages with its persistance on dry
ridges,

We have made many species-site studies using
sources of seed or seedlings often with no know-
ledge of their edaphic or geographic home site.

Additional examples of ignoring or failing to
understand the importance of adaptation might be
cited, but let this suffice. These past mistakes will
haunt us for some time to come, but the experience
should be worthwhile, if we learn something solid
about adaptation from it. We may even need to
unlearn some things from our past mainly of the
empirical and frequently uncritical studies. As Sir
Thomas Browne said over 300 years ago: “To pur-
chase a clear and warrantable body of truth, we
must forget and part with much we already know.”

But I should not want to close on so negative a
note. We surely know enough about adaptation
already to avoid making big mistakes. The first
and perhaps the greatest benefit from application
of the genetic viewpoint in forest management
stems from rigorous use of the growing knowledge
about adaptation. The beauty of it is that we need
not wait to start reaping the benefits; they accrue
from the day we employ that knowledge in forest
renewal.

How Much Is Forest Genetics Helping the Forester by
Increasing Growth, Form, and Yield?

John C. Barber

Southeastern Forest Experiment Station
Forest Service, U.S. Department of Agriculture, Asheville, North Carolina

Today we are beginning to measure the increas-
ed yields from forest genetics, and we now have a
basis for estimating the gains expected in the future.
Until recently the principles of forest genetics had
been consciously practiced very little, but good
silvicultural practices incorporate many of the basic
principles. Thinning from below, seed tree cutting,
and shelterwood management are examples of
practices that should leave the best formed, fastest
growing trees for the regeneration of future stands.
The recent developments in forest tree improve-
ment have served to make foresters consciously
aware of the importance of considering the in-
dividual tree in addition to species and race with
the future in mind. We realize that we have to
maintain a positive selection pressure to avoid re-
trograde. We have also become much more aware
of the importance of individual traits in the ulti-
mate use of the tree and recognize the intrinsic
characteristics of the wood itself. We are learning
how important characteristics like straightness can
affect pulp quality and how small differences in
specific gravity can mean important differences in
pulp yield and in yield per acre.

Wakeley (1954) has shown that the choice of the
wrong geographic source of seed can be disastrous
in terms of growth and disease resistance. Perry
and Wang (1958) have translated these volume
differences into monetary values to demonstrate
that we can afford to make large investments in
controlling our seed to prevent the use of wrong
sources. There are numerous similar references
covering various species, but let us leave geogra-
phic origin with the assumption that good land
managers are going to use the best geographic
source available; they will be using local seed
when they do not have reliable data on which to
base their selection of geographic source of seed.
For the remainder of this paper, I will be speaking
with reference to individual trees and local stands.

We haven’t begun to realize many of the gains
from our tree improvement work, but the material
we are now planting and using from seed product-
ion areas and, on a limited scale, from seed orchards
is earning its way in added growth and quality that

will be havested in the not-too-distant future. In
fact, the Ida Cason Callaway Foundation sold for
pulpwood last winter the rogues from a seedling
seed orchard established in 1955. This is probably
exceptional, but it does show that we don’t have
to wait a lifetime to see some of the fruits of our
labors.

The available data on growth, form, and quality
of individual trees and progenies are limited to
rather young age classes. Charlie Webb will tell
you how confident we can be in projecting our
juvenile data and using them as a basis for select-
ing our best trees and progenies. With some degree
of conservatism, here is some current information
and what I think we might realize based upon it.

Growth

Squillace and Bengtson (1961) reported herita-
bilities of 8 to 16 percent for height and 29 to 58
percent for diameter in 14-year-old slash pine. I
have calculated’ heritabilities of 20 to 30 percent
for height and 6 to 34 percent for diameter in 7-
and 8-year-old slash pine progenies.

There are numerous reports in the literature of
significant differences among open and control-
pollinated progenies of our southern pines. Con-
sidering only the older of this material, we see
differences of 10 to 20 percent among means for
both diameter and height. If we use a selection
differential of the top 5 percent of a stand (2.06
standard deviations) and apply a heritability of
about 30 percent for height and diameter, we can
expect our future crop to exceed the present stand
average by about 10 percent. So you see, we can
translate our present knowledge into a rough figure
to show what we might expect in the future by
immediate control of seed source. Gain = herita-
bility »% selection differential, thus both are of
great importance. As you realize, our selection
level for seed orchard clones is much higher than
mentioned and though the heritabilities may be
slightly lower when applied to selections in natural
stands, we can expect greater increases in growth—

‘Barber, John Clark. An evaluation of the slash pine progeny tests of the Ida Cason Callaway Foundation (Pinus elliottii

Engelm.). Ph.D. Diss., Univ. Minn. 206 pp., illus. 1961.

possibly up to 12 or 15 percent in height and dia-
meter.

In the first selection cycle from natural stands,
the selection differential can be at a maximum for
all traits because we have the entire population to
choose from. In subsequent cycles of selection, we
will be selecting from progenies of only a few
hundred trees at most and will be limited in the
selection differential that we may use.

Our data on the southern pines is supported by
studies in several other species. Callaham and Hasel
(1961) have estimated heritability of height growth
from 15-year-old ponderosa progenies of about 39
percent. These progenies averaged about 26 feet in
height. Toda (1958) has reported broad-sense heri-
tabilities of 68 percent for height and 58 percent
for girth in Cryptomeria from seedling (42 years
old) and vegetatively propagated (39 years old)
trees. He concluded that where vegetative propaga-
tion could be used, selection of the top 1 percent
of the stand would show gains of 17 percent in
height and 28 percent in girth. In another study,
Toda et al. (1959) worked with 20-year-old progen-
ies of Cryptomeria averaging 26 feet in height and
found narrow-sense heritabilities of about 26 per-
cent for both height and diameter. He also worked
with some data from Europe on Scots pine and
found a narrow-sense heritability of 24 percent for
height.

Form

One very interesting aspect of form is the pro-
portion of the wood produced by a tree that is in the
stem where it can be harvested. Fielding (1953)
determined the volumes of several Monterey pines
by trunks and branches. Two of his trees were
almost identical in total volume (9.59 and 9.75 cubic
feet), but one had 48.9 percent of its total volume
in the trunk and the other 62.6 percent—a differ-
ence of about 1.5 cubic feet or 28 percent more stem
wood in one tree than the other. As we select for
smaller, shorter limbs we may also select for a
higher proportion of total wood in the stem.

Bob McElwee will give you the details of wood
quality, but I want to mention the part that some
aspects of form play in the production of compress-
ion wood. Zobel and Haught (1962) reported a study
made of loblolly pine compression wood. There are
two important aspects to their work, the effect of
stem straightness on the production of compression
wood and the compression wood associated with
knots. Their “straight” trees had about 6 percent
compression wood, the ‘‘average”’ trees had 9 per-
cent, and their “crooked” trees about 16 percent.
These differences are not only statistically signi-
ficant—they are meaningful. Mergen (1955) report-
ed on the inheritance of crook in controlled crosses
of slash pine where he found 76 percent of the
offspring from one parent were crooked. Perry
(1960) has reported similar data on loblolly pine
resulting from crosses made among crooked parents.

* Ibid.

17

He obtained 88.5 percent crooked trees when both
parents were crooked and 51.8 percent when both
trees were straight. Reversing these figures, we
see that the straight trees produced more than
four times as many straight progeny.

McWilliam and Florence (1955) discussed stem
form of slash pine in Australian plantations. Using
“routine seed’’ from the best 160 crop trees per
acre as their control, they found that open-polli-
nated progenies of selected trees contained twice
as many acceptable trees. Controlled crosses among
selected trees produced four times as many as
“routine seed.’ The best open-pollinated progenies
produced 7 to 10 times as many plus trees per acre.
The best control-pollinated progenies produced 20
to 25 times as many plus trees. These trees were
picked at a lower level of selection than we are
using in our seed orchard programs; consequently,
we can expect substantial improvement in stem
form.

Open-pollinated slash pine progenies at the Call-
away Foundation varied from 30 to 90 percent
crooked stems at 7 and 8 years of age.” These were
subjective ratings but very critical. Many of the
stem deviations considered as crook or sweep will
probably be masked by eccentric stem growth with
its associated compression wood before the trees
reach harvest age.

From these few studies we can see the opportuni-
ty for improving stem form to produce straight
logs, poles, and other products. The accompanying
reduction in compression wood will improve pulp
quality and reduce problems in drying lumber.

As mentioned earlier, Zobel and Haught (1962)
reported on the compression wood surrounding
knots. They found that each knot was surrounded
by an approximately equal volume of compression
wood. Here we have the opportunity to reduce not
only the size and amount of knots but at the same
time to reap the benefits of a parallel reduction in
compression wood.

Many foresters are inclined to think that stand
conditions are of primary importance in determin-
ing branch size and limb retention. Undoubtedly,
stand conditions are very important in controlling
crown length, but sampling of natural stands or
plantations of our southern pines will show great
variation in branch size and number in trees of the
same crown length. Our data on the inheritance of
branching characteristics is limited because most
of our progeny test are relatively young and have
not had an opportunity to exhibit these traits under
closed stand conditions. I have already mentioned
Fielding’s (1953) work on the relative volumes
of stem and branches which emphasizes the impor-
tance of branching habits.

Kiellander (1957) showed that branching and
quality in spruce could not be controlled entirely
by plantation spacing. Finely branched trees plant-
ed at 4.9 feet retained their good branching and
quality while a course source planted at 3.9 feet

remained course branched and of poor quality.
Fielding (1960) has described the number of
whorls as highly heritable in Monterey pine, and
they appear not much influenced by site.

Detailed crown examinations of seven 25-year-
old slash pine trees in a plantation in Georgia show-
ed that the average basal area of branches varied
more than 100 percent from the finest to the coars-
est." A similar range was found among their open-
pollinated progenies, but data were too limited to
establish a reliable parent-progeny regression. Dif-
ferences in crown width from 39 to 51 percent and
heritabilities of 16 to 19 percent have been reported
in slash pine (Barber 1961). Squillace and Bengt-
son (1961) reported heritabilities of crown width in
slash pine of 12 to 48 percent. Trousdell et al.
(1963) have recently published data on crown
differeneces in 7-year-old loblolly pine open-polli-
nated progenies with heritability estimates of 17
to 34 percent. Although we cannot translate these
differences into dollars or quantities harvestable
at maturity, we can expect the gain in form to
represent appreciable value not only to the manu-
facturer of primary products but also to the land-
owner in increased volume and value and to the
harvesting crew in reduced labor for limbing.

Yield

These differences in growth and form add up to
increased yield in quantity and quality. Also, im-
provement in wood quality and disease and insect
resistance will add further increases in forest pro-
ductivity.

Let’s translate some of our height and diameter
values into volume. In Queensland (1962) the best
crosses among slash pine gave 30 percent more
volume at 10 years and showed a ‘“‘substantial
superiority in stem straightness.’”’ Squillace and
Bengtson (1961) reported volumes among 14-year-
old progenies of 6.0 to 8.4 cubic feet; the fastest
growing contained about 40 percent more volume
than the slowest. From these data they estimated
heritabilities of 31 percent from control-pollinated
material and 18 to 35 percent from open-pollinated
progenies.

Peters and Goddard (1961) arrived at an estim-
ate of heritability of ‘“‘vigor’’ in slash pine. This
was the ratio of progeny superiority in height to
parental superiority in volume. Based on controlled
crosses and open-pollinated progenies, they arrived
at a heritability of 15 percent.

Fielding and Brown’s (1961) report on tree-to-
tree variations in health of Monterey pine and
response to fertilizers showed very sharp contrasts
in growth and response. They worked with both
seedling and clonal material. At 15 years they had
clones varying in height from 20 to 50 feet and
with foliage color differences; each clone was
characterized by its own vigor state and set of
systems. These sharp differences in site adaptabil-

‘Ibid.

‘Data of the Southeast. Forest Expt. Sta. on file at Macon, Ga.

ity certainly reflect important selection criteria
and point the way to greater gains where site ex-
tremes are encountered.

Working with 7- and 8-year-old slash pine data
(Barber 1961) and using an approximation of cubic-
foot-volume [cubic foot volume = (d.b.h.)° (ht.)
(0.002315)], I have found that the faster growing
progenies average about 2 times the volume (0.41
to 1.21 cubic feet; 0.87 to 1.59 cubic feet) of the
slower growing ones. If calculated on total plot
volume, the range would have been greater because
the faster growing progenies had the better surviv-
al, resulting in more trees per plot.

Toda (1958) has been very optimistic about the
values to be achieved with selections of Cryptomer-
ia. He calculated that selection of the top 5 percent
of seedling trees, when propagated vegetatively,
would increase volume by 43 percent. This was
based on increases of 8 percent in height and 15
percent in diameter.

I will mention specific gravity only briefly.
There are numerous references on variation and in-
heritance of specific gravity (Dadswell et al. 1961,
Goggans 1961, Van Buijtenen 1962, Thorbjornsen
1961, Zobel 1961). Heritabilities are quoted from
about 20 to 70 percent, depending upon material
examined and methods used. Squillace et al. (1962)
calculated values of 21 to 42 percent among open-
pollinated material and 56 percent for controlled
crosses. Let us assume a heritability of 30 percent—
this means that we can expect yield increases of
4 to 6 percent simply by confining selection to the
top 20 percent of our stands.’

Now let us add to this the increase in yield and
quality associated with the reduction in compress-
ion wood. By appropriate selection, we might im-
prove the straightness of our trees to achieve a
reduction in average compression wood of as much
as 25 percent. I am sure we would be hard-pressed
to place a value on this, but if compression wood
is important, then such a reduction should be mean-
ingful.

Oleoresin yield is another factor I’ve not pre-
viously mentioned. This is the trait about which
we have the best data arrived at from breeding.
Squillace and Dorman (1961) summarized this
work recently and reported heritabilities of 45 to
90 percent for the trait. They used an average
heritability of 55 percent to calculate estimated
gains with various methods and levels of selection.
If a selection level of 200 percent average yield
was taken, then their open-pollinated progenies
would be expected to yield a gain of 27 percent.
In a clonal seed orchard, proven 200-percent yield-
ers should give progenies with a yield 100 percent
above average. In the case of a seedling seed
orchard based on 9 Fy clones from crosses among
200 percent proven high yielders, the expected in-
crease would be 152 percent or 2% times present
yields. They also projected yields for a seed produc-
tion area using the top 10 percent from a stand

of 300 trees per acre. These seedlings should reflect
an increase of about 30 percent for this one trait.

Toda (1956) has introduced the possibility of
increasing total yields by increasing the number
of trees per acre. He calculated that a 17-percent
decrease in crown diameter would permit 50 per-
cent more trees per acre, and though they may
grow slower as individuals, the gross yield would
be higher.

Discussion

Now how can you use this information? There
are several ways, depending on the management
programs for your forest holdings and whether you
have or contemplate having an active tree improve-
ment program as such. No matter what your situa-
tion, you can put certain principles of forest gene-
tics into action.

Let’s consider the situation of the landowner who
uses natural regeneration. He can begin by paying
particular attention to the trees that will produce
the seed for his new stand. He should remember
that his new stand may be established several
years before he makes his final cut and adjust
his marking rules to insure the elimination of all
undesirable parents before regeneration becomes
established. This landowner can expect to make
appreciable improvement in only those traits that
can be readily evaluated by ocular estimate. He
can select for straightness, growth rate, form, and
possibly disease resistance. In some situations he
may be able to make a crude selection for oleoresin
yield. It is not now practical to make quick screen-
ings for the various wood quality traits. If he uses
the shelterwood method, he may be selecting at a
level equivalent to the top 5 to 10 percent of the
stand. Of course, he is limited to the particular
stand on the site, but gains can be appreciable.

The first step for those land managers who use
some form of artificial regeneration is the establish-
ment of seed production areas in the best stands
they have. This will provide seed requirements
for the immediate future and will serve on a
continuing basis or until seed orchards have been
placed into production.

The availability of stands for conversion to seed
production areas determines how effective the
program may be. Those of you who have tried to
find suitable stands with sufficient trees meeting
requirements such as those of the Georgia or South
Carolina Certification Standards know that these
stands are rare and difficult to locate. Where these
top quality stands are not available we must still
take the best we have and work with them. We
would probably all agree that our ‘‘best’’ stands are
much above those from which the majority of State
and commercial seed are usually obtained when
purchasing cones from unrestricted collectors. We
have here a real opportunity to upgrade the gene-
tic quality of our seed by converting our best stands
to seed production areas and realizing the gains
from limiting the parentage to the top 5 to 10 per-

19

cent of the stand. Easley (1963) in a first report
on growth of trees from loblolly seed production
areas, has found a height superiority of 17 percent
above controls at 5 years on a sandy site and 27
percent on a clay site.

We must realize that the amount of gain for
any single trait is going to depend upon its herita-
bility and the selection differential used. As we add
additional traits to our selection criteria, the select-
ion level at which we work for each trait must
be reduced sharply if we are to retain sufficient
trees for seed production. As an example, if we
wish to retain only trees that are within the top
10 percent of the population for one trait, we could
keep an average of 1 out of 10. When we go to the
second trait and select at the same intensity, we
could keep 1 out of 100, assuming traits are in-
dependent. The third trait drops it to 1:1,000 and
the fourth to 1:10,000. This means that where more
than two traits are involved we must reduce our
standards or have too few trees remaining for
practical use. If we lower our standards of select-
ion to maintain sufficient trees for efficient seed
production, then we sacrifice some of the gain we
might have achieved from a single trait. To many
people, the standards for individual trees on seed
production areas seem rather liberal, but when you
consider height, diameter, straightness, branching,
natural pruning, and pest resistance, you find a
reason to remove most trees.

For single traits we might expect appreciable
gains from seed production areas, such as the 30-
percent increase in oleoresin yield calculated by
Squillace and Dorman (1961) for selection of the
top 10 percent. Other individual traits, such as
height and diameter, might produce gains of 5 to
10 percent, but when several traits are considered
we must sacrifice part of the gains possible for
each because of the limited population and area
concerned.

With immediate seed needs stop-gapped by seed
production areas, and recognizing their limitations,
the next step for a land manager is the seed
orchard. He must project his needs to determine
what characteristics are important to his goals and
then draw up selection criteria to evaluate plus-tree
selections. The same rules of probability apply here
where many traits are rated, but the individual
tree may be found anywhere without regard for
frequency per unit area. Clonal establishment of
an orchard means that a broad spectrum of select-
ions may be assembled to interbreed freely—each
parent possibly representing the best among many
thousands of trees. Each parent must be tested to
insure that it will transmit the desirable traits for
which it was selected and that it does not transmit
any undesirable trait.

When these clones have been tested and the poor
ones rogued, we will be producing seed that should
eventually yield appreciable gains in several traits
simultaneously. I believe we can conservatively
think in terms of increased volumes of 10 to 15 per-
cent, gains in specific gravity of 4 to 6 percent, and

reduction in compression wood of several percent.
Add to this increased quality value for straightness,
form, and pruning, further increased yield achieved
by disease resistance and improvement of other
wood quality traits, and you can recognize the
worth of an aggressive tree improvement program.

I expect someone to raise the question of which
type of orchard to use—clonal or seedling? Both
have merits—I do not believe there is any single
answer. We have recommended both. Quickly,
I might say that seedling orchards are somewhat
cheaper to establish, but remember that parent
selection costs the same and control pollinations
on widely scattered trees are expensive and time
cansuming; at least 3 years are needed after
selection to get seedlings in the field and by then
you could have 2- or 3-year-old grafts. My obser-
vations are that clonal orchards will ‘‘flower”
earlier and more abundantly than seedlings of slash
and loblolly pines. Seedling slash pine orchards
planted in 1955 by the Callaway Foundation have
produced no ‘‘flowers.’’ Clonal orchards established
by the Georgia Forestry Commission since then
have been ‘‘flowering”’ well for several years.

Possible inbreeding (selfing) in clonal orchards
has been raised in objection, but the effects may be
low. In seedling orchards, we have the risk of mat-
ing the full-sibs and half-sibs, but less risk of
selfing. The effect of this is unknown. Where the
usual 6 to 10 traits are rated, we have the problem
of probabilities in seedling orchards. How many

20

trees per acre will we have left if we keep only
the top 10 percent for each of six or more traits—
hardly enough to recognize the area as a seed
orchard. In clonal orchards we select the parents
at whatever level we wish without particular con-
cern for the probabilities or frequency of occur-
rence per unit area or per unit of population.

Theoretically, the idea of seedling orchards is
good when considering a limited number of traits
and when juvenile-mature correlations are high.
Practically, the idea is sound under similar restric-
tions and it has a place—but when time is of es-
sence, I personally prefer to place the added invest-
ment in clonal orchards. However, until we have
seedling orchards established on at least a pilot-
plant scale, we will not be able to make a sound
comparison with the extensive clonal orchards now
beginning to produce seed.

In closing, let us look at the values Perry and
Wang (1958) placed on seed of varying yield
potentials. A meager 2-percent increase in yield
over a 25-year pulpwood rotation is equivalent to
$18.93 per pound of seed when used in the nursery
for seedling production. A 10-percent yield increase
would amount to $90.63 in value per pound of seed;
a per acre per year yield increase of $1.05. These
values are what we can afford to spend to improve
our seed. We cannot afford to lag any longer. We
should be aggressively pursuing our tree improve-
ment programs now.

Genetics in Wood Quality Improvement

R. L. McElwee
School of Forestry, North Carolina State College, Raleigh

The incorporation of genetic principles into a
forest management program can be, and has been,
justified for one or a combination of several dif-
ferent reasons. The papers already presented on
this panel have developed ideas which show that
genetic principles applied to management of forest
lands will produce tangible gains in the form of
increased growth and yields. By the application of
tree improvement principles, including use of prop-
erly adapted seed sources, additional benefits can
be gained by increasing volume and improving
form of trees. The speaker immediately following
me will show how increased benefits can be derived
through reduction of losses to diseases and insects.

In this paper I shall attempt to emphasize bene-
fits derived from a genetically oriented forest man-
agement program through improving the quality
of wood produced. Other speakers have stressed
improvement of yields of wood per acre but have
not been primarily concerned with the anatomical
characteristics of the wood or the overall benefits
that can be gained from the control of the wood
quality.

Benefits gained through improvement of wood
quality are not of a type that can be credited back
to forest management in the same manner as can
an increase in yield. Evaluation of benefits to be
gained from wood quality improvement is based on
higher value of the end product. Final evaluation
of benefits must be made by mill technologists and
chemists rather than by foresters; thus, the respon-
sibility of determining the direction and kind of
wood quality improvement must be shared both
by those who grow the trees and by those who con-
vert them into the final product.

Methods of Achieving Improvement

Tree improvement practices for use on commer-
cial forests have varied from judicious marking
rules on areas to be restocked naturally, through
seed production areas to seed orchards. For signifi-
cant improvement in wood quality, very little can
be accomplished below the level of seed orchards;
this method is the only one through which sufficient
control of the wood quality of parents and progeny
can be feasibly determined and maintained. Selec-
tive marking for natural regeneration or seed pro-
duction areas does not afford the opportunity to
find desired qualities in large enough percentages
of stems to achieve significant improvement.

21

This shortcoming can perhaps best be illustrated
by a hypothetical situation in which it is desired
to improve one wood characteristic, specific gravity,
through the use of seed production areas. For
simplicity, assume that it is desired to increase
specific gravity in the next generation of trees
obtained from the seed production area. It has
been found that specific gravity is distributed norm-
ally among the individuals of the stands from which
the seed production area has been established; thus,
about 50 percent of the trees in the stand would
have less than average specific gravity. Experience
has shown that few of the remainder will have
suitable form and growth, leaving too few trees
with both desired wood and desired form to make
an adequate seed production area. The difficulties
of achieving desired goals in improvement of wood
characteristics through seed production areas are
thus limited. This example concerns only one prop-
erty; if a second criterion is added the rigor of
selection is increased several fold, and significant
improvement of wood quality tnrough means less
intensive than a seed orchard becomes much more
difficult.

Seed orchards, the most commonly applied breed-
ing method, contain only individuals having the
desired wood properties; however, this is always
within the framework of the features of fast growth
and good form. Each additional wood characteristic
included results in a manifold increase of the diffi-
culties of locating seed orchard material. However,
the benefits to be gained outweigh these difficulties
if wood quality is of any importance to manufactur-
ing operations and quality control of the final
product.

Measurement of Improvement

Wood properties, as other characteristics, vary
in their expression among individuals. This varia-
tion is the result of the genetic makeup of the indi-
vidual and the action of the environment on the
genotype.

The genotype is defined by Johannsen (1911) as
the reaction norm and is the modification of expres-
sion set within the limits of the genotype. This
simply means that most characteristics of the indi-
vidual, including wood characteristics, are con-
trolled to some degree by the genetic makeup of the
individual. Additionally, all environmental modifi-
cations influencing expression of the characteristic

operate within the framework of this genetic make-
up.

The magnitude of the genetically determined
limits can be expressed mathematically by use of
the term heritability, which is a numerical expres-
sion falling between 0 and 1. The exact value for
any characteristic indicates the degree to which
variation is received from the parents; thus, a heri-
tability of 1.0 means that the expression of the
characteristic is determined entirely from the pa-
rents, the environment playing no part in the
expression. A heritability of 0 connotes that the
characteristic is governed entirely by the environ-
ment, inheritance having no influence.

Heritability is but one part of the ‘‘equation”’
necessary to determine genetic gain, the other being
the selection differential. Genetic gain in the final
analysis is the factor which expresses the amount
of improvement possible. Care must be exercised
in discussions involving heritabilities; some of the
points to be considered have been pointed out by
Zobel (1961) as:

1. What type of heritability is being referred
to? Broad sense heritabilities are those which in-
clude all of the genetic variances (additive, dom-
inant, and epistatic) normally present in biological
material. It is impossible to take advantage of all
three types of heritability in forest trees unless
working with clonally propagated trees such as
the hybrid poplars.

Narrow sense heritabilities are derived using only
additive genetic variances. Most wood properties
in which we are interested are inherited quantita-
tively (several genes are conditioning the inheri-
tance of a particular characteristic). Trees having
the type of wood in which we are interested demon-
strate the possibility of having the capability of
producing the wood quality sought. By allowing
many such individuals having this capability to
interbreed in the seed orchard, we enhance the
opportunity to produce trees having the desired
wood quality. The amount of improvement in wood
quality is partially determined by the narrow sense
heritability for the particular wood property.

2. To what age material does the heritability
apply? Heritabilities change as plants get older.
A characteristic may have high heritability at a
young age and be lower nearer maturity of the
tree, or the converse.

3. Under what environmental conditions were
the heritabilities estimated? Heritabilities will dif-
fer for the same plant material under different
environmental situations in which it might be
planted.

4. How was the heritability estimated? Several
methods of estimating heritability are available,
which may give considerably different estimates.

Statistical procedures for determining heritabili-
ties are complicated and beyond the scope of this
paper, but results (i.e., the heritability figures) will
be cited as indication of genetic control. In general,
the characteristics considered to be of importance

22

to wood quality improvement have heritabilities
within the range of 0.2 to 0.7.

Heritabilities are not the same for each wood
property in which we are interested, and improve-
ment possible in one character will be different
from another character using the same breeding
procedure in a given period of time. In an improve-
ment program for wood quality, it behooves the
forester to work with production technologists to
decide which characters are most important to the
finished product, as well as develop a system of
priorities allowing most rapid progress toward
achieving improvement.

Amounts of Improvement Possible

The following discussion will emphasize heri-
tability estimates in pines, principally the southern
pines; these estimates will be used as indicative
of possible genetic gain.

Specific gravity.—Much interest has centered on
improvement of wood qualities by varying specific
gravity. Such improvement has been approached
in three ways: (1) increase in average specific
gravity, (2) decrease in average specific gravity,
and (3) no change in the average out elimination
of the extremely high and extremely low gravities.
Such manipulation of specific gravity is needed to
increase quality of pulp, such as tearing strength,
bonding strength, or burst, or to provide a more
uniform raw material for production of a more
uniform end product. Table 1, based upon the

TABLE 1.—Heritabilities of specific gravity of certain species of

pines
eae sense | sense Source
herita- herita-
ab cia | bility bility
Slash pine:
14-year open-polli- Squillace et al.
nated seedlings 0.2 (1962)
14-year contro}-
pollinated seedlings i) Do.
12- to 14-year grafts 0.7 Do.
Monterey pine:
20-year grafts 5 Fielding and
Brown (1960)
6-year open-polli-
nated seedlings ‘2 Do.
8-year grafts a Dadswell and
Wardrop (1960)
Grafts (rings 7 and 8) a Do.
Loblolly pine:
l-year grafts .2-.6 van Buijtenen
(1962)
5-year grafts .6-.8 Do.
2-year control-polli-
nated seedlings .4- 5 Do.
6-year open-polli-
nated seedlings 6-1.0 Do.

Core wood—open-

pollinated seed-

lings 5 years old 8
Core wood—open-

pollinated seed-

lings 2 and 3 years

old At

Goggans (1962)

Stonecypher '

"R.W. Stonecypher, personal communication.

results of several different workers, emphasizes
the high degree of heritability of this complex
character.

