Historic, Archive Document

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

United States
Department of
Agriculture

Forest Service

Intermountain
Research Station
Ogden, UT 84401

General Technical
Report INT-203

April 1986

Ms

to// 9*

Proceedings-
Conifer Tree Seed in
the Inland Mountain
West Symposium

Missoula, Montana, August 5-6, 1985

8B6B55

FOREWORD

The symposium "Conifer Tree Seed in the Inland
Mountain West" started, developed, and matured
in much the same manner and time frame as many
conifer cone crops. Like the biological stress-
es that often stimulate cone production, the
stress of knowledge needs prompted the initia-
tion of this symposium.

The first suggestion that a symposium be held to
consolidate the available cone and seed knowl-
edge came at a January 27, 1983, meeting of
Forest Service Intermountain Station and North-
ern Region scientists discussing cone and seed
production information needed to enhance natural
and artificial regeneration in the Northern
Rockies. Like the initiation of a cone bud, the
idea for a symposium was nothing really new, it
just came at the right time and fell into a
fertile environment of silviculturists, geneti-
cists, and entomologists.

A steering committee, chaired by Dr. Raymond C.
Shearer of the Intermountain Research Station,
was formed and proceeded with the planning.
Committee members were: Dr. George Blake, School
of Forestry, University of Montana; Mr.
Jerald E. Dewey, Pest Management, Northern
Region, Forest Service; Dr. A. K. Helium,
Department of Forest Science, University of
Alberta; Dr. George E. Howe, Division of Timber
Management, Northern Region, Forest Service; and
Mr. Wayne Maahs, Forestry Division, Champion
International Corporation. The School of
Forestry and Center for Continuing Education of
the University of Montana handled many of the
logistical details for the symposium and the
Intermountain Research Station edited and pub-
lished the proceedings.

Numerous meetings, letters, announcements,
coordination sessions with authors, and a host
of other details preceded the very successful
symposium, August 5-6, 1985. Included were 40
papers presented by silviculturists, geneti-
cists, entomologists, physiologists, patholo-
gists, nurserymen, and others. The objective of
the symposium — to consolidate information and
research conducted on conifer tree cones and
seeds native to the Inland Mountain West of the
United States and Canada — was met, and 164
attendees left with a much better idea of what
is known about western conifers. This symposium
culminated efforts of many people, and, like a
cone that survives all the hazards and produces
seed at the end of its development period, suc-
ceeded in producing a very viable proceedings.

Wyman C. Schmidt, Silviculture Project Leader
Intermountain Research Station
Forest Service, U.S. Department of Agriculture
Bozeman, MT

Proceedings— Conifer Tree
Seed in the Inland Mountain
West Symposium

Missoula, Montana, August 5-6, 1985

Compiler:

RAYMOND C. SHEARER, Research Forester, Intermountain Research Station,
Missoula, Montana

Conference sponsored by:

Forest Service, U.S. Department of Agriculture

Intermountain Research Station

Northern Region
University of Montana

School of Forestry

Center for Continuing Education
Canadian Institute of Forestry
Inland Empire Reforestation Council
Inland Empire Tree Improvement Cooperative
Society of American Foresters

SAF 85-10

CONTENTS

Page

Keynote Address

Seed Conservation Problems: Natural
and Unnatural

J. Derek Bewley and Andrew

D. Powell 1

Section 1. Cone and Seed Biology 13

Review Paper:

Cone and Seed Biology

John N. Owens 14

Contributed Papers:

Idaho Western Redcedar (Thuj a plicata
Donn.) Conebearing Status in an Unusually

Dry Year

William R. Gall and Allen

S. Rowley 32

Effects of Cold Stratification and Seed
Coat Sterilization Treatments on Pinyon
(Pinus Edulis) Germination

Gerald J. Gottfried and

L. J. Heidmann 38

Acetone is Unreliable as a Solvent for
Introducing Growth Regulators Into
Seeds of Southwestern Ponderosa Pine
(Pinus Ponderosa var. Scopulorum)

L. J. Heidmann 44

Are You Getting Ponderosa Pine and
Douglas-fir Cone Crops at High
Elevations?

Glenn L. Jacobsen 47

Seed-Dispersal Characteristics of
Conifers in the Inland Mountain West

Ward W. McCaughey, Wyman C. Schmidt,

and Raymond C. Shearer 50

Cone Production on Douglas-fir and
Western Larch in Montana

Raymond C. Shearer 63

Cone Production in Pinus Albicaulis
Forests

T. Weaver and F. Forcella 68

Section 2. Cone Prediction, Collection, and

Processing 77

Review Paper:

Cone Prediction, Collection, and
Processing

D. G. W. Edwards 78

Contributed Papers:

Can We Attain British Columbia Seed

Production Goals for Interior Spruce?

Paul J. Birzins 103

Cold Stratification for Lodgepole
Pine Seed

A. K. Helium and I. Dymock .... 107
Effect of Stratification Time and
Seed Treatment on Germination of
Western White Pine Seed

R. J. Hoff 112

Seed Dormancy in Three Pinus Species
of the Inland Mountain West

Carole L. Leadem 117

Page

Net Retrieval Seed Collection

James L. McConnell and

Jerry L. Edwards 125

Douglas-fir Seed Collection and Handling

Richard M. Schaefer III and

Daniel L. Miller 129

Aerial Cone Harvesting in
British Columbia

D. P. Wallinger 133

Reconditioning Stratified Seed at
Pine Ridge Forest Nursery

Katherine A. Yakimchuk 140

Section 3. Seed Orchard and Seed Production

Area Management 144

Review Paper:

Review of Seed Production Area and
Seed Orchard Management in the
Inland Mountain West

Jenji Konishi 145

Contributed Papers:

Production of Improved Western White
Pine and Douglas-fir at Potlatch
Corporation's Cherry lane Seed Orchard

Roger L. Blair 159

Potential and Actual Seed Yields from
a Southern Pine Seed Orchard

David L. Bramlett 162

Effect of Nitrogen Fertilizer and
Girdling on Cone and Seed Production
of Western Larch

Russell T. Graham 166

Flowering of Pinaceae Family Conifers
With Gibberellin A^/y Mixture: How
to Accomplish it, Mechanisms and
Integration with Early Progeny Testing

Richard P. Pharis and

Stephen D. Ross 171

Potential for Container Seed Orchards

Stephen D. Ross, Andrea M. Eastham

and Ralph C. Bower 180

Managing Black Spruce Seedling Seed
Orchards for Cone and Seed Production

Ronald F. Smith 187

Variation in Cone and Seed Production
from Natural Stands, Plantations, and
Clonal-Banks of Lodgepole Pine

Cheng C. Ying and Keith

Illingworth 191

Section 4. Effects of Biological Factors

On Seed Production 200

Review Paper:

Influence of Vertebrates on Conifer
Seed Production

Curtis H. Halvorson 201

Contributed Paper:

Squirrel Behavior Influences Quality
of Cones and Seeds Collected from
Squirrel Caches — Field Observations

Jeanne L. Pedro White and

Monte D. White 223

Page

Review Paper:

Insects and Conifer Seed Production
in the Inland Mountain West: A Review

Gordon E. Miller 225

Contributed Papers:

Insects Destructive to Ponderosa

Pine Cone Crops in Northern Arizona

Elizabeth A. Blake, Michael R.

Wagner and Thomas W. Koerber. . . . 238
Western Spruce Budworm Impact on
Douglas-fir Cone Production

Jerald E. Dewey 243

Protection of Blister Rust-Resistant
Western White Pine Cones from Insect

Damage with Permethrin and Fenvalerate
Michael I. Haverty and Patrick

J. Shea 246

Insect Problems and Control Efforts
Associated with Selected Douglas-fir
Plus Trees

William F. Johns 251

Implantation and Injection of Systemics
to Increase Seed Yield in Douglas-fir

Richard C. Reardon and Larry

E. Stipe 252

Impact of Insects on Cone/Seed
Production in Three Blister
Rust-Resistant Western White
Pine Seed Orchards

Patrick J. Shea 256

Review Paper:

Influence of Diseases on Seed
Production

Jack R. Sutherland 260

Contributed Papers:

Diseases of Conifer Seedlings Caused
by Seed-Borne Fusarium Species

R. L. James 267

Monoclonal Antibodies Recognizing
the Seed-Borne Fungal Pathogen
Sirococcus strobilinus

Leslie Ann Mitchell 272

Poster Paper Abstracts 279

Symposium Summary: Conifer Tree Seed in the
Inland Mountain West

Stanley L. Krugman 282

Attendance Roster 283

h

SEED CONSERVATION PROBLEMS: NATURAL AND UNNATURAL

J. Derek jJBewley and

ABSTRACT: The majority of seeds respond to
storage conditions in an "orthodox" manner, in
that a low seed moisture content and low tempera-
ture prolong their storage life. Deterioration
can take place in the seemingly "dry" stored
seeds , perhaps largely due to physico-chemical
processes. The consequences of deterioration to
metabolic integrity of the seeds are various, but
macromolecular damage, leading to defective mem-
branes and damaged organelles are manifestations.
Storage of seeds in the imbibed state, a condi-
tion in which many seeds find themselves in
"natural" storage conditions in seed-soil banks,
may prolong their longevity. Hydration may allow
the seeds to put into operation the repair
mechanisms essential for the maintenance of their
metabolic integrity. The cost, however, is the
consumption of reserves to provide energy, which
may eventually become depleted. Recalcitrant
seeds cannot withstand drying and must be stored
in a wet state. These latter seeds might persist
better as seedlings, and the preservation of
difficult seed species (both orthodox and
recalcitrant) in seedling banks is a possible
strategy.

INTRODUCTION

The seed is the dispersal stage of the plant life
cycle and, in many instances, it is the only
stage which can survive severely adverse
environmental conditions. A low water content
(usually some 5-15%), combined with an inherent
ability of the cells to withstand severe water
loss, facilitate the seed's survival. In the dry
state, many seeds can withstand extremes of cold
(storage in liquid nitrogen is becoming an
increasingly common practice for genetic stocks
of seeds) and heat (on the desert floors
temperatures may exceed 75°C). As a general
rule, commercial storage of seeds takes advantage
of their ability to retain metabolic quiescence
in the dry state, although other factors must be
considered, as will be discussed later.

Paper presented at the Conifer Tree Seed in the
Inland Mountain West Symposium, Missoula, MT,
\ugust 5-6, 1985.

J. Derek Bewley is Professor, Dept. of Biology,
University of Calgary, Calgary, AB, T2N 1N4,
Canada. Present permanent address: Dept. of
Botany, University of Guelph, Guelph, ON,
NIG 2W1, Canada. Andrew D. Powell is presently
a Postdoctural Research Associate in the Dept.
of Crop Science, University of Guelph.

H1

Andrew D. Powell

Seeds which retain their viability in the dry
state, and can be stored thus, are regarded as
conforming to the general "rules" for storage
conditions; such seeds are described as showing
orthodox viability characteristics. The seeds of
all common agricultural and horticultural crop
species which are annual or biennial exhibit this
orthodox storage behaviour. There is another
group of species, however, which produce seeds
that normally never dry out on the mother plant,
which are dispersed in the moist condition, and
which are killed if their moisture content de-
clines below some relatively high critical value.
Since these seeds do not conform to the rules of
storage that are applicable to the orthodox ones,
such desiccation-tolerant seeds have been de-
signed as recalcitrant (Roberts 1973).

In this paper we will consider some of the
problems associated with the storage of orthodox
seeds in the dry state — a common practice, but
an unnatural condition for many seeds to find
themselves in for protracted periods. More
naturally, when many seeds are shed from the
mother plant they remain buried in the substrate,
often in an intermittently wet state. Such seeds
can withstand drying, however, which is in con-
trast to the situation in recalcitrant seeds, for
these must be kept constantly in a near-hydrated
state .

ORTHODOX SEEDS AND DRY STORAGE

Premature Drying of Seeds for Storage

Acquisition of desiccation-tolerance during seed
development occurs approximately at the midway
stage, e.g. in Ricinus communis (Kermode and Bew-
ley 1985a), Pisum sativum (Rogerson and Matthews
1977), and Glycine max (Adams and others 1983).
Prior to this time, premature desiccation gener-
ally results in poor recovery of germinability of
the seed upon subsequent rehydration. There is a
concomitant reduction in the metabolic capacity
of the seed, e.g. for protein synthesis, when one
attempts to germinate seeds following drying dur-
ing this desiccation-intolerant stage (Dasgupta
and others 1982). Immature seeds exhibit a
great variation in the rate of water loss that
they can tolerate. Thus, when we refer to a de-
veloping seed acquiring desiccation-tolerance, we
must define also the rate of water loss toler-
ated. For example, Phaseolus vulgaris seeds can
tolerate extreme rates of drying, i.e. rapid des-
iccation over activated silica gel (Dasgupta and
others 1982), from about the 26th day of devel-
opment (of a 42-day developmental cycle), whereas
in R. communis tolerance of such extreme rates

1

of water loss is not acquired until the 55th day
of a 60-day developmental cycle (Kermode and Bew-
ley 1985a). On the other hand, this latter seed
can withstand slow water loss, even to complete
desiccation, at a stage less than half-way
through its development, and can germinate fully
upon subsequent rehydration. Seeds of Glycine
max also require slow dehydration (which can be
achieved while they are in the pod) to survive
premature drying; those shelled before drying
lose their viability (Fjerstad and others 1981;
Adams and others 1983) . The advantage of the
slow water loss is that it may permit membranous
structures within cells to undergo certain rever-
sible conformational changes during desiccation,
thus allowing them to retain their integrity
(Bewley 1979; Fjerstad and others 1981). Thus,
upon rehydration, metabolism within reasonably
intact cells can recommence, and the appropriate
germination mechanisms can be put rapidly into
effect. Rapid drying of immature seeds, on the
other hand, may lead to irreversible cellular
changes which are too extensive for repair to be
effected in the germinating seed, with a conse-
quent decline in, or loss of, viability.

Premature harvesting of seeds while in the
desiccation-tolerant stage of development might
be advantageous, in some instances, for the sub-
sequent establishment and growth of the progeny.
Studies on the endosperms of developing seeds of
R. communis (Kermode and Bewley 1985b) show, for
example, that synthesis of certain protein frac-
tions (including the soluble and major insoluble
storage proteins) reaches a peak at about the
midway point of development (fig. 1A) . Later in
development protein synthetic activity declines;
after 45-50 days of development this decline is
because the seed is drying out as it reaches the
latter stages of maturation. Hence the seed's
synthetic potential changes with time during
development. When the seed is dried prematurely
at 30 days of development it survives, and upon
rehydration its rate of synthetic activity is
considerably higher than in germinating mature
dry seeds or in 40-day dried and rehydrated seeds
(fig. IB). Hence there is a strong correlation
between the competence of the seed to conduct
protein synthesis prior to premature drying and
its competence to carry out this synthesis upon
subsequent rehydration — even though the products
of protein synthesis formed during development
and germination are very different (Kermode and
Bewley 1985b; Kermode and others 1985). If the
high synthetic capacity also has a bearing on the
survivability and germinability of seeds (i.e. if
there is increased viability because of an
increased potential for synthetic activity) then
it might be of value to harvest certain species
of seeds prematurely, during the desiccation-
tolerant stage, when the capacity for cellular
activity of the seeds is highest. Following the
appropriate drying regimes, and under suitable
storage conditions, the chances of maintaining
seed viability might be greater.

25

20

15-

10-

1 A

INCORPORATION

24 48 72 96 120

Figure 1. — Incorporation of H-leucine into (A)
developing seeds of Ricinus communis harvested at
20-50 days after pollination (DAP) and (B)
following mature seed imbibition 5-120 hours
after imbibition (HAI) (A); following rehydration
of desiccated 40 DAP seeds, 5-120 HAI (A); and
following rehydration of desiccated 30 DAP seeds,
5-120 HAI (X). Note that the high synthetic
activity of 30 DAP seeds is maintained upon
rehydration following premature desiccation.
After Kermode and Bewley (1985b).

Some studies have been carried out to determine
if the maturity and quality of tree seeds at
the time of collection have an effect upon
subsequent germinability. The general consensus
appears to be that early harvesting is not
necessarily advantageous and, indeed, in some
instances it may be disadvantageous (Wang and
others 1982). In no case, however, were
correlations made with metabolic activity during
development, nor was storeability of the seeds a
factor under consideration.

Viability of Seeds in Dry Storage:
Conditions

Optimizing

The majority of seed species are of the orthodox
kind in relation to storage and conform to
certain general rules that predict the pattern of
loss of viability in relation to their storage
environment (Harrington 1973).

2

Rule 1 states that for each 1% decrease in seed
moisture content the storage life of the seed is
doubled ;

Rule 2 states that for each 10°F (5.6°C) decrease
in seed storage temperature the storage life of
the seed is doubled;

And Rule 3 combines these observations to state
that the sum of the storage temperature in
degrees F and percent relative humidity (RH) must
not exceed 100, with no more than 50% of the sum
being contributed by the temperature.

At high seed moisture contents (>30%) non-dormant
seeds will germinate, and from 18-30% moisture
content rapid deterioration by microorganisms may
occur, particularly in the presence of oxygen.
Fungi will grow on seeds at 10-18% moisture
content, and above 18% the stored seeds may
respire and deplete essential reserves. An
additional problem arises under conditions of
poor aeration, for both seed and fungal
respiration could generate enough heat to kill
the seeds. Below about 10% moisture content
insect activity is diminished, and at 5-7%
moisture content the seeds are metabolically
quiescent. A low moisture content of 4-5% seems
to be less desirable for long-term storage than
one of 6-7%, for reasons which are not abundantly
clear .

The use of cold temperatures in storage has a
generally beneficial effect on longevity. At
liquid nitrogen temperatures seeds may survive
indefinitely (Stanwood and Bass 1981). Longevity
at -20°C for many seeds is considerably prolonged
— the required regeneration interval for Pisum
sativum stored at -20°C and 5% moisture content
may exceed 1000 years, for example (Roberts and
Ellis 1977). Even storage at 0-5°C will improve
the longevity of seeds, as would be predicted by
Rule 2, provided that certain precautions are
taken — particularly ensuring that the stored
seeds do not pick up moisture resulting from in-
creased atmospheric humidity at low
temperatures .

The extension of the interval between successive
regenerations of seed stocks in storage has been
a major aim. This not only eliminates the need
to devote time (and money) into seed stock regen-
eration but also eliminates the rate of genetic
drift which might take place in the small seed
batches planted out from the stored stocks. Re-
peated regenerations can also put the seed stock
at risk of diseases and other environmental
stresses which might prove to be fatal.

While considerable attention has been paid to
results from germination tests, more recently
there has come the appreciation that seed batches
with the same germinability can perform differ-
ently in the field, i.e. seed batches with simi-
lar viabilities have different vigour levels.
The International Seed Testing Association has
defined seed vigour as "the sum total of those
properties of the seed which determine the poten-
tial level of activity and performance of the
seed or seed lot during germination and seedling

emergence" (Perry 1978). Vigour tests have been
developed (Perry 1981) which when used in con-
junction with the standard germination test allow
the identification of a seed batch which might
not establish a good stand after germination even
though germination test values were the same.
These tests identify the fact that deterioration
in the performance of a seed can take place in
storage even though high germination levels are
maintained. As we shall see later, some metabol-
ic lesions may contribute to a loss in vigour as
well as viability, but handling of the seeds pri-
or to harvest is also important. The long-term
storage of a seed lot with an initial low vigour
is a waste of precious resources.

The drying of mature seeds for storage can be
problemmatical , particularly in cases where com-
plete loss of water has not occurred on the
mother plant. In the previous section we out-
lined how drying of prematurely harvested seeds
may have to be carefully controlled. Such is
also the case for some seeds harvested at 15-20%
moisture content and then dried to some 5-7%
moisture content for storage. Elevated tempera-
tures during drying or drying too quickly or exces-
sively can reduce viability and vigour, although
under good management this should not occur
(Rampton and Lee 1969). Care must also be taken
during mechanical harvesting and processing to
reduce the damage to the outer seed structures
since this might lead to damage when imbibition
takes place. The moisture content of seeds will
play an important role in their ability to
withstand harsh mechanical handling (Powell, A. A.
and others 1984) .

The viabilities expressed by populations of ma-
ture dry seeds before and during storage are very
variable; for many agricultural and horticultural
species it is usually, and acceptably, very high
— in the range of 95-98%. Viabilities of other
species, including forest species, are frequently
much lower. In some instances relationships are
to be found between mature seed characteristics
and viability. In several Pinus spp., for
example, there may be a correlation between ma-
ture seed weight and germination percentage, and
also correlations with seedling vigour (Ghosh and
others 1976; Kandya 1978), although this may not
always be the case (Chauhan and Raina 1980).
From a practical point of view, sorting of seeds
by weight before planting may not be desirable,
and it may be best to use all viable seeds for
reforestation (Helium 1976). The weight of white
spruce seeds seems to be highly specific for an
individual tree and thus there might be a reduc-
tion in genetic variability if seeds are sorted
by weight.

Metabolic Changes Occurring During Seed Storage

We may wonder how a dry seed, in a state of
metabolic torpor, can undergo deterioration in
storage — and how this can be prevented. Under
conditions where the cells of the seed are
partially hydrated in storage a limited amount of
metabolism might take place, which could be
detrimental. Partial completion of anabolic or

3

catabolic pathways could lead to the accumulation
of undesirable metabolites, resulting in a
potentially fatal disequilibrium, which cannot be
rectified upon subsequent reimbibition.

The metabolic manifestations of deterioration in
dry storage are many, and have been detailed in
several reviews (Bewley and Black 1982; Osborne
1980; Roberts 1975, 1981). A summary of the
various metabolic lesions that occur during
storage and which result in a loss of viability
is presented in fig. 2. Any single lesion could
result in a complete loss of germinability or
simply a reduction in vigour, depending upon its
quantitative expression within the cell. It is
more likely, however, that aging elicits a number
of distinct lesions (perhaps none of them fatal
by themselves) which together reduce the seed's
ability to survive.

Of the cellular changes which take place in the
dry state, those to membranes have attracted the
most interest. This is because the integrity of
the plasmalemma and associated membranes
(endoplasmic reticulum) is an integral
requirement for normal cellular function, and
efficient mitochondria are required for the
maintenance of a favourable energy status. A
loss of integrity of the plasmalemma and
tonoplast in aged seeds is implied from
observations that more substances leak into the
imbibition medium from such seeds than from
unaged ones (Ching and Schoolcraft
1968; Parrish and Leopold 1978). Excess leakage
from deteriorated seeds may represent a loss of
respirable substrate, but quantitative correla-
tions between this loss and subsequent germin-
ability have not been made. In some instances,
increased leakage of organic metabolites might
lead to a secondary damaging effect by stimulat-
ing growth of contaminating microorganisms on the
seed surface.

The cause of the changes in membrane integrity
during seed aging in storage has not been
elucidated, and there is little agreement between

workers on the deteriorative changes suffered by
phospholipids, the major membrane component. In
some seeds there is reasonable evidence that a
decline in unsaturated fatty acids, e.g. linoleic
and/or linolenic acids, can be correlated with
loss of germinability, e.g. in slowly aged
Glycine max (Priestley and Leopold 1983) and
Trif olium subterraneum (Flood and Sinclair 1981) .
Seeds of Acer platanoides when subjected to an
accelerated aging treatment (i.e. by being placed
in a saturated atmosphere at 30°C) exhibit a
marked decline in germination, which is
accompanied by a loss of phosphatidyl choline and
phosphatidyl glycerol from the membrane fraction
(Pukacka 1983). The cause of the changes in
fatty acid characteristics of dry seeds, and
hence their loss of viability, has been
attributed to physico-chemical perturbations of
these molecular components of the membranes. In
particular, peroxidation has been singled out as
a major causative agent, with the auto-catalytic
oxidation of unsaturated fatty acids (e.g.
linoleic acid and linolenic acid) by atmospheric
oxygen in a free-radical chain reaction occurring
in the dry seed. However, the evidence for lipid
peroxidation in aging seeds is tenuous, and that
which is in its favour is largely circumstantial
(Harman and Mattick 1976; Koostra and Harrington
1969).

There are enzyme (e.g. superoxide dismutase) and
scavenger systems present within seeds which can
reduce the activity of free radicals.
Correlations between endogenous superoxide
dismutase activity and viability of seeds have
been attempted, but no definitive role can be
assigned to this enzyme as far as the restriction
of peroxidation is concerned (e.g. Stewart and
Bewley 1980). Tocopherols are organic free
radical scavengers and it is estimated that one
molecule of this compound can afford antioxidant
protection to several thousand fatty acid
molecules. Attempts to correlate tocopherol
levels with susceptibility to peroxidation damage
have been without success; in aged soybean seeds,
for example, tocopherol levels are the same as in

4

unaged controls (Priestley and others 1980;
Fielding and Goldsworthy 1980). On the other
hand, it appears to be beneficial to add
anti-oxidants to isolated membrane preparations,
or to seeds (Fielding and Goldsworthy 1980; Basu
and Dasgupta 1978).

Some restorative treatments have been attempted
to increase germinability of seeds subjected to
loss of viability in dry storage. Hydration of
seeds in atmospheres of high relative humidity
before being placed in water is beneficial to
unaged seeds (Simon and Raja Harun 1972; Powell
and Matthews 1978), and may be of value in
encouraging germination of aged ones. Improved
germinability or growth of seeds with reduced
viability or vigor can be achieved, for example,
by subjecting them to conditions that permit low,
or slow, water uptake, e.g. by "flash imbibition"
of wheat (Goldsworthy and others 1982); by
placing seeds in an atmosphere of high relative
humidity (Basu and Pal 1980; Sanchez and Miguel
1983); or by imbibing seeds initially in an
osmoticum, such as polyethylene glycol
( Brocklehurs t and Dearman 1983; Burgass and
Powell 1984). Presumably the initial slow
hydration of the seeds allows for the ordered
rearrangement or repair of membranes before the
inrush of water associated with imbibition from
the dry state, which might otherwise be too
sudden and disruptive.

Whatever the real causes of deterioration in dry
storage are, it is evident that they take place
while the seed is in a state of quiescence or
near-quiescence, and while it has little chance
to accommodate the changes which are taking place
within its cells. Only inherent defense
mechanisms, such as those against free-radical
damage, can be brought to bear. It is unlikely
that seeds can be treated in advance of storage
to prevent deterioration, although storage
conditions can be devised which minimize damage
(e.g. low temperatures and a dry atmosphere).
But as we will consider now, maintenance for long
periods in the dry state is not the natural way
for many seeds to be stored. Rather, as they lie
in the ground, they are intermittently to

Table 1. — Some examples of viable seeds present in

frequently wetted, and hence undergo periods of
both metabolic activity and relative quiescence.

SEEDS IN THE SOIL: THE NATURAL SEED BANK
The Soil Seed Bank

Seeds buried in the soil for long periods must be
able to cope with a series of problems, not least
of which are the necessity to sustain themselves
metabolically when hydrated and to remain in a
dormant state until appropriate germination and
growth conditions are available.