The average for all of the above heritabilities
is around 0.7 for broad sense and 0.5 for narrow
sense. The heritabilities are for young trees, and
indications are that they may increase in older
material so that narrow sense heritabilities are ap-
proaching the broad sense values. Assuming a
differential in paper yield of 25 pounds per cord
for each increase of 0.01 in specific gravity, 90
pounds green weight (Taras 1956), selection for
higher specific gravity can increase paper yields
from 25 to 80 pounds per cord of wood cut, the
amount of the increase depending upon the selec-
tion differential employed in the selection. Assum-
ing the moderate increase of 40 pounds of paper
per cord for all the wood used, a 400-ton-per-day
paper mill would realize an annual increase of
4,200 tons of paper using no greater volume of
wood.

Increase in specific gravity is attained principally
by an increase in the percent summerwood growth
of the annual ring and an increase in the cell wall
thickness of the individual tracheids. Such prop-
erties are not consistent with an increase in certain
qualities; for example, increasing specific gravity
will increase tearing strength but reduce bursting,
tensile, and folding strength (Watson and Dadswell
1962).

It is unfortunate that certain factors of cell and
tree anatomy which contribute to higher weight
yields may also contribute to a reduction in quality
of the paper produced. It is in this realm of deter-
mining where the balance will be made between
gross fiber yield per cord and quality of product
that mill technologists must aid in decisions in-
fluencing tree improvement programs.

Tracheid length.—A second important wood prop-
erty for products which is under strong genetic
control is that of tracheid of fiber length. This
characteristic is important in the areas of strength
properties of both pulp and paper.

Many authors have reported tracheid length of
progenies to be intermediate between those of the
parents; these include Chowdhury (1931), Jackson
and Greene (1958), and Echols (1955). However,
there are few reports of actual heritabilities for
this characteristic. Dadswell et al. (1961) found
gross heritabilities for tracheid length to be about
0.75 in Pinus radiata and Goggans (1962) obtained
an even greater value in the narrow sense for
Pinus taeda. High heritability for tracheids will
produce rapid progress from selection resulting
in an increase of the mean fiber length. By includ-
ing only long-fibered parents in the seed orchards,
it is possible to obtain an increase up to one-half
millimeter in the first generation of selection.

There is little reason to believe that short fibers
per se will ever be desirable, since bonding strength
is an asset to any product. It is possible to produce
short tracheids in the mill—thus, it would appear
that in the future longer fibers and tracheids will

23

be sought. For fiber length, as well as for specific
gravity, the objective is to improve the average for
the characteristics. Inherent within tree variations
in wood properties and between-tree variation will
always dictate variability in the wood coming into
the mill.

Cell wall thickness.—This characteristic of fibers
and tracheids affects specific gravity as well as
being important in influencing bonding properties.
Thin-walled cells collapse, become ribbon-like, and
provide a strong bond with adjacent fibers. Thick-
walled fibers, on the other hand, tend to retain
a round shape, do not collapse, and provide a poorer
bonding surface, reducing the bonding strengths of
pulp.

Inheritance of wall thickness was found by
Goggans (1962) to be 0.84 in summerwood and
only 0.13 in springwood. Improvement in wall
thickness will go hand-in-hand with improvement
in specific gravity.

Percent summerwood.—Percent summerwood,
like cell wall thickness, influences pulp properties
much as does specific gravity. According to Dads-
well and Wardrop (1959), tearing strength and
bulk density increase with higher percentage late
wood; but bursting strength, tensile strength, and
fold endurance decrease with increase in late wood
percent.

Narrow sense heritability of summerwood in the
entire core in young stems was found to be around
0.8 by Goggans (1962). This value is somewhat
higher than the broad sense value of 0.5 found by
Dadswell et al. (1961). Progress toward improve-
ment of percent summerwood is tied very closely
with progress in specific gravity, just as is cell wall
thickness.

Percent reaction wood.—Reaction wood such as
compression wood in conifers reduces quality in
all types of products. Poor yields and strength
properties result in paper produced from such wood
while lumber cut through zones of compression
wood is subject to a high degree of shrinkage, warp,
and crook.

The type of material produced in reaction wood
is inferior in both softwoods and hardwoods. Qual-
ity of compression wood is such that low yielding,
short-fibered wood, unsatisfactory for high-grade
products, is produced.

Reaction wood is produced whenever a tree stem
grows out of the vertical plane. Auxin balances are
upset, the result being the ‘‘abnormal” type wood.
Extent of such wood, therefore, is closely tied to
straightness of stem. Amount of compression wood
in 50-year-old loblolly pine has been found by
Zobel and Haught (1962) to vary from 6 percent
in essentially straight trees to 16 percent in a
crooked tree, with one exceptionally crooked tree
having 67 percent of its bole volume in compres-
sion wood.

The actual heritabilities of straightness are un-
known; however, Perry (1960) believes straight-
ness to be under the control of several genes. Evi-
dence that straightness is strongly inherited has

also been shown by Mergen (1955) and McWilliam
and Florence (1955). Even though straightness of
stem and consequently percentage of reaction wood
is subject to many environmental influences, it is
apparent that heriditary influences are also im-
portant, offering the opportunity to reduce amount
of this inferior material through producing more
straight trees. It is because of the phenomena of
compression wood production that straightness of
stem is given so much weight in selecting material
to be used in tree improvement programs.

Other characters.—Several other wood character-
istics have been studied by Goggans (1962) to
determine heritabilities in loblolly pine. Several
of the characteristics Goggans worked with are
of practical importance to wood users; economic
importance of others is not recognized at present,
principally because their influence on end products
is not known. Table 2 is a modified ranking, taken
from Goggans’ paper, listing the ease with which
improvement can be made in wood characteristics
in an improvement program.

Table 2.—Relative ranking of wood characteristics
according to the ease with which pro-
gress may be made in a selection pro-
gram.

Numerical rank Characteristic

Summerwood tracheid length
Percent summerwood

Specific gravity

Springwood tracheid width
Springwood tracheid length
Summerwood tracheid width
Summerwood tracheid wall thickness
Springwood tracheid wall thickness

DAKDUPWNHH

Several of the characteristics listed above are
interrelated, and any work toward changing one

24

will result in change of another, i.e., increasing or
decreasing the percent summerwood will have a
similar effect on specific gravity.

Summary

Most tree improvement programs, in addition to
bettering form, growth rate, insect and disease
resistance, etc., have as one of their major objectives
the improvement of wood quality. These wood
quality objectives are directed toward the produc-
tion of trees containing types of wood most benefi-
cial to maintenance of yield and quality of the
final product.

That such objectives are possible is shown by the
results of several workers on the variation, inheri-
tance, and heritabilities of wood properties. Studies
on heritabilities are not numerous but enough have
been reported to indicate that progress toward
improvement of wood quality is possible. Many
other studies are now under way and more con-
crete evidence of the amount of improvement that
can be achieved will be available soon.

Information to date indicates that in seed or-
chards of loblolly pine it is possible to produce
“strains” having one or more of the following char-
acteristics, except where two characteristics are
diametrically opposed:

Increased pulp yields of 40 or more pounds per
cord.

Increased average fiber or tracheid length up to
0.5 mm.

Improved tearing strength.
Improved bonding strength.
Increased bulk density.
Improved bursting strength.
Improved folding ability.
Improved tensile strength.

How Can Genetic Control of Diseases Aid the Forest Manager?
FF. Jewell

Institute of Forest Genetics, Southern Forest Experiment Station
Forest Service, U.S. Department of Agriculture, Gulfport, Mississippi

Forest diseases caused mortality and growth losses
during 1952 of more than 5 billion cubic feet of
growing stock and almost 20 million board feet of
sawtimber in the continental United States, accord-
ing to the Forest Service’s Timber Resource Re-
view (1958). Diseases were responsible for 45 per-
cent of the losses from all causes, including fire
and insects. In the South, fusiform rust alone ac-
counted for 97 million cubic feet of growing stock
and 281 million board feet of sawtimber. The situ-
ation is probably not any better today. Obviously,
we must improve our control of forest diseases if
we are to obtain maximum production on our forest
lands.

The control of tree diseases in the forest by chem-
ical or cultural means has been historically diffi-
cult, usually being temporary, expensive, and gen-
erally unsatisfactory. The application of chemicals,
even antibiotics (Lemin et al.), gives at best short-
term and expensive protection. The one cultural
method in general use in southern forests, i.e., burn-
ing for the control of the brown spot needle disease
of longleaf pine, is a drastic treatment that often
has questionable results. An important avenue of
attack on the overall control problem is through
genetics and tree breeding. Develop a resistant tree
and you will have built-in control with no further
manipulation required.

With agricultural crops, the geneticists and plant
breeders have been able, through selection, progeny
testing, and use of plant hybrids, to produce dis-
ease-resistant varieties that have been a major
force in revolutionizing agricultural production in
this country. For example, of 90 new crop varieties
released by the U.S. Department of Agriculture
and the State experiment stations in 1959, more
than half were developed with specific disease
resistance in mind (U.S. Agr. Res. Serv. 1960).
Some of the findings in disease-resistance research
have virtually saved valuable crops from becoming
lost to commercial production.

So the principles of disease control through re-
sistance have been proven and are available for
application to forest trees. We have begun that
application at the Institute of Forest Genetics in
our attack on the fusiform rust of slash and loblolly
pines. Our first efforts were aimed at determining
if resistance could be incorporated into the sus-
ceptible slash pine by crossing it with the naturally
resistant shortleaf pine. Selection, breeding, and

25

progeny tests under conditions of both artificial
and natural infection have shown that this hybrid
does indeed carry a considerable amount of resist-
ance to fusiform rust (Jewell 1961; Jewell and
Henry 1961).

Our next efforts were to find whether or not
resistance to this rust exists naturally in individual
trees of the susceptible species, slash pine. Open-
and control-pollinated progenies of selected rust-
free parents were artificially inoculated with the
rust along with progenies from check parents. The
open-pollinated progenies of certain of the selected
parents exhibited significantly fewer galled indi-
viduals than the progenies from check parents.
When the selected rust-free parents were crossed
with one another, the progenies showed still greater
resistance (Jewell 1961). So resistance does exist
in individuals within the susceptible slash species.

Therefore, there are two sources of resistance to
fusiform rust. Resistance can be bred into slash
pine by crossing it with shortleaf, or resistant
strains of slash itself can be developed by crossing
individual trees whose progenies have been shown
to be resistant. Both these methods appear promis-
ing.

The concept of individual-tree resistance to fusi-
form rust is actually already in practice in the
South, thanks to the foresight of many early
workers in tree improvement programs. By their
insistence that slash and loblolly pine selections be
free of fusiform rust, an appreciable amount of re-
sistance is apparently already incorporated into the
clonal seed orchards (N.C. State Col. School For-
estry 1963). Future progeny tests for rust reaction
and subsequent roguing should result in a still
higher percentage of resistant material in these
orchards.

The discussion so far has dealt with resistance to
only one forest disease. However, we have evidence
that the same principles of control by resistance
can be applied to others as well. The crossing of
western white pine trees selected for resistance to
blister rust yields a high percentage of resistant
progenies (Bingham 1963). The first-generation,
Fj, progenies are put into seed orchards and the
next generation, Fy , will be used as planting stock.

Another disease that possibly will be susceptible
to genetic control is the brown spot of longleaf pine.
Under field conditions the open-pollinated progeny

from a selected longleaf has consistently shown less
infection than control progenies (Derr 1963). Pro-
geny from a cross of this selected parent and non-
resistant longleaf were far less susceptible than the
open-pollinated progenies from the nonresistant
parents. He concludes that resistance to brown spot
is genetically controlled and that there are distinct
possibilities for developing and producing resistant
longleaf pines.

The three examples just mentioned illustrate the
prospects of controlling forest tree diseases through
resistance. The prospects appear good: not only
does it seem likely that research will be able to
find and produce resistant trees, but control of
diseases may well be among the earliest practical
results of genetics programs.

26

Now for the question that forms the title of
this paper, “‘How can genetic control of diseases
aid the forest manager?” In essence, it can elimin-
ate one of his most plaguing problems—nhaving to
plan a management program in the face of disease
losses that must be expected, but in unknown quan-
tity. It can enable’ him to establish the species he
wants on the site he wants without regard to dis-
ease hazard; it can free him from having to estab-
lish and maintain heavy stocking to compensate for
disease losses. With genetic control he can have
the thinning regime he wants, rather than one dic-
tated to him by the necessity of removing trees
made infirm by disease. And he can carry his stand
on to maturity without fear of disease loss, because
resistance lasts for the life of the tree.

Breeding Methods in Tree Improvement

Franklin C. Cech

Southlands Experiment Forest, International Paper Company, Bainbridge, Georgia

The two previous papers have presented a very
thorough account of current information on heri-
tabilities for wood and morphological characteris-
tics. A primary purpose for determining herita-
bility figures is to better orient breeding methods.
How, then, can this information be used in the
current tree improvement programs and how will it
affect silviculture in the long run?

Ten years ago the descriptive words ‘“‘phenotype”’
and “genotype,” commonly used in genetics, had
very little meaning to management foresters. These
terms are relatively common today and are em-
ployed freely in discussions of tree improvement
programs. They exemplify the new set of terms
used in advancing techniques in genetics and breed-
ing procedures. It will be necessary to define and
develop some other new terms as they are used
throughout this paper.

Heritability, a term so freely used in previous
papers, is defined in the tree improvement glossary
(Snyder 1959) as a ‘‘measure of the relative degree
to which a character (or characteristic) is influ-
enced by heredity as compared to environment.”
Variation among trees can be expressed by formula
as follows:

Total Variation = Hereditary Variation +
Environmental Variation

A properly designed heritability experiment can
separate variation due to heredity from the varia-
tion due to environment. Heritability is usually
expressed as a decimal or percent figure. For
example, if a broad sense heritability of 0.7 or 70
percent is found for a characteristic such as specific
gravity, it means that 70 percent of the total varia-
tion in specific gravity is due to heredity and 30
percent is due to environmental influences.

What elements are included in the heritability
portion of the estimate and what do they mean?
The term ‘broad sense heritability,’ or the total
hereditary variation, refers to a numerical estimate
including three values: namely, additive variance,
dominance variance, and variation due to epistasis.
Additive variance is due to the average or additive
effect of a gene or genes. Each gene contributes a
small addition to the overall effect. Dominance
variance is due to the interaction of alleles. Epista-
tic variance is due to the interaction between non-
alleles. The term ‘‘narrow sense heritability” refers
to ratio of additive to phenotypic variation.

Broad sense heritability is determined by divid-
ing total genetic variance (Vg) by phenotypic var-

27

jiance (Vp) and is expressed mathematically by

> V
the equation h~ = <=
broad sense Vp

heritability is obtained by dividing additive vari-
ance (Vy, ) by phenotypic variance (Vp ) and can be
Via

narrow sense Vp

. Narrow sense

expressed by the formula h?

If the narrow sense heritability figure is high,
then the mass selection method of breeding will be
most productive. On the other hand, if the dom-
inance or epistatic variation is very high, the great-
est gain can be accomplished by an intraspecific
hybridization approach.

Figures on inheritance, represented by heritabil-
ity values, are essential to the development of any
tree breeding program. If, for example, specific
gravity were controlled to a considerable extent by
environment, the field forester would be the one
who could most easily increase or decrease the
specific gravity of stands by silvicultural manipula-
tions. The tree breeder would be able to add very
little by genetic or breeding techniques. If, on the
other hand, as Mr. McElwee has shown, specific
gravity is controlled to a considerable extent by
heredity, the silviculturist can do less to affect it,
and management techniques must be quite drastic
to bring about major changes. However, the tree
breeder can accomplish a great deal.

Initiation of a Tree Improvement Program

Assuming that a tree improvement program is
about to be initiated, the ideal procedure is to make
a thorough study of the species to be improved.
Detailed information concerning variation between
and within individuals must be gathered and a
complete study made of this species, with an assess-
ment of causes of variation over the entire range.
Heritability of important characteristics and the
mode of inheritance must be determined. Con-
trolled crosses with related species should be made
and their progenies studied. Armed with this infor-
mation the tree breeder could then devise a breed-
ing scheme and proceed with improving the species.

Since one rarely approaches ideal conditions,
chances are that a tree breeder will find himself
with the situation that confronted forest geneticists
approximately 10 years ago, when the first southern
pine tree improvement programs were initiated.
Little of the necessary information was available
as a base for an efficient breeding design. There

was not time to make the required surveys and
conduct experiments that would provide knowledge
needed to choose among the several immediate
avenues of approach available. These had already
been explored, developed, and exploited in various
crop improvement activities with other plant spe-
cies. The problem in forestry in the Southeast was
to determine which breeding method would result
in the greatest improvement in the shortest time
so as to provide improved seed for the very large
and expanding planting program then underway.

Methods to Produce “New Creations”

Polyploidy.—The ‘‘go for broke” approach was
characterized by polyploidy adherents. In this
breeding system, the fundamental structure of the
cell is manipulated to change its genetic constitu-
tion. Each cell in every species has two sets of a
fixed number of heredity units or chromosomes.
Pine species have 12; aspen, 19; human beings, 23.
Normally this number is characteristic to the spe-
cies, but occasionally something occurs that upsets
the normal condition and then the resulting pro-
geny have an abnormal number of chromosomes.
There are certain plants which are improved by
such a change in chromosome number; some per-
sons have suggested that this might be true for
forest trees.

Probably the most intensive search for polyploid
trees has been made in the aspen improvement pro-
gram, where individuals were located that were
extremely vigorous and had exceptionally large
leaves. A microscopic examination of cells from
these selections demonstrated that they had 3 sets of
19 chromosomes instead of the normal 2 sets. Since
these trees were so very vigorous, it was hoped
that such triploids might be especially desirable.

Geneticists artificially recreated triploid indi-
viduals with colchicine, a chemical that interrupts
the natural processes of cell division (Inst. Paper
Chem. 1955). When a seed is placed in a solution
of colchicine at the time of germination, a few indi-
viduals with twice the normal number of chromo-
somes develop. An aspen seed with 2 sets of 19
chromosomes can be forced to develop twice this
number. This individual is partly fertile and, when
crossed to a normal aspen, develops seed which
grow into very vigorous seedlings with 3 sets of 19
chromosomes. These triploids are normally sterile
or have very low fertility so could not be repro-
duced by seed. However, it might be possible to
produce triploids in sufficient number by establish-
ing seed orchards of alternating rows of normal
diploid and tetraploid individuals. If so, triploid
individuals could be used in practical forest man-
agement if desired.

Although examples of polyploidy in pines have
been reported, (Hyun 1954; Mergen 1958, 1959)
in every case the seedlings are malformed and grow
poorly, with undesirable form; thus polyploid seed-
lings in the pines are of little value. Polyploidy as
a breeding method in pines must be relegated to
the laboratory, at least for the foreseeable future.

28

Induced chromosomal changes.—Another means
for producing artificial genetic variation is to sub-
ject seed or plants to X-ray (Snyder et al. 1961)
or other radiation (Beers 1962). This causes actual
physical disturbances in one or several chromo-
somes. These can be large changes involving whole
sections of chromosomes or minute “‘gene changes”
which are reflected in changes in the developing
seedlings; such seedlings with artificially induced
changes or mutations have up to now been very
difficult to keep alive, and this technique can also
be considered in the laboratory stage as a practical
breeding tool.

Interspecific hybridization.—Crossing two species
of pine is another method that sometimes produces
seedlings with spectacular characteristics. The re-
sulting progeny are compared to each parent to
determine what improvement, if any, has been
gained. This phase of forest genetics has been
extensively pursued at the Institute of Forest Ge-
neties at Placerville, Calif. Control crossing tech-
niques were developed here in the early 1930’s,
and the major effort of this station has been di-
rected into the methodology and usefulness of inter-
specific hybridization. Such crosses between species
have produced some remarkable hybrids (Calla-
ham 1957), but for one reason or another very few
of them have been of practical use. Two of the
hybrids, the Jeffrey x Coulter and the knobcone
x Monterey, have been quite successful. Jeffrey
x Coulter hybrids are now being planted in Cali-
fornia forest plantations and the Jeffrey Coulter
hybrid backcrossed to Jeffrey is also being planted
in the Jeffrey pine range where it is resistant to
the pine reproduction weevil (Righter 1960).

The most impressive practical use of the hybridi-
zation technique has been developed in Korea
(Hyun 1961) where in 1 year over 1 million con-
trol-pollinated seedlings of a pitch pine x loblolly
pine hybrid were produced. This hybrid has much
of the growth habit of loblolly pine and some of
the frost resistance of the pitch pine parent. The
labor cost of such hand-produced hybrids would be
prohibitive in this country, but a few have been
produced to be used on an experimental basis.
Hybridization, as a completely practical method,
cannot be used until some inexpensive system of
control pollination is developed.

Some research has been done to develop practical
means of mass producing hybrids. Wakeley and
Campbell (personal communication) have tested
a method of applying slash pine pollen to unbagged
longleaf strobili, but with indifferent success. Brown
and Greene (1961) and Hyun (1961) are working
with chemicals that will cause male sterility.

If Wakeley’s or some other simple method can
be developed so that successful hybrids are pro-
duced consistently it would be relatively simple to
make a mass collection of pollen and dust a large
number of trees in the seed orchard inexpensively.
On the other hand, if male sterility can be induced,
orchards could be so designed that the species to
be crossed would be planted in alternating rows.

The pollen of the seed parents would be rendered
sterile so that all of the seed would be hybrid.
These techniques are still in the experimental stages
but may hold some promise.

Improving Existing Species

Intraspecific hybridization.—The technique of
making controlled crosses between members of the
same species may be used to combine desirable
characteristics from different individuals. It always
must be remembered that some selection is implied
regardless of the breeding scheme used. However,
the difficulty of making selections varies consider-
ably with the type of breeding method in use, and
the size of the population that must be examined.
As the size of the population available for selection
and the number of characteristics being considered
increases, selection difficulties are compounded.

The problem of selecting usable breeding stock
is minimized with the intraspecific hybridization
method of plant improvement. Usually in this sys-
tem individuals with one particular outstanding
trait are selected. The original selections are crossed
in an attempt to combine the outstanding features
of each into one hybrid. The system is time-con-
suming, though, especially with species that have
considerable time lapse between seed germination
and the development of reproductive organs. After
the original crosses are made, progeny must be
grown and new selections made which contain the
desirable characteristics. These are then cross bred
to fix the characteristics in a high proportion of
the progeny and to establish a seed source. In order
to prevent inbreeding depression, several selections
must be made and carried on concurrently. The seed
source is eventually developed from these selections.

Mass selection.—This is a system of breeding for
improvement that promises slow steady gains.
Normally, the increase one can expect is limited by
the genetic capability of the most outstanding indi-
vidual in the population. With a large amount of
natural variation of the additive type, large in-
creases can be expected; conversely, with little
natural variation, there can be little improvement
by this method.

As the number of characteristics to be improved
increases, selection difficulty increases. In order
to make the greatest possible gain a high intensity
of selection must be practiced—and suitable indi-
viduals are difficult to locate. On the other hand,
once the selections are made, seed production prob-
lems are simplified as a continuous supply of large
quantities of seed can be produced from vegeta-
tively propagated ramets of the original selections.

Practical Tree Improvement

With these methods available, the pioneer south-
ern pine tree breeders took a long, hard look at
their species. Little or no information was avail-
able from previous data to guide them. A few pre-

liminary surveys indicated that variation was pres-
ent and seed source important. Interspecific hybrid-
ization had not been particularly promising. The
problem at hand was to develop a genetically im-
proved source of seed as soon as possible. The
decision was: mass selection.

The data now available testify to the validity of
this choice for most characteristics. It must be re-
membered that many of the experiments from
which these data have been drawn were designed
for other purposes and that much is from relatively
young material, but the significance cannot be
denied. Heritability values made from measure-
ments of immature material can be expected to
increase as the plantings come closer to maturity,
at least for some species. Let us now, as the poli-
ticians say, examine the record.

Barber and McElwee have quoted heritabilities
for many characteristics and indicated the amount
of improvement that can be expected based on these
figures. They have urged the immediate pursuit of
an aggressive tree improvement program based on
the mass selection method of breeding, or some
variation thereof.

A brief review of some of the heritability figures
quoted will emphasize the soundness of the recom-
mendation.

Some of the oldest plantations which can be used
for estimating heritabilities were established by
personnel of the Southeastern Forest Experiment
Station at Lake City, Fla. Planting consisting of
grafted clones, open pollinated progeny, and con-
trol pollinated progenies are available for estima-
ting the genetic improvement possible. Squillace
and Bengtson (1961) have reported heritability
figures for several characteristics from these 10- to
14-year-old plantations. Narrow sense heritabilities
of 56 percent for specific gravity were obtained with
control pollinated progeny while the broad sense
or total heritability was estimated to be 73 percent
from clonal data. Fairly high heritabilities for
diameter growth (25 to 58 percent), stem volume
(18 to 35 percent), crown width (24 to 48 percent),
and bark thicknéss (33 to 67 percent) were ob-
tained. A fairly low heritability for height growth
of 5 to 10 percent was obtained. A tentative hypo-
thesis advanced by Squillace indicates that this is
probably due to the wide spacing (20 by 20 feet)
of the plantation and the attendant lack of compe-
tition. This is somewhat substantiated by Barber’s '
figure of 27 to 37 percent for material planted at
a spacing of 10 by 10 feet, van Buijtenen’s figure
of 20 percent for material planted at an 8 by 8
foot spacing (personal communication), and by re-
cent results obtained by Stonecypher (personal
communication) from seedlings planted also at an
8 by 8 foot spacing. It is interesting to note that
the broad sense heritability of oleoresin yield is
estimated at 90 percent while the narrow sense
estimates vary from 45 to 90 percent depending on
the methods used.

‘Barber, John Clark. An evaluation of the slash pine progeny tests of the Ida Cason Callaway Foundation (Pinus elliottii

Engelm.). Ph.D. Diss., Univ. Minn. 206 pp., illus. 1961.

van Buijtenen (1962) estimated broad sense
heritability of 64 percent and 84 percent from
5-year-old grafts for specific gravity, and narrow
sense values of 37 and 49 percent for 2-year-old
control pollinated material. He also reported dia-
meter heritability of 20 percent for 6-year-old lob-
lolly pine progenies. He estimated that an improve-
ment of 10 percent for each of these characteristics
may be expective from one cycle of selection and
that an increase of 25 percent in total wood produc-
tion for the several factors combined would not
be out of reason (personal communication).

Stonecypher (personal communication) found a
low narrow sense heritability for first year height
growth, which increased with second year measure-
ments. He noted that narrow sense specific gravity
heritability estimates for 1- and 2-year-old pro-
genies approach the actual values given by van
Buijtenen and that they remain constant for the
2 years.

Selection severity can be varied from slight to
intense, with selection of low to medium intensity.
Forestry practices can be guided so that any land-
owner can achieve a moderate amount of genetic
improvement in his forest stands. Some gain will
be achieved, if, in a seed tree cutting, selection of
seed trees from which cones will be collected is
made as the first step in harvesting. Improvement
here is controlled by the nature and number of
trees in the stand being cut. There is no doubt that
a minimum amount of improvement can be ex-
pected when the selections represent. only 10 to
20 percent of the stand, which is normally the case.
As soon as the area is sufficiently stocked, the seed
trees are harvested during a year of heavy cone
production. Such areas are designated as seed
collection areas. Careful planning is necessary as
harvesting of the seed trees must be scheduled to
coincide with the optimum period for cone collec-
tion. Cost of cone collection from trees which have
been cut is relatively inexpensive, especially when
large crops are present.

The next highest level of selection severity is
represented by the seed production area, where
better than average stands are selected, carefully
rogued ot undesirable individuals, and managed
for a continuing supply of genetically improved
seed. Since mature trees must be climbed for cone
collection it would seem at first that the expense
would be prohibitive, but collection costs as low
as $3 to $5 per bushel have been attained as com-
pared to an average cost of $2.50 per bushel when
cones are purchased on the open market (Goddard
1958; Cole 1962). According to Easley (1963),
seedlings from one seed production area had a
height advantage of 24 percent at the age of 8
years on sandy soil as compared to nursery run
seed; this same seed production area produced
seedlings with a height advantage of 7 percent on
heavy clay soil. He concludes that the collection
of seed from a local source of selected parent stock
is advantageous.

The most severe selection that can be commerci-

30

ally applied is through seed orchards. Individual
elite selections are made only after the examination
of thousands of acres of forest land. These selec-
tions are rigidly graded in comparison to a number
of surrounding dominant specimens, and included
in the orchard only after they indicate a maximum
amount of advantage. The selections are vegeta-
tively propagated and planted in a central orchard
location, or a seed orchard of seedlings from the
selected parents is established. The orchard is de-
signed to insure a minimum amount of inbreeding.
Here the trees are cultivated as intensely as in
fruit orchards and will serve as a source of high
quality seed for the future.

Based upon the figures quoted today and if we
take into consideration improvement due to _ in-
creased vigor, finer limbs, straighter boles, and less
disease, it seems probable that the yield increase
of 10 percent suggested by Barber can be obtained
with ease and the suggested figure of 25 percent
advanced by van Buijtenen is within reach. As-
suming no change in pulpwood stumpage values
from the figure advanced by Perry and Wang
(1958) and figuring a 20 percent increase in yield
due to tree improvement efforts, we can realize a
gross increased profit of some $600 per pound of
seed at a 25-year rotation, or an extra $2.10 per
acre per year. This profit, they say, justifies the
expenditure of $181.27 per pound of seed, allowing
5-percent interest on the invested money.

Perry’s figures are based on yield alone, and no
attempt was made to include other advantages
which cannot be represented easily by monetary
values—for example, the morphological character-
istics of bole straightness. It is difficult to deter-
mine how much more solid wood content would be
delivered per cord by minimizing the amount of
crook, spiral, and sweep. It is difficult to learn
how much more cellulose is in each cord and how
much less cooking liquor will be needed at the
mill. One can only estimate the increase in usable
fiber contained in the straight pulpwood stick. Add
to these figures a decrease in knot wood volume, a
concomitant decrease in compression wood asso-
ciated with knots, and an increase in the number
of seedlings growing to maturity by virtue of in-
creased disease resistance. Consider also the pulp
increase reported earlier by McElwee due to in-
creases in specific gravity and the probable im-
provement in paper sheet formation due to having
wood with more uniform fibers. All these advan-
tages, nebulous as they may be, are to be gained
as a result of the activities now taking place. Esti-
mates of possible improvements made several years
ago covered the entire scale from a minimum 1, 2,
and 5 percent made by the conservative members
of this group to a maximum 100 percent by the
most optimistic. Actual values today indicate an
intermediate expected increase of 15 to 25 percent
based on yield, plus the additive increment due
to quality. Perry’s figures, which seemed so un-
obtainable 5 years ago, are becoming more realistic
as we gain additional knowledge of the inheritance
patterns in the species we are using.

How Can We Improve Southern Hardwoods Through Genetics?

James R. Wilcox

Institute of Forest Genetics, Southern Forest Experiment Station
Forest Service, U.S. Department of Agriculture, Gulfport, Mississippi

Before discussing the genetic potential of hard-
woods we must define the objectives in hardwood
tree improvement. We can safely generalize and
say that any tree-improvement program is going to
have as its major goal the increase in genetic poten-
tial for rapid growth and superior tree quality.
Rapid growth is an easily understood characteristic,
but tree quality is a more complicated concept.

The quality of a hardwood log is determined by
its size and shape, and particularly by degree to
which it is free from defects like knots, holes, bark
pockets, stain, and rot (Lockard, Putnam, and Car-
penter 1950). In general, freedom from defects is
more important than species in determining the
value of an individual tree. Increasing the propor-
tion of high-grade material within the tree there-
fore will be an important facet of tree improvement,
and will require evaluation for tree form, branching
characteristics, and resistance to insects and dis-
eases.

When hardwoods of improved strains have been
developed, they will probably be grown in planta-
tions. We don’t know how most species will re-
spond to this kind of culture. The performance of
cottonwood has generally been good, but unex-
pected problems may arise such as the stem disease
which is attacking cottonwood plantings along the
Mississippi River. Undoubtedly, hardwoods grown
in plantations are going to require different man-
agement procedures from those for natural stands,
and tree breeders and silviculturists will have to
work in full cooperation to develop such proce-
dures. Among other things, the possibility of plant-
ings of mixed species as well as single species
should be explored.

Beyond general concepts we must cease speaking
of hardwoods as a group and concentrate on indi-
vidual species. Some 140 hardwood species are
native to the South. Of these, between 60 and 70
are of some commercial importance (Putnam, Fur-
nival, and McKnight 1960). These 60 to 70 species
represent 25 genera that differ in morphology, site
requirements, and utilization. Are we considering
a species of oak or ash? These species commonly
grow in mixed stands, a fact which complicates
comparisons among individual trees. Or are we
considering cottonwood, which usually occurs in
pure, even-aged stands in which comparisons of
individual trees with their neighbors are quite

31

easy? Some species, such as sweetgum, occur in
both mixed and pure stands. How should this affect
our criteria of individual tree selection?

In discussing seed production, we must again
relate our comments to individual species, for they
differ in floral morphology and pollinating agent.
Cottonwood is a dioecious species; yellow-poplar
has perfect flowers. Yellow-poplar is insect-pollin-
ated while many other hardwoods are wind-pollin-
ated. Differences such as these will influence the
development of techniques for controlled pollina-
tion and eventual methods of producing commercial
quantities of seed of improved stock.

Finally, characters to be improved and their rela-
tive importance will vary from one species to
another, depending upon species characteristics and
utilization practices.

When we have designated the species we want
to work with and defined our objectives for that
species, we can begin systematic research to attain
these objectives. You have heard the various steps
in tree improvement discussed in detail at previous
meetings. The initial step is to assess the variation
present in the species for characters of interest.
The next job is to estimate heritability for these
characters, i.e., determine how much of the varia-
tion is under genetic control and how much is due
to environmental effects. The modification of that
portion of the variation which is under genetic
control is the final phase of the improvement pro-
gram.

Considering these steps in tree improvement,
where do we stand now with some of the major
commercial hardwoods of the South? I would like
to review what has been done to date on five repre-
sentative southern hardwood species, or genera,
as applicable to the South. These include sweetgum,
yellow-poplar, cottonwood, several oaks, and green
ash.

Sweetgum and yellow-poplar occur throughout
much of the Eastern United States. Their wood is
used for a variety of products including lumber,
veneer, cabinetwork, and, especially in sweetgum,
pulpwood. Cottonwood reaches its maximum de-
velopment on the bottomlands adjacent to the lower
Mississippi River. At the present time the greatest
quantity of cottonwood goes into veneer and saw
logs, with pulpwood a promising outlet for the
future. This species is planted in greater quantities

than any other hardwood in the South today. Sev-
eral species of oak grow rapidly, generally develop
good form, and have a variety of uses including
lumber, veneer, cooperage, and pulpwood. Ash is
a high-value species used primarily in the manu-
facture of handles, sporting goods, and furniture.

Sweetgum

The literature on variation in sweetgum is limited
to variation in leaf morphology (Holm 1930; Dun-
can 1959). There is no information on inheritance
patterns for any characteristics. Studies on pat-
terns of variation in wood quality characteristics
are under way at North Carolina State College.
Other work there and at the Institute of Forest
Genetics, Southern Forest Experiment Station, in-
cludes studies of flowering habits and pollination
techniqves, and progeny tests from individual pa-
rent trees. This research will provide information
on geographic variation as well as on inheritance
patterns for economically important characters.

Characters which are receiving major attention
include resistance to epicormic sprouting and, since
sweetgum is the leading hardwood pulp species in
the South, wood quality. Epicormic sprouting fre-
quently occurs along the upper and middle portions
of the bole in response to environmental disturb-
ances such as extreme release from competition.
Although careful management can reduce the ex-
tent of sprouting, there is an immediate need for
some method of evaluating individual-tree variation
in this character during the juvenile stage. Other-
wise it will not be possible to evaluate progenies
for resistance until they have attained considerable
size, and improvement for this characteristic will
be an extremely slow process.

Wood quality is a complex characteristic involv-
ing density, fiber length, and cellulose content.
Since hardwood pulps are commonly used in blends
with longer fibered pulps, increasing fiber length
should increase the proportion of hardwood pulps
in these blends. Sweetgum is particularly important
in the manufacture of dissolving pulps and, since
cellulose is the desired product, high cellulose con-
tent would be a desirable feature of improved
strains. Any speculation on the improvement po-
tential for these and other characters will have to
await results of the research mentioned previously.

Yellow-Poplar

Seed source studies with yellow-poplar show a
general north-to-south trend in resistance to cold
(Funk 1958; Sluder 1960). Sluder (1960) reported
better survival of the local source than of sources
from several other geographic locations. Height
growth of local yeliow-poplar has been reported
to be as good as or better than that of seedlings
from other locations (Sluder 1960; Lotti 1955).
In one study (Limstrom and Finn 1956) significant
differences (0.05 level) appeared among six geo-
graphic sources in average height of 1-year seed-
lings. Variation in height of progenies from dif-

32

ferent trees was equally significant (0.05 level).
Often the variation in height among seedlings from
a single tree exceeded the mean difference in pro-
geny height among trees and seed sources.

The data just mentioned are from young stands,
5 years or less in.age. Whether the patterns will
persist to maturity and be evident for additional
characters is problematical.

Thorbjornsen (1961) reported variations in wood
quality characteristics in natural stands. He found
that wood density varied considerably among indi-
vidual trees within stands but discovered no dif-
ferences among stands. He concluded that rela-
tively rapid improvement could probably be made
by selecting and breeding for high density. In
fiber length, the variation within trees exceeded
both the variation among trees and among stands,
an indication that selection for long fibers would
probably not result in a marked increase in fiber
length.

Several studies on variation in seed quality and
effects of pollination are well worth mentioning,
since they will probably influence the methods and
techniques used in improvement programs. Lim-
strom (1959). comparing seed quality from five
trees in each of six stands over a 3-year period,
found as much variation among individual seed
trees within a stand as among stands. Seed quality
varied considerably from year to year.

There is general agreement that self-pollination
results in markedly reduced seed set as compared
to cross-pollination. Self-pollinations yield up to
11 percent filled seed while cross-pollinations yield
up to 60 percent filled seed (Boyce and Kaeiser
1961b; Carpenter and Guard 1950; Guard 19438;
Wright 1953).

Efficient control-pollination techniques have been
developed for the species (Taft 1962). Seedlings
from cross-pollinations tend to be more vigorous
than seedlings from wind pollinations. Pollen from
distant trees tends to produce the greatest increase
in vigor (Carpenter and Guard 1950).

Boyce and Kaeiser (1961b) concluded that yel-
low-poplar trees are not freely interbreeding under
natural conditions and that there is a low rate of
gene interchange among stands. Self- and cross-
incompatibilities are important, as adjacent trees
are likely to be closely related and less compatible
than trees a mile or more apart.

The results cited indicate that yellow-poplar has
a high potential for improvement in growth rate
as well as in wood density. The most immediate
gains will probably result from interpollinations
among stands within broad geographic areas rather
than from the use of open-pollinated seed from
selected trees. Orchards of outstanding trees from
several stands should produce seed of good viability
and subsequent stands with good vigor.

Cottonwood

Several studies have been made on wood quality
variation in eastern cottonwood. Kaeiser (1956)

reported that fiber length within trees increased
as number of rings from the pith increased.

In one study stem diameter and number of annual
rings accounted for an estimated 50 percent of the
fiber-length variation within and among cotton-
woods. The genetic variance was estimated to be
about 30 percent. High correlation coefficients
were reported for mean fiber length among rings
of the same tree. The twentieth ring from the pith
had the highest correlation with all other rings and
was the best ring for comparing fiber length among
trees (Boyce and Kaeiser 196la).

Statistically significant differences in specific
gravity and fiber length have been reported be-
tween clones of cottonwood (Gabriel 1956).

In studies being carried on by the Southern Hard-
woods Laboratory in cooperation with the Institute
of Forest Genetics, progenies from individual parent
trees have differed significantly (0.05 level) in
growth rate, branching, and resistance to Melamp-
sora rust. When the progenies are clonally propa-
gated under commercial planting conditions, they
are expected to yield much valuable information
on inheritance patterns of these and other charac-
ters.

Probably the bulk of the cottonwood planted in
the South will be from unrooted cuttings. This
asexual method of propagation will undoubtedly
influence the kind of breeding program followed
and the final product of any improvement program.
Once favorable genotypes are developed they can
be multiplied with ne genetic change, and selected
clones can be planted together over extensive
acreages for maximum production with no danger
of diluting the superior genetic stock.

Cottonwood is very sensitive to apparently minor
site differences, and one improvement problem will
be the isolation of genotypes that will perform
consistently well throughout plantations.

Cottonwood shows a high potential for improve-
ment in growth rate and wood quality characteris-
tics. Its rapid growth on good sites combined with
the relatively short rotation age will mean that
progenies can be evaluated for economically im-
portant characteristics sooner than most other hard-
wood species. Hence improvement should be more
rapid than with other hardwoods.

Oaks

As with the species already discussed, much of
the data on variation is from reports of juvenile
characteristics manifested in seed-source studies.
Genetic differences among seedlings of four seed
collections of Shumard oak have been reported
(Gabriel 1958). Seed sources ranged from Illinois
to Florida. In plantations in Pennsylvania, seed-
lings of northern origin suffered less dieback and
grew more rapidly during the early part of the
growing season than seedlings of southern origin.

Two-year data on seed-source studies with Shu-
mard oak, bur oak, and water oak indicated that

33

these species should be classed as geographically
variable (Wright 1957). All three species exhibited
variation in growth rate and autumn coloring asso-
ciated with seed sources.

Single-tree progenies of five white oak and five
red oak species from several locations were evalu-
ated for earliness of germination, 1-year and 3-year
height, and survival (Santamour and Schreiner
1961; Schreiner and Santamour 1961). From these
studies the authors concluded that individual-tree
selection appears to offer more promise for genetic
improvement than ecotypic or racial selection in
these species.

Variation in wood properties within and among
trees of southern red oak have been reported (Ham-
ilton 1961). Specific gravity, percentage of late-
wood, and toughness were inversely related to both
height in the bole and age from the center. Fiber
length was directly related to height and age. Some
among-tree variation existed, but in most instances
more variation was observed within individual
trees.

A statistical comparison of the distribution of
forkedness and straightness among the two largest
stems of 132 sprout oak clumps indicated that the
tendency to fork may be hereditary (Downs 1949).

An improvement program with any species of
oak will have to overcome a number of technical
problems. Especially needed are techniques for
making controlled pollinations. The oaks are no-
toriously difficult to control-pollinate, and the most
informative genetic studies as well as the accumu-
lation of favorable genes into a superior genotype
are dependent upon such pollinations.

The limited data cited indicate that early im-
provement should be possible in tree form as well
as growth rate. Bringing together superior indi-
viduals within broad geographic areas would be
the initial step. Wind pollinations among these
individuals should result in progeny with good
seedling vigor and good tree form.

Green Ash

Wright (1959) has summarized much of the
information on the genetic variation of green ash.
He differentiated (1944) a northern ecotype and
a Coastal Plain ecotype on the basis of growth rate,
petiole color, and winter hardiness. Meuli and
Shirley (1937) distinguished three ecotypes on the
Great Plans on the basis of drought-resistance.

Information on individual-tree variation is very
limited. Wright (1959) demonstrated clonal dif-
ferences in hardiness of leaves to growing-season
frosts among the open-pollinated offspring of a
single female parent. He surmised that randomly
distributed variation in pubescence and samara
shape is under genetic control.

The extremely limited information on this species
does not permit an estimate of progress to be made
in tree improvement.

Summary

Beyond these few species, or genera, data on
variation and patterns of inheritance in important
southern hardwoods are either extremely limited
or completely lacking. For example, there is no
information on the extent of variation or on pat-
terns of inheritance for important characteristics
of sycamore, willow, black cherry, and the tupelos.

In summary, then, just what answers do we
have to the questions, “What do we know about
inheritance patterns of southern hardwoods and
how can we get improvement through genetics?”

We actually have very little information about
inheritance patterns of southern hardwoods. We
have good evidence from seed-source studies that
several juvenile characteristics, including survival
and early growth, are under genetic control. The
information at this stage is limited, but the studies
will yield more data as they mature. Research on

34

wood quality of yellow-poplar, cottonwood, and
oak indicates that enough of the variability 1s under
genetic control to make selection profitable in these
species.

In spite of the limited information on inheritance
patterns, we Know we can get improvement through
genetics. The genetic principles that apply to other
plant and animal species apply to southern hard-
woods as well. Selection of individuals with desir-
able characteristics and their propagation, clonally
or from seed, should result in timber stands that
are better than the average forest of today. As
results of current and future research become
known, optimum ways of combining germ plasm
from selected individuals will bring further bene-
fits. The genetic potential is available to us. The
development and application of breeding techniques
can improve southern hardwoods to their full po-
tential.

Physiology of Trees as Related to Forest Genetics

Wm. H. Davis McGregor

Department of Forestry, Clemson College, Clemson, South Carolina

After defining some terms, I shall present some
specific examples which I hope will illustrate the
relationship between genetics and physiology and
show that an understanding of the physiological
processes of forest trees is important to forest genet-
icists. It is also important to progress in the every-
day task of forest management, but that is not my
subject. In this discussion I have borrowed heavily
from ideas introduced to me by Dr. Paul J. Kramer
at Duke University, and from the book ‘“‘Physiology
of Trees” (Kramer and Kozlowski 1960).

What is forest tree physiology, and are there any
unusual attributes of trees that make tree physi-
ology a special area in the general field of plant
physiology? Webster’s Third New International Dic-
tionary defines physiology as ‘‘a branch of biology
dealing with the processes, activities, and phenom-
ena incidental to and characteristic of life or of
living organisms.” B.M. Duggar (1911, p. 3) made
this somewhat more specific: ‘‘Plant physiology

. concerns itself with plant responses and plant
behavior under all conditions; that is, with relations
and processes readily evident or obscure, simple or
complex, which have to do with maintenance,
growth and reproduction of plants.”’

This seems to cover all angles, but are trees any
different from other plants in their physiology?
Kramer and Kozlowski (1960) describe clearly the
differences that make trees distinctive:

“The peculiar characteristics of trees are a
matter of degree rather than of kind, however.
They go through the same stages of growth
and carry on the same processes as other plants,
but their larger size, slower maturity, and
longer life accentuate certain problems as com-
pared with smaller plants with a shorter life
span. The most obvious difference between
trees and herbaceous plants is the great dis-
tance over which water, minerals, and food
must be translocated in the former. Also,
because of their longer life span, they usually
are exposed to greater variations and extremes
of temperature and other climatic and soil con-
ditions than annuals or biennials. Thus, just
as trees are notable for their large size, they
are also notable for their special physiological
development.”

The only thing I would add to this is that in
forestry we are primarily interested in the stem
of the plant, rather than in the fruit, which is the

35

object of main interest to scientists interested in
field crops. This different emphasis may change
somewhat the direction of physiological research
on the part of those interested in forest tree physi-
ology.

Keeping in mind these definitions of forest tree
physiology, how then is it related to forest genetics
or species improvement? To illustrate the asso-
ciation I would like to refer to the concept which,
according to Kramer and Kozlowski (1960), was
developed by the German physiologist Klebs and
refined by others in this country. This concept
emphasizes the principle that hereditary or environ-
mental factors can affect the growth of a living
organism—be it an alga, cotton plant, or tree—only
by affecting the plant’s internal processes and
conditions; in other words, its physiology. These
relationships are illustrated by the accompanying
diagram.

ENVIRONMENTAL
FACTORS
forest ecology

HEREDITARY POTENTIALITIES
OF TREES
forest genetics
and tree breeding and forest soils

ST i

PHYSIOLOGICAL PROCESSES AND CONDITIONS
tree physiology

TREE GROWTH
forestry and horticulture

(After Kramer 1956)

Thus, in order to understand how genetic factors
may affect tree growth, wood quality, or other
important features, we must learn how the factors
affect the physiological processes involved.

Now I hope that this concept does not offend
the geneticists present, since I seem to be saying
that they cannot get anywhere without knocking
on the physiologist’s door. I do not intend to imply
that physiology is more important than genetics.

In reality you can bypass physiology temporarily
and, for example, develop a hybrid which grows
faster than either parent species, without knowing
why or how this growth increase occurs. For the
greatest progress and for the widest application of
our results, however, we eventually would have
to try and determine what processes or conditions
in the tree were changed to bring about an increase
in growth. Rather than to attempt a comprehensive
literature review of physiology-genetics work under
way, let us just illustrate the relationship with some
specific examples in several areas of forest genetics.

Selection

The phase of forest species improvement with
which most of us are somewhat familiar here in
the Southeast is selection. To illustrate how selec-
tion for a desired trait is related to tree physiology,
let us use the example of selection for high oleo-
resin yield which has been conducted by the U.S.
Forest Service at Olustee, Fla., since 1941 (Squil-
lace and Dorman 1961; Squillace and Bengtson
1961).

The first step in this work was the selection of
12 slash pine trees for high gum-yielding potential
from natural stands in nortn Florida and south
Georgia. The yield of these trees was about double
that of comparable non-selected trees. Subsequent-
ly, crosses were made among nine of these selected
trees, and between the selected trees and average
and low-yielding trees. By using a micro-chipping
technique on young trees stemming from _ these
crosses, it was shown that oleoresin yield is in-
herited, with a heritability of about 55 percent.
These studies also showed that of the original nine
rigid selections used, only three were outstanding
in passing on their high gum-yield qualities to their
progeny.

Now, how does physiology enter this picture?
Schopmeyer et al. (1954) suggested that gum yield
should be related to certain anatomical and physi-
ological characteristics, namely, number and size
of resin ducts, gum exudation pressure, and gum
viscosity. Mergen et al. (1955) demonstrated that
gum yield was inversely related to gum viscosity,
and Bourdeau and Schopmeyer (1958) were able
to prove that oleoresin exudation pressure was
directly correlated with oleoresin yield. They con-
cluded that the ratio of pressure to viscosity could
be used for predicting yield potential of young
trees.

So, what has happened here? The geneticist has
selected high-yielding trees and proven that some
of them pass to their progeny this high-yielding
trait. The physiologist, working hand-in-hand with
the geneticist (indeed, often they have been the
same individual), has discovered why some trees
yield more oleoresin than others. Now they have
a tool which is available to improve selection tech-
niques and to improve progeny testing. If tech-
niques can be developed so that exudation pressure
and gum viscosity can be measured on a seedling,

36

the testing program can be speeded up consider-
ably.

This research has revealed some physiological
differences, but has raised many new problems for
the physiologist to consider. Why do some trees
have a higher exudation pressure? Why do some
produce low-viscosity oleoresin? These questions
will carry the researcher back toward more funda-
mental processes, e.g., photosynthesis, cell metabo-
lism, gum synthesis. When you answer one “Why?”
you generate a dozen new ‘“‘Whys?”’.

Before we leave the subject of selection, let me
just mention the physiologically complex problem
of selection for fast growth rate. What are we
really selecting for? Efficient photosynthesis, effi-
cient utilization of water or minerals, some differ-
ence in cell metabolism that allows one plant to
convert to cellulose more of the products of photo-
synthesis.

Breeding

Now let us move on from selection and consider
for a few minutes the subject of breeding for de-
sired traits and multiplication of genetically identi-
cal individuals once a desired strain is available.

One major deterrent to rapid progress in forest
genetics is the flowering habit of most commercial
tree species. They do not normally begin producing
the organs for sexual reproduction in appreciable
numbers until they are 10 or more years old, so
breeding and progeny testing are delayed. Compare
this with, say, corn breeding, where four crops of
a 90-day maturing variety can be raised in one
year by using a greenhouse during cold weather.
The forest geneticist Knows or can work out the
techniques of breeding in the various species, but
he cannot do breeding without flowers or strobili.

Some treatments, such as fertilizing, strangling,
and root pruning, have been successful in stimu-
lating precocious flowering, but there remains
abundant opportunity for further advancement.
Here again the physiologist can perhaps help. The
U.S. Forest Service’s Dr. R.L. Barnes and associ-
ates at the Research Triangle near Durham, N.C.,
are trying to determine internal physiological fac-
tors governing flowering. They are studying the
biochemical changes which bring about ‘“‘readiness
to flower,’ and are attempting to identify the
basic processes which initiate the changes. This
work is far from finished, but when the controlling
physiological processes have been identified and
the biochemical steps determined, one can then
make some logical ‘‘guesses’’ about methods of
manipulating flowering with more hope of success
in producing flowers or strobili on young saplings
or even seedlings.

In this same area of genetics, it is desirable to
have some reliable means of clonal reproduction
of individuals with desirable traits. This requires
grafting or some form of rooting, and raises many
problems of a physiological nature. For example,
with loblolly pine and many other species, root-

ability declines with age (McAlpine and Jackson
1959). What is the basic cause of this decline?
Can rooting potential be restored? Also, cuttings
from one part of a tree may root better than those
from another (Grace 1939), and the resulting ram-
ets may even have different growth characteristics
(Libby and Jund 1962). What physiological pro-
cesses of a cutting are affected by age of parent
tree, or by its original position in the crown?

As for grafting, several techniques have been
used successfully to establish the initial graft union,
but in many instances a large number of the grafts
later die. Some physiological difference between
stock and scion causes an incompatibility which
prevents normal functioning of some essential pro-
cess. What causes the incompatibility, and can it
be overcome? When the physiologist can answer
some of these questions, the geneticist can make
more rapid progress in species improvement and
will be able to assess more precisely true genetic
differences.

Application

As a final example, let us turn to the essential
step of utilizing superior strains once they have
been tested and proven. Let us suppose that selec-
tion, breeding, and testing at the University of
Georgia have produced a strain of shortleaf pine
highly resistant to littleleaf disease (Zak 1955).
Now disease resistance is not of much benefit unless
the tree also makes satisfactory growth. Can we
expect good growth from this new strain through-
out the natural range of shortleaf pine from New
Jersey to Texas and in areas where it has been
planted outside its natural range? We know that
geographic races exist within species, and so we
would assume that this new strain would be limited
in the geographic range over which it could be
expected to make good growth. We could make
trial plantings of the new strain throughout the
range of shortleaf pine and wait 20 to 40 years to
assess the pertinent growth results. But as our
knowledge of physiology of trees increases, per-
haps we can arrive at some valid estimates of
potential range by making certain physiological
tests. For example, drought resistance of some
species has been found to be related to stomatal
control and rate of transpiration (Polster and
Reichenbach 1957). Could we make estimates of
soil moisture or rainfall limits of the new strain
by measuring these characteristics on seedlings in
the laboratory? Cold hardiness has been associated
with the concentrations of certain cell constituents
(Parker 1962). Could temperature limits for the

37

new strain be determined by measuring cell sugars?
The optimum temperature for maximum net photo-
synthesis has been established for some _ species
(Decker 1944) and could be established for the
new strain. Some species and races within species
can tolerate shorter daylengths than others and
still make satisfactory growth (Pauley and Perry
1954; McGregor et al. 1961; Allen and McGregor
1962; Watt and McGregor 1963). These limits also
could be established for the resistant shortleaf pine
strain.

By measuring the rates of the various physiologi-
cal processes and determining the limits of optimum
operation of these processes under various condi-
tions, perhaps we could predict how well our new
strain would perform in a certain locality in com-
petition with other tree species. With continued
research, this will be possible.

I seem to have raised many questions and given
very few answers. However, I hope that I have
contributed to your better understanding of physi-
ology and of its relation to forest genetics. Let me
summarize by saying that physiologists are inter-
ested primarily in how trees grow, while forest
geneticists are interested in changing the way in
which trees grow. The greatest progress will be
made when the two work together to solve the
many remaining problems.

Bibliography

Mirov, N. T.
1937. Application of plant physiology to the
problems of forest genetics. Jour. For-
estry 35:840-844.

Pauley, Scott S.

1958. Photoperiodism in relation to tree im-
provement. In Thimann, Kenneth V.,
ed. The physiology of forest trees, pp.
557-571. New York: The Ronald Press
Co.

Perry, T.O.
1953. The genetics of the photoperiodic response

in poplar tree species. Genetics 38:681-
682.

Richardson, S. Dennis.

1960. The role of physiology in forest tree im-
provement. Fifth World Forestry Con-
gress Proc. 2:733-741.
Wilde, S.A.
1954. Soils and forest tree breeding. Jour. For-

estry 52: 928-932.

How Far Can Seed Be Moved?
Philip C. Wakeley

Institute of Forest Genetics, Southern Forest Experiment Station
Forest Service, U.S. Department of Agriculture, New Orleans, Louisiana

How far it pays to move forest tree seed has Maryland loblolly of comparable ages. At 35 years,
been a serious question for nearly two centuries. the mean annual increments of three loblolly stocks
Baldwin (1942) traces discussion of it back to originating 350 to 450 miles from a planting site
an anonymous Swedish author writing in 1769. at Bogalusa, La., were from 47 percent to as little

as 20 percent of the mean annual increment of
local Louisiana stock (fig. 1).

Use of seed from the wron¢ source can eliminate
any chance of profit from a plantation. Weidman

(1939), for example, reports a northern Idaho test In extreme cases like these, when stock of dis-
of 20 races of ponderosa pine, one of which, after tant geographic origin produces only a fifth as
9 years of successful growth, suffered 100 percent much wood as local stock, or no wood whatever,
mortality in a colder-than-average winter. Leon it is easy to name specific sources from which seed
Minckler (personal communication) reports exten- should not be obtained.
sive to complete killing of North and South Caro- Most of the evidence from studies of racial vari-
lina races of loblolly pine in the Central States by ation is, however, less clear and more difficult to
winter temperatures that did negligible harm to interpret. The immediately practical questions
3 100 1

80

Nw

60

ROUGH CORDS

40

PERCENT (LA. =100%)

20

LA. TEX. GA. ARK.

{REMOVED IN 22-YEAR THINNING
STANDING AT AGE 35 OF SeaaNG TEX GA. ARK.

FicurRE 1.—Absolute and relative mean annual increments per acre, at 35 years, of loblolly pines of four
geographic origins, planted at Bogalusa, La. The four sources, from left to right, were 50, 350, 450, and
350 miles from the planting site.