There is an appreciable reserve of viable seeds
in the soil underlying a wide range of plant
communities. The species composition of a seed
bank reflects the differing strategies of past
and present components of the vegetation, and a
great diversity is often apparent (Roberts 1981).
Studies of the litter and surface soil layers of
North American forests have shown the presence of
viable seeds of species not represented in the
vegetation, thus indicating long-term survival of
species even from previous successional stages
(Livingston and Allesslo 1968). An indication of
the seed banks present in some western forest
stands is presented in table 1. In coniferous
forests there is a virtual absence of viable
seeds of the dominant tree species, owing to low
seed inputs and rapid losses from the surface
seed bank (Kellman 1974; Whipple 1978). The
predominant seeds present are those which arise
after clear-felling. In the U.S.S.R. the number
of viable seeds present in the soil of a
100-year-old southern tiaga forest ( Picea ables
and Vaccinium myrtillus) was in the order of
1200-5000 m~2 (Karpov 1960). But little
correspondence was found between the floristic
composition of the ground flora and that of the
seed bank, the latter being comprised of species
involved in early successional stages, or present
in clearings. In old and comparatively
undisturbed Picea or Picea/Plnus forests in the
Moscow area (Petrov 1977) species of seeds of
herbs and shrubs were present even though the
vegetative stages of these species were virtually
absent from the vegetation.

soils of Western North American forests

Province or state Major forest species Age of stand Depth of burial Seeds per m

(years) (cm)

British Columbia Tsuga heterophylla

Pseudotsuga menziesii

Oregon

Abies

grandis

Pinus

contor ta

Abies

grandis

Picea

engelmannii

Abies

grand i s

Lar ix

occidentalis

Colorado

Picea

engelmanni i

Abies

lasiocarpa

Excerpted from Roberts (1981)

100 0-10 1016

130 litter + 4 421

150 litter + 4 1863

175 litter + 4 3447

325 0-5 53

5

In northern hardwood stands the situation is
somewhat different. Analysis in the spring of
such stands in Pennsylvania dominated by Prunus
serotina, Acer saccharum or Acer rubrum showed
that they contain seeds of P. serotina, P.
pensylvanica and Betula spp. in the soil (Marquis
1975) — about 370 m~2 . Seeds of some species
(A. saccharum, Tsuga canadensis and Fagus grandi-
flora) germinated mainly in the first spring, and
hence were only transiently present in the seed
soil bank. Other species survived for up to 5
years ( Fraxinus americana , P. serotina and Betula
spp.), while high numbers of P. pensylvanica
seeds remained viable in the soil for in excess
of 30 years after this species had died out of
the overstory. Evidence for even more persistent
seeds in a forest understory has been presented:
seeds of Comptonia peregrina have survived for 70
years or more beneath a Pinus strobus forest in
Connecticut (Del Tredici 1977)T

In general, though, most of the shade-tolerant
true forest species do not produce seeds which
enter the buried seed bank. This is also gener-
ally true for the dominant trees of mature for-
ests. The greatest number of seeds present in
the soil beneath closed forest or woodland are
those of species commonly found in more open hab-
itats such as clearings or the early stages of
secondary successions. Such seeds may persist in
the soil for 30 years or more.

Contrasting this situation in forest soils is
that which exists in arable and grassland soils.
In arable soils the average number of viable
seeds at plough depth (to approx. 20 cm) usually
exceeds 4000 m-2 , and in very weedy fields
may be around 75000 m-2 (Jensen 1969). The
main contributors to banks of arable seeds are
the annual weeds, which often account for at
least 95% of all seeds; those of perennial weeds
and crop species are only poorly represented.
Quite often there are one or two species which
have seed numbers much greater than the rest.

Seed Burial and Dormancy

The probability that a seed will remain viable in
the buried state is enhanced by the presence of
dormancy mechanisms which delay germination in
the period immediately following seedfall. Some
seeds are dormant when they are shed from the
mother plant, and are said to exhibit an innate
or primary dormancy (Harper 1957; Bewley and
Black 1982). Others acquire dormancy after
shedding and are said to exhibit enforced
dormancy or secondary dormancy (Harper 1957;
Bewley and Black 1982). Some examples of primary
dormancy are characteristic of the embryo, and
include the need for an extended period of
incubation in warm moist conditions to allow
maturation of the embryo, inhibition of germina-
tion by light or a specific requirement for
stimulation by light, or a requirement for
chilling. Another type of dormancy mechanism is
coat-imposed, due to some sort of mechanical
restraint set up by the structures surrounding
the embryo. The secondary dormancy which
develops in already dispersed, mature seeds is

enforced when seeds are prevented from germinat-
ing by the unfavourable conditions prevailing in
the habitat in which they find themselves .

Germination of buried seeds may be held in check
for many years, often until some disturbance
occurs. If cultivated soil is turned over, or if
a natural disturbance takes place as in a river
bank, many of the buried seeds germinate and a
flush of seedlings results. A major factor in
the promotion of germination is light, and as
little time as half a second exposure to full
sunlight can cause many seeds to germinate and
produce seedlings (Wesson and Wareing 1969a, b;
Sauer and Strulk 1964). It has been reasonably
concluded, therefore, that many seeds which are
light-requiring normally fail to germinate
because burial in soil excludes the light. But
many of the species that appear after soil
disturbance are known to produce seed which, when
freshly shed from the mother plant, are capable
of germinating in darkness. Hence seeds of such
species must acquire a light-requirement when
buried (Wesson and Wareing 1968; Taylorson 1972).
This is because burial induces a secondary
dormancy by mechanisms which still remain to be
explained. Light sensitivity may also be lost
during burial, e.g. as in Barbarea vulgaris, the
seeds of which are light sensitive when freshly
dispersed, become insensitive to light after a
period of burial, and then acquire light sensiti-
vity once more (Taylorson 1972).

The effects of light in relation to burial might
act to secure the most favourable position in the
soil for germination. It would be undesirable
for small seeds to commence germination at too
great a depth in the soil since, because of their
limited food reserves, they may be unable to grow
sufficiently to reach the light and become
autotrophic .

Besides absolute exposure to light, exposure to
light of appropriate wavelengths is also
important for germination. The spectral energy
distribution of sunlight passing through a canopy
of leaves is greatly reduced in the
dormancy-breaking red region, and is relatively
enriched in the far-red region (Holmes and
McCartney 1975). Some seeds, therefore, will
become dormant as a result of irradiation with
canopy light and will germinate only later when a
dormancy-releasing factor, such as chilling or
light (usually direct sunlight) has been
experienced (Fenner 1980). This mechanism
explains the paucity of seedlings on forest
floors and the flush that follows clearing in the
leaf canopy, brought about when individual trees
die and fall, or when tree removal occurs.

The second major environmental factor which
promotes the germination of seeds in the soil
bank is temperature. Seeds whose dormancy is
broken by chilling will germinate when tempera-
tures begin to rise in early spring, which has
dual advantages. Firstly, the emergence of
seedlings is prevented immediately before
inhospitable winter conditions; secondly,
seedlings can establish themselves prior to
extensive shading by leaves. Hence species

6

Inhabiting deciduous woodlands, including the
nascent tree seedlings themselves, derive benefit
by germinating before the leaf canopy appears.
Germination of buried seeds in situ is often
brought about by a response to diurnal fluctua-
tions in temperature which may result from the
removal of the insulating layers, such as canopy
foliage, litter or humus (Grime 1979). Different
sizes of gaps in the overlying cover influence
the amplitude of the temperature fluctuation,
which in turn appears to correlate with the
germination percentages. The capacity to respond
to particular amplitudes of temperature fluctua-
tion in darkness may act as a depth-sensing
mechanism, such that the dampening of the
amplitude can be used by dormant seeds to
determine the depth at which they are buried.
Clearance of vegetation can therefore have
important effects in addition to an alteration of
the light environment in the top layer of the
soil; thus buried seeds can use both temperature
sensitivity and their phytochrome system to give
them positional information (Thompson and Grime
1978).

Seeds of any species represented within a seed
soil bank exhibit polymorphism with respect to
their germination requirements. Seeds produced
by the same population of plants (Grousiz and
others 1976) or even by one individual plant
(Cavers and Harper 1966) may show considerable
differences in their response to environmental
cues, and there may be very wide variation with
respect to the amplitude of diurnal fluctuation
in temperature required to initiate germination.
This is of great value in environments where the
risks of seedling mortality are high and
synchronous germination of the seed bank could
put the population in peril of extinction.

Although seeds in soil banks display polymorphism
in their germination requirements, which is of
considerable importance for the maintenance of
their genetic diversity and for recolonization ,
the recurrent germination of seeds from the soil
seed bank is a financially draining problem for
farmers, who must combat persistent weeds like
wild oats, wild mustard and redroot pigweed.
Efforts have been made to "flush" the soil with
promoters to induce the stored population to
germinate simultaneously. Success has been very
limited, although application of ethylene to the
soil does break the dormancy of some buried
species (Schonbeck and Egley 1981).

Seedling Banks
Species?

A Form of Persistence for Forest

For many forest species the time between seed
shedding and germination is relatively short (van
der Pijl 1972), although among temperate forest
trees it is not uncommon for germination itself
(e.g. Fagus sylvatica) or plumule extension (e.g.
Quercus petraea) to be temporarily delayed. Such
seeds do not accumulate in persistent seed banks,
however. In mature temperate forests a common
regenerative strategy is for tree seedlings and
saplings to persist for long periods in a stunted
or etiolated condition (table 2). A small
percentage survive to maturity by expanding into

canopy gaps resulting from natural or man-made
clearings. Many forest trees do not produce
seeds each year, and the ability of seedlings to
survive unfavourable conditions for long periods
ensures that the potential for regeneration is
maintained. One factor which may contribute to
survival in the seedling state is the size of the
seeds of temperate forest species (Ng 1978).
They are often large in relation to seedling
growth and may provide an effective nutrient
source during the initial establishment in
heavily shaded and/or nutrient-deficient
environments (Grime and Jeffrey 1965).

Table 2. — Forest species which can regenerate by
means of persistent seedlings

Species

Hemlock (Tsuga canadensis )
Norway spruce (Picea abies)
White oak (Quercus alba )
Holly ( Ilex aquifolium)
Sugar maple (Acer saccharum)
Beech ( Fagus grandif lora)

After Grime (1979)

Location of study

N. America

Sweden

N. America

U.K.

N. America
N. America

STORAGE IN THE IMBIBED STATE

Almost any combination of time, temperature and
moisture content will result in loss of viability
of dry stored seeds and will result in some
genetic damage in the survivors. Increasing
chromosome damage occurs with increased periods
of storage, and this is manifested in an
increased number of aberrant cells (Roberts
1975). While accumulation of genetic damage in
stored seed lots that are planted for food or
feed production may be of little consequence,
unless expressed in the first generation of
planting, such damage in seeds grown for
germplasm stock may have more serious long-term
implications. A stored seed lot with little
reduction in viability could harbour a
considerable number of mutations, which would not
be expressed immediately in the generation grown
from that seed, but which would begin to
segregate in subsequent generations, and in all
those thereafter.

Why aging seeds accumulate chromosome damage is
not understood, but it is apparent that it can be
circumvented if seeds are stored in the imbibed
state. When seeds of lettuce (Lactuca satlva)
and ash (Fraxinus excelsior) are maintained in a
fully imbibed state, in a dormant condition, they
maintain their full capacity for germination for
years, and yet sustain very little chromosome
damage (table 3). Loss of viability from seeds
stored at low moisture content (approx. 10%)
might result from inactivity of enzyme systems
capable of repairing storage-induced damage to
DNA, and also to other essential macromolecules
(and organelles in the cytoplasm). Assuming that
repair can only occur successfully in the
fully imbibed state, then any damage suffered by

7

the wet-stored seeds will be continuously
repaired and will not accumulate. In dry-stored
seeds, the repair mechanisms will become acti-
vated upon imbibition after storage, but by this
time the damage might be so extensive and beyond
restitution that viability would be lost.

Table 3. — Effects of a six-month storage period
at different water contents on the
germinability and extent of chromosome
damage in radicle tips of germinated
lettuce seeds

% Moisture % Germination Aberrant

content of nuclear
stored seed divisions

9.7

17

45

7.0

80

17

5.1

100

10

100

100

2

Not stored

100

2

After Villiers (1974)

While storage of seeds in an imbibed state (or
even intermittently imbibed) may be of advantage,
it is pertinent to ask: how do such seeds main-
tain themselves metabolically without exhausting
their energy sources? Recent work on imbibed-
stored lettuce seeds (Powell, A. D. and others
1983, 1984) shows that when these are maintained
in darkness at supra-optimal temperatures (25°C)
they enter a condition of secondary
( skoto-dormancy ) dormancy from which they cannot
be aroused by conventional dormancy breaking
treatments (e.g. red light or gibberellic acid).
This loss of response occurs over a 1-3 week
period. Initially, as the seeds remain in the
primary dormant condition there is an increase in
oxygen consumption and protein synthesis (fig.
3A,B) revealing vigorous metabolic activity —
even in seeds which have not received a
germination stimulus. With entry into
skotodormancy , however, metabolic rates fall
precipitously, and are maintained at a low level
during many months of storage (fig. 3A,B) . The
lowered oxygen consumption is the result of an
inherent change in the respiratory mechanism
within the seeds, and is not a consequence of
restrictions in oxygen permeability due to
changes in the surrounding seed coat (Powell, A.
D. and others 1984) . Even as .protein synthesis

Figure 3. — Changes in (A)
oxygen uptake and (B) in-
corporation of 3n-leucine
into protein by imbibed
lettuce seeds stored in
a dormant state for up to
10 months. After Powell
and others (1983) .

* II ■ >*

-93kd

? 1

Figure 4. — Protein synthetic pro-
-26 files of lettuce seeds stored

imbibed in a dormant state for up
to two years. Proteins were lab-
elled with JH-leucine prior to ex-
traction, separation by one-dimen-
sional electrophoresis and prepar-
lm 2m 6m 12m *24™ ation of the fluorograph. From

Powell (1985).

8

declines there are only a few obvious changes in
the actual proteins being synthesized (fig. 4).
Presumably, then, the proteins required for the
normal maintenance of the cells and the
metabolic "well-being" of the seeds are produced
in adequate amounts.

The energy for maintenance of metabolism in the
skotodormant condition probably comes from
utilization of the stored reserves within the
lettuce cotyledons. Mobilization is relatively
slow, with a decline in lipid and protein being
approx. 20% and 25%, respectively, over a
ten-month storage period (Powell and others
1983). Extrapolating such findings to the
situation of seeds in the soil bank, it is
possible that eventually the constraint on
viability could be nutritional rather than a
consequence of metabolic and/or structural
deterioration. Such mobilization within seeds
buried in the soil, however, would be at a much
reduced level compared with that in the
artificial conditions of high temperature and
moisture used in the studies on lettuce seeds.
Seeds achieving long-term viability in the soil
would be subjected to generally lower
temperatures, lower moisture contents, and
intermittent drying, all of which could prolong
the time over which the storage reserves would be
retained .

Storage of seeds in the imbibed state may not be
economically feasible if carried out on a large
scale. The seeds must be maintained under con-
stant temperature and moisture conditions; and
contamination by fungi or other microorganisms
must be prevented. Hence, they will require
attention at regular intervals if the seed stocks
are to be held in optimal conditions.

RECALCITRANT SEEDS: SPECIAL STORAGE PROBLEMS

There is a group of species that produce seeds
which normally never dry out on the mother plant;
they are shed in the moist condition and are
killed if their moisture content is reduced below
some relatively high critical value. Hence these
seeds do not conform to the rules which can be
applied to the orthodox group, and they have been
called recalcitrant (Roberts 1973; Roberts and
King 1980). Even when these recalcitrant seeds
are stored under moist conditions their longevity
may only extend from several weeks to several
months. Included in the list of species that
produce recalcitrant seeds is a number of
large-seeded hardwoods (e.g. Corylus , Castanea ,
Quercus , Aesculus , Salix and Juglans) , aquatic
species, and a number of important tropical
plantation crops such as Coffea, Cola , Theobroma
and Hevea .

The inability to store seeds of recalcitrant
species is a serious problem, for while vegeta-
tive propagation is possible for some species the
retention of viable seed stocks is desirable in
order to preserve maximum genetic diversity.
While dry storage is obviously out of the
question, other methods which are suitable for
orthodox seeds are also undesirable, e.g. low-
temperature storage is inappropriate for seeds of

tropical plants, although for those of temperate
woodland species it may be of value. Even within
the same genus some seeds appear to be more
sensitive to low temperatures than others, e.g.
Shorea ovalis seeds have to be stored above 15°C,
whereas S^. talura seeds can be maintained at 5°C
(Sasaki 1976).

Determination of optimal storage conditions for
recalcitrant seed species is generally empirical,
and little has been done to define the quantita-
tive relationship between environmental
parameters and viability. Perhaps an alternative
strategy for storage, other than as the seed
itself is appropriate. We noted earlier that
large-seeded forest species can establish
themselves as stunted seedlings on the forest
floor under very low light conditions. Recalci-
trant seeds normally possess a rapid and uniform
germination strategy, with no dormancy. The
seeds are generally large. Hence it may be
appropriate to store recalcitrant species as
seedlings or young plants, and not as seeds. As
suggested by Hawkes (1980) research on recalci-
trant seeds might best be directed away from seed
storage problems, and towards seedling survival
at low light intensities. The most appropriate
soil and moisture conditions should be elucidat-
ed, the most efficacious intensity and quality of
light determined, as well as the requirement for
other specific factors (e.g. mycorrhizal
associations). Methods for seed storage in soil,
or even in culture tubes, could be devised, and
it can be reasonably expected that many seedlings
could be conserved per unit area as tubes of
explants. Removal of seedlings from storage,
accompanied by an increase in light intensity
should lead to the resumption of normal growth.
Work on Cryptomeria seedlings in culture has
indicated that seedlings will survive for up to
two years in a healthy condition at 10°C under
short-day illumination (Isikawa 1978).

SUMMARY

Methods for seed storage are many, and diverse,
but for orthodox seeds storage at low tempera-
tures (0 to -20°C) and low moisture content
(5-7%) seems to be satisfactory for most species.
Such conditions are not particularly "natural" as
far as the seed is concerned, but it seems well
adapted, both structurally and physiologically to
survive them. Problems with deterioration are
exacerbated by poor storage conditions, but even
in the dry state seeds undergo deterioration in a
manner which cannot be combatted by their natural
defence mechanisms, which appear to require
hydration in order to be effective. For some
seeds with a low viability in storage, or a low
recovery rate of metabolism upon subsequent
rehydration, there may be advantages in harvest-
ing prematurely.

Many seeds persist naturally in the soil as
seed-soil banks for long periods, and therein may
be in at least a partially hydrated state for
extended periods. Dormancy mechanisms ensure
that full germination of the seed bank does not
occur, and the slow release from dormancy,

9

usually following an appropriate environmental
due is of advantage.

Recalcitrant seeds cannot be stored in conditions
of low relative humidity, and many do not
withstand a low temperature treatment. Little is
known about the optimum conditions required for
storing such seeds, but the use of seedling
banks, rather than seed banks might be most
appropriate. Such seedling banks might be
considered for more widespread use for conserva-
tion of difficult forest species which do not
exhibit dormancy mechanisms, and which do not
persist in seed-soil banks, but which can survive
under sub-optimal growth conditions.

REFERENCES

Adams, C. A.; Fjerstad, M. C; Rinne, R. W.
Characteristics of soybean seed maturation:
necessity for slow dehydration. Crop Sci . 23:
265-267; 1983.

Basu, R. N.; Dasgupta, M. Control of seed
deterioration by free radical controlling
agents. Indian J. Exp. Biol. 16: 1070-1073;
1978.

Basu, R. N.; Pal. P. Control of rice seed
deterioration by hydration-dehydration
pretreatments . Seed Sci. & Technol. 8:
151-160; 1980.

Bewley, J. D. Physiological aspects of desicca-
tion tolerance. Ann. Rev. Plant Physiol. 30:
195-238; 1979.

Bewley, J. D.; Black, M. Physiology and biochemi-
stry of seeds in relation to germination. 2.
Viability, dormancy and environmental control.
Springer-Verlag, Berlin; 1982. 375 p.

Brocklehurst , P. A.; Dearman, J. Interactions
between seed priming treatments and nine seed
lots of carrot, celery and onion. II.
Seedling emergence and plant growth. Ann.
Appl. Biol. 102: 585-593; 1983.

Burgass, R. W.; Powell, A. A. Evidence for repair
processes in the invigoration of seeds by
hydration. Ann. Bot. 53: 753-757; 1984.

Cavers, P. B.; Harper, J. L. Germination
polymorphism in Rumex crispus and Rumex
obtusifolius. J. Ecol. 54: 307-382; 1966.

Chauhan, P. S.; Raina, V. Effect of seed weight
on germination and growth of chir pine (Pinus
roxburgii Sargent). Indian Forester. 106:
53-59; 1980.

Ching, T. M.; Schoolcraft, I. Physiological and
chemical differences in aged seeds. Crop Sci.
8: 407-409; 1968.

Dasgupta, J.; Bewley, J. D.; Yeung, E. C.

Desiccation-tolerant and desiccation-intoler-
ant stages during the development and
germination of Phaseolus vulgaris seeds. J.
Exp. Bot. 33: 1045-1057; 1982.

Del Tredici, P. The buried seeds of Comptonia
peregrina , the sweet fern. Bull. Torrey Bot.
Club. 104: 270-275; 1977.

Fenner, M. The induction of a light requirement
in Bidens pilosa seeds by leaf canopy shade.
New Phytol. 84: 103-106; 1980.

Fielding, J. L.; Goldsworthy, A. Tocopherol

levels and ageing in wheat grains. Ann. Bot.
46: 453-456; 1980.

Fjerstad, M. C; Adams, C. A.; Rinne, R. W.
Soybean seed maturation requires slow
dehydration. Plant Physiol. 67(suppl.): 47;
1981.

Flood, R. G.; Sinclair, A. Fatty acid analysis of
aged permeable and impermeable seeds of
Trifolium subterraneum (subterranean clover).
Seed Sci. & Technol. 9: 475-477; 1981.

Ghosh, R. C; Singh, B . ; Sharma, K. K. Effect of
seed grading by size on germination and growth
of pine seedlings. Indian Forester. 102:
850-858; 1976.

Goldsworthy, A.; Fielding, J. L . ; Dover, M. B. J.
"Flash imbibition": a method for the
re-invigoration of aged wheat seed. Seed Sci.
& Technol. 10: 55-65; 1982.

Grime, J. P. Plant strategies and vegetational
processes. John Wiley & Sons, Chichester;
1979. 222 p.

Grime, J. P.; Jeffrey, D. W. Seedling

establishment in vertical gradients of
sunlight. J. Ecol. 53: 621-642; 1965.

Grovzis, M.; Berger, A.; Heim, G. Polymorphisme
et germination des graines chez trois especes
anuelles du genre Salicornia . Oecol. Plant.
11: 41-52; 1976.

Harman, G. E . ; Mattick, L. R. Association of
lipid oxidation with seed ageing and death.
Nature. 260: 323-324; 1976.

Harper, J. L. The population biology of plants.
Academic Press, London; 1977.

Harrington, J. F. Problems of seed storage. In:
Heydecker, W., ed . Seed ecology. Butterworths ,

London; 1973. 251-263.

Hawkes, J. G. Genetic conservation of

"recalcitrant species" - an overview. In:
Withers, L. A.; Williams, J. T. , eds. Crop
genetic resources. The conservation of
difficult material; 1980. 83-92.

10

Helium, A. K. Grading seed by weight in white

spruce. Tree Planter's Notes. 27: 16-23; 1976.

Holmes, M. C; McCartney, H. A. In: Smith, H.,
ed. Light and plant development. Butterworths ,
London; 1975. 467-476.

Isikawa, H. Storage maintenance of Cryptomer la
seedlings utilizing photoperiodic responses.
In: Akihama, T; Nakajima K., eds. Long-term
preservation of favourable germplasm in
arboreal crops. Fruit Tree Research Station,
M.A.F., Japan; 1978. 148-152.

Jensen, H. A. Content of buried seeds in arable
soil in Denmark and its relation to the weed
population. Dansk Bot. Ark. 27: 1-56; 1969.

Kandya, A. K. Relationships among seed weight and
various growth factors in Pinus oocarpa
Schlede seedlings. Indian Forester. 104:
561-567; 1978.

Karpov, V. G. On the species composition of the
viable seed supply in the soil of spruce -
Vacclnium myrtillus vegetation. Trudy mosk.
Obshch. ispyt. Prir. 3: 131-140; 1960.

Kellman, M. C. Prelminary seed budgets for two
plant communities in coastal British Columbia.
J. Biogeogr. 1: 123-133; 1974.

Kermode, A. R.; Bewley, J. D. The role of

maturation drying in the transition from seed
development to germination. I. Acquisition of
desiccation-tolerance and germinabillty during
development of Ricinus communis L. seeds. J.
Exp. Bot. - in press; 1985a.

Kermode, A. R.; Bewley, J. D. The role of

maturation drying in the transition from seed
development to germination. II.
Post-germinative enzyme production and soluble
protein synthetic pattern changes within the
endosperm of Ricinus communis L. seeds. J.
Exp. Bot. - in press; 1985b.

Kermode, A. R.; Gifford, D. J.; Bewley, J. D. The
role of maturation drying in the transition
from seed development to germination. III.
Insoluble protein synthetic pattern changes
within the endosperm of Ricinus communis L.
seeds. J. Exp. Bot. - in press; 1985.

Koostra, P. T.; Harrington, J. F. Biochemical
effects of age on membrane lipids of Cucumis
sativus L. seed. Proc . Int. Seed Test. Assoc.
34: 329-340; 1969.

Livingston, R. B.; Allessio, M. L. Buried viable
seed in successional field and forest stands,
Harvard Forest, Massachusetts. Bull. Torrey
Bot. Club. 95: 58-69; 1968.

Marquis, D. A. Seed storage and germination under
northern hardwood forests. Can. J. For. Res.
5: 478-484; 1975.

Ng, F. S. P. Strategies of establishment in
Malayan forest trees. In: Tomlinson, P. B.;
Zimmerman, M. H., eds. Tropical trees as
living systems. Proc IV Cabot Symp.,
Petersham, Mass.; Cambridge Univ. Press, U.K.;
1978. 129-162.

Osborne, D. J. Senscence in seeds. In: Thimann,
K. V. ed., Senescence in plants. CRC Press,
Boca Raton, Florida; 1980. 13-37.

Parrish, D. J.; Leopold, A. C. On the mechanism
of aging in soybean seeds. Plant Physiol. 61:
365-368; 1978.

Perry, D. A. Report of the vigour test committee
1974-1977. Seed Sci. Technol. 6: 159-181;
1978.

Perry, D. A. (Editor) Handbook of vigour test
methods. International Seed Testing
Association, Zurich; 1981.

Petrov, V. V. Reserve of viable plant seeds in
the uppermost soil layer beneath the canopies
of coniferous and small-leaved forests. Moscow
Univ. Biol. Sci. Bull. 32: 33-40; 1977.

Powell, A. A.; Matthews, S. The damaging effect
of water on dry pea embryos during imbibition.
J. Exp. Bot. 29: 1215-1229; 1978.

Powell, A. A.; Matthews, S.; Oliveira, M. De A.
Seed quality in grain legumes. Adv. Applied
Biol. 10: 217-285; 1984.

Powell, A. D. Long-term storage of lettuce seeds
in the imbibed state. Ph.D. Thesis, University
of Calgary; 1985.

Powell, A. D.; Dulson, J.; Bewley, J. D. Changes
in germination and respiratory potential of
embryos of Grand Rapids lettuce seeds during
long-term imbibed storage, and related changes
in the endosperm. Planta. 162: 40-45; 1984.