[

38

of how far one may venture from the planting
locality to get seed when the local seed crop is
inadequate, and of which of several moderately
distant sources to choose, are hard to answer. Fur-
thermore, the best answers we can give today
cannot be considered final. They will require re-
vision and amendment as new studies are estab-
lished and reported and as trends in existing studies
change with the passage of time.

Surveys of variation in the morphological char-
acters or wood specific gravity of native stands,
such as those Thor (Thorbjornsen 1961) and
Wheeler and Mitchell (1959) reported at the Sixth
and Fifth Southern Conferences, are of little prac-
tical help in choosing a source of seed for a plant-
ing program. Often such surveys deal with char-
acters (like shape of seeds or size of pollen grains)
that, while important in basic research, have no
direct bearing on the survival, growth, or form of
planted trees. In any event they fail to distinguish
between the effects of the genetic makeup of a race
and the effects of the environment in which the
race occurs. Growth-chamber and laboratory stud-
ies are sometimes more helpful guides. Character-
istically, though, growth-chamber and_ similar
studies cover such a brief portion of a tree’s life
cycle that they supply only a fraction of the infor-
mation needed. With few exceptions, therefore,
the practical guides to choice of seed source have
been conventional provenance tests, in which stocks
representing several different geographic origins
have been planted together in one place and ob-
served for a number of years under field conditions.

Provenance tests are not all equally reliable or
useful, however.

To justify generalization about racial variation
within a species, a provenance test must include
stocks representing a considerable portion of the
species’ range—preferably all of it. To distinguish
races clearly and to indicate their geographical
distributions, the test must include stocks repre-
senting numerous sources not too widely or irregu-
larly spaced.

To yield dependable information, the stocks rep-
resenting the various sources must be replicated
in the plantation and planted in random arrange-
ment within replications; the planting site must be
relatively uniform; all stocks must be planted at
essentially the same time; and the nursery treat-
ment, lifting, packing, and shipping of all stocks
must be as nearly identical as possible. A proven-
ance test is a specialized form of progeny test and
should adhere to the same exacting standards as
other progeny tests (Wakeley et al. 1960). Close
scrutiny of the records, however, will show that
very few provenance tests, and practically none
of the older ones, have don2 so.

Finally, conclusions must be drawn cautiously,
if at all, from the earliest remeasurements of a
provenance test, lest later developments show them
to have been both premature and misleading. Let
me illustrate briefly what I mean.

39

In the ponderosa pine study reported by Weid-
man (1939), the stock of Coconino origin grew
fastest the first few years, and at 10 years excelled
all other stocks but one in height, and equalled
that one. At 10 years it might easily have been
selected as best for planting in northern Idaho. By
the 20th year, however, its average height was
less than that of 12 of the 18 other stocks still
surviving in the study.

Similar reversals have occurred in southern pine
provenance tests. In the study established at Boga-
lusa, La., with four loblolly pine stocks from the
1925 seed crop, the Texas stock was very signifi-
cantly taller than the Georgia stock at 15 and 22
years, and taller even at 28 years. By the 35th
year, however, the Georgia stock had overtopped
the Texas stock (fig. 2).

AVERAGE HEIGHTS OF ALL TREES,
LOBLOLLY AT BOGALUSA

70

o
Oo

ol
{e)

ff
Oo

Os
Oo

MEAN HEIGHTS (FEET)
i)
Oo

S)
BE WUINNED sa

[5 ee Ont, 20) e190) 350
IN PLANTATION

3) 10
YEARS

FIGURE 2.—Mean heights of all trees of stocks rep-
resenting four sources of loblolly pine seed from
the 1925 crop, planted at Bogalusa, La.

Through the 15th year of this same study at
Bogalusa, the Texas and Arkansas stocks survived
better than the local Louisiana stock. By the 35th
year, the survival of the Texas stock had fallen
slightly below, and the survival of the Arkansas

stock had fallen significantly below, that of the
Louisiana stock (fig. 3).

SURVIVAL BASED ON ALL TREES
PLANTED, LOBLOLLY AT BOGALUSA
1e]@)

a o 00
Oo Oo 2)

SURVIVAL (PERCENT)

)
{e)

5
YEARS

10 15 20 25 30 35

IN PLANTATION

FicuRE 3.—Mean survival percents of stocks rep-
resenting four sources of loblolly pine seed from
the 1925 crop, planted at Bogalusa, La.

Only four southern-pine provenance tests of ma-
jor importance were installed before the Southwide
Pine Seed Source Study (Wakeley 1959; 1961), and
of these only the loblolly study established at Boga-
lusa, La., with seed from the 1925 crop, has gone
through a full pulpwood rotation—35 years. All
four of these earlier provenance tests suffered from
various defects of design, execution, or both. In
the study of loblolly from the 1925 crop, the
extreme contrast between the Louisiana and Arkan-
sas stocks from age 15 onward must be discounted
somewhat because of the nonrandom arrangement
of sources in the replicated rows. Except for the
results obtained with seed sent to the Union of
South Africa (Sherry 1947), an ambitious study
established with seed from the generally abundant
1935 crop was practically a total loss.

Ten or more conventional southern-pine proven-
ance tests of potentially major importance have
been established since the Southwide Study, and
reports on them are appearing with increasing

40

frequency. Several are superior to the Southwide
Study in design, sampling, or execution, but none
is as broad in scope, and it is not beyond possibility
that some of the first conclusions drawn from them
will have to be revised.

We have, in short, an insufficient basis on which
to lay down any final rules for the movement of
the forest tree seed principally used for reforesta-
tion in the South.

I feel, however, that we are in far better position
to lay down tentative rules than we were 10 or
even 5 years ago. The 10th year analyses of the
Southwide Study, plus forthcoming publications on
other studies, may enable us to improve such tenta-
tive rules even within the next 12 months.

Personally, I have no doubt whatever that eco-
nomically important racial variation associated with
geographic location exists in all four principal
species of southern pine.

Such variation is clearly very great in loblolly
and shortleaf. Stocks from opposite extremes of
the ranges of these two species differ conspicuously
in their requirements for optimum survival and
growth. There is good evidence that, even within
individual States, loblolly pine varies in suscepti-
bility to fusiform rust, and, toward the western
limit of its range, in drought resistance.

Racial variation, though present, seems to me
to be least in slash pine, particularly in those por-
tions of the species’ range in which seed is collected
commercially.

The picture of racial variation in longleaf pine
is still somewhat obscure. The species is difficult
to plant successfully, slow to commence height
growth, and prone to brown-spot infection. For
these reasons, results of provenance tests take
longer to obtain than with other species, and tend
to be erratic. My personal impression is that long-
leaf exhibits less racial variation than loblolly and
shortleaf, but considerably more than slash pine.

Certain extremely long movements of seed have
had catastrophic results. They obviously should
be avoided in practice, especially when they have
been tried several times. Longleaf from seed col-
lected in Hillsborough County, Fla., has twice made
a very poor showing in States north and west of
Florida. Shortleaf seed from the central and south-
ern Atlantic States and the Gulfcoast States has
been tried three times in Pennsylvania without
suecess. North and South Carolina loblolly stocks
have succumbed to cold in Central States locations
in which Maryland loblolly has survived.

Noncatastrophic but still economically serious
setbacks have occurred when longleaf, loblolly, and
shortleaf stocks have been tested at shorter but still
considerable distances from their points of origin.
In a majority of instances in the Southwide Pine
Seed Source Study and other studies, the setbacks
have taken one of two forms. Either the stock from
a distant source has survived well but grown poorly,
or the survivors, although fairly rapid in growth,
have been few in number.

There are indications, though there is hardly as
yet conclusive proof, that a few geographic races
of southern pines are capable of both good survival
and good growth, even at very great distances from
their points of origin. Longleaf pine from Baldwin
County, Ala., and loblolly pine from Onslow
County, N.C., have exhibited such wide adaptabil-
ity to varied conditions, each in two sets of planta-
tions established with different seed lots collected
from different stands in different years.

For the first 3 to 5 years, shortleaf and loblolly
stocks of northern origin have generally outgrown
stocks of southern origin when planted with them
in the northern portions of the species’ ranges,
while in the southern parts of the ranges southern
stocks have generally outgrown northern stocks
(figs. 4 and 5). In some cases, though not in all,
the tendency has persisted through the 10th year
(figs. 6 and 7). There seems to me to be good
evidence that variations in both temperature and
day length, each of which is strongly correlated
with latitude, are involved in this pattern of growth
behavior.

As a rule, though again with some exceptions,
an east-and-west movement of southern pines in

SHORTLEAF SERIES 4—AT 3 YEARS

£

HIEIGH Tea (FEES) +)
ol

2 BURLINGTON COUNTY, N.J.
(r=.94)

30

32

EATITUDE OF SEED SOURCE
(DEGREES N)

34 36 38 40

FicurE 4.—Mean heights of shortleaf stocks at 3
years, over latitudes of seed sources, in northern
and southern plantations.

41

LOBLOLLY SERIES 1—AT 5 YEARS
12

WORCESTER COUNTY, MD,

> 8
ly
ly
Ww
= 6
ts
x=
S ,
a
PEARL RIVER COUNTY, MISS.
(r=-.48)
2

30

32

LATITUDE OF SEED SOURCE
(DEGREES N.)

34 36 38 40

F1GuRE 5.—Mean heights of loblolly stocks at 5
years, over latitudes of seed sources, in northern
and southern plantations.

the same general latitude seems to affect growth
less than does movement for an equal distance
north and south.

The susceptibility of loblolly pine to fusiform
rust does, however, vary conspicuously with longi-
tude of seed source. While variations in suscepti-
bility occur even within individual States, they
seem to be overshadowed by a general tendency
for susceptibility to decrease from east to west.
The lower susceptibility of western stocks has been
dramatically illustrated by a Southeastern Station
study in Georgia, in which, at 5 years after plant-
ing, the percent trunk-infected in each of 14 Georgia
and 3 north Florida stocks was from 4 to 10 times
the percent trunk-infected in a single Arkansas
stock planted among them.

If I were a land manager or company executive
and had to decide in favor of one as against some
other nonlocal source of seed, I would follow
these 10 guides in making my choice.

1. I would assume that the farther I moved seed
in any direction, the greater would be the risk of
its being poorly adapted to the planting locality,

LOBLOLLY PINE AT 9 YEARS AT
LATITUDE 31°S IN SOUTH AFRICA
40

30

HEIGHT (FEET)
i)
Oo

S)

30

32
LATITUDE OF SEED SOURCE
(DEGREES N)

34 36 38 40

Figure 6.—Mean heights of loblolly stocks at 9
years, over latitudes of seed sources, in a planta-
tion at a low latitude in South Africa (data from
Sherry 1947).

and the more serious the maladaptation might be.
The evidence to date does not justify saying it is
always safe to move seed of a certain origin thus
far and never safe to move it any farther.

2. I would avoid moving seed of any of the
southern pines, even slash pine, over extreme dis-
tances, lest I duplicate one of several catastrophes
already demonstrated. To avoid such extreme
moves, I would go to considerable lengths to store
seed of suitable origin in years of abundant produc-
tion. As a last resort, I would suspend planting or
seeding till seed of a suitable source became avail-
able.

3. I would be more cautious about moving seed
of any of the southern pines a given distance north
or south than about moving it an equal distance
east or west. Going north or south involves a
greater change in temperature, to which racial
variation evidently is strongly related, and also a
greater change in day length, to which loblolly
and shortleaf races seem delicately adjusted and
to which races of the other species may be adjusted
also.

42

4. Other things being equal, I should prefer to
move seed east rather than west, and would con-
sider moving it farther to the east than to the west,
especially if I were planting on droughty sites.
Longleaf, loblolly, and shortleaf pines from western
sources may be somewhat slower growing than
those from eastern sources, but do seem to be more
drought resistant and hence to be capable of better
survival in dry years and on dry sites.

5. I should be particularly cautious about mov-
ing loblolly very much to the west. Maryland lob-
lolly has incurred relatively light rust infection
wherever planted, but other eastern provenances,
from North Carolina south to Florida, have gener-
ally proved markedly more rust susceptible than
more westerly provenances from corresponding lat-
itudes.

6. Even within these limitations, I would try to
get seed from a source (such as Baldwin County,
Ala., for longleaf or Onslow County, N.C., for lob-
lolly) that had proved widely adaptable in at least
two tests.

LOBLOLLY SERIES 1—AT IO YEARS

WORCESTER COUNTY, MD.

PEARL RIVER COUNTY, MISS.
(r=-.05)

HEIGHT (FEET)

30 32 34 36 38.740

LATITUDE. OF SEED SOURCE
(DEGREES N.)

FicuRE 7..——Mean heights of loblolly stocks at 10
years in the same plantations as those shown in
fig. 5. The relation of growth to latitude of seed
source has become intensified in the northern
plantation but dissipated in the southern one.

7. Although supporting evidence is not yet con-
clusive, I should be strongly inclined to limit plant-
ing of longleaf on the Carolina or Florida sandhills
to stock grown from seed from the corresponding
sandhill areas.

8. Within the range from the central Florida
peninsula north to southern South Carolina and
west to eastern Louisiana, I should be less appre-
hensive about unrestricted movement of slash pine
seed than about similar movement of seed of the
other three principal species. Even here, however,
I should feel less free to move slash seed north or
south than to move it east or west, and I should
avoid getting seed from coastal-strip slash pine of
a typical form for the species.

9. I should by no means depend upon correct
provenance alone to insure good growth in my
plantation, but should take care also to avoid

43

getting seed from high-graded, inbred, or otherwise
minus stands within the provenance chosen.

10. Though there is as yet no experimental evi-
dence to support me, I believe I should risk moving
genetically superior seed from plus stands, elite
trees, or tested seed orchards slightly farther than
I would move “run-of-the-woods” seed. Loss in
growth resulting from the movement might be
offset, at least in part, by a gain in growth resulting
from selection. Under no circumstances, however,
would I move seed-orchard or other improved seed
over extreme distances. It is questionable, for
example, whether any degree of selection and breed-
ing would enable Maryland loblolly to equal the
growth of ordinary Texas loblolly if both were
planted in Texas. The same would be true of any
other genetically improved southern pine seed
moved an excessive distance from its geographic
origin.

Management of Pine Seed Production Areas

Donald E. Cole

Continental Can Company, Inc., Savannah, Georgia

A great deal of planting and direct seeding is
being done with the southern pines and there is
every indication that this will be the case for a
long time to come. Since the seed used has such
a profound effect on the harvest and since we are
planting and seeding on such a grand scale, it is
vital that we use the best seed available as long
as its cost is not excessive.

The fastest way of mass-producing southern pine
seed is by means of seed production areas. Where
suitable stands are available, substantial quantities
of seed may be produced in from two to five years
from the time a seed production area is established.
This is much quicker production than is possible
from grafted or seedling seed orchards and although
the degree of improvement from seed production
area seed is not as great as may be expected with
seed orchard seed, we feel that the combination of
rapid seed production and a modest improvement
in quality is sufficient to make the establishment
of seed production areas worthwhile.

Having thus stated our basic premise, let us con-
sider in more detail what is involved in the estab-
lishment and management of seed production areas.

A seed production area is a stand managed specif-
ically for the production of seed; its purpose is to
provide, in quantity, seed of known origin from
the best phenotypes available. The establishment
of seed production areas is a stop-gap measure
designed to produce seed of the best possible quality
until our seed orchards begin to bear.

The most important single factor in the establish-
ment of a seed production area is the quality of
the stand; there is no method by which fertilization,
spraying, etc., can produce first class seed from
second class trees. Therefore, it is essential that
the stand chosen be of the best possible quality
(quality being used here primarily with reference
to vigor, freedom from fusiform cankers, and form
—good bole and crown characteristics—not site).
Site quality isn’t too important as long as it will
permit fair growth and cone production (Thorb-
jornsen 1960) and the site is fairly representative
of the area where the seedlings are to be planted.

The next point to consider is stocking; the larger
the number of trees per acre, the more selective you
can be regarding the trees you leave and this again
has an effect on quality. Therefore, only well-
stocked stands are suitable for conversion to seed
production areas; 100 trees 10 inches in diameter

44

or 50 feet of basal area per acre should be the
minimum acceptable stocking.

The size, in area, of the stand has a bearing on
the practicability of the operation; as the size of
the area increases, management costs per acre are
reduced and the proportion of the total area tied
up in the isolation zone decreases. For instance,
a 5-acre seed production area will have about 21
acres in its isolation zone; a 20-acre seed production
area will have about 48 acres in its isolation zone.
In addition, the number of trees on the area has an
effect on the frequency with which the area can
be harvested economically (if the number of trees
is 100, and 20 percent have a crop of harvestable
size, it is almost sure to be more expensive to
collect them than it would be on an area where the
total number of trees is 500, and 20 percent have
a crop of harvestable size). We now feel that 10
acres in the seed production area proper is the
least that is worth developing and we prefer larger
areas.

Tree size has a strong influence on cone produc-
tion, of course; we try to choose stands where the
average diameter of the leave trees will be at least
12 inches. Stands of smaller trees can be used but
they will take longer to produce cone crops of
harvestable size.

Having selected a stand that meets our require-
ments for quality, stocking, acreage, and average
diameter, the leave trees are marked and every-
thing else is cut. We follow the Georgia Crop
Improvement Association’s Standards for the Certi-
fication of Forest Tree Seed in selecting leave trees
even in States which have no provision for the
certification of the seed. These give rather stiff
specifications for bole and crown characteristics,
freedom from fusiform cankers, width of the iso-
lation zone, ete. We have found that these standards
give us a good set of reference points to follow in
establishing the areas. And the examination of
the areas by the Association inspectors, with the
attendant culling of sub-standard trees, puts the
areas in very good shape. Generally about 10-15
trees are left per acre. This often seems like a very
sparse stand, but heavy culling is necessary if much
improvement in quality is to be attained.

Matthews (1963) cites research by Florence and
McWilliams which showed that the density giving
maximum cone production per tree is much lower
than the density giving maximum cone production

per acre; this has an important effect on the eco-
nomics of seed-production area management since
the size of the cone crop per tree is so closely
correlated with cost of cone collection. Pollen pro-
duction is also greater at wider spacing and is re-
flected in a higher number of viable seeds per cone.
It is possible to have too few trees, of course; eight
fair-sized trees per acre is probably close to the
lower limit for good cone production and seed-set.

The release furnished by such heavy cut has a
stimulating effect on the remaining trees (Allen
and Trousdell 1961; Allen 1953; Bilan 1960; Easley
1954; Phares and Rogers 1962). The third season
after release they usually will begin producing
larger cone crops. This may continue for two or
three seasons or longer, depending on the density
of the stand on the seed production area.

We do not yet have sufficient data or experience
to estimate accurately the number of trees needed
to produce a given volume of cones. I am less
optimistic in this regard than I once was, however.
I now feel that about five trees are needed for each
bushel of cones that is required annually. This is
necessary because of the irregularity of good cone
crops, because many trees do not produce cones in
harvestable quantities, and because a buffer is
needed against the loss of trees to insects, storms,
lightning, etc.

Once a seed production area has been established,
we have found that additional cultural practices
are beneficial.

Fertilization has been found to be an effective
method for increasing cone and seed production
by a number of workers (Allen 1953; Hoekstra
and Mergen 1957; Timofeev 1959). B.F. Malac,
Union Bag-Camp Paper Corporation, is experiment-
ing with the effect of different amounts of a com-
plete fertilizer on cone production on a seed pro-
duction area. He reported in a personal communi-
cation that (1) fertilized trees produced approxi-
mately twice as many cones as unfertilized trees
in the same seed production area, and (2) approxi-
mately 50 percent of the cones were lost between
the time of pollination and the time of harvest with
no difference in the rate of loss between fertilized
and unfertilized trees. However, it has been re-
ported by Asher (1963) that squirrels prefer cones
from fertilized trees, that cone losses from all causes
were significantly greater on fertilized plots, and
that this suggests insects also may prefer cones from
fertilized trees. And Hughes and Jackson (1962)
say that fertilization, especially with phosphorous,
markedly increased damage from Dioryctria and
Cronartium in young slash plantations. Fertiliza-
tion may have other effects; Mergen and Voigt
(1960) found that seed from fertilized slash pines
produced larger and more vigorous seedlings than
did seeds produced on unfertilized control trees.

The cost of fertilizing 1,665 trees on our seed
production areas was 63 cents per tree per applica-
tion of 20 pounds of 8-8-8 (NPK, with sulfate of
potash-magnesia) at $47.75 a ton. Of this, 15 cents
was for labor, at a rate of $1.50 per hour. It took

45

almost exactly 1 man hour, including loading,
unloading, and travel, to fertilize 10 trees. The
only seed production areas fertilized are those pre;
pared according to the Georgia Crop Improvement
Association Standards where we feel that the extra
cost is justified by the quality of the seed to be
harvested. The fertilizer is applied in the spring,
no later than mid-May.

Root pruning, girdling, and strangulation have
been used to increase seed production (Bilan 1960;
Grano 1960; Hoekstra and Mergen 1957; Timofeev
1959) but the results have been erratic and have
even been reported to give fewer cones, in the
long run, than no treatment (Bergman 1955; Girg-
idov 1960; Klir et al. 1956). Speaking of these prac-
tices, Matthews (1963) says “The girdling of stems
of fruit trees was in common use one hundred years
ago as also was root pruning; both techniques have
been superceded in general practice by the use of
fertilizers, shoot pruning, and clonal rootstocks.
It appears certain that similar treatments will be
of greater benefit than root pruning and stem gird-
ling or strangulation in increasing seed production
in forest trees.”

The control of seed and cone insects is very im-
portant to the continued successful management of
seed production areas. Thrips, Laspeyresia seed-
worms and Dioryctria coneworms seem to be the
worst offenders; they can cause drastic losses of
cones and seed from the time of pollination right
on up to the time of harvest. And it is in this
phase of seed production management, the econom-
ical control of cone and seed insects, that I believe
the greatest opportunity lies for increasing the
yield of our seed production areas.

Edward P. Merkel, located at the Olustee, Fla.,
unit of the Southeastern Forest Experiment Station,
in a personal communication recommends the fol-
lowing formulations for the control of coneworms
(Dioryctria spp.): BHC (gamma isomer) at 4
pounds of active toxicant per 100 gallons of water
or Guthion at 1% pounds of active toxicant per
100 gallons of water. Applications should be made
during each of these periods: March 15-31, May 1-
15, June 1-15, July 10-20. To lower costs the July
application can be omitted with very little loss in
cone protection. At these concentrations the cost
of the chemicals is about the same for both BHC
and Guthion and since the May application of
Guthion alone gives good control of seedworm
(Laspeyresia) it would seem to be the preferred
material at least for the May application; it is
more toxic to humans than BHC, however. His
work was done with a hydraulic sprayer which
would reach trees 50 feet tall; about 8.5 gallons
of spray was used per tree at a cost of 80 cents per
tree for chemicals alone.

More recently Merkel has compared the relative
effectiveness of hydraulic sprayers and mist blowers
for applying insecticides. In a personal communi-
cation he reports that he made applications on
April 10, May 5, and June 8 of the following formu-
lations: (1) 0.5 percent BHC hydraulic spray, (2)

2.5 percent BHC mist blower application, and (3)
1 percent Guthion mist blower application. Treat-
ment 1 gave 93 and 85 percent control of Dioryctria
on first and second-year cones. Treatment 2 gave
only 50 and 69 percent control of Dioryctria on
first and second-year cones. Treatment 3 gave 88
and 69 percent control of Dioryctria on first and
second-year cones and 70 percent control of Las-
peyresia (slash pine seedworm). BHC has no effect
on Laspeyresia. The cost for both the BHC and
Guthion mist treatments was 48 cents per tree per
application for the chemicals alone. About 1 gallon
of spray was applied per tree with the mist blower.

John F. Coyne, of the Institute of Forest Genetics,
Gulfport, Miss., in a personal communication re-
ports a cost of $1.72 per tree per application where
he was treating individual parent trees with a 0.5
percent BHC water emulsion. He used a Buffalo
turbine mist blower mounted on a two-wheel trac-
tor; his cost figures include chemicals, labor, and
depreciation of equipment. Three applications were
made each season for a total cost of $5.16 per tree
per season. Cone survival was 70-80 percent in
treated trees and about 30 percert in untreated
trees. It is quite likely that these costs could be
reduced where similar work was being done on a
seed production area or seed orchard where the
trees are closer together.

Cone rust can cause heavy losses of slash and
longleaf pine conelets on the Gulf Coast and in
North Florida. If it is not possible to locate seed
production areas outside of the areas where cone
rust losses are likely to be heavy, the rust may
be controlled by spraying at 5-day intervals during
the time of pollination with Ferbam at the rate of
2 pounds per 100 gallons of water plus a Du Pont
spreader-sticker (Matthews and Maloy 1960). Add-
ing heptachlor (1% pints of a 2-pound-per-gallon
emulsifiable concentrate of heptachlor per 100 gal-
lons of ferbam suspension) gave significant control
of both cone rust and thrips (Southeastern Forest
Expt. Sta. Ann, Rpt. 1961, p. 30).

Regardless of the original condition of the stand,
control of understory vegetation sooner or later
becomes necessary because the release and fertiliza-
tion stimulates the understory vegetation as well
as the pines. Such control reduces competition and
makes harvesting and other operations on the area
much easier. The method chosen may be a control
burn herbicidal spray, mowing, or a combination
of these. But it should be suited to conditions in
a given stand and the ideal result would be the
lightest vegetative cover that would keep the soil
in place.

But in spite of all we do, cone crops are extreme-
ly variable. They are not always produced on
schedule the third season after release and they
do not occur consistently on the same areas even
when we fertilize, control competing vegetation,
etc. Apparently the number of flowers produced
is fairly consistent from year to year in a given
stand (but not always), and most of the variation
in cone crops is caused by climatic factors, e.g., too

46

little moisture at the time flowers primordia are
initiated, too much rain at the time of pollination,
untimely freezes and droughts, ete., and variation
in the severity of insect and disease attacks. Since
we can’t control the weather, the control of cone
insects becomes even more important in securing
harvestable crops more frequently.

Harvesting the cones economically has been a
problem in seed production area management. In
this connection the first thing to be decided is
whether or not the cone crop is heavy enough to
be worth harvesting; and an early answer to this
question makes orderly arrangements for harvest-
ing operations much easier. The maturing cones
are large enough to count by about June 1 and
several workers have developed methods of esti-
mating cone yields (Hoekstra 1960; Wenger 1953a).
In deciding whether or not to harvest a particular
seed production area, we base the decision on the
number of trees with a crop worth collecting; on
certified slash seed production areas we set a cone
count of 100 sound cones per tree as the minimum;
on the other slash seed production areas and all
loblolly seed production areas the minimum count
is 150 cones (the actual number of cones collected
is usually about twice the number counted). And
we don’t collect in areas where less than 20 percent
of the trees have a crop of this size. The minimum
acceptable cone count can be varied as seems desir-
able considering the size of the crop, how badly
cones are wanted, etc. For most purposes this count
need not be precise; all you need to know is the
number of trees with a harvestable crop. With a
little practice most trees can quickly be judged
as harvestable or unharvestable and only borderline
trees need be checked carefully.

We have done all of our cone collecting from seed
production areas by climbing for the cones, rather
than by cutting the trees. It is considerable trouble
to prepare the areas and we want to Keep them in
production as long as possible. We feel that the
extra cost of collection from standing trees is justi-
fied by the continued production of quality seed.

We have tried several methods of collection;
climbing with aluminum tree climbing ladders,
with a trailer mounted extension ladder, and with
spurs and ropes. Climbing with spurs and ropes is
the best method; it is cheaper, it is quicker, and
so far, after two seasons, there has been no tree
mortality that we can attribute to the use of spurs.
Any trees which are buggy are removed at the
time of harvest, however, so as to get rid of poten-
tial sources of infestation. With this system, the
men climb into the trees on their spurs and descend
on their ropes; this is fastest and minimizes dam-
ages to the trees. The trees that have been climbed
are marked and are being watched for beetle at-
tacks and to see how often the same trees produce
worthwhile crops.

On slash seed production areas, the cones are
pushed off with a cone hook with little difficulty
or damage to the following season’s crop. Loblolly
presents more of a problem, however, and on the

loblolly seed production areas a pruner is used and
the whole twig is clipped off. This means the loss
of the next season’s cones and the flowers for the
following season on these twigs but it seems to
be the only economically feasible way of collecting
loblolly cones from standing trees.

It is very important that the cones be ripe when
collection starts. Immature cones produce less seed
and the germination may be reduced (Speers 1962)
which increases the cost of seed.

The first two seasons that we collected from our
seed production areas, the collection was done by
contract with a tree surgery company. The first
season climbing was by means of aluminum tree
climbing ladders. Our men and the climbers were
just learning how to harvest cones; the cost of
collecting slash cones was $5.65 per bushel (table
1). This figure includes climbing, moving ladders,
picking up, sacking, and loading the cones for ship-
ment to the cone warehouse. Collection costs were
considerably lower when the second area (Meadows
Tract) was collected that year because climbers
and ground crew were more familiar with the job
and some excess ground crewmen had been elimin-
ated; but because of the marked difference in seed
yield between the two, the cost per pound of seed
was nearly the same on both areas.