Powell, A. D.; Leung, D. W. M.; Bewley, J. D.
Long-term storage of dormant Grand Rapids
lettuce seeds in the imbibed state:
physiological and metabolic changes. Planta.
159: 182-188; 1983.

Priestley, D. A.; Leopold, A. C. Lipid changes
during natural aging of soybean seeds.
Physiol. Plant. 59: 467-470; 1983.

Priestley, D. A.; McBride, M. B. ; Leopold, A. C.
Tocopherol and organic free radical levels in
soybean seeds during natural and accelerated
aging. Plant Physiol. 66: 715-719; 1980.

Pukacka , S. Phospholipid changes and loss of
viability in Norway maple (Acer platanoldes
L.) seeds. Z. Pf lanzenphysiol . 112: 199-205;
1983.

Rampton, H. H.; Lee, W. 0. Effects of windrow
curing vs. quick drying on pre-harvest
development of orchardgrass (Dactylis
glomerata L.) seeds. Agron. J. 61: 483-484;
1969.

11

Roberts, E. H. Predicting the storage life of
seeds. Seed Sci. & Technol. 1: 499-514; 1973.

Roberts, E. H. Problems of long-term storage of
seed and pollen for genetic resources
conservation. In: Crop genetic resources for
today and tomorrow. I.B.P., Cambridge Univ.
Press; Vol. 2; 1975. 269-296.

Roberts, E. H. Physiology of ageing and its

application to drying and storage. Seed Sci. &
Technol. 9: 359-372; 1981.

Roberts, E. H.; Ellis, R. H. Prediction of seed
longevity at sub-zero temperatures and genetic
resources conservation. Nature. 268: 431-433;
1977.

Roberts, E. H.; King, M. W. The characteristics
of recalcitrant seeds. In: Chin, H. F.;
Roberts, E. H., eds. Recalcitrant crop seeds.
Tropical Press, Malaysia; 1980. 1-5.

Roberts, H. A. Seed banks in soils. Adv. Applied
Biol. 6: 1-55; 1981.

Rogerson, N. E. : Matthews, S. Respiratory and
carbohydrate changes in developing pea (Pisum
sativum L.) seeds in relation to their ability
to withstand desiccation. J. Exp. Bot. 28:
304-313; 1977.

Sanchez, R. A.; Miguel, L. C. de . Ageing of

Datura ferox seeds embryos during dry storage
and its reversal during imbibition. Z.
Pf lanzenphysiol. 110: 319-329; 1983.

Sasaki, S. The physiology, storage and

germination of timber seeds. In: Chin, H. F.;
Enoch, I. C; Raja Harun, R. M., eds. Seed
technology in the tropics. Universiti
Pertanian Malaysia, Selanger, Malaysia; 1976.
11-115.

Sauer, J.; Struik, G. A possible ecological
relation between soil disturbance,
light-flash, and seed germination. Ecology. 45:
884-886; 1964.

Schonbeck, M. W.; Egley, G. H. Changes in

sensitivity of Amaranthus retroflexus L. seeds
to ethylene during preincubation. 2. Effects
of alternating temperatures and burial in
soil. Plant Cell Environ. 4: 237-242; 1981.

Simon, E. W.; Raja Harun, R. M. Leakage during
seed imbibition. J. Exp. Bot. 23: 1076-1085;
1972.

Stanwood, P. C; Bass, L. N. Seed germplasm

preservation using liquid nitrogen. Seed Sci.
& Technol. 9: 423-437; 1981.

Stewart, R. R. C; Bewley, J. D. Lipid

peroxidation associated with accelerated aging
of soybean axes. Plant Physiol. 65: 245-248;
1980.

Taylorson, R. B. Phytochrome controlled changes
in dormancy and germination of buried weed
seeds. Weed Sci. 20: 417-422; 1972.

Thompson, K. ; Grime, J. P. Seasonal variation in
herbaceous seed banks in ten contrasting
habitats. J. Ecol. 67: 893-921; 1979.

van der Pijl, C. Principles of dispersal in

higher plants. Springer-Verlag , Berlin; 1972.

Villiers, T. A. Seed aging: Chromosome stability
and extended viability of seeds stored fully
imbibed. Plant Physiol. 53: 875-878; 1974.

Wang, B. S. P.; Pitel, J. A.; Webb, D. P.

Environmental and genetic factors affecting
tree and shrub seeds. In: Thomson, J. R., ed.
Advances in research and technology of seeds.
Part 7 . Centre for Agricultural Publishing and
Documentation, Wageningen, The Netherlands;
1982. 87-135.

Wesson, G.; Wareing, P. F. The role of light in
the germination of naturally occurring
populations of buried weed seeds. J. Exp. Bot.
20: 402-413; 1969a.

Wesson, G.; Wareing, P. F. The induction of light
sensitivity in weed seeds by burial. J. Exp.
Bot. 20: 414-425; 1969b.

Whipple, S. A. The relationship of buried,
germinating seeds to vegetation in an
old-growth Colorado subalpine forest. Can. J.
Bot. 56: 1505-1509; 1978.

12

Section 1. Cone and Seed Biology

13

CONE AND SEED BIOLOGY,
John N. j Owens

ABSTRACT: The time between cone initiation and
seed maturity of conifers from the Inland
Mountain West varies from 16 to 27 months. Cone
and seed production are affected at many stages
during these long reproductive cycles. Eight
patterns of cone initiation occur which help
explain how environmental factors affect cone
initiation and aid in timing cone induction.
Four pollination mechanisms occur. Pollination
at the optimal time for each pollination
mechanism can significantly increase seed set.
Most pre- and post-fertilization factors
affecting cone and seed development are poorly
understood but can affect seed production.

INTRODUCTION

Conifer reproduction begins with the initiation
of reproductive buds after a variable period of
juvenile growth. The time between cone
initiation and seed maturity varies from 16 to
27 months. The reproductive cycles are well
known for a few conifers from the Inland
Mountain West but many details are not known for
most species. Generalization may be made about
species within a genus but exceptions may occur.
Factors can promote or reduce cone and seed
production at many stages during the long
reproductive cycles. Little is known about many
of these factors and how they affect seed or
cone development. Understanding the
reproductive cycles and the factors affecting
development will help increase seed production
and seed quality for reforestation. The purpose
of this report is to describe the status of our
knowledge of cone and seed development of
conifers of the Inland Mountain West, point out
areas where more research is needed and suggest
some new approaches which might be used to study
cone and seed development.

REPRODUCTIVE CYCLES

Matthews (1963) was the first to generalize that
in temperate-zone trees reproductive buds are

Paper presented at the Conifer Tree Seed in the
Inland Mountain West Symposium, Missoula, MT,
August 5-6, 1985.

John N. Owens is Professor of Biology, University
of Victoria, Victoria, BC.

initiated in the growing season preceding the
spring in which cones or flowers appear and
anthesis occurs. This generalization has held
over the years with few exceptions.

Reproductive buds undergo early development
before winter dormancy and overwinter at various
stages. During the second and in some species
the third or fourth growing season variations
occur affecting the length of the reproductive
cycle. Three primary types of reproductive
cycles represent the variation found in most
conifers. Descriptions of many species are
brief and include only a few aspects of the
reproductive cycle.

The 2-Year Cycle

The most common reproductive cycle in conifers
is represented in figure 1, depicting
Picea glauca (Owens and Molder 1984c).
Pollination occurs in the spring or early summer
of the second year. The time between
pollination and fertilization is brief, usually
only a few weeks. Following fertilization,
embryo and seed development are rapid and
continuous. Seeds are mature and may be
released as early as late summer the year of
pollination. Retention of seed beyond that time
is often determined by climatic or biotic
requirements unique to a species and its method
of seed dispersal.

Detailed descriptions of the complete
reproductive cycles of conifers with this
pattern are few and include Picea (Owens and
Molder 1984c) , Pseudotsuga (Allen and Owens
1972; Owens 1973), Thuja (Owens and Molder
1984a) and Tsuga (Owens and Molder 1984d) .

The 3-Year-Cycle — I

t

A second reproductive cycle is found in most
species of Pinus , and several other conifers.
Unfortunately, this cycle is used in general
textbooks as the "typical" conifer reproductive
cycle (fig. 2). Pollination occurs in the
spring or early summer of the second year,
pollen tubes and ovules partially develop but
then stop usually in mid-summer. Development
resumes the following spring, fertilization
occurs and embryos and seeds are mature in the
fall. Seeds are usually shed in the year they

14

/

Figure 1. — The reproductive cycle of white spruce
(Picea glauca) (from Owens and Molder 1984c).

Q..JV MtCKJl SCAil

Figure 2. — The reproductive cycle of lodgepole
pine (Pinus contorta) (from Owens and Molder
1984b)~ ~ ~

mature. Serotinous seed cones may remain closed
for many years before opening, commonly in
response to extreme heat from fires, releasing
many years of accumulated seeds at one time.
This is a minimum 3-year cycle (commonly about
27 months) from reproductive-bud initiation to
seed maturity. Complete descriptions of this
type of reproductive cycle are limited to Pinus
(Ull 1974; Owens and Molder 1984b). Similar
life cycles are shared by a few other conifers

from the northern hemisphere (e.g. Sequoia) and
several from the southern hemisphere (e.g.
Araucaria , Podocarpus) (Singh 1978).

The 3-Year-Cycle~II

A third reproductive cycle is found in a few
conifers in the Cupressaceae . Pollination
occurs in the spring or early summer of the
second year, and fertilization occurs within a
few weeks. Embryo and seed development begin
but become arrested in late summer or fall. The
seeds and cones overwinter in a dormant
condition, then resume development in the spring
of the third year (fig. 3). This reproductive
cycle has been completely described for
Chamaecypar is nootkatenis (Owens and Molder
1984a) and partially described for several
species of Juniperus (Johansen 1950).

Variations of the Basic Cycles

General silvics (Fowells 1965) and seed manuals
(Schopmeyer 1974) refer to the time between
pollination and seed release. However, where
this time extends over more than one year the
reason is not given and it is usually uncertain
if the extended cycle results from the second or
third type of life cycle.

A combination of the two 3-year reproductive
cycles occurs in Juniperus communis (Ottley
1909; Kotter 1931) and three species of Pinus
(P_. plnea, P. leiphylla and P_. torreyiana)
(Dallimore and Jackson 1966; Francini 1958). In
J_. communis cone initiation occurs before winter
dormancy and pollination occurs the following

1 5

Figure 3. — The reproductive cycle of yellow
cedar (Chamaecyparis nootkatensis) (from Owens
and Molder 1984a).

spring. Pollen tube growth and ovule
development become arrested and overwinter, with
fertilization occurring in the third year. The
immature embryos overwinter, then complete
development during the fourth growing season.
In the three species of pine, pollination occurs
in the spring but pollen tube and ovule
development remain arrested for two years.
Fertilization, and embryo and seed maturation,
occur in the fourth year.

In the long reproductive cycles there is
tremendous scope for variation in the phenology,
even though the sequence will remain unchanged.
We must not assume that a species fits the
textbook example, especially if we are
attempting to control seed production. Also, as
the length of the reproductive cycle increases
so does the possibility of something going
wrong. Therefore, it is not surprising that
many species have rather low seed yields. To
determine the causes of poor seed yield we must
determine what went wrong at what stage of
development .

CONE INITIATION

The time of cone initiation in the life of a
tree and during the growing season may vary from
one species to another as may the sites of cone
buds in the crown and on the shoot.
Understanding those variations is a prerequisite
to understanding factors which affect cone
initiation and for cone induction.

Cone buds are borne terminally or laterally (in
axils of leaves) on the branch, and, except for
the Cupressaceae, they are enclosed by bud
scales. As a conifer reaches reproductive age
seed cones are produced first, usually
on vigorous lower order shoots, followed by
pollen cones, usually on less-vigorous higher
order shoots. There are several reviews of
times and methods of cone initiation (Owens
1973, 1980; Puritch 1972; Owens and Molder 1977,
1979; Eis and Craigdallie 1981). Figure 4 is a
summary of the most recent information. Several
species from the Inland Mountain West have been
studied and times and patterns within genera are
usually similar.

Abies (true firs)

Several species have been studied including:
A. amabilis (Ritchie 1966; Owens and Molder
T977a); A. balsamea (Powell 1974, 1977a, b);
A. grandis (Owens 1984b); A. lasiocarpa (Owens
and Singh 1982) and A. procera (Ritchie 1966).
A summary is given by Owens and Molder (1985).
Potential seed-cone buds are initiated in the
axils of leaves on the upper surface of
elongating primary or secondary shoots in the
upper few whorls of the crown. Seed-cone buds
occur most frequently on vigorous nodal shoots
or on less-vigorous internodal shoots.
Potential pollen-cone buds are initiated in the
axils of leaves on the lower surface of
elongating shoots in the mid- and lower regions
of the crown. There is usually little overlap
between seed-cone and pollen-cone bearing
regions and seldom do the two types of cones
occur on a branch. All axillary buds are
initiated during early shoot elongation, at
about the time of vegetative bud flush.
Axillary buds do not become determined as
pollen-cone, seed-cone or vegetative buds until
the end of bud-scale initiation (fig. 5) .
Microsporophylls, bracts, ovuliferous scales,

jan| fes|mar

APK -VA> IIJN

jul|aug| SEP

OCT NOV DEC

AN F t B MAR

APR |mAy|jUN

Abiei

A

i

— P —

Pieudoiiuoo

A

,<?-
•?-

— P—

Piceo

T
A
I
A

— P—

-?

I
A

T

-P-

■ ¥

E

Loin

D

— P—

Q

¥

Pinui
Diplojylor.

- --<*.

A

¥-

— P —

Hoplo.ylon

A

A-

— P —

¥

Thujoand
Chomoccypom

T-.