The next time (1961) we collected from our seed

and ropes (detailed costs and yields are shown in
table 2). Collection costs totaled $4.03 per bushel
on the certified slash seed production areas and
$3.50 per bushel on the uncertified slash seed pro-
duction areas; the combined average cost was $3.71
per bushel. Costs differed because there was a
lower minimum number of cones on the certified
areas. Collection costs were $4.98 per bushel on
the loblolly seed production areas (costs for certi-
fied and uncertified loblolly areas were lumped
together since too small a part of the total came
from certified seed production areas to permit an
accurate comparison). Loblolly collection costs
were higher than those for slash because of the
greater difficulty of collecting loblolly cones and
because the loblolly areas were generally more
brushy. Costs were 20 to 60 percent higher when-
ever climbing methods other than spurs and ropes
were used. Seed yields averaged 0.86 pound per
bushel for cones from certified slash seed produc-
tion areas, so collection costs per pound were $4.69.
For uncertified slash seed production areas the
figures were 0.80 pound of seed per bushel and a
collection cost of $4.36. By contrast, the yield from
more than 1,000 bushels of purchased slash cones
was 0.71 pound per bushel and the cost per pound
was $1.97 (purchase price was $1.25 per bushel
and supervision, transportation, etc., added about

production areas, climbing was done with spurs $0.15 per bushel). The difference in seed yield
TABLE 1.—Slash pine cone collection costs and seed yield, 1958
Costs | Seed yields
Trees Quantity Climbing | Labor All Per | Collection
Tract collecte Per Per Per Per Total costs
from | Potal [ee Mota | bushel | otal | bushel | Total | bushel | bushel per pound
Number Bu. Bu. Dollars Dollars Dollars’ Dollars Dollars Dollars Lbs. Lbs. Dollars
Newman 136 150 Iai 879.37 5.86 199.50 1.33 1,078.87 7.19 175 17, 6.16
Meadows 185 200 leat 760.63 3.80 139.95 -70 900.58 4.50 133 66 6.77
Total or
average 321 350 1.1 1,640.00 4.69 339.45 .97 1,979.45 5.65 308 .88 6.43
TABLE 2.—Slash and loblolly pine collection costs and seed yield, 1961
SLASH PINE
| Costs Seed yields
Rees Quantity Climbing Brush control Picking up, loading, etc. All Collection
Tract collecte Per Per Per Per Per | Total Per cost per
Total| tree| Time] Total bushel] Time| Total | bushel] Time Total | bushel} Total bushel bushel] pound
Number Bu. Bu. Hours Dollars Dollars Hours Dollars Dollars Hours Dollars Dollars Dollars Dollars Lbs. Lbs. Dollars
Certified:
Robinson 189 228 1.2 139.5 592.87 2.60 15.5 43.71 0.19 114.0 171.00 0.75 807.58 3.54 194.0 0.85 4.16
Blundale 52 46 9 45.0 191.25 4.16 4.0 11.28 .25 31.0 46.50 1.01 249.03 5.41 44.0 .96 5.64
H. and P. 88 95 1.1 76.5 325.13 3.42 12.0 33.84 36 47.0 70.50 .74 429.47 4.52 81.0 85 5.32
Total 329 369 11 261.0 1,109.25 3.01 31.5 88.83 24 192.0 288.00 .78 1,486.08 4.03 319.0 86 4.69
Uncertified:
Meadows qi 65 9 113.0 503.63 7.74 - é ; 48.0 72.00 HBL 575.63 8.86 68.0 1.05 8.05
Blundale 334 528 1.6 274.5 1,166.62 2.21 12.0 33.84 06 201.5 302.25 Om) LOZ 2.85 407.0 tit 3.70
Total 405 593 1.5 387.5 1,670.25 2.82 12.0 33.84 .06 249.5 374.25 .63 2,078.34 3.50 475.0 80 4.36
Total, ;
all slash
areas 734 962 1.3 648.5 2,779.50 2.89 43.5 122.67 13 441.5 662.25 69 3,564.42 3.71 794.0 83 4.47
LOBLOLLY PINE
Uncertified 225 306 1.4 260.0 1,121.25 3.66 33.0 90.75 .30 208.0 312.00 1.02 1,524.00 4.98 288.0 94 5.30
Certified 28 25 9 : ° 3 4.98 31.0 1.24 4.02

47

between purchased cones and collected cones is due
to the better control over cone quality (ripeness,
freedom from insect injury, etc.) which is possible
on acompany job. The difference in yields between
certified and uncertified areas is probably due to
the fertilization and spraying for cone insects which
was done on the certified seed production areas.
Similar trends were evident on the loblolly seed
production areas; on the certified loblolly seed pro-
duction areas the seed yield was 1.24 pounds per
bushel and collection costs were $4.02 per pound.
On the uncertified areas the yield was 0.94 pound
per bushel and collection costs were $5.30 per
pound of seed (no loblolly cones were purchased
so a comparison with the yield from purchased
cones is not possible).

In 1962 climbing again was with spurs and ropes
but the contract was on a per tree basis rather than
a straight weekly rate for the crew as had been
the case in the past. The Seelbach Company, of
Atlanta, was the successful bidder with a bid of
$3.12 per tree for slash and $3.50 per tree for
loblolly.

The details of the costs of collection per bushel
of cones and per pound of seed for 1962 are given
in table 3. Climbing costs ranged from $1.96 to
$3.93 per bushel and total collection costs ranged
from $2.88 to $5.02 per bushel, depending on the
bushels per tree. The collection cost per pound
of seed varied from $2.81 per pound to $8.66 per
pound; this is a reflection of the combined effect
of bushels per tree and pounds of seed per bushel
(pounds per bushel varied from $0.58 to $1.16). The
yield from ordinary slash cones purchased by Con-
tinental Can Company in 1962 was 0.48 pound
per bushel; at a cost of $1.40 per bushel (price of
cones was $1.25 per bushel plus $0.15 per bushel
for transportation, supervision, etc.) the purchased
seed cost $2.92 per pound. The combined average
figures for the seed production areas are 1.4 bushels
of cones per tree, $2.24 per bushel for climbing,
total collection costs $3.16 per bushel, average
pounds per bushel 0.84, average cost per pound
$3.76. There wasn’t enough of a crop to make
collection worthwhile on the loblolly areas so we
haven’t any figures on loblolly for 1962. Climbers
can collect the cones from about 5 to 12 trees per

TABLE 3.—Slash pine collection costs and seed yield, 1962

day depending on the cone crop per tree and
whether they are working in slash or loblolly pines.

So far we have only one set of data regarding
cone collection from a loblolly seed production area
in successive years (from our Hodge, La., district)
but it is very interesting (table 4). There are 153

TABLE 4.—Cone yields'in successive years from a loblolly pine
seed production area (Hodge, La.)

Trees r Cones Cones | Repeaters '
Year Collected Av. per} per Cones
Total from Total tree | bushel |Trees per tree
Number Number Bu. Bu. Number Number: Bu.
1961 153 95 221.0 2'3 * 600 84 2.4
1962 153 134 437.0 3.3 420 84 3.2

‘Trees from which successive cone crops were collected.
* Estimated.

trees on the 1ll-acre seed production area, and in
1961 an average of 2.3 bushels of cones were col-
lected from 95 trees by clipping the twigs with a
pruner. In 1962, on the same area, 437 bushels
were collected from 134 trees for an average of 3.3
bushels per tree. Of the 95 trees from which col-
lections were made in 1961, 84 were collected from
again in 1962; the average cone yield from these
trees was 2.4 bushels per tree for 1961 and 3.1
bushels per tree for 1962. From an examination
of yield data from the individual trees, it appears
that when 3.5 or more bushels were collected from
a given tree in 1961 the 1962 yield from that tree
was reduced; but even so, the average 1962 yield
from those high yielding trees was 2.9 bushels per
tree. Thus it appears that two successive crops of
cones may be collected from a loblolly seed produc-
tion area even when the cones are clipped off. It
will be very interesting to see when these trees
will produce a crop of harvestable size again; they
look rather like plucked chickens now.

We like contracting for cone collection on a per
tree basis. It is the cheapest method we have yet
developed and since payment is on a per tree basis,
the pressure to keep the climbers moving is on the
contractor which makes supervision easier for us.
The contractor was well enough satisfied with the
arrangement to have expressed an interest in doing
it again; I feel that the costs are reasonable, con-
sidering the size of the cone crop.

Costs _ | Seed yields
Trees Quantity Climbing Brush control | Picking up, loading, etc. | All ollection
Tract collected Per Per | Per | [ Per | Per | Total | Per | cost per
Os frotal | tree | Total | bushel |Time | Total | bushel Time Total | bushel |Total bushel jbushel| pound
Number Bu. Bu. Dollars Dollars Hours Dollars Dollars Hours Dollars Dollars Dollars Dollars Lbs. Lbs. Dollars
Robinson 34 27 0.8 106.08 3.93 2.6 11.23 0.42 12s1 18.15 0.67 135.46 5.02 15.7 0.58 8.66
Blundale 54 56 1.0 168.48 3.01 4.1 V7 .32 25.2 37.80 .67 223.99 4.00 32.5 58 6.90
H. and P. 58 82 1.4 180.96 2.21 5.0 21.60 26 36.9 55.35 67 257.91 3.15 92.0 112 2.81
Sav. Town 34 44 1.3 106.08 2.41 3.0 12.96 29 19.8 29.70 67 148.74 3.38 51.0 1.16 2.92
Total :
certified 180 209 1.2 561.60 2.69 14.7 63.50 30 94.0 141.00 .67 766.10 3.67 191.2 91 4.03
Blundale
(uncertified) 212 9337 1.6 661.44 1.96 16.3 70.42 21 151.6 227.40 67 959.26 2.85 207.0 61 4.67
Grand
total 392 546 1.4 1,223.04 2.06 31.0 133.92 25 245.6 368.40 67 1,725.36 3.16 398.2 .73 4.33

If the difference in cost between seed from pur-
chased cones and those collected from our seed pro-
duction areas seems alarming, in view of the com-
bined effect of the costs of fertilization, spraying,
and collection, it should be remembered that the
cost of seed is only a small fraction of the costs of
planting an area, especially if mechanical prepara-
tion of the site is required. On areas where site
preparation is necessary, and the planting is done
by machine, the cost of seed from purchased cones
represents less than 1 percent of the total cost of
machine planting. Thus a large increase in the cost
of seed has only a small effect on planting costs.

On the other hand, a small increase in volume
or quality will, by the end of a rotation, have a
pronounced effect on the “dollar harvest’ from the
plantation. On an “average” slash site (70 foot site
index at 50 years) a 1 percent increase in volume
yield over a 35 year rotation would mean that the
cost of seed could be increased about five times and
the planter would still break even (this assumes
all of the increase is considered to be in sawtimber
at $35.00 per M bd. ft. at the end of the period and
5 percent interest is charged). And Perry and
Wang (1958) have presented calculations to show
that seed is only one-half of 1 percent superior
to the average, would, under the conditions they
have assumed, be worth an extra $4.52 per pound.
There are other factors which eventually should
reduce the cost of seed from seed production areas.
On our own seed production areas and others
certain trees produce most of the cone crop year
after year (Hagner 1958; Matthews 1963; Thorb-
jornsen 1960; Timofeev 1959; Wenger 1953b). Thus,
after three or four good cone crops on a seed
production area, it should be possible to identify
the good producers. Cultural operations could then
be concentrated on the cone-producing trees with
a proportional reduction in the cost of such opera-
tions.

In any case, the most important point is the
degree of improvement provided by the seed from
seed production areas. Easley (1963) has reported
a field test of loblolly pine seedlings from a seed
production area in comparison with ordinary seed-
lings on both sand and clay soils: “After five years
in the field the seed production area stock produced
17 percent more height growth than the nursery
run seedlings on deep sand. On the heavy clay soil
the seed production area stock produced 27 percent
more height growth than seedlings from nursery
run stock .... This study so far indicates that

49

the collection of seed from a local source of selected
parent stocks can very well be worth the effort,
time, and care required to manage a seed production
area.’”’ More recently in a personal communication
he said that after 8 years in the field, the seed
production area stock on the deep sand site was
25 percent ahead of the nursery run seedlings in
height growth. However, the difference between
the two types of stock was decreasing on the heavy
clay soil, indicating that the growth of the seed
production area seedlings was beginning to level
off on that site. He adds, ‘‘This is not unexpected
on the heavy clay soil. Slash pine seedlings in the
same test on the clay soil are superior in height
growth to both sources of loblolly seedlings; heavy
clay savannah soil is the only place where I recom-
mend slash pine over loblolly pine in the George-
town area.”

Results such as this lend a most reassuring sub-
stance to all the theoretical arguments that have
been advanced to justify the establishment of seed
production areas and seed orchards. However, it
cannot safely be assumed that the establishment
of seed production areas will automatically assure
us of a 20 percent increase in growth (or any in-
crease at all). Each seed production area and seed
orchard is a separate case and must be tested. For
this purpose, our company has made test plantings
of seedlings from our certified and uncertified
seed production areas in comparison with nursery
run seedlings on a number of sites and soil types.
It will be some time before any definite results can
be expected; and the results, for good or ill, will
depend on the quality of the stand originally chosen
for the seed production area and the care exercised
in marking the trees to be left on the area. But
we have enough confidence in the outcome that
we are continuing to establish seed production
areas, and we expect that this seed will be in
demand for a long time to come.

In summation, we can say that seed production
areas offer the quickest means of producing large
quantities of good seed and the cost of such seed
is probably quite reasonable if the stand and trees
chosen for seed production are of good quality and
good cultural practices and methods of harvest are
used to maximize cone crops and minimize collec-
tion costs. But cone crops are extremely variable
and more economical control for cone insects and
diseases is needed. Finally, each seed production
area needs to be tested to see if it is producing seed
worth the extra cost.

Management of Seed Orchards
Paul J. Otterbach |

International Paper Company, Mobile, Alabama

Managing seed orchards in the South today is
big business. The size of the seed orchard establish-
ment increases from year to year; 2,380 acres are
now devoted to producing forest seed from elite
sources in the region extending from Virginia to
Texas, 1,500 acres are scheduled to be producing
by 1966, and another 750 acres are now in planning
stages.

The development and management of seed or-
chards tasks the abilities of many foresters and
forest workers. Interest in the program reaches
high levels because of possible future benefits in
forest growth and quality. This fact, plus the lack
of full answers to the many involved problems
which arise, provides common ground for valuable
communication between seed orchard workers.

This presentation provides evidence of the coop-
erative work by foresters in tree improvment. The
results of a questionnaire, answered by many of
you in the audience, are summarized to give an
overall picture of present-day seed orchard man-
agement and development. The answers come from
37 out of 38 organizations which received the ques-
tionnaires. Among the respondents were 34 com-
panies or agencies having orchards installed or
planned as of May 15, 1963.

Size and Design

Apparently the first grafting work for possible
commercial seed orchards began during the winter
of 1954-1955. Seed orchards have been started
each year since 1955. The year 1956 marked the
high spot in initiation of orchards with 11 orchards
begun.

Today, individual orchards range in size from 4
acres to over 600 acres. The 40- to 70-acre size
class has the largest number of orchards. The
average size of all orchards approximates 65 acres.

Orchard outplanting design varies considerably.
Rectangular, diagonal, square, and combinations
of these designs plus random or mechanical loeation
of clones furnish the patterns for outplanting. The
majority of the orchards have a square design with
random location of clones. Included in the orchards
are subsections for species or geographic sources.

Loblolly pine is grown by 26 organizations; slash
pine is grafted in 18 orchards; Virginia pine is in-
cluded in 6; shortleaf pine in 4; and pond pine in
2. Two orchards plan to produce seed from plus

50

sand pine trees, and four orchards plan to develop
longleaf pine seed. The average orchard has three
subsections for species or geographic sources.

The number of clones and trees planted per acre

varies somewhat from orchard to orchard, as shown
in the following tabulation:

Orchards
(number)
Clones per acre:
1-9 5
10-14 0
15-25 22
26-40 5
41-60 2
Ramets per acre:
0-50 18
51-100 8
101-150 1
151-200 8
201-250 2

Orchard Sites

The majority of seed orchards are located on old
field sandy loams or sandy clays. Two orchards
are on clay land, and two orchards are on alluvial
soils. Clear-cut and site-prepared forest areas make
the base sites for 10 orchards. Seven areas are
located on deep sands. The average site index for
all orchards falls around 75 feet for slash pine and
85 feet for loblolly.

Grafting

Most orchard managers use the cleft and side
graft methods for propagating their elite material.
Two organizations supplement their normal graft-
ing with bottle grafting and three orchards include
attempts by air-layering. Grafting from outplanted
trees in the orchard to root stock (inarching) takes
place in eight orchard organizations. The grafting
parties use 22 shade houses of sorts, 20 field areas,
and 12 nursery beds as their locations for grafting
work.

Grafting success varies from orchard to orchard
and from man to man. Table 1 is a compilation of
grafting and transplanting success by species.

Individual clone grafting success varies from 100
percent failure to almost 100 percent success.

—————————

TABLE 1.—Success of grafting and transplanting,

based on orchard totals
Grafting success
Range | Average’

Grafting success
Range | Average *

Species
Percent Percent Percent Percent
Loblolly pine 41-80 64 60-97 85
Slash pine 60-95 "(4 60-97 85
Virginia pine 52 55 90-95 92
Shortleaf pine 51-90 65 90-95 94
Sand pine 76 76 90 90
‘ Unweighted.
Growth

The growth of grafts from select trees definitely
shows the vigor associated with the parent material.
The unweighted average age of all trees in all
orchards approximates three growing seasons. The
average grafted tree has a stem diameter of 3
inches and is 12 feet tall. The largest tree in the
southern orchards stands over 31 feet tall and
reaches 9 inches d.b.h. after seven growing seasons.
It is at Georgetown, S.C., and has had top pruning.
Table 2 gives you an inkling of the growth of plus
grafted material.

TABLE 2.—Average growth of largest
trees in seed orchard

Growing seasons P
Gans) i D.b.h Height
Inches Feet
7 6.1 ONG
6 5.5 24
5 3.9 19
4 4.1 20
Phenology

The initiation and amount of strobili develop-
ment has an important bearing on early progeny
testing and seed production for all orchards. The
questionnaire answers give the impression that
loblolly pine has a somewhat earlier seed produc-
tion potential than slash. The following tabulation
indicates year of first “flowering,” as reported by
individual orchards:

Loblolly
(year)

Slash
(year)

Earliest reported

Male Within year planted 1

Female do. 1
Average of orchards

Male 2 3

Female 2 3
Latest reported

Male 3) 4

Female 3 4

To date, clonal “flowering” varies by orchards
from 1 to 100 percent of the clones in the orchard
for male strobili production (averaging 44 percent
of clones) and from 20 to 100 percent (averaging

51

59 percent) for female strobili production. The
“sexiest”? clonal orchards, excluding the Georgia
Forestry Commission, are (1) Western Region,
International Paper Co., (2) College Station, Texas
Forest Service, (3) Georgetown Region, Interna-
tional Paper Co. The first two of the three pre-
ceeding loblolly pine seed orchards lie west of
the Mississippi River. Individual ramet ‘‘flowering”’
as measured by the number of ramets having more
than 100 male clusters or female strobili separates
the men from the boys in the following manner:

Male strobili.i—1. University of Florida, + 300
slash pine ramets, 2. College Station, Texas Forest
Service, + 200 loblolly pine, 3. Savannah, Union
Bag-Camp Paper Corp. + 200 slash pine.

Female strobili.m1. Chesapeake Corp., + 400 lob-
lolly pine, 2. Gulf Region, International Paper Co.,
80 slash pine, 3. College Station, Texas Forest
Service, 62 loblolly pine.

Almost half of the orchards do not have ramets
with more than 100 hundred male clusters or female
strobili as yet. However, controlled pollination for
progeny material is taking place in 27 orchards.
Six orchard organizations pollinate for hybrid
study, and one orchard makes crosses for disease
resistant material. Buckeye Cellulose Corp. experi-
ments with ultra violet ray treatment of plus seed
to obtain possible mutations having worthwhile
characteristics.

Mortality After Outplanting Success

Hottes (1934) quotes Dr. L.H, Bailey (circa
1890) saying, in essence, that grafting accomplishes
one of the four following things: (1) increases
flowering, (2) decreases flowering, (3) has no
effect, (4) shortens the life of the plant.

We notice the fourth rule today in the consider-
able mortality apparently caused by phloem block-
age due to stock-scion incompatibility and/or poor
grafting technique. The questionnaire results show
that an average of 11 percent of the clones in the
present orchards suffer from phloem blockage and
about 7.5 percent of the total ramets are dead due
to graft union incompatibility (excluding the Geor-
gia Forestry Commission). Bailey’s axiom appar-
ently holds for the mortality angle. I wonder what
influence grafting has on increasing or decreasing
“flowering” in our southern pines.

Drought (480 trees), insects (+ 430), wind
(+ 400), disease (+ 360), and temperature (+ 260),
reduce the survival rate of grafted trees by about
2 to 5 percent overall. Ice (19 trees) and lightning
(6) add some insult to the mortality picture. Again,
data from the Georgia Forestry Commission is not
included and could change the information as
presented.

Cultural Practices

Orchard culture varies with the site and prob-
lems of each individual orchard. The following
information and some side comments give an idea

of the different current practices of orchard organ-
izations (number of organizations in parentheses):

1. Mulching (14): sawdust (8), pine straw (5),
hay or straw (2), plastic (2), used papermakers felt
(1), paper (1).

Most of the mulching is done for the first 1 or 2
years after outplanting.

2. Watering (17): mostly during critical periods
in the first year of outplanting.

The Panama City Region, International Paper
Co., supplements rainfall to insure a minimum of
1 inch of water per week on 2 acres in an attempt
to evaluate the effect of balanced rainfall.

3. Mowing (28): frequency is shown in the fol-
lowing tabulation:

Mowings
per year Organizations
(number) (number)

a 2

2 6

3 4

3-4 13

4 3

None 4

4. Weed control (14): Weeds can be a problem
in certain areas and increase fire hazard if not
checked. Seven organizations use chemicals to
combat weeds, seven apply the old proven method
of hoeing by hand, one scalps once a year to remove
vegetation. Most orchards rely on mowing and
grass invasion to keep weeds under control.

5. Discing (4): Four orchard managers disc
their areas with harrows or furrow discs every year
to till the soil and turn in the vegetation.

6. Cover Crops (11): crimson clover (3), Ber-
muda grass (3), Bahia grass (2), Dutch clover (1),
hairy indigo (1), Korean lespedeza (1), ryegrass
(1). Cover crops reduce the frequency of moving,
can choke weeds to some extent, and provide nu-
trients to the soil.

7. Pruning (most orchards): Many orchard man-
agers prune a portion or all of the stock branches
of each outplant to force scion growth. Disease
infected branches are removed or pruned in all
orchards. Two small orchards have a_ pruning
study underway to determine the effect of various
pruning techniques on ‘flower’ production.

8. Fertilizers—The value of fertilizing orchard
trees is still nebulous, except for height growth,
according to the 26 answers from orchards using
fertilizers. Balanced NPK fertilizers are applied by
21 orchards; ammonium nitrate-limestone is applied
by 7 organizations. Muriate of potash (1), lime (1),
0-14-14 NPK (1), and dolomite (1) are also applied
in certain orchards.

Fertilizer quantities per application range from
1 ounce to 2 pounds per tree depending on tree
size and fertilizer type. One pound of balanced
fertilizer per 100 square feet of horizontal crown
surface is in use as a guide for fertilizer amounts

52

in quite a few orchards. Heaviest application in any
one orchard amounts to 2,000 pounds of balanced
fertilizer per acre, per year. The average dosage
per orchard seems to be 400 pounds of 8-8-8 or
10-10-10 NPK per acre spread around each tree
area once or twice a year.

Only seven orchard managers make mention of
any noticeable tree’ response to nutrient additives.
The following tabulation lists the results of the
answers on fertilizer for 8 orchards out of the 26
that use them.

Organizations
(number)
Clonal 2
Individual trees 2,
Increased flowering 7
Increased height growth 5
Decreased height growth 1

9. Fire Protection (24): Most orchard men
maintain fire lines around their orchards as a pre-
ventive fire measure. Four managers strengthen
the fire lines with strip burning. One organization
burns the orchard each year, a practice, although
hazardous, offering maximum security.

Insect Problem and Control

The following insects listed in order of magni-
tude make seed orchards their habitat and pose a
sometimes serious or nuisance problem for the
orchard manager: 1. Nantucket pine tip moth,
2. coneworms (Dioryctria spp.), 3. red spider
mites, 4. aphids (Cinara spp.), 5. thrips (mostly
Gnophothrips piniphilus), 6. midges (gall midge
larvae), 7. seedworms (Laspeyresia spp.), 8. scale
insects (Chionaspis spp.), 9. ants, 10. sawfly (Neo-
diprion spp.), 11. black turpentine beetle, 12.
grasshoppers, 13. Japanese beetle.

Table 3.—Use of insecticides in seed orchards

Insecticides Organizations | Range in Strength

Number Percent
Malathion 22 0.1-1.0
DDT 18 .5-1.0
BHC IEF .25-.75
Guthion 3 “2,
Heptachlor 2 al
Sevin i 8
Di-syston i 10
Thimet 1 10
Kelthane 1 t
Aldrin 2 .2-.5
Volck 1

Experience to date seems to indicate that mala-
thion, DDT, and Guthion give excellent control of
tip moth when applied at the proper times. Gu-
thion appears to have a longer residual effect but
is much more dangerous to use. Heptachlor, Gu-

thion, and BHC appear effective on “flower” in-
sects but not much is known about their phytotoxic
effects. BHC works well in controlling bark bee-
tles, coneworms and seedworms, depending again
on the right time for application. Aldrin should
be effective against most larvae and beetles but
has a high human toxicity number. The systemics,
Di-syston and Thimet, seem to hold some promise
for tip moth control.

Application times for chemicals range from once
a week to once a year depending on the need and
the chemical. Eighteen orchards spray on a regu-
larly scheduled basis. Several orchard workers
rear Nantucket pine tip moth in small enclosures
and spray their orchards when the captive adults
emerge and begin flight. I think the same thing
can be done with other insects, especially Dioryctria
spp.

A number of orchard managers feel that the
coneworms, thrips, and red spider mites pose a
large problem for the future. Eight orchards list
Dioryctria spp. today as their number one insect
problem. I think the Nantucket pine tip moth will
continue to plague the orchards because of its
affinity for fresh branchlets which could produce
“flowers.”

Disease

Fusiform rust (Cronartium fusiforme) heads the
list of diseases striking seed orchards with cone
rust (Cronartium strobilinum) far down in second
place. Most of the fusiform rust infection appears
in the branches on the stock of the grafted material
with some attacks occurring in the graft unions.
Cone rust strikes conelets in seven orchards located
in Georgia and Florida. One orchard finds a prob-
lem with needle cast. Fomes annosus in one or
possibly two orchards causes concern.

Pruning infected areas and spraying with ferbam
(usually at the rate of 2 pounds ferbam per 100
gallons water during periods of high telial spore
formation and flight) for containing and preventing
fusiform rust and cone rust forms the basic control
practice in the majority of orchards. Radical sur-
gery of the stem canker of fusiform rust seems
to be successful in certain instances. At least the
trees involved either survived, or died very quickly
instead of slowly. Champion Paper Co. applies
phytoactin and grafting wax to the cut area of
the bole in their radical surgery method. Ferbam
sprays are in use for the control of needle cast.
In dealing with Fomes annosus, Hiwassee Land Co.
treats the area around each uprooted tree with
methyl bromide gas at the rate of 11 pounds per
square inch.

A recent recommendation of considerable worth
calls for applying borax powder or spray to freshly
cut pine stumps within the orchard area. The
complete lifting of the dead stem and root system
is suggested as a sanitation procedure for Fomes
annosus in orchards where dead grafts or failures
in the field are problems.

53

Birds

Two orchards have dealt with the yellow-bellied
sapsucker who enjoyed nibbling on elite material.
Several orchards have placed poles or bamboo
canes as bird perches in an attempt to reduce pine
leader breakage or disfigurement by bluebirds and
other winged orchard visitors. I am not sure if
the results were impressive.

Equipment

Apparently tractors and mowers form part of
the basic equipment of each orchard. Sixteen organ-
izations furnish trucks for use in their orchard
programs. Supplemental water is applied to 13
orchards from portable tanks. Two orchards have
sprinkler systems and six orchards make use of
ditches or dikes for irrigation or drainage. Pres-
sure sprayers of varied manufacture, including
homemade, furnish the means of distributing chem-
icals for 15 orchards. Eight orchards use back
pack mist blowers and two orchards rely on the
truck-mounted mist blower.

The development of rigs for reaching high branch-
lets is quite interesting and needs more study and
design.

The types of reaching equipment now in use
(and the number of orchards using them) are:
ladder (7 plus others not answering questions);
ladder with bicycle wheels (2); ladder, truck
mounted (4); ladder, trailer mounted (3); ladder,
tractor mounted (4); and elevated platforms on
truck (4).

Costs

The costs of managing seed orchards can be
hidden from view by the simple expediency of not
mentioning them. The limited number of answers
furnished by the questionnaires prove that seed
orchard costs are too trivial to list, too high to
recall, or simply unknown. The range of rough
cost estimates furnished by the questionnaires is
from 15 to 350 dollars per acre per year. My esti-
mate of annual expenses for the average seed
orchard would be in the neighborhood of 40 to 50
dollars per acre which, in turn, could be higher
depending on the size of the spraying and cross
pollination programs.