T

■-¥

— P —

Ch.
-P—

1

A A.illoryApe. Bud Stole Troni.lion or Eorly Cone Development -P- Poll.nol.on

I Terminol Ape. Iniliolion Diflerenriahon

0 Dwarf Ape.

Figure 4. — Times and methods of cone
initiation.

16

Figure 5. — Potential pathways of terminal and
axillary bud development in Abies . Lower line
shows vegetative bud development (from Owens and
Molder 1985).

PRIMORDIUM

SEEO CONE

VEGETATIVE

POLLEN CONE

Figure 6. — Alternative pathways of axillary bud
development in Douglas-fir (from Allen and Owens
1972).

late in the fall (fig. 4). All microsporophylls ,
microsporangia, bracts, fertile ovuliferous
scales, and leaves are initiated before buds
become dormant. As in Abies , terminal buds rarely
become reproductive and axillary buds may abort
or become latent. In Pseudotsuga the number of
cone buds which develop is determined not by the
number of axillary buds initiated (although this
does vary) but primarily by the proportion of
these buds which differentiate into cone buds
(fig. 6) (Owens 1969).

and leaves begin to be initiated about mid-July.
Pollen-cone buds complete development in about 2
months, whereas seed-cone and vegetative buds
continue development into autumn. All
microsporophylls, bracts, fertile ovuliferous
scales and leaves are initiated before winter
dormancy (fig. 4). In addition to the above
three alternative pathways of bud development,
axillary buds may abort during early development
or become latent before becoming determined
(fig. 5). Aborted buds often degenerate before
forming many bud-scales, whereas latent buds
initiate many bud scales and retain a living
apical meristera capable of future growth.

Pseudotsuga (Douglas-fir)

Douglas-fir has been studied extensively (Owens
1969; Allen and Owens 1972). It is similar to
Abies in most respects, except the position of
cone buds is less rigorous. There Is
considerable overlap of seed-cone and
pollen-cone bearing regions within the crown and
both types of cone buds often occur on the same
shoot. In the latter case, seed-cone buds are
more distal. Both cone-bud types occur
primarily on the lateral and lower surfaces of
shoots. All axillary buds are initiated at the
onset of vegetative bud growth and have
initiated several bud scales before vegetative
bud flush. The earliest stages of axillary bud
determination can be recognized by using
hlstochemical tests in early June (Owens 1969).
All axillary buds become anatomically determined
at the same time early in July, when lateral
shoot elongation is nearly complete (Owens and
others 1985). Microsporophylls, bracts and
leaves begin to be initiated in early July.
Pollen-cone buds complete development by late
summer and become dormant early in the fall.
Seed-cone and vegetative buds become dormant

Picea (spruces)

The time and method of cone initiation has been
determined for P. glauca (Fraser 1962; Eis 1967;
Owens and Molder 1977e), £. engelmannli
(Harrison and Owens 1983), P_. mariana (Fraser
1966; G. Caron, personal communication) and
P_. sitchensis (Owens and Molder 1976) and Is
summarized by Owens and Molder (1984c). Spruce
cone buds may develop from terminal apices which
have been vegetative for one or more years or
from newly-initiated axillary apices on
elongating shoots (fig. 7). Except for the
presence of terminal cones, cone distribution in
the crown and on branches is similar to that
described above for Pseudotsuga . In
Ptcea engelmannli when cone buds are abundant,
they occur mostly in the axillary but also in
the terminal position. When cone buds are few
(less than 35 percent of total buds) they are
about equally distributed in terminal and
axillary positions (Harrison and Owens 1983).
This also appears to be true for P_. glauca
(Owens and Molder 1977e) and P. mariana
(G. Caron, personal communication).

The time of cone-bud determination is remarkably
similar for most species of Picea which have
been studied. Cone buds become anatomically
determined at the end of bud-scale initiation
which occurs near the end of the period of
lateral shoot elongation (Owens and others 1977;
Owens and Molder 1977e; Harrison and Owens 1983;
Dunberg 1979). Dunberg (1979) stressed that
biochemical differentiation must occur before
shoot elongation stops and anatomical
differentiation begins. Pollen-cone, seed-cone
and vegetative buds in terminal and axillary
positions become determined at essentially the
same time In all regions of the crown of a tree.

17

•4T 1 ATION

1 *" SEED CONE

POLLEN CONE

VEGETATIVE

LATENT

ABORTED

DIFFERED

— SEED CONE

POUEN CONE

ABORTED

' — POLLEN CONE

Bod

DORMANT ^ BUD - SCALE

NITIAIION **■

~* OEVE

LOPMENT OORMANT

AUGUST

SEPTEMBER | OCTOBER

Figure 7. — Potential pathways of terminal and
axillary bud development in Picea. Lower line
shows vegetative bud development (from Owens and
Molder 1984c).

In _P. sitchensis (Owens and Molder 1976) and
P. glauca growing at low elevations (Fraser
T9581 Owens and Molder 1976, 1977e), bud
determination begins about mid-July. In
P. engelmannii growing at higher elevations
determination begins in mid- to late July
(Harrison and Owens 1983). The time of cone-bud
determination within a species may vary with
elevation and latitude, therefore provenance
differences may be significant. Pollen-cone
development is completed first, usually by
late-September or early-October. Seed-cone and
vegetative bud development may continue until
mid-October in interior species (Owens and
Molder 1977e; Harrison and Owens 1983) (fig. 4).
All microsporophylls and microsporangia ,
bracts and functional ovuliferous scales, and
leaves are initiated before buds become dormant
(Owens and Molder 1984c). Axillary buds have
the same potential pathways as in Pseudotsuga
(fig. 6) and the abundance of cone buds is
determined more by the pathways along which buds
develop than by the number of axillary buds
initiated. Terminal cone buds halt the future
growth of a shoot and abundant terminal cones
can reduce the cone-bud production and crown
expansion of a tree.

Tsuga (hemlocks)

Only J_. heterophylla (Owens and Molder 1974a)
and TT. mertensiana (Owens 1984a) have been
studied and these are summarized by Owens and
Molder (1984d). There is considerable overlap
of seed- and pollen-cone buds in the crown and
on branches. Seed-cone buds are terminal (some
exceptions occur during cone-induction) and
develop from apices which have been vegetative
for 1 or more years. They form on vigorous
lateral shoots in distal portions of branches.
Pollen-cone buds usually develop from
newly-initiated axillary buds on short lateral
shoots in proximal portions of branches. They
commonly form a cluster of buds at the base of

the shoot but the terminal apex may also develop
into a pollen cone. Cone position is
essentially the same in both species (Owens and
Molder 1984d). In coastal low elevation
J_. heterophylla , pollen-cone buds become
determined in late June and seed-cone buds in
mid-July (fig. 4) . In coastal high elevation
J_. mertensiana, both pollen- and seed-cone buds
become determined in late July (fig. 8).

Terminal, potential seed-cone apices have
limited pathways of development — they may
remain vegetative or differentiate into seed
cone buds after bud scales are initiated.
Axillary, potential pollen-cone buds may abort,
become latent or differentiate into pollen-cone
buds. In J_. heterophylla, cone buds continue
development until late fall, whereas in
T. mertensiana development stops by mid- to late-
October. All microsporophylls and
microsporangia, bracts and functional
ovuliferous scales are initiated before
dormancy. Prolific terminal seed-cone
development is common and may limit subsequent
vegetative growth of the branch.

Larix (larches)

Larix has dwarf (short) and long shoot buds.
L^. occidentalis (Owens and Molder 1979b) and
L. laricina (Powell and others 1984) have been
studied in detail and the position of cone-buds
is generally considered to be similar in other
species of Larix (Dallimore and Jackson 1966).
Pollen- and seed-cone buds normally
differentiate from vegetative terminal apices on
dwarf shoots that are at least 1-year-old
(fig. 9). Occasionally cone buds develop from
terminal-long-shoot buds on suppressed branches.
Pollen-cone buds are commonly proximal on
nonvigorous, often pendant, long shoots.
Seed-cone buds are commonly distal on vigorous,
but only slightly pendant to upswept long

BUD - SCALE

AXILLARY

BUD
NIDATION

BUD SHOOT
ENLARGEMENT j ELONGA-
VEGETATlVE T|0N

INITIATION

DORMANT
VEGETATIVE
BUD

OVULIFEROUS
SCALE
INITIATION

MICROSPOROPHYLL

SHOOT
SATURATION

DORMANT
SEED - CONE
BUD

DORMANT
POLLEN - CONE
BUD

AUG SEPT. OCT NOV DEC

Figure 8. — Phenology of bud differentiation and
development in mountain hemlock (redrawn from
Owens 1984a).

18

• uo - KM!

INITIATION

5 >
i -

A1IAL lfAl>
INIT 1 AT ION

DORMANCY

s

■ AflO

SlOW

z

.3

INITIATION

BRACT

• DEVELOPMENT

Z

o

9

55

<

* <

ovuhterous

SCALE

DORMANCY

o„

1 3
O-

>

O

INITIATION

Z
<

z

o

o

MIC ROSPOBANC At

z
o

o

o

a

0

i

z

DEVELOPMENT

DORMANCY

pi

USMING

LEAF

SI T 1 AT 1 ON

D

S I Ow

DORMANCY

s

... | ... |

M At | A Pfi 11

|«a, 1

UNI

AUG | SEPT | OCT

| MOV | DEC

Figure 9. — Phenology of vegetative short shoot
(VSS), long shoot (VLS), pollen-cone and
seed-cone bud development in L. occidentalis
(from Owens and Molder 1979b)7

shoots. Considerable overlap occurs in
distribution of cone-buds in the crown and along
the branches (Owens and Molder 1979b).
In L. occidentalis pollen-cone and seed-cone
buds become determined in mid-June.
Microsporophyll initiation is complete in about
6 weeks and is followed by microsporangial
development. Bracts and ovuliferous scales are
initiated until early November when both
seed and pollen-cone buds become dormant (fig. 9)
(Owens and Molder 1979b). Similar detailed
studies have not been made for other Lar ix
species. Related genera, Cedrus and
Pseudolarix, within the Laricoideae of the
Pinaceae have similar distributions of cones but
details of development are not known.

Axillary buds are initiated at the base of the
LSB in the spring or early summer. They
initiate bud scales then differentiate Into
pollen-cone or dwarf-shoot buds during the
summer. A more-distal group of axillary buds
differentiate into dwarf shoots bearing the
characteristic number of needles for the
species. The most-distal axillary buds are
initiated in late summer and after a period of
bud-scale initiation, differentiate into either
lateral long-shoot or seed-cone buds (fig. 10).
In hard pine (diploxylon) , lateral long-shoot
and seed-cone buds differentiate in late summer
or early fall in north temperate regions (Sacher
1954; Duff and Nolan 1958; Gifford and Mirov
1960; Van den Berg and Lanner 1971; Curtis and
Popham 1972; Owens and Molder 1975a) (fig. A).
In soft pines ( haploxylon) , distal axillary
primordia do not differentiate before the LSB
becomes dormant. Immediately following dormancy
they differentiate into either lateral LSBs or
seed-cone buds (Sacher 1954; Owston 1969; Owens
and Molder 1977b, c) (fig. 4).

In pines having polycyclic LSBs, more than one
series of seed-cone buds may occur in a LSB but
the first series usually bears most of the
seed-cone buds (Lanner and Van den Berg 1975;
Owens and Molder 1975a; Sweet 1979; Greenwood
1980). Polycyclic LSBs occur in many hard pines
and are most common in those from southern
latitudes. They have not been reported in soft
pines .

Cupressaceae (cypress family)

The time and method of cone initiation has been
determined only for Cupressus arizonica (Owens
and Pharis 1967), Thuja pllcata (Owens and
Pharis 1971) and Chamaecyparls nootkatensis
(Owens and Molder 1974b, 1984b)~ In the

P inus (pines)

In pines, cone buds are initiated in a complex
vegetative long-shoot bud (LSB) rather than on
an elongating shoot or a dwarf shoot. The LSB
consists of a series of scale leaves
(cataphylls) which are initiated throughout the
growing season. Most cataphylls have an
axillary apex which initiates a series of bud
scales, then differentiates into a dwarf shoot,
pollen-cone or lateral LSB (fig. 10).
Therefore, axillary bud initiation and
differentiation occur throughout the growing
season. The time when an axillary bud
differentiates is determined by its position in
the LSB — the proximal buds which were initiated
first differentiate before the more distal
axillary buds (Doak 1935; Sacher 1954; Duff and
Nolan 1958; Owston 1969; Van den Berg and Lanner
1971; Sucoff 1971; Curtis and Popham 1972;
Lanner and Van den Berg 1975; Owens and Molder
1975a, 1977b, c) . LSB's may be monocyclic
consisting of one complete sequence (fig. 10) or
polycyclic consisting of two or more sequences.

SEED CONE POLLEN CONE

BEARING LSTB BEARING ISTB

Figure 10. — The development of monocyclic LSBs
of _P. monticola . The columns on the left and
right Indicate the time of initiation and
proportion of cataphylls and axillary structures
initiated during this time. The center columns
Indicate the approximate times of
differentiation of the axillary structures on
each LSTB (from Owens and Molder 1977b).

19

Cupressaceae , cone buds form by the transition
of a vegetative apex into a reproductive apex.
Buds are not enclosed by bud scales. Seed cones
are terminal on short lateral shoots located on
distal portions of vigorous shoots. Pollen
cones are terminal on proximal, less-vigorous
lateral shoots. There is considerable overlap
in cone distribution on branches and in the
crown. Shoot elongation and leaf and axillary
bud initiation occur over most of the growing
season. Transition from a vegetative to a
pollen-cone apex begins in early June in
T. plicata, and in late June in C. nootkatensis .
Transition to seed-cone apices follows in 4
weeks and 1 week, respectively (Owens and Pharis
1971; Owens and Molder 1974b, 1984a) (fig. 4).
Only pollen cone initiation as a result of
gibberellin Aj treatments, rather than under
natural conditions, was studied in
Cupressus arizonica (Owens and Pharis 1967). In
the Cupressaceae which have been studied, all
microsporophylls and microsporangia,
bract-scales and ovules are initiated before
cone-buds become dormant in the fall.

FACTORS AFFECTING CONE INITIATION

Knowledge of time and method of cone initiation
is useful in interpreting environmental or
developmental factors which may affect floral
initiation and in determining the correct time
to attempt cone enhancement or induction.

Many studies have tried to demonstrate a
relationship between environmental factors and
cone initiation in regions, stands and
individual trees (see reviews by Mathews 1963;
Jackson and Sweet 1972; Puritch 1972; Lee 1979;
Owens and Blake 1985). Correlation between seed
crop and weather data are tempting. However,
Rehfeldt and others (1971) cautioned that, "In
any attempt to correlate previous cone crops and
previous weather, problems arising from
intercorrelations among the dependent and
independent variables will be encountered; it is
difficult to identify causal mechanism because
of these intercorrelations."

Despite the problems, several factors have been
identified. The very old concept that high
summer temperatures favor increased floral
initiation has been demonstrated in several
conifer genera which grow in the Inland Mountain
West, including Pinus (Maguire 1956; Daubenmire

1960) , Pseudotsuga (Lowry 1966; Van Vredenburch
and La Bastide 1969; Eis 1973), Abies (Eis
1973), Picea (Fraser 1958) and Larix (Yanagihara
and others 1960). Increased light intensity
through thinning has enhanced cone production in
Pinus ponderosa (Barnes 1969) and Pseudotsuga
(Reukema 1961). Photoperiod may effect
sexuality of cones in P. contor ta (Longman

1961) , Tsuga heterophylla (Owens and Molder
1974a), Chamaecyparis nootkatensis (Owens and
Molder 1974b) and Thuja plicata (Pharis and Morf
1967). Studies of moisture have given
conflicting results primarily because times of
cone-bud differentiation were not known.
Generally moisture stress has enhanced seed cone

initiation in Pinus monticola (Rehfeldt and
others 1971), Pseudotsuga (Eis 1973; Lowry
1966), Abies grandis (Eis 1973) and Picea glauca
(Fraser 1958). Frost resulting in lesions and

girdling has increased cone production in
Pseudotsuga (Ebell 1971).

The lack of positive correlations with
environmental factors may result from endogenous
factors within the trees. Cone-bud
differentiation occurs during shoot elongation
and maturation of subtending cones (Allen and
Owens 1972; Owens 1984b) which are strong
metabolic sinks (Ching and Ching 1962). No
matter how favorable environmental factors might
be, they rarely override the effects of bearing
a heavy cone crop and consecutive heavy cone
crops do not occur. Bumper crops may occur when
endogenous and favorable environmental factors
are in phase.

CONE INDUCTION

Cone induction in juvenile or otherwise
non-reproductive conifers is a valuable tool in
genetic tree improvement programs and In seed
production for reforestation. Successful
treatments are numerous and results variable
depending upon the species, growing conditions
and time of treatment. Any cone induction
treatment must be applied before the time of
anatomical differentiation of buds (fig. 4).
How long before and for what duration are
unknown for most species.

Various cultural treatments have been reviewed
by Puritch (1972), Jackson and Sweet (1972),
Brazeau and Veilleux (1976), Lee (1979) and
Owens and Blake (1985). Successful treatments
of genera from our region include: (1)
fertilizers in hard pines (Lee 1979), soft pines
(Schubert 1956; Barnes and Bingham 1963),
Pseudotsuga (Ebell and McMullan 1970; Ebell
1972a, b) and Picea (Hoist 1971); (2) girdling,
banding and strangulation in Pseudotsuga (Ebell
1971) and Larix (Melchior 1960, 1961a, b); (3)
moisture stress, usually applied by root
pruning, in Pinus (Stephens 1961, 1964),
Pseudotsuga (Melchior 1968; Ross and others
1985), Larix (Heitmuller and Melchior 1960) or
withholding water in potted trees of Tsuga,
Picea and Pseudotsuga; (4) growth of seedlings
under continuous light to reduce the age to
flowering in Picea (Young and Hanover 1976) and
Pinus (Wheeler and others 1982); (5) altered
photoperiod which may affect sexuality of cones
in Pinus (Longman 1961), Picea (Durzan and
Carabell 1979), Larix (Yokoyama and Asakawa
1973), Thuja (Owens and Pharis 1971) and
Chamaecyparis (Owens and Molder 1977f); and, (6)
shoot pruning in Pinus (Coffen and Bordelon
1981) and Pseudotsuga (Copes 1973). Many of
these cultural treatments are most successful
when used in conjunction with growth
regulators .

About 100 research papers describe results
utilizing a variety of species, growth
regulators, concentrations of growth regulators,

20

times and methods of application plus several
adjunct treatments. These are reviewed or
tabulated by Owens and Blake (1985). The most
successful treatments have used glbberellins
(GAs). GA j was first shown to induce cones in
some members of the Cupressaceae and Taxodlaceae
in the late 1950's (Kato and others 1959).
Since then 19 species in 10 genera within these
families have responded positively to GA ^
treatments (Owens and Blake 1985). More
recently less-polar GAs, usually a mixture of
GA4/7, have induced cones in 17 species of the
Pinaceae representing five genera (Lar ix, Picea,
Pinus, Pseudotsuga and Tsuga) (Owens and Blake
1985) . Best results have been achieved when
GA4/7 was applied with some adjunct cultural
treatment. Progress has been slower in the
Pinaceae because: less-polar GAs are not as
available and are more expensive than GA3;
treatments are not as easily applied (foliar
sprays may not be effective); the timing of
treatment is more critical since it must
correlate with stages of bud development; the
length of treatment may be longer (commonly 6
weeks or more); adjunct treatments are often
necessary; and sexuality of cones is not easily
manipulated. Much more research is needed on
all of these problems in many species. Recent
research on containerized (potted) rooted
cuttings, grafts and seedlings (S.D. Ross,
personal communication) shows promise for the
development of containerized seed orchards of
some species. Basic research on the mode of
action of GAs must continue if cost-effective
treatments are to be developed.

Most cone induction researchers are confident
that seeds resulting from induced cones are
comparable in quality to those from noninduced
trees because cone induction treatments are of
short duration and precede seed maturation by
one or more years. However, excessive flowering
can result in ovule, seed and cone abortion and
perhaps seed of poor quality due to competition
for resources. This may be especially true in
small trees but may be most easily overcome in
containerized trees.

CONE-BUD DEVELOPMENT

Seed-Cone Development

Seed cones of most conifers are preformed within
the bud and any ovuliferous scales initiated
after dormancy are sterile (Owens and Blake
1985). Exceptions occur in Pinus. In hard
pines, one-third to all ovuliferous scales are
initiated before winter dormancy, depending upon
the position of the cone-bud on the shoot (Owens
and Molder 1975a). In soft pines (see fig. 4).
all bracts and ovuliferous scales are initiated
after winter dormancy (Owens and Molder 1977b).
We know little about the effect of overwintering
conditions on seed-cone buds. Seed-cone buds
resume development before vegetative buds, often
during harsh late winter weather conditions.
Cone-bud abortion is a common but poorly
documented phenomenon in some species.

Pollen-Cone Development

Pollen cones initiate all microsporangla (pollen
sacs) before winter dormancy (see fig. 4).
However, the stage of microsporangial
development reached before dormancy varies
between genera (fig. 11). Overwintering may
begin when pollen cones are at the sporogenous
stage in Pinus (Kupi la-Ahvenniemi and others
1978; Owens and Molder 1977c; Owens and others
1981b) or at the pre-melotlc pollen mother cell
(PMC) stage as In Picea and Abies (Owens and
Molder 1977d, 1979a; Singh and Owens 1981a, b;
Harrison and Owens 1983). In these genera,
raeiosis and pollen development occur after
winter dormancy. Ultrastructural studies of
overwintering sporogenous cells of Pinus
(Kupila-Ahvenniemi and others 1978; Cecich 1984)
and PMCs of Pseudotsuga (Singh and others 1983)
have shown that a true dormant period does not
occur, rather nuclear and cytoplasmic changes
occur throughout the winter. This period Is
more correctly called a period of reduced
activity rather than dormancy. Similar studies
have not been made for Abies or Picea .

In Lar ix, Pseudotsuga , Thuja and Tsuga, meiosis
begins In the fall, then becomes arrested when
PMCs reach either the pachytene or diffuse
diplotene stages of meiosis (Eriksson 1968a;
Eriksson and others 1970a, b; Owens and Molder
1971a, b; Hall 1982). After dormancy meiosis is
rapidly completed, followed by pollen
development .

In Chamaecypar is and Juniperus , meiosis and
pollen development occur before winter dormancy
(Owens and Molder 1974b). Overwintering pollen
cones contain mature, dry pollen. No structural
changes have been observed during winter.

Pollen-Cone Ml crosporophyl 1 Mlcroapor<inKl<il SporoRenous

Initiation Development Development Tissue Development

/ Dornancy
/ (Pinup

Pollen Mother

Melotlc — •» Microspores •» Microspore — m» Cell Division In — »»M4tjrc

Division Development Pollen Devclopocnt Pollen

Pollen Shed
(Anthcsls)

Figure 11. — Stages of pollen and pollen-cone
development and times when dormancy may occur in
different genera (from Owens 1982).

21

Winter temperatures and the stage at which
pollen cones overwinter may affect pollen-cone
buds and pollen quality (Eriksson and others
1970a). A higher incidence of pollen
abnormalities occurred in Larix growing in
Sweden, when PMC's developed beyond the diffuse
stage before winter dormancy (Ekberg and others
1967; Eriksson 1968a, b) . Consequently, trees
moved to seed orchards in colder areas may have
a higher incidence of pollen inviability or
abnormalities .

Following meiosls, the tetrads of haploid
microspores remain within the PMC wall for a
brief time. They soon swell, burst out of the
PMC wall and become suspended in fluid within
the microsporangium where subsequent pollen
development occurs (Singh 1978).

Two patterns of cell division occur during
pollen development. In the Pinaceae (fig. 12A) ,
prothallial cells form which have no known
function. Pollen may be shed at the 4- or
5-celled stage. The body cell forms the two
male gametes after pollination. Pollen is
generally large, sacci (wings) are present in
some genera and storage products are in the form
of starch (Owens 1982; Owens and Molder 1971b,
1975b, 1977c, d, 1979a, b; Owens and others
1981b; Singh 1978; Singh and Owens 1981a, b) .

In the Cupressaceae, Taxodiaceae and Taxaceae
(fig. 12B) , pollen has no prothallial cells and
is shed at the 1- or 2-celled stage. The
generative cell forms two male gametes after
pollination. In these families, pollen is
small, lacks sacci, is sculptured with orbicules
and the storage products are oil droplets (Owens
1982; Owens and others 1980; Singh 1978).

The time sequence of meiosis and pollen
development varies between species but can be
separated into five stages of development:
pre-meiotic division, meiotic division,

microspore development, cell divisions, and
anthesis. The time spent at each stage varies
between species and with weather. The time from
the end of pollen-cone dormancy to anthesis may
be as little as 1 week in C. nootkatensis (Owens
and Molder 1974b), to 12 weeks in
Tsuga mertensiana (Owens and Molder 1975b). The
effects of temperature on the rate of different
stages of development are not known. However,
increased temperatures will shorten the total
time to anthesis (Sarvas 1962, 1965; Winton
1964; Boyer and Woods 1973) and forcing pollen
for early pollen extraction is possible.
Generally, the long period of post-meiotic
pollen development makes it possible to control
the rate of development in many species but the
extent to which pollen can be manipulated before
pollen development, vigor or viability are
affected has not been determined.

POLLINATION MECHANISMS

'Pollination mechanism' is the term used to
describe the process of pollen capture and
entrance of pollen or pollen tubes into the
ovules (Doyle 1945). The subject has been
reviewed by Doyle (1945), Dogra (1964), Singh
(1978), Owens (1980) and Owens and Blake (1985).
All conifers are wind pollinated but differences
occur in the pollination mechanisms.

Two mechanisms involving a pollination drop
exist. In the Cupressaceae, Taxodiaceae and
Taxaceae, ovules are flask-shaped with a narrow,
short neck out of which a pollination drop is
exuded. The pollination drop is a clear, dilute
solution of sugars (McWilliam 1958). Conelets
open in the spring exposing the ovules. Within
a few days pollination drops are exuded from
some of the ovules at night and withdrawn during
the day for 2 or 3 days. Pollination drops then
remain exuded throughout the day and night for a
few days. This is followed by a few days when

First

Prothallial
Cell

Prothallial
Cell

Stalk Cell

Pollen Hothe
Cell Meiosls

tetrad of
- microspores

Embryonal
• Ce 1 1

Antherldlal Gener
.Initial -»-Cell

- Pollination

Pollen Mother
Cell Meiosls -

Tetrad of
Microspores

Microspores

(1-celled pollen)

Generative Cell
(2-celled pollen)

Pollination

Figure 12. — A. Pollen development in Abies, Larix, Picea,
Pinus, Pseudotsuga, and Tsuga (from Owens 1982). B. Pollen
development in Chamaecyparis , Juniperus , Taxus , and Thuja.