Research

Research aids the seed orchard manager in finding
answers to some of his problems. For your con-
sideration and thought the following is a partial
listing of some specific areas needing basic or ap-
plied research:

A. Physiology
1. Stimulation of “flowering”’

a. Water requirements

b. Nutrient requirements

ec. Other physiological developments or re-
quirements
(1) Climate

(2) Mechanical bending,
pruning
(3) Catalysts (biochemical)
d. Genetic influences
2. Graft incompatability
a. Healing process and tissue development
(1) Translocation of water and nutri-
ents
(2) Genetic influence
(3) Catalysts (biochemical control)
b. Graft techniques
(1) Effects of different types of grafts
(2) By-pass methods
B. Phenology
1. Reasons for seasonal variations in time and
amounts of ‘‘flowering”’
a. Climatic
b. Soil
C. Propagation
1. Improvements in propagation and trans-
planting methods
a. Rooting
b. Grafting
c. Catalysts (hormones)
D. Methods of breeding
1. Efficient and simple handling,
and storage of pollen

girdling or

extraction,

54

2. Efficient low cost isolation material

3. Development of equipment to mix and dis-
tribute large quantities of pollen if necessary

4. Development of multi-use equipment for
reaching

E. Insect and disease control

1. Continued research for non-phytotoxic chem-
icals with high residual effects

Systemic chemicals for insects or disease
Application systems

Natural enemies of insects and disease
Knowledge of genetic control of insects and
disease.

on Bo ty

Perhaps, in time, enough answers and develop-
ments will furnish aid to the orchard manager for
maximum yearly seed harvests. A conservative
estimate, based on the 4,500 acres of seed orchards
which will exist after 1967 and a future yield of
15 bushels of cones per acre per year, will forecast
the availability of enough plus seedlings to plant
almost 700,000 acres per year in 20 to 30 years or
less.

The answers to the questionnaires definitely
emphasize the fact that we are entering the large
scale area of elite forest seed production. Forest
seed orchards are big business now,

Juvenile-Mature Tree Relationships
Charles D. Webb

Southeastern Forest Experiment Station
Forest Service, U.S. Department of Agriculture, Asheville, North Carolina

The best criterion for selecting geographic races,
progenies, or individual trees will be their perform-
ance record based on periodic measurements from
the time of planting through rotation age. Many
years and great expense will be required to obtain
this record. Our present individual tree selections
are based on performance at maturity, which is a
very valuable record in genetic evaluation, but
these selections must be tested to reaffirm their
superiority if this degree of accuracy is required.
Tree breeders have recognized the opportunity for
and have written much about speeding up this test-
ing process by predicting mature tree performance
from juvenile performance. Little concrete evi-
dence has been presented yet to show the exact
nature and strength of juvenile-mature tree rela-
tionships, but records being made now of juvenile
performance in progeny tests will soon provide
valuable information on them.

The objectives of this paper are: (1) to discuss
briefly the problems presented by juvenile selec-
tion, i.e., how the stages of plant growth affect
the reliability of juvenile evaluation, and (2) to
outline some of the current information on juvenile-
mature tree relationships for the southern pines.

The Problem

The Stages of Plant Growth

Sax (1958) has divided the stages of plant growth
into embryonic differentiation, juvenile develop-
ment, maturity, and old age. Excepting the end
of embryonic development, there are no clear-cut
distinctions between successive stages, and all parts
of the plant do not change from one stage to the
next simultaneously.

The existence of these stages in trees and shrubs
has not always been recognized, and there are in-
stances in taxonomy where mere juvenile forms
have been erroneously elevated to specific or
varietal status (Schaffalitzky 1954). Cuttings from
plants in the juvenile stage typically root and graft
more easily than cuttings from older plants (Schaf-
falitzky 1954, Sax 1962). Generally, trees and
shrubs are sterile during juvenility, although some
species and individuals within other species produce
flowers as early as 1 or 2 years (Greene and Porter-

‘Personal communication from A.E. Squillace, Apr. 29, 1963.

field 1963). The wood near the pith of the tree has
been variously referred to as “juvenile wood” or
“core wood,” and its properties differ from those
of wood produced farther from the pith (Zobel et
al. 1959). We are all familiar with the S-shaped
height-growth curve and its typical trend during
the years immediately following establishment.

Obviously, the stages represent varying physi-
ological processes in the life of the tree. Many dif-
ferent gene complexes control these processes, and
their expressions are in turn affected by environ-
mental variations.

Reliability of Juvenile Evaluation

There are numerous factors affecting the reliance
that we can place on the evaluation of juvenile
performance. Among these are (1) the number of
progenies being tested in relation to the accuracy
required, (2) the strength of the offspring-parent
heritability estimate, and (3) the age at which
expression of a particular trait is critical.

Number of progenies in relation to accuracy re-
quired.—Numerous statistically significant differ-
ences among progenies and seed sources have been
reported for embryonic and juvenile characters,
such as embryo size (Vincent 1957), height growth
(Wakeley 1961), rooting habits (Snyder 1961),
photosynthetic rate (Reines 1963), and dry-matter
content of needles (Schmidt 1957). However, suf-
ficient time has not elapsed to allow many of these
differences to be related to meaningful values near
harvest age.

Furthermore, a statistically significant juvenile-
mature tree relationship is not necessarily a useful
one. If a small number of progenies are being com-
pared for juvenile performance, a higher degree
of accuracy is required and mere statistical signifi-
cance may not be strong enough. However, if a
large number of progenies are being compared, a
statistically significant juvenile-mature tree rela-
tionship can be used but with the realization that
some poor progenies will be accepted and some
good progenies rejected.

Strength of offspring-parent heritability.—Squil-
lace’ suggested that “... if parent tree evaluations
are available, including good controls, such infor-
mation can and should be used in genotypic evalu-

ation as well as juvenile performance.” For traits
exhibiting a strong offspring-parent heritability,
careful measurements made on mature parent trees
can add greater reliability to the evaluation of
juvenile performance within 10 years or so after
planting.

Age of expression.—Certain traits are critical
during the juvenile stage and can be evaluated then
on the assumption that they will exhibit a strong
juvenile-mature tree relationship. Resistance to
drought, brown spot, and fusiform rust are ex-
amples of such traits. Height growth and many
other traits are expressed over a much longer period
and will require continued scrutiny.

At this point, the true value of juvenile perform-
ance may be stressed by coining the phrase ‘‘ju-
venile rejection and mature tree selection.” The
juvenile performance of progenies and races is an
integral part of a test. If a progeny or race does
not perform well for traits critical at an early age,
then it can be eliminated quickly from further
costly consideration.

But what are the risks involved in using juvenile
performance to predict traits critical at a later age?
Many of us are faced with making decisions that
cannot wait for the 20-year results. A review of
current information on some traits may partially
answer this.

Current Information

Current information on juvenile-mature tree re-
lationships will be discussed primarily from the
standpoint of evaluating progenies or groups of
individuals rather than selecting individual trees
within progenies. Callaham and Duffield (1963)
postulate that ‘‘juvenile-mature correlations in
height growth (of ponderosa pine) may be quite
significant for predictions of growth of progenies
but not for predictions of growth of individuals
within progenies.’ This will probably be true for
other traits as well as height growth.

For the purpose of this discussion, most of the
traits in which we are interested can be grouped
into two broad classifications: those affecting vol-
ume per acre, and those affecting wood quality.

The variables affecting volume per acre are:

1. Height
2. Diameter breast height
3. Form class
4. Competitive ability
a. Photosynthetic efficiency
b. Crown form
c. Root competition
5. Disease and insect resistance
6. Drought, heat, and cold resistance

The variables affecting wood quality are:

1. Straightness

2. Specific gravity

3. Tracheid and fiber lengths
4. Cellulose content

? Personal communication with Keith W. Dorman April 29, 1963.

ol

Percent summerwood
6. Self pruning

a. Branch angle

b. Branch diameter

c. Number of branches

d. Photosynthetic efficiency
7. Oleoresin yield

Volume Per Acre

Height growth.—The juvenile-mature tree rela-
tionship for height growth has received more atten-
tion than any other characteristic, but an exact
relationship has not been established. Numerous
juvenile-mature correlations have been reported
for height (table 1), but r° provides a gauge for

TABLE 1.—Juvenile-mature correlations for height

Material | Ages 1 2
studied | Authors | correlated| 7 us a

Races:
Ponderosa pine (Squillace and 2-30 0.85 0.72 10
Silen 1962) 3-30 .75 56 10
4-30 48 323 10
5-30 65 42 10
6-30 .69 48 10
11-30 86 74 10
20-30 86 .74 10
Scotch pine (Schreiner 2-13 61 .37 25
et al. 1962)
European larch (Genys 1960) 4-12 74 155 36
Progenies:
Slash pine (Squillace *) 8-14 .79 -62 8
8-14 82 67 8

Individual trees:

Slash pine (Barber 1961) 1-7 53 .28 1058
2-7 12 52 1058

2-8 67 45 1433

3-8 .78 61 1433

Ponderosa pine (Callaham and 12-20 83 .69 --
Duffield 1963) 12-20 15 57 --

12-20 65 -42 --

12-20 67 45 --

‘All values of r are significant at the 1 percent level except those
for progenies, which are significant at the 5 percent level.
“Squillace, A.E. Unpublished data of the Southeast. Forest Expt.
Sta. on file at Olustee, Fla. April 1963.

the reliability which can be placed on these corre-
lations. Although the correlation coefficients are
statistically significant, the amount of the variation
in mature height which is accounted for by juvenile
height is still relatively small. Some of the authors
cited presented several correlations at different
ages but only the earliest ages showing a significant
juvenile-mature tree relationship are given.

Dorman’ has expressed the opinion that the
length of time required to reach a height equiva-
lent to two logs (i.e. 32 feet) may provide a usable
compromise for some purposes to long-term evalu-
ation of height growth and tree quality. By 8 to
12 years the tree has attained a height which will
represent a majority of its final volume. This
will probably be more applicable for short-rotation
predictions than for sawtimber, although there
seems to be no generally accepted rotation age for
southern pines.

Diameter breast height.—Although differences in
diameter growth are significant among young slash
pine progenies, evaluation cannot be considered
reliable until after the base of the crown has pro-
gressed above 4.5 feet and crown competition has
begun.’ Squillace * reported correlation coefficients
of 0.72 and 0.96 between d.b.h. at 8 years and d.b.h.
at 14 years for slash pine progenies. These were
significant at the 5 and 1 percent levels respectively
(Gia 56)):

Form class.—I have found no references that
mention the juvenile-mature tree relationship for
form class, absolute or Girard. Certainly absolute
form class will mean little until crown closure
has been in effect for some time, and measurement
of Girard form class should be delayed until the
base of the live crown is above the first 16-foot log.

Competitive ability—Competitive ability is a
composite character, and it will be difficult to
ascribe to it a juvenile-mature tree relationship.
Yet, it may eventually be possible to relate com-
petitive ability to maturity with demonstrated
juvenile differences among progenies in photosyn-
thetic rate (Reines 1963; Wyatt and Beers’) and
crown form (Barber;’ Trousdell et al. 1963). Root
competition will be one of the most complicated
and difficult components of competitive ability to
determine. Unless differences in rooting habits of
seedlings can be related to mature tree perform-
ance, root competition cannot be practically evalu-
ated.

Disease and insect resistance.—Southern forest
geneticists are interested in combating diseases and
insects that are critical early in the life of the
tree; e.g., fusiform rust, brown spot, and tip moth.
Because they are critical in the juvenile stage,
the mature tree performance may be relatively
unimportant.

In contrast, littleleaf disease usually does not
appear until a later age. Although Zak (1961)
found differences in susceptibility of loblolly and
shortleaf seedlings to Phytophthora cinnamomi
Rands, his laboratory tests have not yet been
related to actual field performance.

Drought, heat, and cold resistance.—Resistance
to drought, heat, and cold are among the character-
istics which are critical during the juvenile stage.
The short-term approach can be taken to test the
ability of progenies to become established in spite
of drought, heat, and cold. However, testing for
growth rate under adverse conditions, especially
drought and heat, will probably require the long-
term approach.

Of the variables affecting volume per acre, the
very important factors of resistance to certain dis-
eases and insects and to drought can be evaluated

at an early age, if the proper environmental condi-
tions exist.

Wood Quality

There is more room for optimism in early testing
for wood quality than is presently evident for
volume per acre. The ability to correlate wood
produced in early years with mature wood has
provided an excellent opportunity to establish reli-
able relationships.

Straightness.—Perry (1960) reported striking
differences in straightness and crook among lob-
lolly pine progenies as early as 2 years in the field.
Barber‘ found the same to be true for slash pine
up to 8 years of age. However, for all the southern
pines, straightness probably can be evaluated best
on young trees around 30 feet tall, representing
ages from 8 to 12 years. As mentioned with respect
to height growth, this height is the first two 16-foot
logs, and represents a major part of the volume of
the tree. If a tree is straight up to around 30 feet,
it will very likely continue to be straight to ma-
turity. However, slightly crooked trees may smooth
out and apparently become straighter by eccentric
growth with resulting compression wood.

Specific gravity.—The juvenile-mature tree cor-
relation for specific gravity is complicated by the
presence of compression wood in seedlings. How-
ever, Brown and Klein (1961) found highly signifi-
cant differences in specific gravity among 2-year-
old seedlings of loblolly pine produced by crossing
various combinations of high and low specific
gravity parent trees. These seedlings reflected, to
a high degree, the specific gravity of the parent
trees. Zobel et al. (1960) reported highly signifi-
cant correlations between juvenile or core specific
gravity at breast height and the specific gravity of
whole loblolly and slash pine trees. Forty slash
pines showed a correlation coefficient of 0.798 and
14 loblolly pines showed a correlation of 0.890, both
being significant at the 1 percent level. Together,
these results indicate that a rough screening of
progenies for specific gravity can be carried out
on young seedlings, and a fairly accurate evaluation
of specific gravity in progeny tests can be obtained
at about 10 years of age.

Tracheid length.—Kramer (1957) presented data
on loblolly pine that strongly discouraged the use
of breast height tracheid length at two or three
rings from the pith for evaluating the tracheid
length of the mature tree. He suggested using the
10th or a later ring from the pith to denote the
tracheid length of a particular tree. This means
that progeny tests should be at least 10 years of
age before any rough screening for tracheid length
is attempted. Preferably, tracheid lengths should

*Barber, John Clark. An evaluation of the slash pine progeny tests of the Ida Cason Callaway Foundation (Pinus elliottii

Engelm.). Ph. D. Diss., Univ. Minn. 206 pp., illus. 1961.

4 Squillace, A. E. Unpublished data of the Southeast. Forest Expt. Sta. on file at Olustee, Fla. April 1963.
*Wyatt, W.R., and Beers, W.L., Jr. Growth chamber analysis of wind-pollinated plus tree progeny—slash pine (P. elliottii

Engelm., var. elliottii). Unpublished manuscript. 1963.
°Op. cit.
‘Ibid.

be measured from several rings to account for
year-to-year environmental variations.

Cellulose content.—For the present use of tree
improvement, the complex field of cellulose chem-
istry can be divided into water-resistant carbohy-
drates and alpha cellulose. These are two important
constituents used to characterize wood chemically
in pulp and paper manufacture (Forest Biology
Subcommittee 1960). There was a significant cor-
relation between water-resistant carbohydrates in
the core wood zone at breast height and water-
resistant carbohydrates for the whole tree (Zobel
et al. 1960). However, the relationship between
core wood alpha cellulose and alpha cellulose for
the whole tree was consistently non-significant.
This indicates that a rough screening for high pro-
ducers of water-resistant carbohydrates might be
attempted at 10 years of age, but evaluation of
alpha cellulose must be delayed until later.

Percent summerwood.—Measurements of per-
cent summerwood during the juvenile stage have
little meaning because of the erratic and poorly
defined summerwood in the core wood zone (Zobel
et al. 1959). The seventh or eighth ring is prob-
ably as early as this characteristic can be meaning-
fully measured, and measurement should extend
over several rings. Consequently, evaluation of a
progeny test for percent summerwood will have
to be delayed until 12 or 13 years of age.

Self-pruning.—Self-pruning is another complica-
ted composite character, and the juvenile-mature
tree correlations of its components are still un-
certain. Photosynthetic efficiency was mentioned
earlier. It is very obvious in the field that 3- to
7-year-old progenies differ widely in branch angle,
diameter, length, and number of branches ( Barber;*
Trousdell et al. 1963). This may be the best age to
evaluate these traits, but their exact relationship
to self-pruning remains unknown. The existence
of an auxin gradient is suggested by Barber’s
report’ that certain slash pine progenies begin to
show self-pruning well before crown closure. This
is further indicated by the results of VanHaverbeke
and Barber (1961) showing that branches bent to
a horizontal and to a downward position showed

8 Ibid.
®* Ibid.

58

50 percent less elongation than branches left in
their normal upward position.

Oleoresin yield.—Oleoresin yield is included
under wood quality because it is closely related to
certain anatomical variations. A micro-chipping
method, developed for determining oleoresin yield
on young trees, shows heritabilities of yield ranging
from 45 to 90 percent based on 14-year-old pro-
genies (Squillace and Dorman 1961). Results
should soon be available to show the relationship
between yield at various ages according to this
micro-chipping method.

Summary

Selection for mature characteristics on the basis
of juvenile performance cannot be as accurate as
selection based on the mature tree. Yet, indications
are that certain time-saving and essential informa-
tion can be obtained from juvenile development.
The fact that plants pass through different stages
of growth affects the reliability that we can place
on juvenile performance. In comparing a small
number of progenies, a strong, accurate juvenile-
mature tree relationship is needed, but a weaker
relationship can be successfully used in screening
larger numbers of progenies. Characteristics exhib-
iting a strong offspring-parent heritability can be
selected for or against fairly soon, and other charac-
teristics that are critical at an early age can also
be evaluated at an early age.

Current information on factors affecting volume
per acre indicates that resistances to certain dis-
eases and insects and to drought, which are critical
at an early age, can be evaluated during the
juvenile stage. Other components of volume per
acre such as height, diameter, form class, and
competitive ability will require scrutiny for a
longer period of time.

Of the variables affecting wood quality, the very
important traits straightness and specific gravity
offer encouragement for testing on the basis of
juvenile performance. Other traits, such as tracheid
length, cellulose content, percent summerwood,
self-pruning, and oleoresin yield, will require eval-
uation at 8 to 10 years of age or later.

Economic Considerations of the Genetic Approach

laa OF

Virginia Division of

I feel that the following quote, from a company
forester in the proceedings of a tree improvement
conference, is apropos: “We have spent consider-
able time and money on a tree improvement pro-
gram. We have no assurance that this will pay off;
however, we believe in it firmly enough to continue
every effort in this direction.”’ I must admit I view
this quote with mixed feelings. However, this
forester may be speaking for many others and is
honest enough to express his true feelings. I do not
happen to agree with his statement concerning
whether or not the tree improvement program
will pay off because I happen to believe that all
this work being done in the field of tree improve-
ment is worth the time and money. I would like
at this time to present some facts and figures to
substantiate why I so believe.

At the outset the speaker wants to point out that
the genetic approach used and referred to in this
talk will be the so-called clonal seed orchard ap-
proach widely used in the South today whereby
good phenotypes are selected for parents and
grafted into the seed orchards. It is assumed that
progeny testing will start immediately so that the
genetic worth of these selected parents may be
tested and assessed as soon as possible. The progeny
test results will then be used to remove or rogue
those parents proven inferior. Quite frankly if it
were not for a direct action program such as this
it is doubtful whether or not the organization I
work for would be involved in a tree improvement
program. We have a critical need for seed right
now and want production orchards as soon as pos-
sible. We recognize that the immediate gain or
improvement from these production orchards will
not be as great as later gains. However, by moving
ahead with these production orchards we will be
producing seed at an earlier date and in the speak-
er’s opinion this seed will offer considerable im-
provement over the seed in present day use. Qual-
ity improvement in seed orchards is really a never
ending job and as time goes on we will be con-
tinually upgrading the genetic quality of our seed
produced by our seed orchards.

We do know that establishing and managing seed
orchards is not an inexpensive operation. Most of
us have learned a long time ago that we do not
get something for nothing, so costs attendant to
seed orchard establishment and management must
be expected. I will elaborate no further on seed
orchards costs; each organization must figure its

Marler
Forestry, Charlottesville

59

own. However, it should be remembered that seed
orchard costs must be prorated over the entire life
expectancy of the orchard and since this may be
as long as 60 years (Dimpflmeier 1954) this will
reduce the cost of seed per pound considerably.

Scarcity of Seed

We in the Virginia Division of Forestry are
becoming increasingly aware of seed shortages.
Each year, for us, loblolly pine seed is becoming
more difficult to obtain. Let me cite our experience
in 1962. Our reforestation division, in order to
supply loblolly pine seed for our nursery (we have
a loblolly pine nursery production of approximately
30 million seedlings annually) and direct seeding
needs, wanted to collect a minimum of 7,000 bushels
of loblolly pine cones. In order to secure these
cones the State Forester asked that cone collection
be placed second in work priority only to fire. An
all out effort was made to secure these cones but
in spite of this we were only able to collect a dis-
appointing 1,200 bushels of loblolly pine cones.
Seed yield per bushel of these cones was only 0.8
pound whereas normally we expect a seed yield of
approximately 1 pound per bushel. Not only were
loblolly pine cones scarce in Virginia last year
but what seed was available was eagerly sought
by other organizations. Competition for existent
seed is keen in Virginia and from all indications
will continue to be keen for many years to come.
Our reforestation division estimates that our lob-
lolly pine seed needs for Virginia Division of
Forestry use alone projected for the next 5 years
will be nearly 10,000 pounds each year. Nursery
demands and direct seeding demands have made it
necessary to carefully limit seed sales and as a
result many Virginia landowners are unable to buy
local seed from our organization for direct seeding.
You may be interested to know that we sell repel-
lant-treated loblolly pine seed for $6 per pound
(based on dry weight) and that last year we were
unable to fill orders for hundreds of pounds of seed.
One alternative for those persons wanting locai
seed but unable to obtain it is to buy and use non-
local seed, which many Virginia landowners did
last year. We do not think this advisable and I
personally shudder at not only the immediate but
long-term implications of using non-local seed, but
this is what is happening. I should like to make
a point: for us, local seed is scarce and is in strong
demand. Seed orchards, once in production, will
assure ample supplies of local seed for our use.

Cone Collection Costs

Once seed orchards are established and producing
seed, will it prove costly to collect cones? Evidence
is accumulating which indicates that seed orchard
cone collection costs will not be exorbitant and
may compare favorably with seed collection under
present-day methods. This evidence is provided by
those who have kept seed production area seed col-
lection costs. Before mentioning these costs, I
would like to tell you of our Virginia Division of
Forestry present cone collection and purchase meth-
od. We buy from local cone collectors and pay
$2.50 per bushel for loblolly pine cones. Under
normal conditions a bushel of loblolly pine cones
will yield us 1 pound of seed. However, this $2.50
per bushel or per pound of seed, exclusive of ex-
traction costs, etc., does not represent a true cost of
seed per pound, which in some instances is much
more. Not included in this cost are certain overhead
costs, cost of locating cuttings and securing per-
mission from the owner to collect, cost of inspecting
the tops for cone ripeness and quality (used only
in the sense of the cones not being damaged),
inspection costs at pick-up points and cone trans-
portation costs. Also, oftentimes it has been our
experience that cones will be collected before ripe
enough, thereby increasing extraction cost and de-
creasing seed yield per bushel of cones. I should
like to emphasize that we try to maintain rigid
standards with respect to our cone collections and
that this adds to our costs. In 1962, exclusive of
other costs mentioned above, loblolly pine seed cost
us $3.12 per pound for what seed we could obtain.

Seed collected from seed production areas has
not proven to be overly costly and provides us with
some notion of what seed orchard seed collection
costs may be. Quoted below are collection costs
from standing trees; the cones were collected by
climbing. Goddard (1958) reports that in Texas
the average cost of collecting from tall standing
loblolly pines was $4.77 per bushel of cones and
that these cones yielded approximately 1% pounds
of seed. Therefore, the cost of seed per pound was
$3.18. Cole (1962) reports that in collecting from
standing loblolly pines in a seed production area in
Georgia, seed cost was $3.88 per pound from certi-
fied areas and $4.58 per pound from uncertified
areas. Cole further points out that superior seed
yields from slash seed production areas were ob-
tained versus cones purchased on the open market.
Sweetland in private communication reported that
in 1961 from a 65-acre seed production area in
Prince George County, Va., 520 bushels of loblolly
pine cones were collected from 311 pines. The seed
yield was 679 pounds, and the cost per pound of
clean seed amounted to $5.41. This cost figure
includes charges for picking (through contract with
a tree expert company), measuring, sacking, thresh-
ing and cleaning, supervision, and transportation
incidental to the harvesting. Sweetland went on
to say, ‘““we think these costs can be lowered con-
siderably by improving harvesting techniques.” It
should be remembered that the costs reported
above are for climbing pines of considerable height

60

and that the seed production areas had remaining
some 18 to 20 trees per acre. Within our seed
orchards many more trees per acre will be avail-
able for climbing and collecting purposes, thereby
travel time to the tree should be less. Also, we
should be able to better control when to start col-
lection so that cones will be mature when harvested
and this, in turn, should result in lower seed extrac-
tion costs and higher seed yield per bushel of cones
collected. These are all very real economic con-
siderations for us to keep in mind. Furthermore,
I have confidence that we will develop and devise
more efficient and easier means of collecting cones
from standing pines. This will tend to lower seed
costs even more.

Gains or Improvements

What basis do we have for making any claims
for immediate gains through seed orchards? Evi-
dence is accumulating daily which indicates that
considerable improvement may be expected through
seed orchards. Some means of providing improve-
ment are:

ie Through better adaptation.— In a classic
loblolly pine study (Wakeley 1944) it was found
that stock from seed collected within 50 miles of
the planting site produced 1.8 to 2.7 times as much
merchantable pulpwood in 22 years as did stock
from seed collected 350 to 450 miles from the
planting site. The potential growth lost by using
Arkansas seed instead of local Louisiana seed was
1.2 cords per acre per year.

Zobel and Goddard (1955) demonstrated the
presence of pronounced differences in seedling sur-
vival among local strains of loblolly pine. Any-
thing which affects tree survival must be consid-
ered economic. If a seedling fails to live it certainly
will not produce wood and it costs as much money
to plant this seedling which doesn’t live as the
one which does.

So that seed might be better adapted to its proper
site most of us in the seed orchard business are
establishing separate orchards for different geo-
graphic areas. This will enable us to use local seed
and capitalize on these benefits mentioned.

2. Through improved disease resistance.—Bar-
ber (1961) found in Georgia open pollinated slash
pine progenies highly significant differences in
freedom of fusiform rust canker when comparing
parents. The 1952 plantings varied from 19 to 88
percent of the trees free of rust comparing various
parents. Wakeley (1961) also found significant
differences in susceptibility to fusiform rust; the
Georgia seed source had a much higher degree of
infection than the other sources represented. Derr
(1963) found that wind pollinated seedlings from
a brown-spot resistant longleaf pine growing in
central Louisiana have demonstrated a high level
of resistance to the disease. This finding indicates
the genetic control of this trait, and suggests the
possibility of selection for resistant strains of long-
leaf pine. There are other references in the litera-
ture pointing toward the fact that susceptibility

or resistance to disease appears to be hereditary
and that by selecting disease-resistant parents the
chances of producing disease-free offspring are im-
proved considerably. If a tree dies before it be-
comes merchantable it costs us money, and every
merchantable tree which can be added to our
harvest cut adds income. The selection of disease-
resistant parent trees for seed orchard use is an
important economic consideration.

3. Through wood quality improvement.—Zobel
and Haught (1962) found that the total merchant-
able volume of moderately straight trees contained
less than 10 percent compression wood (compres-
sion wood affects the properties of both pulp and
lumber), while more crooked trees commonly had
over 15 percent of the total volume as compression
wood. In excessively crooked trees compression
wood exceeded 50 percent of the total bole volume.
Compression wood lowers actual pulp yield and
also lowers quality for sawtimber purposes. Several
studies on inheritance of bole straightness have
been reported; some of these will be mentioned
later. The substance of these studies is that straight-
ness is controlled genetically. Straight parent trees
in seed orchards should produce straighter off-
spring which in turn result in improved wood
quality. I believe that all of us are stressing
straightness in the selection of trees for our seed
orchards.

Evidence is accumulating concerning the _ heri-
tability of wood specific gravity. Fielding and
Brown (1960) and Dadswell et al. (1961) found
definite evidence of heritability of wood specific
gravity in Monterey pine. Brown and Klein (1961)
by regression analysis found a real association
between parent tree wood specific gravity and
progeny wood specific gravity in the crosses of
loblolly pine tested.

Squillace et al. (1962) found high heritability
of specific gravity in slash pine comparing specific
gravity of parent and specific gravity of 14-year-
old controlled and open pollinated progeny.

A high specific gravity correlation between 6-
year-old open pollinated loblolly pine progeny and
the female parent was found by van Buijtenen
(1962). From one selection for specific gravity he
had an estimated progress of approximately 4 per-
cent, based on a selection differential of one stand-
ard deviation.

Zobel points out in a private communication
that we should be able to increase specific gravity
by about 50 to 300 pounds a cord green weight
from seed orchards. Assuming an increase of 150
pounds per cord this amounts to approximately a
3-percent improvement for weight alone.

4. By increasing growth, form and yield.—
Mergen (1955) found that certain slash pine par-
ents produced better stem form than others. One
female slash pine parent’s progeny included 51.6
percent trees with sweep; another female slash
parent’s progeny included 40.9 percent with sweep.

Barber (1961) found that trees containing stem
crook varied from 30 to 89 percent among progen-

61

ies of different slash parents; that ‘parents that had
a greater amount of crook had progenies that were
among those having the greatest percentage of
crooked stems.’ For young trees of loblolly pine
Perry (1960) found that bole straightness has a
fairly strong inheritance pattern. Progeny from
crooked parents were significantly more crooked
than those from straighter parents. Try as we
might we cannot escape the importance of having
straight trees. Too much depends upon it and
evidence indicates that straightness is genetically
controlled.

Peters and Goddard (1961) report a heritability
of vigor of very roughly 15 percent in slash pine
based on measurements 5 years after the progeny
were outplanted.

McWilliam and Florence (1955) tested slash pine
progeny in Australia in which open pollinated
progeny were selected from the outstanding slash
pine phenotypes in 1932 plantations. A limited
number of controlled pollinated progeny were also
included. For comparison purposes, a routine plant-
ing (representing the general plantation stock,
resulting from seed collected from the best 160
pruned trees per acre) was included in the study.
These progeny were assessed for both vigor and
form. Vigor included both height and volume.
Form included all other visual characteristics of
the tree such as straightness, branch size and angle,
and appearance. A difference of 5 percent in form
represents a big improvement.

The results of the open pollinated progeny test
were as follows:

Parent Best Routine Worst
Acceptable stems per acre PAT OD WLP? 80
Form percent 47 40 36
Plus stems 21 1 —
Minus stems 43 151 248

The results of the controlled pollinated progeny
were as follows:

Acceptable Form Plus stems Minus stems
Parent stems per acre (percent) per acre per acre
CB 74 selfed 520 62 184 --
CB 76 selfed 496 56 80 24
CB 74 x CB 76 440 56 64 32
CB 74 open pollinated 216 46 21 45
CB 76 open pollinated 176 44 7 63
Routine 112 40 1 151

Note the superiority of the controlled pollinated
progeny over the routine progeny. Not including
the “selfs” the controlled pollinated cross CB
74 CB 76 progeny exhibited a difference of 16
percent in form compared to routine progeny and
had 64 plus stems per acre versus 1 for the routine
progeny.

McWilliam and Florence further found that a
considerable improvement in the straightness of
stems was obtained in comparing controlled pollin-
ated progeny with routine plantings. They had
twice the number of acceptable stems per acre com-

paring controlled pollinated with routine progeny.
Because of its great economic importance in forest
management stem form must be of considerable
concern to forest managers. An undesirable tree
of poor form not only yields less usable wood sub-
stance but also occupies just as much space in a
forest (perhaps more) than a straight, well-formed
tree.

Nikles (1962) in Queensland reports that volume
production of slash pine was increased by at least
30 percent by crossing superior phenotypes. Nikles
compared the controlled pollinated trees with rou-
tine plantings (routine plantings were progeny of
trees selected for high pruning) and found nearly
three times as many acceptable trees (trees having
superior growth and straightness) among the con-
trolled crosses versus the routine trees. A tabular
summary prepared by Nikles comparing volume
production and numbers of acceptable trees in
7¥2-year-old slash pine progeny follows:

Mean number

Progeny Mean volume’ acceptable trees ~*
Gill»~x 15 60.9 16.5
G 34 « 16 57.9 12.5
G15 x 13 Died 20.0
G 34 x 11 56.2 12.0
G8x9 53.4 18.0
G9 x 15 51.3 16.0
G17~x 15 49.1 16.75
Routine * 40.5 6.5
G 3 self x G 2 self 37.6 10.0

1 Total volume of 25 trees in cubic feet; means of four plots
per treatment.

*A tree scoring at least a certain minimum of points for
straightness as well as reaching a minimum level of volume
production.

‘Progeny of trees selected for high pruning.

Nikles further points out that these crosses by
producing a larger number of straight offspring
will result in a higher recovery of sawn timber.
Juvonen (1961) corroborates this. Nikles sums up,
‘In view of this, and evidence from other trials up
to 16 years of age, it would be conservative to
claim an increase in recoverable volume of more
than 30 percent by the 10th year as a result of selec-
tion and cross breeding.”

Economic Implications of Expected
Improvement

Just a few studies have been mentioned which
indicate the many different areas in which im-
provement is possible through genetic control.
Considering these studies and improvements noted
it seems most reasonable that we may expect at
least a 5-percent improvement as a result of our
seed orchard programs. It is assumed that this 5-
percent improvement will manifest itself in 5 per-
cent more wood substance or yield than is being
obtained today using routine nursery stock grown
from seed collected by present-day collection meth-
ods.

62

A 5-percent improvement in yield might not
sound impressive to some but the economic impli-
cations are tremendous. Here is what a 5-percent
improvement could mean to my organization’s tree
planting program in Virginia assuming that the
planted pines would be harvested by a clear-cutting
operation 20 years after being planted. We found
that our loblolly pine plantations were growing,
on the average, 1.64 cords of pulpwood per acre
per year. Using this 1.64 cords per acre per year
as a base growth rate, in 20 years the average
acre would contain 32.8 cords of pulpwood. If a
$6 per cord pulpwood stumpage price is assumed
at the end of 20 years the average acre would have
a gross pulpwood value of $196.80. If a 5-percent
improvement in yield is realized as a result of
using improved planting stock from our seed or-
chards 20 years after being planted the average
acre would have a gross pulpwood value of $206.64
or an increase of $9.84 per acre. Each year the
Virginia Division of Forestry distributes for plant-
ing approximately 30 million loblolly pine seed-
lings. Our average planting space is 6 by 8 feet or
approximately 900 seedlings per acre. We, there-
fore, plant approximately 33,333 acres of loblolly
pine annually in Virginia. If a 5-percent increase
in total pulpwood yield results at the end of the
first 20-year period (assuming all 33,333 acres were
planted using improved planting stock) landowners
stand to gain $327,996.72 over what their returns
would have been had routine nursery planting stock
been used. Once our seed orchards are producing
enough seed to fully supply our nurseries it should
be remembered that each year improved planting
stock is used thereafter in a planting program
that these benefits will accrue and become avail-
able at harvest time. It should be kept in mind
that it costs just as much to plant a routine nursery
stock seedling as it does an improved seedling; and
it costs just as much to prepare land for planting
routine nursery stock seedlings as to prepare land
for planting improved seedlings. It also costs just
as much to release an acre planted with routine
planting stock seedlings as it does an acre on which
improved planting stock has been planted. As a
matter of fact, presupposing a $9.84 increase per
acre in 20 years as a result of planting improved
planting stock and charging a 5-percent interest
rate we could afford to spend an additional $3.70
per acre for site preparation, release, etc.

Some of us may be concerned with seed orchard
establishment costs because they may seem high.
However, since we expect to gain considerable
improvement in seed used for our reforestation
programs this should not unduly concern us. An
example is provided using the same set of condi-
tions as mentioned earlier, i.e. assuming a 5-percent
increase in yield and clearcutting plantations 20
years after planting, which would result in a total
increase of $327,996.72 realized from an annual
planting program of 33,333 acres. Let us assume
that it will take 15 years before our seed orchards
furnish enough seed for our reforestation programs
(planting only) and that an additional 20 years

will elapse before we are able to harvest our first
pulpwood by clearcutting. We will further assume
that we will recover $327,996.72 each year for a
total of 6 years. Therefore, from the time of seed
orchard establishment to time of harvesting our
sixth successive annual pulpwood crop a period of
40 years will have elapsed. At the end of 40 years,
using a 5-percent interest rate, $2,230,377.70 will
have accumulated which represents the increase in
returns alone resulting from using improved plant-
ing stock. Therefore, again charging a 5-percent
interest rate one could afford to spend some
$316,815.01 in seed orchard establishment and de-
velopment costs and still break even 40 years after
beginning the seed orchard program. In practice
this will not be the case, however, since we will be
collecting some quantities of improved seed from
our seed orchards before the end of 15 years and
this presents a more favorable financial picture
because we could start to amortize our investment
sooner. Also, once our seed orchards are in produc-
tion, each year we use improved seed our benefits
accrue and it is reasonable to expect these benefits
to be available for many years to come—more years
than in the example above. Furthermore, the cost
of our seed orchards should be prorated over the
entire life expectancy of the orchard, which may
be 50 years or longer.

In all of the calculations used above only ex-
pected gains or improvement in plantations are
noted. It is assumed that until seed becomes abun-
dant in seed orchards the first seed produced will
be used for planting and not for direct seeding.
It should be remembered that economic gains will
be realized using improved seed in direct seeding
programs as well.

Cole (1962) computes improvement in another
manner using slash pine on sawtimber rotations.
Cole states that on an ‘average’ slash pine site
(site index 70 feet at 50 years) a 1-percent increase
in volume yield over a 35-year rotation would
mean that the cost of seed could be increased about
5 times and the planter would still break even
(this assumes all of the increase is considered to

63

be in sawtimber at $35 per M bd. ft. at the end
of the period and 5-percent interest is charged).

Perry and Wang (1958) provide evidence that
genetic improvements of as little as 1 or 2 percent
more than justify the extra costs involved in pro-
grams of seed orchard establishment. They point
out that frequently because of improper geographic
origin or inferior genetic quality, the only seedlings
available for planting will yield growth rates and
profits 4 percent or more below average.

Percent improvements of a small magnitude may
seem small and inconsequential. However, when
one considers all the wood harvested each year
in our respective states and the economic implica-
tions of using improved seed in our direct seeding
programs and using genetically improved planting
stock for our planting programs these small per-
centage figures become very impressive indeed. I
have heard one company forester make the state-
ment that if only a 1-percent improvement is real-
ized that this would amount to more than a million
dollars a year to one mill!

In summary I believe our seed orchards, once in
production, will assure us of ample supplies of
seed to supply our reforestation programs. It will
cost no more to collect this seed and we will be
able to verify its origin.

The different types of improvement possible and
noted by others and reported were: (1) better
adaptation of seed to site, (2) better disease resist-
ance, (3) better wood quality, and (4) straighter,
more vigorous trees of better form. In view of these
I believe it entirely realistic to expect at least a
5-percent overall improvement from our seed or-
chard programs—this 5-percent improvement to
manifest itself in increased wood yields.

It should be remembered that a small percent
gain or improvement has tremendous economic
implications. We stand to be amply repaid many
times over for our time and expense spent on our
seed orchard programs. We must be careful not
to oversell our seed orchard programs but we must
not be guilty of underselling either!

Literature Cited

Allen, Peter H., and Trousdell, Kenneth B.

1961. Loblolly seed production in the Virginia-
North Carolina Coastal Plain. Jour. For-
estry 59: 187-190, illus.

Allen, Robert M.

1953. Release and fertilization stimulate longleaf

pine cone crop. Jour. Forestry 51: 827.

and McGregor, W.H.D.

1962. Seedling growth of three southern pine spe-
cies under long and short days. Silvae
Genet. 11: 43-45, illus.

Asher, William C.

1963. Squirrels prefer cones from fertilized trees.
U.S. Forest Serv. Res. Note SE-3, 1 p.
Southeast. Forest Expt. Sta., Asheville,
NSEC;

Auganbaugh, J. E,

1950. Shortleaf pine in southern Pennsylvania.
Pa. Forests and Waters 2: 82-85, 93, illus.

Baldwin, Henry Ives.

1942. Forest tree seed of the north temperate
regions with special reference to North
America. 240 pp., illus. Waltham, Mass.:
Chronica Botanica Co.

Barber, John C.
1961. Growth, crown form, and fusiform rust
resistance in open-pollinated slash pine
progenies. Sixth South. Forest Tree Impr.

Conf. Proc. 1961: 97-104,

Beers, Walter L., Jr.

1962. The effect of cobalt-60 gamma radiation
upon germination of slash pine seed and
initial growth of resulting seedlings. For-
estry Dept., Buckeye Cellulose Corp. Res.
Note 1, 4 pp., illus.

Bergman, F.

1955. Forsook med tvangsfruktificiring av _ tall,
gran och bjork (Research on inducing
fruiting of pine, spruce, and birch.) (Ab-
stract.) Svenska Skogsvforen Tidskr. 53
(3): 275-304.

Bilan, M. Victor.

1960. Stimulation of cone and seed production in
pole-size loblolly pine. Forest Sci. 6:207-
220, illus.

Bingham, R. T.

1963. New developments in breeding western
white pine. I. Breeding for blister rust
resistance. Forest Genet. Workshop Proc.
1962, pp. 69-70. South. Forest Tree Impr.
Com, and Soc. Amer. Foresters Tree Impr.
Com.

Bourdeau, Phillippe F., and Schopmeyer, C.S.
1958. Oleoresin exudation pressure in slash pine:

64

its measurement, heritability, and relation
to oleoresin yield. In Thimann, K. V., ed.,
The Physiology of Forest Trees. Pp. 313-
319, illus. New York: The Ronald Press
Co.

Boyce, Stephen G., and Kaeiser, Margaret.

1961a. Environmental and genetic variability in
the length of fibers of eastern cotton-
wood. TAPPI 44: 363-366, illus.

and Kaeiser, Margaret.

1961b. Why yellow-poplar seeds have low viabil-

ity. U.S. Forest Serv. Cent. States Forest
Expt, Sta. Tech. Paper 186, 16 pp., illus.

Briegleb, Philip A., and McKnight, J.S.
1960. The place of hardwoods in forest manage-
ment. The Unit, News Letter 83, pp. 10-11.

Broadfoot, W. M.
1958. Study effects of impounded water on trees.
Miss. Farm Res. 21(6): 1-2, illus.

1960. Field guide for evaluating cottonwood sites.
U.S. Forest Serv. South, Forest Expt. Sta.
Occas. Paper 178, 6 pp., illus.

1961. Guide for evaluating cherrybark oak sites.
U.S. Forest Serv. South. Forest Expt.
Sta. Occas. Paper 190, 9 pp., illus.

1963. Guide for evaluating water oak sites. U.S.
Forest Serv. Res. Paper SO-1, 8 pp., illus.
South. Forest Expt. Sta., New Orleans, La.
and Krinard, R. M.
1959. Guide for evaluating sweetgum sites.
Forest Serv. South. Forest Expt.
Occas. Paper 176, 8 pp., illus.
Brown, C.L., and Greene, J.T.
1961. A preliminary report on the development
of male gametocides for southern pine.
Ga. Forest Res. Paper 6, 7 pp, Ga. Forest
Res. Council, Macon, Ga.

and Klein, Jerome.
1961. Observations on inheritance of wood specific
gravity in seedling progeny of loblolly
pine. Jour. Forestry 59: 898-899.
Callaham, R. Z.
1957. Better trees through systematic breeding.
Thirty-third Natl. Shade Tree Conf. Proc.
1957: 272-275.
and Duffield, J. W.
1963. Heights of selected ponderosa pine seedlings
during 20 years. Forest Genet. Workshop
Proc. 1962, pp. 10-13. South. Forest Tree
Impr. Com. and Soc. Amer. Foresters Tree
Impr. Com.

WHS:
Sta.

and Hasel, A. A.
1961. Pinus ponderosa height growth of wind-pol-
linated progenies. Silvae Genet. 10: 33-42,
illus.

Carpenter, I. W., and Guard, A. T.

1950. Some effects of cross-pollination on seed
production and hybrid vigor of tuliptree.
Jour. Forestry 48: 852-855, illus.

Chowdhury, Kafiluddin A.
1931. Anatomical studies of the wood of a hybrid
larch. Jour. Forestry 29: 797-805, illus.

Cole, D. E.
1962. Harvesting the first certified pine cones.
Forest Farmer 21(5): 6-8, illus.

Dadswell, H. E., and Wardrop, A. B.

1959. Growing trees with wood properties desir-
able for paper manufacture. APPITA 12:
129-136.

and Wardrop, A. B.

1960. Some aspects of wood anatomy in relation
to pulping quality and to tree breeding.
APPITA 13: 161-173.

Fielding, J.M., Nicholls, J.W.P., and

Brown, A. G.

1961. Tree-to-tree variations and the gross heri-
tability of wood characteristics of Pinus
radiata. TAPPI 44: 174-179, illus.

Decker, John P.
1944. Effect of temperature on photosynthesis
and respiration in red and loblolly pines.
Plant Physiol. 19: 679-688, illus.

Derr, Harold J.

1963. Brown-spot resistance among F, progeny
of a single resistant longleaf parent. Forest
Genet. Workshop Proc. 1962, pp. 16-17.
South. Forest Tree Impr. Com, and Soc.
Amer. Foresters Tree Impr. Com.

Dimpflmeier, R.

1954. Pfropfungen an Waldbaumen durch Prof.
Dr. Heinrich Mayr vor 55 bis 60 Jahren.
(Grafts of forest trees made by Prof. Dr.
Heinrich Mayr 55 to 60 years ago.) Ztschr.
f. Forstgenet u. Forstpflan. 3(6): 122-125.

Downs, Albert A.
1949. Low forking in white oak sprouts may be
hereditary. Jour. Forestry 47: 736.

Duggar, B. M.
1911. Plant physiology. With special reference to
plant production. 513 pp., illus. New
York: The MacMillan Co.
Duncan, W.H,
1959. Leaf variation in Liquidambar styraciflua
L. Castanea 24(3): 99-111, illus.

Easley, L. T.
1954. Loblolly pine seed production areas. Jour.
Forestry 52: 672-673, illus.

1963. Growth of loblolly pine from seed produced
in a seed production area vs. nursery run
stock. Jour. Forestry 61: 388-389.

Echols, R. M.

1955. Linear relation of fibrillar angle to tracheid
length, and genetic control of tracheid
length in slash pine. Trop. Woods 102:
11-22, illus.

Engler, Arnold.
1905. Einfluss der provenienz des samens auf die
eigenschaften der forstlichen holzge-
wachse. Mitt. der Schweiz. Cent. Anstalt f.
das Forstl. Versuchsw. 8: 81-236, illus.

Fielding, J. M.
1953. Variations in Monterey pine. Austral. For-
estry and Timber Bur. Bul. 31, 43 pp., illus.

1960. Branching and flowering characteristics of
Monterey pine. Austral. Forestry and
Timber Bur. Bul. 37, 59 pp., illus.

and Brown, A. G.

1960. Variations in the density of the wood of
Monterey pine from tree to tree. Austral.
Forestry and Timber Bur. Leaflet 77, 28
pp.

and Brown, A. G.

1961. Tree-to-tree variations in the health and
some effects of superphosphate on the
growth and development of Monterey pine
on a low-quality site. Austral. Forestry
and Timber Bur. Leaflet 79, 19 pp., illus.

Forest Biology Subcommittee No. 2 on Tests and

Quality Objectives.

1960. Pulpwood properties: response of proces-
sing and of paper quality to their varia-
tion. TAPPI 43(11): 40A-64A.

Funk, David T.

1958. Frost damage to yellow-poplar varies by
seed source and site. U.S. Forest Serv.
Cent. States Forest Expt. Sta., Sta. Note
115, 2 pp.

Gabriel, William J.

1956. Preliminary report on clonal differences
in the wood and phloem of Populus del-
toides and P. trichocarpa. Third Northeast.
Forest Tree Impr. Conf. Proc. 1956: 33-34.

1958. Genetic differences in juvenile Shumard
oak. U.S. Forest Serv. Northeast. Forest
Expt. Sta. Forest Res. Notes 81, 3 pp.

Genys, John B.

1960. Geographic variation in European larch.
Caroline A. Fox Research and Demonstra-
tion Forest Bul. 13, 100 pp., illus. Hills-
boro, N. H.

Girgidov, D. Ja.

1960. [Methods of increasing the seed-bearing of
Pinus sylvestris.] (Abstract.) Voprosy
lesovedenijailesodstava. Pp. 157-170.

Goddard, Ray E.

1958. Additional data on cost of collecting cones
from a seed production area. Jour. For-
estry 56: 846-848, illus.

Goggans, James F.

1961. The interplay of environment and heredity
as factors controlling wood properties in
conifers with special emphasis on their
effects on specific gravity. Forest Tree
Impr. Program, N.C. State Col. School
Forestry Tech. Rpt. 11, 56 pp.

1962. The correlation, variation, and inheritance
of wood properties in loblolly pine. N.C.
State Col. School Forestry Tech. Rpt. 14,
155 pp., illus.

Grace, N. H.

1939. Vegetative propagation of conifers. I. Root-
ings of cuttings taken from the upper and
lower regions of a Norway spruce tree.
Canad. Jour. Res. Sect. C, 17: 178-180.

Grano, C. X.

1960. Strangling and girdling effects on cone
production and growth of loblolly pine.
Jour. Forestry 58: 897-898, illus.

Greene, James T., and Porterfield, H. D.

1963. Selection and progeny testing for early cone
production in loblolly and slash pines.
Forest Genet. Workshop Proc. 1962, pp.
9-10. South. Forest Tree Impr. Com. and
Soc. Amer. Foresters Tree Impr. Com.

Guard, A. T.

1943. The development of the seed of Lirioden-
dron tulipifera L. Ind. Acad. Sci. Proc.
53: 75-77, illus.

Hagner, S.

1958. Om kott-och froproduktionen i svenska barr-
skogar. [The production of cones and seed
in the conifer forests of sweden.] (Ab-
stract.) Meddel SkogsforsknInst. 47(8):
120. Stockholm.

Hamilton, John R.

1961. Variation of wood properties in southern
red oak. Forest Prod. Jour. 11: 267-271,
illus.

Hoekstra, P. E.

1960. Counting cones on standing slash pine. U.S.
Forest Serv. Southeast. Forest Expt. Sta.
Res. Notes 151, 2 pp.

and Mergen, Francois.
1957. Experimental induction of female flowers
on young slash pine. Jour. Forestry 55:
827-831, illus.

Holm, Theo.
1930. Leaf-variation in Liquidambar styraciflua
L. Rhodora 32: 95-100, illus.
Hottes, Alfred Carl.
1934. Plant propagation. 228 pp., illus. New York:
Delamare Co.

Hough, A. F.
1945. Frost pockets and other microclimates in

forests of the northern Allegheny Plateau.
Ecol. 26: 235-250, illus.

Hughes, Ralph H., and Jackson, James E.
1962. Fertilization of young slash pine in a culti-

66

vated plantation. U.S. Forest Serv. South-
east. Forest Expt. Sta., Sta. Paper 148,
14 pp., illus.

Hyun, Siu Kyu.

1954. Induction of polyploidy in pines by means
of colchicine treatment. Ztschr. f. forst-
genet. u. Forstpflan. 3(2): 25-33.

1961. Tree seed orchards, breeding for disease
resistance, racial variation. In Recent Ad-
vances in Botany II: 1675-1681. Univ.
Toronto Press.

Institute of Paper Chemistry.

1955. Forest genetics at the Institute of Paper

Chemistry. 12 pp., illus. Appleton, Wis.
Jackson, L. W.R., and Greene, J. T.

1958. Tracheid length variation and inheritance
in slash and loblolly pine. Forest Sci. 4:
316-318, illus.

Jewell, F. F.

1961. Artificial testing of intra- and interspecies
southern pine hybrids for rust resistance.
Sixth South. Forest Tree Impr. Conf. Proc.
1961: 105-109.

and Henry, B. W.
1961. Breeding for resistance to southern fusiform
rust. In Recent Advances in Botany II:
1694-1695. Univ. Toronto Press.

Johannsen, W.
1911. The genotype conception of heredity. Amer.
Nat. 45: 129-159, illus.

Johnson, Robert L.

1962. Weed control and site preparation for na-
tural regeneration of cottonwood. Fif-
teenth South. Weed Control Conf. Proc.
1962: 181-184.

Juvonen, Kauko S.
1961. Recovery in sawmilling. Jour. South Afri-
can Forestry Assoc. 37: 16-21.

Kaeiser, Margaret.
1956. Variations in fiber length of eastern cotton-
wood. U.S. Forest Serv. Forest Prod.
Lab. Rpt. 2047, 6 pp., illus.
Kiellander, C. L.
1957. Forbandet och rasen. [Planting distance
and race.] Skogen. 44: 578, 581. [Plant
Breeding Abs. 28: 154, 966 (1958).]

Klir, J., Kozak, J., and Novotny, M.
1956. [The technique of seed harvesting from tall
trees.] (Abstract.) Lesnica Prace 35(12):
537-547.
Kramer, Paul J.
1956. The role of physiology in forestry. Forestry
Chron. 32: 297-308, illus.
and Kozlowski, Theodore T.
1960. Physiology of trees. 642 pp., illus.
York: McGraw-Hill Book Co., Inc.
Kramer, Paul R.
1957. Tracheid length variation in loblolly pine.
Tex. Forest Serv. Tech. Rpt. 10, 22 pp.,
illus.

New

Landsberg, Hans H., Fischman, Leonard L., and

Fisher, Joseph L.

1963. Resources in America’s future. Patterns of
requirements and availabilities, 1960-2000.
1017 pp., illus. Baltimore: The Johns
Hopkins Press.

Lemin, A. J., Klomparens, W., and Moss, V. D.

1960. Translocation and persistence of cyclohexi-
mide (Acti-Dione) in white pines. Forest
Sci. 6: 306-314.

Libby, W.J., and Jund, E.
1962. Variance associated with cloning. Hered.
17: 533-540.

Limstrom, G. A.
1959. Yellow-poplar seed quality varies by seed
trees, stands, and years. U.S. Forest Serv.
Cent. States Forest Expt. Sta., Sta. Note
134, 2 pp.
and Finn, Raymond F.
1956. Seed source and nursery effects on yellow-
poplar plantations. Jour. Forestry 54: 828-
831, illus.
Lockard, C.R., Putnam, J. A., and Carpenter, R. D.
1950. Log defects in southern hardwoods. U.S.
Dept. Agr., Agr. Handb. 4, 37 pp., illus.

Lord, William B.
1963. A reconsideration of the farm forestry prob-
lem. Jour. Forestry 61: 262-264.

Lotti, Thomas.

1955. Yellow-poplar height growth affected by
seed source. U.S. Forest Serv. Tree Plant-
ers’ Notes 22, p. 3, illus.

McAlpine, Robert G., and Jackson, L. W. R.

1959. Effect of age on rooting of loblolly pine
air-layers. Jour. Forestry 57: 565-566,
illus.

McClurkin, D.C.

1963. Soil-site index predictions for white oak in
north Mississippi and west Tennessee. For-
est Sci. 9: 108-113, illus.

McGregor, Wm. H. Davis, Allen, Robert M., and
Kramer, Paul J.

1961. The effect of photoperiod on growth, photo-
synthesis, and respiration of loblolly pine
seedlings from two geographic sources.
Forest Sci. 7: 342-348, illus.

McKnight, J.S.

1963. On the way to intensive culture of cotton-
wood. Soc. Amer. Foresters Proc, 1962:
44-46.

McKnight, Virginia.

1962. How money grows on trees—if you know
where and how to plant them. Miss. Val.
Stockman-Farmer 16(12): 16-17, illus.

McWilliam, J. R., and Florence, R.G.
1955. The improvement in quality of slash pine
plantations by means of selection and cross
breeding. Austral. Forestry 19(1): 8-12.

Matthews, Fred R., and Maloy, Otis C.
1960. What to do about cone rust. Forest Farmer
19(4): 8, 14-15.

67

Matthews, J.D.
1963. Factors affecting the production of seed by
forest trees. Forestry Abs. 24(1): i-xiii.
Mergen, Francois.
1955. Inheritance of deformities in slash pine.
South. Lumberman 190(2370): 30-32, illus.

1958. Natural polyploidy in slash pine. Forest
Sei. 4: 283-295, illus.

1959. Colchicine-induced polyploidy in pines.
Jour. Forestry 57: 180-190, illus.

Hoekstra, P. E., and Echols, R. M.

1955. Genetic control of oleoresin yield and vis-
cosity in slash pine. Forest Sci. 1: 19-30,
illus.

and Voigt, G. K.

1960. Effects of fertilizer on two generations of
slash pine. Soil Sci. Soc. Amer. Proc. 24:
407-409.

Meuli, Lloyd J., and Shirley, H. L.

1937. The effect of seed origin on drought re-
sistance of green ash in the prairie-plains
states. Jour. Forestry 35: 1060-1062, illus.

Munger, Thornton T., and Morris, William G.

1936. Growth of Douglas fir trees of known seed

source. U.S: Dept: Agr. Tech. Bul. 537,
40 pp., illus.
Nikles, D. G.

1962. Tree breeding in Queensland 1957 to 1962.
Eighth British Commonwealth Forestry
Conference, East Africa. 27 pp. Queens-
land Dept. Forestry, Brisbane.

North Carolina State College School of Forestry.

1963. North Carolina state-industry cooperative
tree improvement program seventh annual
report. 27 pp., illus.

Parker, Johnson.
1962. Seasonal changes in cold resistance and free
sugars of some hardwood tree barks. For-
est Sci. 8: 255-262, illus.

Pauley, S.S., and Perry, T. O.

1954. Ecotypic variation of the photoperiodic re-
sponse in Populus. Jour. Arnold Arbore-
tum 35: 167-188.

Perry, Thomas O.
1960. The inheritance of crooked stem form in
loblolly pine (Pinus taeda L.). Jour. For-
estry 58: 943-947, illus.

and Wang, Chi-Wu.
1958. The value of genetically superior seed. Jour.
Forestry 56: 843-845, illus.

Peters, W.J., and Goddard, R. E.

1961. Inheritance of vigor in slash pine. Sixth
South. Forest Tree Impr. Conf. Proc. 1961:
80-84, illus.

Phares, Robert E., and Rogers, Nelson F.

1962. Improving shortleaf pine seed production
in Missouri. Jour. Forestry 60: 322-324,
illus.

Polster, H., and Reichenbach, H.

1957. Ein verfahren zur prognose der vitalen dur-
reresistenz durch ermittlung des stomata
regulations vermogens abgeschnittener
pflanzensprosse. Biol. Zbl. 76: 700-721.

Putnam, John A., Furnival, George M., and
McKnight, J. S.

1960. Management and inventory of southern
hardwoods. U.S. Dept. Agr., Agr. Handb.
181, 102 pp., illus.

Queensland Department of Forestry.

1962. Department of Forestry Annual Report 1961-

1962. 17 pp., illus.
Reines, M.

1963. Photosynthetic efficiency and vigor in pines:
variation. Forest Genet. Workshop Proc.
1962, pp, 14-15. South. Forest Tree Impr.
Com. and Soc. Amer. Foresters Tree Impr.
Com.

Righter, Francis I.

1960. Forest tree improvement through inbreeding
and intraspecific and interspecific hybridi-
zation. Fifth World Forestry Cong. Proc.
2: 783-787.

Santamour, F.S., Jr., and Schreiner, E. J.
1961. Juvenile variation in five white oak species.
Morris Arboretum Bul, 12: 37-46.

Sax, Karl.
1958. The juvenile characteristics of trees and
shrubs. Arnoldia 18: 1-6, illus.

1962. Aspects of aging in plants. Ann. Rev. Plant
Physiol. 13: 489-506.

Schaffalitzky, De Muckadell, M.
1954. Juvenile stages in woody plants.
ogia Plant. 7: 782-796, illus.

Schmidt, Werner.

1957. Waldbaumzuchtung, die sicherung von fruh-
diagnosen bei langlebigen gewachsen.
[Breeding of forest trees, value of early
diagnosis in long lived plants.] Der
Zuchter 4: 39-69.

Schopmeyer, C.S., Mergen, Francois, and Evans,

Thomas C.

1954. Applicability of Poiseuille’s law to exuda-
tion of oleoresin from wounds of slash
pine. Plant Physiol. 29: 82-87, illus.

Schreiner, Ernst J., Littlefield, E. W., and Eliason,
1
1962. Results of 1938 IUFRO Scotch pine proven-
ance test in New York. U.S. Forest Serv.
Northeast. Forest Expt. Sta., Sta. Paper
166, 23 pp., illus.

and Santamour, F.S., Jr.
1961. Juvenile variation in five red oak species.
Morris Arboretum Bul. 12: 65-70.

Sherry, S. P.

1947. The potentialities of genetic research in
South African forestry. Brit. Empire For-
estry Conf., 11 pp., illus. City Printing

Physiol-

68

Works, Ltd. Pietermaritzburg, South Af-
rica.
Sluder, Earl R.

1960. Early results from a geographic seed source
study of yellow-poplar. U.S. Forest Serv.
Southeast. Forest Expt. Sta. Res. Notes
150, 2 pp., illus.

Snyder, E. B., ed.

1959. Glossary for forest tree improvement work-
ers. U.S, Forest Serv. South. Forest Expt.
Sta. for Soc. Amer. Foresters. 22 pp.

1961. Racial variation in root form of longleaf
pine seedlings. Sixth South. Forest Tree
Impr. Conf. Proc. 1961: 53-59.
Grigsby, H.C., and Hidalgo, J. U.
1961. X-radiation of southern pine seed at various
moisture contents. Silvae Genet. 10: 125-
131, illus.
Speers, Charles F.
1962. Fraser fir seed collection, stratification, and
germination. Tree Planters’ Notes 53: 7-8.

Squillace, A. E., and Bengtson, G. W.
1961. Inheritance of gum yield and other char-
acteristics of slash pine. Sixth South. For-
est Tree Impr. Conf. Proc. 1961: 85-96.

and Dorman, Keith W,

1961. Selective breeding of slash pine for high
oleoresin yield and other characters. In
Recent Advances in Botany II: 1616-1621,
illus. Univ. Toronto Press.

Echols, R. M., and Dorman, Keith W.
1962. Heritability of specific gravity and sum-
merwood per cent and relation to other
factors in slash pine. TAPPI 45: 599-601,

illus.

and Silen, Roy R.
1962. Racial variation in ponderosa pine.
Sci. Monog. 2, 27 pp., illus.

Taft, KA. dt:

1962. The effect of controlled pollination and
honey-bees on seed quality of yellow pop-
lar (Liriodendron tulipifera L.) as assessed
by X-ray photography. N.C. State Col.
School Forestry Tech. Rpt. 13, 21 pp.

Taras, Michael A.

1956. Buying pulpwood by weight: As compared
with volume measure. U.S. Forest Serv.
Southeast. Forest Expt. Sta., Sta. Paper 74,
11 pp., illus.

Forest

Thorbjornsen, Eyvind.

1960. A cone production study in loblolly pine
on the coastal plain of North Carolina.
Jour. Forestry 58: 543-547, illus.

1961. Variation patterns in natural stands of lob-
lolly pine. Sixth South. Forest Tree Impr.
Conf. Proc. 1961: 25-44, illus.

1961. Variation in density and fiber length in

wood of yellow poplar. TAPPI 44: 192-

195, illus.

Timofeev, V. P.

1959. [Structure of seed harvests in Scots pine
stands.] (Abstract.) Lesn. Z., Arkangel’sk
2(3): 22-28.

Toda, Ryookiti.

1956. [On the crown slenderness in clones and
seedlings.] Ztschr. f. Forstgenet. u. Forst-
pflan. 5(1): 1-5.

1958. Variation and heritability of some quanti-
tative characters in Cryptomeria. Silvae
Genet. 7: 87-93.

Nakamura, Kentaro, and Satoo,
Taisitiroo.

1959. The heritability of tree height and stem
girth in Cryptomeria through sexual re-
production. Silvae Genet. 8: 43-49.

Trousdell, K. B., Dorman, Keith W., and Squillace,
A. E.

1963. Inheritance of branch length in young lob-
lolly pine progeny. U.S. Forest Serv. Res.
Note SE-1, 2 pp., illus. Southeast. Forest
Expt. Sta., Asheville, N.C.

U.S. Agricultural Research Service.
1960. Resistance. Agr. Res. 8(12): 2.

U.S. Forest Service.

1958. Timber resources for America’s future. U.S.
Dept. Agr. Forest Resource Rpt. 14, 713
pp., illus.

van Buijtenen, J.P.

1962. Heritability estimates of wood density in

loblolly pines. TAPPI 45: 602-605.

Van Haverbeke, David F., and Barber, John C.

1961. Less growth and no increased flowering
from changing slash pine branch angle.
U.S. Forest Serv. Southeast. Forest Expt.
Sta. Res. Notes 167, 2 pp., illus.

Vincent, G.

1957. Unterscheidungsmerkmale der fichten -und
kiefertypes in ihrer fruhen jugend. [A
few differentiating characteristics of spruce
and pine varieties in their early youth.]
Der Zuchter 4: 88-93.

Wakeley, Philip C.

1944. Geographic source of loblolly pine seed.

Jour. Forestry 42: 23-32, illus.

1952. Working plan for cooperative study of geo-
graphic sources of southern pine seed.
U.S. Forest Serv. South. Forest Expt. Sta.
for the Com. on South. Forest Tree Impr.
35 pp., illus.

1954. Planting the southern pines. U.S. Dept.
Agr., Agr. Monog. 18, 233 pp., illus.

1959. Five-year results of the southwide pine seed
source study. Fifth South. Forest Tree
Impr. Conf. Proc. 1959: 5-11.

69

1961. Results of the southwide pine seed source
study through 1960-61. Sixth South, For-
est Tree Impr. Conf. Proc. 1961: 10-24,
illus.

Zobel, Bruce J., Goddard, Ray E., et al.
1960. Minimum standards for progeny-testing
southern forest trees for seed-certification
purposes. U.S. Forest Serv. South. Forest
Expt. Sta. for the Com. on South. Forest

Tree Impr., 19 pp.

Watson, A. J., and Dadswell, H. E.

1962. Influence of fibre morphology in paper prop-
erties. II. Early wood and late wood.
APPITA 15(6): 116-128.

Watt, Richard P., and McGregor, Wm. H. Davis.

1963. Growth of four northern conifers under
long and natural photoperiods in Florida
and Wisconsin. Forest Sci. 9: 115-128,
illus.

Weidman, R. H.

1939. Evidences of racial influence in a 25-year
test of ponderosa pine. Jour. Agr. Res.
59: 855-887, illus.

Wenger, Karl F.

1953a. How to estimate the number of cones in
standing loblolly pine trees. U.S. Forest
Serv. Southeast. Forest Expt. Sta. Res.
Notes 44, 2 pp., illus.

1953b. The effect of fertilization and injury on

the cone and seed production of loblolly
pine seed trees. Jour. Forestry 51: 570-
573, illus.

Wheeler, P. R., and Mitchell, H. L.
1959. Specific gravity variation in Mississippi
pines. Fifth South. Forest Tree Impr. Conf.
Proc. 1959: 87-96, illus.

Wilde, S.A.
1954. Soils and forest tree breeding. Jour. For-
estry 52: 928-932.

Wright, Jonathan W.
1944. Ecotypic differentiation in red ash. Jour.
Forestry 42: 591-597.

1953. Summary of tree-breeding experiments by
the Northeastern Forest Experiment Sta-
tion 1947-1950. U.S. Forest Serv. North-
east. Forest Expt. Sta., Sta. Paper 56, 47
pp., illus.

1957. Studies to improve shade trees. Amer. Nurs-
eryman 106: 56-57, 60-63.

1959. Silvical characteristics of green ash. U.S.
Forest Serv. Northeast. Forest Expt. Sta.,
Sta. Paper 126, 18 pp., illus.

and Bull, W. Ira.
1963. Geographic variation in Scotch pine. Silvae
Genet. 12(1): 1-25, illus.

Wylie, John E.

1961. Cottonwood in northeastern Missouri. Jour.
Forestry 59: 518.

Yoho, James G., and Muench, John, Jr.

1962. Regional economic factors bearing on the
future of the lumber and woodpulp indus-
tries in the south. Jour. Forestry 60: 312-
319, illus.

Youngberg, C. T.

1952. Effect of soil fertility on the physical and
chemical properties of tree seed. Jour.
Forestry 50: 850-852.

Zak, Bratislav.

1955. Inheritance of resistance to littleleaf in
shortleaf pine. U.S. Forest Serv. South-
east, Forest Expt. Sta. Res. Notes 88, 2 pp.

1961. Aeration and other soil factors affecting
southern pines as related to littleleaf dis-
ease. U.S. Dept. Agr. Tech. Bul. 1248,
30 pp., illus.

Zobel, Bruce.

1961. Inheritance of wood properties in conifers.
Silvae Genet. 10: 65-70.

70

and Goddard, Ray E.
1955. Preliminary results on tests of drought
hardy strains of loblolly pine (Pinus taeda
L.). Tex. Forest Serv. Res. Note 14, 23
pp., illus.

Goggans, J. F., Maki, T. E., and
Henson, F.
1961. Some effects of fertilizers on wood proper-
ties of loblolly pine. TAPPI 44: 186-192,
illus.

and Haught, Aldin E., Jr.
1962. Effect of bole straightness on compression
wood of loblolly pine. N.C. State Col.
School Forestry Tech. Rpt. 15, 13 pp., illus.

Henson, Faye, and Webb, Charles.
1960. Estimation of certain wood properties of
loblolly and slash pine trees from breast
height sampling. Forest Sci. 6: 155-162,
illus.

Webb, Charles, and Henson, Fay [Faye].
1959. Core or juvenile wood of loblolly and slash
pine trees. TAPPI 42: 345-356, illus.

Adams, Paul M.
Allen, Robert M.
Aylin, E. M.
Baggett, Daniel E.
Barber, John C.
Barres, Herster
Bateman, Wallace R.
Baxley, W.R.
Beers, Walter L., Jr.
Bercaw, T. E.

Bilan, M. Victor
Bonninghausen, R. A.
Brown, Claud L.
Brown, W. Lester
Bryan reels.
Buckles, Oliver W.
Burns, Edmond B.
Byrd, Roger R.
Campbell, Robert K.
Carter, Mason C.
Cech, Franklin C.
Chaiken, L. E.
Clark, Philip M.
Cole, Donald E.
Cook, Denton R.
Cox, Bert

Coyne, J.F.

Craig, James W.
Crocker, John F.
Darby, Sanford
Dantzler wl. E:
Dorman, Keith W.
Dorward, R. E.
Draper, Lee, Jr.
Driver, Charles H.
Estes, Leon F.

Farmer, Robert E.
Ference, George M.
Finger, G. P.

Fogg, Peter
Folweiler, Al

Fox, George F.
Franson, John E.
Fuentes, Juan
Futrell, L. M.
Gaddis, Billy T.
Gash, Dan
Goddard, Ray
Goggans, James F.
Griffin, John W.
Grigsby, Hoy C.
Guilkey, Paul C.
Gunter, Erin R.

Hargreaves, L.A., Jr.
Harris, Clinton E.

Registrants

Arkansas Forestry Commission, Little Rock, Ark.
U.S. Forest Service, Gulfport, Miss.
International Paper Co., Bay Minette, Ala.

International Paper Co., Poplarville, Miss.

U.S. Forest Service, Macon, Ga.

U.S. Forest Service, Box 577, Rio Piedras, Puerto Rico
International Paper Co., Bay Minette, Ala.
International Paper Co., Marianna, Fla.
Buckeye Cellulose Corp., Foley, Fla.

Crown Zellerbach Corp., Bogalusa, La.

Stephen F. Austin State College, Nacogdoches, Tex.
Florida Forest Service, Tallahassee, Fla.
University of Georgia, Athens, Ga. .

Bowaters Carolina Corp., Catawba, S.C.

L. N. Dantzler Lumber Co., Perkinston, Miss.
U.S. Forest Service, Moncks Corner, S. C.
Louisiana Forestry Commission, Alexandria, La.
Louisiana Forestry Commission, De Ridder, La.
Weyerhaeuser Co., Centralia, Wash.

Auburn University, Auburn, Ala.

International Paper Co., Bainbridge, Ga.

Duke University, Durham, N. C.

U.S. Forest Service, Asheville, N.C.
Continental Can Co., Inc., Savannah, Ga.
Warrior Land and Timber Co., Tuscaloosa, Ala.
U.S. Forest Service, Oxford, Miss.

U.S. Forest Service, Gulfport, Miss.

Forestry Suppliers, Inc., Jackson, Miss.
International Paper Co., Monroeville, Ala.
Georgia Forestry Commission, Macon, Ga.
International Paper Co., Mobile, Ala.

U.S. Forest Service, Asheville, N.C.

Hiwassee Land Co., Vonore, Tenn.

Container Corp. of America, Fernandina Beach, Fla.
International Paper Co., Bainbridge, Ga.
Mississippi Forestry Commission, Jackson, Miss.
U.S. Forest Service, Stoneville, Miss.

Brunswick Pulp and Paper Co., Brunswick, Ga.
Weyerhaeuser Co., Plymouth, N. C.

Louisiana State University, Baton Rouge, La.
Texas Forest Service, College Station, Tex.

U.S. Forest Service, Jackson, Miss.

U.S. Forest Service, Jackson, Miss.

International Paper Co., Natchez, Miss.

U.S. Forest Service, Dry Prong, La.

156 Pemberton Dr., Jackson 8, Miss.

International Paper Co., Moss Point, Miss.
University of Florida, Gainesville, Fla.

Auburn University, Auburn, Ala.

International Paper Co., Jackson, Ala.

U.S. Forest Service, Crossett, Ark.

U.S. Forest Service, Asheville, N.C.

Louisiana Forestry Commission, Baton Rouge, La.
University of Georgia, Athens, Ga.

Rayonier, Inc., Fernandina Beach, Fla.

71

Henry, Berch W.
Hilbourn, Ted
Hill, Harold J.
Hitt, Robert G.
Holman, Jack
Hood, W. L.
Hyde, Robert S.

Jewell, F. F.
Johnson, Howard
Johnson, J. H.
Johnson, John W.
Jones, Ed

Jones, LeRoy

Jones, Robert D.
Jordan, Rufus Aaron

Kaufman, C. M.
Keller, Eugene
Kellison, Robert C.
Kelso, W. C.

King, Johnsey
Kite, Johnny
Knight, Vernon J.
Koenig, Robert L.
Kok, H.R.

Kraus, John F.

MacClendon, Travis
McClure, Clinton
McElwee, R. L.
McGregor, Wm. H. Davis
McKnight, J.S.
McMahen, James V.
Maki, T. Ewald
Manchester, Edwin H.
Mannion, E. G.
Marler, Ray

Merrick, E. T.

Miller, Bill

Mirams, Rex V.
Moberg, T. R.

Monk, John R., Jr.
Morse, William E.

Namkoong, Gene
Nelson, Harold A.
Nelson, John A.
Nelson, Tom
Newton, Philip A.
Nichols, C. R.
Nicholson, G. W. E.
Norris, T. L.

O’Gwynn, Claude H.
Orr, Leslie W.
Otterbach, Paul J.

Parrish, Kelly L.
Patton, Earl
Peevy, Orion J.
Peters, William J.
Plyler, R. P.

Ralston, Jim
Reines, M.

Renfro, Jim
Rogers, Charley J.
Rosdahl, Dave

U.S. Forest Service, Gulfport, Miss.
Weyerhaeuser Co., Plymouth, N. C.

Marathon Southern Corp., Butler, Ala.

U.S. Forest Service, Macon, Ga.

Mississippi Forestry Commission, Jackson, Miss.
International Paper Co., Quitman, Miss.

St. Regis Paper Co., New York, N. Y.

U.S. Forest Service, Gulfport, Miss.
International Paper Co., Natchez, Miss.
Chesapeake Corp. of Virginia, West Point, Va.
Union Bag-Camp Paper Corp., Savannah, Ga.
N.C. State College, Raleigh, N.C.

U.S. Forest Service, Macon, Ga.
International Paper Co., Poplarville, Miss.
Florida Forest Service, Tallahassee, Fla.

University of Florida, Gainesville, Fla.
Marathon Southern Corp., Butler, Ala.

N.C. State College, Raleigh, N.C.

Gulfport Creosoting Co., Gulfport, Miss.
U.S. Forest Service, Brooklyn, Miss.
International Paper Co., Vancleave, Miss.
Kimberly-Clark Corp., Coosa Pines, Ala.
Union Bag-Camp Paper Corp., Savannah, Ga.
Florida Forest Service, Tallahassee, Fla.

U.S. Forest Service, Lake City, Fla.

Olin Mathieson Chemical Corp., West Monroe, La.

Allison Lumber Co., Bellamy, Ala.

N.C. State College, Raleigh, N.C.

Clemson College, Clemson, S. C.

U.S. Forest Service, Stoneville, Miss.

U.S. Forest Service, Mt. Ida, Ark.

N.C. State College, Raleigh, N.C.

U.S. Forest Service, Murphy, N. C.

Queensland Forestry Dept., Brisbane, Queensland, Australia
Virginia Division of Forestry, Charlottesville, Va

U.S. Forest Service, Asheville, N.C.

Rayonier, Inc., Fernandina Beach, Fla.

New Zealand Forest Products, Ltd., Auckland, New Zealand
Ozan Lumber Co., Prescott, Ark.

L. N. Dantzler Lumber Co., Perkinston, Miss.

Buckeye Cellulose Corp., Foley, Fla.

N.C. State College, Raleigh, N.C.

Weyerhaeuser Co., Plymouth, N.C.

International Paper Co., Camden, Ark.

U.S. Forest Service, Asheville, N.C.

U.S. Forest Service, Cayce, S.C.

S.C. State Commission of Forestry, Columbia, S.C.
Tennessee River Pulp & Paper Co., New York, N. Y.
St. Marys Kraft Corp., St. Marys, Ga.

International Paper Co., Mobile, Ala.
U.S. Forest Service, New Orleans, La.
International Paper Co., Mobile, Ala.

Georgia Kraft Co., Rome, Ga.

Allison Lumber Co., Butler, Ala.

Weyerhaeuser Co., Washington, N.C.

628 NE 2nd St., Gainesville, Fla.

Arkansas Forestry Commission, Bluff City, Ark.

International Paper Co., Georgetown, S.C.
University of Georgia, Athens, Ga.
Container Corp. of America, Brewton, Ala.
St. Regis Paper Co., Jacksonville, Fla.
U.S. Forest Service, Hot Springs, Ark.

72

Saylor, LeRoy C.
Sentell, N. W.
Sewell, Carl S.
Simmons, B. M.
Slade, W. L.
Sluder, Earl R.
Smith, Lloyd F.
Sosbe, E. H.
Squillace, A. E.

Stevenson, Donald D.

Strickland, Ray K.
Swearingen, J. W.
Sweetland, Ted
Switzer, Geo. L.

Swofford, Thomas F.

Taylor, Robert E.
Trowbridge, K.S.
Tucker, R. E.

Wiha SCe

van Buijtenen, J. P.
Vande Linde, Frank
Verrall, Arthur F.
Vickery, Windell W.
Viele, Donald H.
Wakeley, Philip C.
Weaver, S. W.
Webb, Charles D.
Wells, Osborn O.
Wenger, Karl F.
Wheeler, Edwin Y.
White, Buford

Whitman, J. Dennis, Jr.

Wilcox, James R.
Williams, Richard A.
Wynens, Jim
Zarger, Thomas G.

Zimmerman, Richard H.

Zobel, Bruce

N.C. State College, Raleigh, N.C.

Tennessee River Pulp & Paper Co., Counce, Tenn.

St. Marys Kraft Corp., Madison, Fla.
International Paper Co., Bay Minette, Ala.
U.S. Forest Service, Pollock, La.

U.S. Forest Service, Asheville, N.C.
U.S. Forest Service, Gulfport, Miss.
Georgia Kraft Co., Rome, Ga.

U.S. Forest Service, Lake City, Fla.
Buckeye Cellulose Corp., Foley, Fla.
University of Florida, Gainesville, Fla.
Scott Paper Co., Mobile, Ala.
Continental Can Co., Inc., Hopewell, Va.

Mississippi State University, State College, Miss.

U.S. Forest Service, Atlanta, Ga.
International Paper Co., Chunchula, Ala.
Weyerhaeuser Co., Plymouth, N.C.
Continental Can Co., Hodge, La.

Hudson Pulp & Paper Corp., Palatka, Fla.
Texas Forest Service, College Station, Tex.
Brunswick Pulp & Paper Co., Brunswick, Ga.
U.S. Forest Service, New Orleans, La.

Tennessee River Pulp & Paper Co., Savannah, Tenn.

International Paper Co., Camden, Ark.

U.S. Forest Service, New Orleans, La.
U.S. Forest Service, Oxford, Miss.

U.S. Forest Service, Macon, Ga.

U.S. Forest Service, Gulfport, Miss.

U.S. Forest Service, New Orleans, La.
Oklahoma State University, Stillwater, Okla.
International Paper Co., Bay Minette, Ala.
St. Regis Paper Co., Cantonment, Fla.
U.S. Forest Service, Gulfport, Miss.
Georgia-Pacific Corp., Crossett, Ark.
Georgia Forestry Commission, Macon, Ga.
Tennessee Valley Authority, Norris, Tenn.
Texas Forest Service, College Station, Tex.
N.C. State College, Raleigh, N.C.

73

10.

11.

Previous Publications and Reports Sponsored by
The Committee on Southern Forest Tree Improvement

Report of the first southern conference on
forest tree improvement, held in Atlanta, Geor-
gia, January 9-10, 1951. U.S. Forest Serv.,
Atlanta, Ga., 65 pp., 1951.

Proposal for a cooperative study of geographic
sources of southern pine seed. Subcommittee
on Geographic Source of Seed, Philip C. Wake-
ley, Chairman. South, Forest Expt. Sta., 16
pp., illus. 1951.

Standardized working plan for local tests of
seed source. Subcommittee on Geographic
Source of Seed, Philip C. Wakeley, Chairman.
South. Forest Expt. Sta., 11 pp., illus. 1951.

Hereditary variation as the basis for selecting
superior forest trees. Subcommittee on Tree
Selection and Breeding, Keith W. Dorman,
Chairman. Southeast. Forest Expt. Sta., Sta.
Paper 15, 88 pp., illus. 1952.

Directory of forest genetic activities in the
south. Subcommittee on Tree Selection and
Breeding, Keith W. Dorman, Chairman. South-
east. Forest Expt. Sta., Sta. Paper 17, 17 pp.
1952.

Working plan for cooperative study of geo-
graphic sources of southern pine seed. Sub-
committee on Geographic Source of Seed,
Philip C. Wakeley, Chairman. South. Forest
Expt. Sta., 35 pp., illus. 1952.

Suggested projects in the genetic improvement
of sourthern forest trees. Committee on South-
ern Forest Tree Improvement, Carl E. Ostrom,
Chairman. Southeast. Forest Expt. Sta., Sta.
Paper 20, 13 pp. 1952.

. Testing tree progeny. A guide prepared by the

Subcommittee on Progeny Testing, E. G. Wiese-
huegel, Chairman. Tenn. Val. Authority Div.
Forestry Relat. Tech. Note 14, 77 pp., illus.
1952.

Report of the second southern conference on
forest tree improvement, Atlanta, Georgia,
January 6-7, 1953. U.S. Forest Serv., Atlanta,
Ga., 92 pp., illus. 1953.

Progress in study of pine races. Philip C.
Wakeley. South. Lumberman 187 (2345): 137-
140, illus. December 15, 1953.

The role of genetics in southern forest man-
agement. Special Subcommittee of the Com-
mittee on Southern Forest Tree Improvement,
Bruce Zobel, Chairman. Pt. 1, Forest Farmer
14(1): 4-6, 14-15, illus. Pt. 2, Forest Farmer
14(2): 8-9, 14-19, illus. Pt. 3, Forest Farmer
14 (3): 8-9, 14-15, illus. 1954. (Reprint, 11 pp.,
illus. 1954.)

74

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

Proceedings of the third southern conference
on forest tree improvement, New Orleans,
Louisiana, January 5-6, 1955. South. Forest
Expt. Sta., 132 pp., illus. 1955.

Better seed for better southern forests. Sub-
committee on Genetic Control of Seed, T. E.
Maki, Chairman. N.C. State Col. School For-
estry Tech. Rpt. 9, 16 pp., illus. 1955.

Forest tree improvement for the south. Com-
mittee on Southern Forest Tree Improvement,
T. E. Bercaw, Chairman. 13 pp., illus. 1955.

Supplement No. 1 to the original working
plan of September 12, 1952, for the southwide
pine seed source study. Subcommittee on Geo-
graphic Source of Seed, Philip C. Wakeley,
Chairman. South. Forest Expt. Sta., 110 pp.,
illus. 1956.

Time of flowering and seed ripening in south-
ern pines. Subcommittee on Tree Selection
and Breeding, Keith W. Dorman, Chairman,
and John C. Barber. Southeast. Forest Expt.
Sta., Sta. Paper 72, 15 pp., illus. 1956.

Proceedings of the fourth southern conference
on forest tree improvement, Athens, Georgia,
January 8-9, 1957. Univ. Ga., 149 pp., illus.
1957,

Pest occurrences in 35 of the southwide pine
seed source study plantations during the first
three years. B.W. Henry and G.H. Hepting.
South. Forest Expt. Sta., 7 pp., illus. 1957.

Proceedings of the fifth southern conference
on forest tree improvement, Raleigh, North
Carolina, June 11-12, 1959. School Forestry,
N.C. State Col., 114 pp., illus. 1959.

Minimum standards for progeny-testing south-
ern forest trees for seed-certification purposes.
Subcommittee on Progeny Testing for Seed
Certification Purposes. Philip C. Wakeley,
Chairman. South. Forest Expt. Sta., 20 pp.
1960.

Proceedings of the sixth southern conference
on forest tree improvement, Gainesville, Flor-
ida, June 7-8, 1961. School Forestry, Univ. Fla.,
187 pp., illus. 1961.

Proceedings of a forest genetics workshop,
Macon, Georgia, October 25-27, 1962. Spon-
sored jointly with the Tree Improvement Com-
mittee of the Society of American Foresters,
and partly supported by the National Science
Foundation through its grant No. GE-395. 98
pp., illus. 1963.

—— Fr

i;