22

TOTAL FINE PROM END Of DORMANCY

EM) OF DORMANCY ( ) UNTIL 1ST NEIOTIC HEIOTIC MICROSPORE

species until Mmusis Division Division development cell divisions

UIEI AMARtLIS
AtlES LASIOCARPA

cha.maecypaa1s
nootkatensis

UUtll OCCIDENTALS

PICEA INCELMANNII

PICEA CLAUCA

PICEA SITCHENSIS

PINVS CONTORT* VAX.
LATIPORIA

PINVS NONTICOLA

THUJA PLACATA
TSUCA MERTEHSIANA

(PARLY APRIL) ' IRS
(LATE MARCH) 8-9 UKS
(EARLT APRIL) 1 UK

(PARLY MARCH) S-7 UKS
(MID-APRIL) »-tl UKS
(MID-APRIL) 7-« UKS
( EARLY MARCH) J-t UKS
(EARLY APRIL) 10 UKS

(MID-MARCH) » UKS

(MID-APRIL) > UKS
(LATE PER) 7-« UKS

(MID-FEi) 6-7 UKS
(LATE MARCH) 12 UKS

- - -» ANTHESI S

— ANTHESIS

• ANTHESI S

- ANTHESIS

ANTHESI S

ANTHESIS

- ANTHESIS

> -.ANTHESIS

_ ANTHESIS

. ANTHESIS

- — ANTHESIS

ANTHESI S

Figure 13. — Phenology of post-dormancy pollen-cone
development in 13 conifers (from Owens 1982).

pollination drops are exuded only at night, then
they are withdrawn permanently. Pollen landing
on a pollination drop immediately sinks into the
drop. Pollen adhering to the edge of the
micropyle, before pollination-drop formation,
is picked up by the pollination drop (Owens and
others 1980).

A pollination drop is also exuded in Pinus and
Picea but here the ovule is inverted and the
ovule tip consists of two micropylar arms
(McWilliam 1958; Lill and Sweet 1977; Owens and
others 1981b; Singh and Owens 1981a; Owens and
Blake 198A). The micropylar arms secrete minute
sticky droplets to which pollen adheres for
several days before a pollination drop forms.
This provides a method by which pollen is
collected and later taken in when the
pollination drop floods the space between the
micropylar arms (Owens and others 1981b; Owens
and Blake 1984). McWilliam (1958) suggested
that the pollination drop was produced by
guttation but a recent study (Owens and Blake
1984) shows the drop is secreted by nectary-like
tissue at the tip of the necellus.

In Abies the integument tip is funnel-shaped and
lobed but no pollination drop forms (Owens and
Molder 1977d; Singh and Owens 1981b). Minute
droplets are secreted on the edge and inner
surface of the funnel to which pollen adheres.
The funnel then crimps inward carrying pollen
closer to the nucellus.

Two pollination mechanisms occur in Tsuga. In
the more primitive T. mertensiana , the two
micropylar arms are broad flaps which secrete
minute droplets to which pollen adheres. The
flaps collapse, entrapping pollen grains which
germinate, forming pollen tubes that grow into
the micropyle (Owens and Blake 1983). In
J_. heterophylla , pollen has spines which adhere
to long web-like cuticular hairs on the abaxial
surface of the bract. After the
ovulif erous-scales overgrow the bracts, the
pollen grains germinate and form very long

pollen tubes that grow over the bracts and into
the raicropyles of ovules of more distal
ovulif erous-scales (Colangeli and Owens 1984).

In Pseudotsuga (Allen 1963; Ho 1980; Owens and
others 1981a) and Larlx (Owens and Molder 1979c;
Villar and others 1984) the stigmatic tip
develops into two unequal lobes covered with
unicellular stigmatic hairs, and the micropyle
is a narrow slit, too small for pollen to enter.
Pollen sifts between the bracts of erect
conelets and becomes entangled in the stigmatic
hairs. This occurs for several days, then cells
around the micropyle collapse and cells on the
surface of the lobes elongate, carrying
stigmatic hairs and attached pollen into the
micropyle. Complete engulfraent of stigmatic
hairs occurs within about 2 weeks. There are no
secretions on the stigmatic hairs and no
pollination drops.

The optimal time for pollination and the
duration of effective pollination vary for
maximum seed yield with the pollination
mechanism. In species having a pollination
drop, most effective pollination occurs when the
pollination drop is present (Owens and others
1981b; Owens and Blake 1984). However, in those
that secrete microscopic droplets on the
micropylar arras, pollen is also effectively
collected on the micropylar arras. Experiments
with Pinus (Owens and others 1981b) and Picea
(Owens and Blake 1984) using stained pollen
applied at various times show that some of this
early pollen which adheres to the arms is taken
into the ovule, but the most effective
pollination occurs when the ovule has an exuded
pollination drop. In these studies, and more
recent studies of P_. glauca (J.N. Owens,
unpublished data), it was shown that not all
ovules in a cone exude pollination drops
simultaneously. Exudation occurs over several
days and progresses acropetally in the cone.
Therefore, different regions of each cone are
most receptive at different times. This has
obvious implications for controlled and

23

supplemental pollinations and implies that
maximum seed efficiency in the field may occur
when pollen arrives at the conelets over several
days .

Experiments using stained pollen applied at
different times throughout the receptive period
show that, in Pseudotsuga, pollen is accumulated
over several days and the most effective time
for pollination is within about 4 days from when
conelets first become receptive (Ho 1980; Owens
and others 1981a). The first pollen to arrive
at the stigmatic surface is taken into the ovule
preferentially over pollen arriving later (Owens
and Simpson 1982). In Picea glauca that was
pollinated at different times with colored
pollen, the pollen applied early was also most
likely to be taken in and accomplish
fertilization (R. Ho, personal communication).

Other pollination mechanisms which rely on a
collection process (Abies and Tsuga) may have
long receptive periods without optimal times.
This has been demonstrated in T. heterophylla
where each day for more than two weeks,
previously unpollinated cones were pollinated,
but the seed set was essentially the same for
all dates (Colangeli and Owens 1984).

POST-POLLINATION FACTORS AFFECTING CONE AND SEED
DEVELOPMENT

The time between pollination and fertilization
is variable according to species (figs. 1-3).
During this period seed cones rapidly enlarge
and increase in dry weight (Ching and Ching
1962), ovules enlarge and female gametophytes
mature (Owens and Blake 1985). After
fertilization embryos and seeds develop. Both
endogenous and exogenous factors may affect
ovule, seed and cone development but these
factors are poorly understood (Owens and Blake
1985).

Cone Abortion

Conelet abortion at and soon after pollination
is common and thought to be primarily caused by
low temperatures (Hard 1963; Hutchinson and
Bramlett 1964; Krugman 1966), although there
have been no experimental studies to prove this
(Owens and Blake 1985). Death of the conelet
commonly begins in the cone axis (White and
Knopp 1978) and ovules, followed by wilting and
browning of the bracts. Most studies of conelet
loss have used Pinus , where this can usually be
traced back to inadequate pollination. Sarvas
(1962) estimated that 80 percent of conelet drop
in P_. sylvestris resulted from Inadequate
pollination and the remaining 20 percent
resulted from damage to the conelet. He also
estimated that when more than 20 percent of
potentially fertile ovules aborted (due to lack
of pollen) the conelet aborted. Strong clonal
differences in amount and timing of conelet drop
occur in Pinus (Sweet and Thulin 1969; Forbes
1971). Sweet (1973) and Owens and Blake (1985)
provide complete reviews of the limited

literature in this area. Generally, the
physiology of conelet drop is poorly understood
but the ability to induce cones on small
containerized trees now makes experimental
studies possible.

Post-fertilization cone loss has been reported
(Brown 1970; Bramlett 1972) but is not common.
Rehfeldt and others (1971) described this in
_P. monticola and showed it to be influenced by
early summer water deficit. Sweet (1973)
suggested the low cone loss at this time was due
to less competition for nutrients since seeds and
cones have approached full size by the time of
fertilization. The seed coat is well developed
at fertilization and seeds and seed wings have
usually separated from the ovuliferous scales.
No vascular connections exist between seed and
cone at this time (Owens and others 1982; Singh
and Owens 1981a, b, 1982) and seeds would no
longer be strong metabolic sinks.

Pref ertilization Factors Affecting Seed
Development

The site and time of pollen germination vary.
In Tsuga heterophylla very long pollen tubes
grow into the micropyle (Colangeli and Owens
1984). In Pinus and Picea the pollination drop
draws pollen to the nucellar tip where it
soon germinates (Sarvas 1968; Owens and others
1981b; Owens and Blake 1984). In Abies (Owens
and Molder 1977d; Singh and Owens 1981b, 1982)
pollen germinates in the short micropylar canal
near or on the nucellus. In Larix and
Pseudotsuga, engulfed pollen elongates down the
micropylar canal but a narrow pollen tube does
not form until contact is made with the nucellus
(Owens and Molder 1979c; Allen and Owens 1972).
In all but Pinus and Picea there is considerable
time between pollination and pollen germination
but the stimulus for germination (or its
inhibition) are not known. Many factors could
prevent or retard pollen germination and pollen
tube growth which would reduce seed set.

Pollen tube growth through the nucellus may take
as little as one week in Pseudotsuga
(J.N. Owens, unpublished data) to one year in
Pinus (Sweet 1973). Competition between pollen
tubes may occur during this period. The
mechanisms which control pollen tube growth and
penetration of the nucellus are not known.
These aspects of conifer reproduction have not
been studied.

Interractions may occur between pollen and
pollen tubes and the nucellar and female
gametophyte tissues. Such systems have been
extensively studied in flowering plants but
intraspecif ic prezygotic self-incompatabillty
has not been reported in conifers (Hagman 1975).
However, recent data shows that up to 20 percent
of failure to set seed occurs before
fertiization (A. Colangeli, personal
communication). Pettitt (1977a, b, 1979, 1982)
has detected proteins ad glycoproteins in the
pollen wall of the primitive gymnosperra Cycas
and has demonstrated inter-species and

24

Inter-tissue precipitation reactions between
various components of ovular tissues. This
suggests that proteins could be Important in
prezygotic incompatability during pollen
germination and pollen tube growth through the
nucellus, raegaspore wall and female garaetophyte,
all of which are genetically different from the
pollen. Seriologically active substances have
been demonstrated in Plnus pollen (Hagman 1975)
and Immunochemical differences have been
detected in female gametophytes of different
£. strobus trees (Eckert and Eckert 1984).

These observations indicate that prezygotic
stages of development have a potentially great
effect on seed set. Consequently, we have
embarked on a light and electron microscopy
study of pollen tube growth and interspecific
pollen competition using controlled pollinations
of ^ labelled pollen in Tsuga heterophylla,
Picea glauca and Pseudotsuga menziesli . In
addition, immunocy tochemical techniques are
being developed to study specific proteins
produced by pollen tubes in Picea.

Deficient ovules may result from ovule abortion
or lack of ovule development. All cones have
ovuliferous scales at the base, and to a lesser
extent at the tip, which bear either no ovules
or rudimentary ovules. Rudimentary ovules
either do not fully develop or develop too
slowly to be pollinated (Owens and others 1981a,
b) . This is most obvious in some species of
Plnus where the majority of scales bear no
fertile ovules (Sarvas 1962; Owens and others
1982). Deficient ovules also occur in fertile
regions of cones but this has not been carefully
studied with regard to cause or frequency within
cones, trees or species. Whatever the causes,
deficient ovules can significantly reduce viable
seed set.

Postfertilization Factors Affecting Seed
Development

The literature on conifer embryogeny is
extensive and most genera have been described
(Roy Chowdhury 1962; Owens and Blake 1985).
Since most conifers are multiarchegoniate , more
than one egg may be fertilized and resultant
proembryos develop at about the same rate.
Fertilization of one egg in a female gametophyte
does not appear to prevent subsequent
fertilization of other eggs; however,
unfertilized eggs rapidly degenerate as adjacent
embryos develop. The fertilization of more than
one egg, simple polyembryony (SPE), is common
but not universal. The resulting embryos have
different genotypes. Rarely does more than one
proembryo develop into an embryo.

The proembryo stage ends when suspensor cells
force the apical cells into the female
gametophyte. Cells of the apical tier divide
and may form a single embryo or separate into
four filaments, each of which may develop into a
separate embryo. This is called cleavage
polyembryony (CPE) and the embryos are
genetically identical.

Embryo development may continue normally (figs.
1, 2), be Inhibited but resume development when
conditions are favorable (fig. 3), or the
embryos may degenerate. Degeneration has been
traced to physiological incompatability between
embryos and the female gametophyte as a result
of self-pollination in Plnus (Hagman and Mikkola
1963), Pseudotsuga (Orr-Ewing 1957), and Picea
(Mergen and others 1965). In other conifers
such as Abies (Sorensen 1982) self-pollination
does not lead to embryo degeneration. It has
generally been thought that selection against
selfing occurs by low self-embryo viability
(Sorensen 1982) and that selection occurs during
the post-zygotic stages (Wlllson and Burley
1983). Some of these concepts may need to be
re-evaluated after prezygotic stages have been
thoroughly investigated.

Irregularities in embryogeny have been
demonstrated (Dogra 1967; Owens and Blake 1985)
but their frequency, causes and final Impact on
seed set are generally not known. Dogra (1967)
observed that the occurrence of any one
Irregularity may not be common but all together
they played a significat role in seed sterility.
Unfortunately, large samples and numerical data
are generally lacking for embryologlcal studies.
Embryo retardation and abortion are common but
the causes must be studied during development
rather than after seeds are mature. Classes of
normal appearing seeds have been described based
on the condition of female gametophyte and
embryos (Dogra 1957; Owens and Blake 1985).
This approach, but with emphasis on the causes,
should be applied to more conifers.

Seed and Cone Maturation

Seed and cone maturation have not been
thoroughly studied developmentally or
physiologically. Most studies have dealt with
assessing the time of cone collection which is
the subject of the next session. Only a few
comments relate to the present topic.

Changes within the female gametophyte and embryo
during seed development have not been well
described for conifers (Owens and Blake 1985).
One of the most complete studies was that of
Ginkgo (Favre-Duchartre 1956, 1958), a
primitive gymnosperm with seeds similar to
conifers. Similar but less complete studies
have been made for the Pinaceae (Hakansson 1956;
Takao 1960). Hakansson (1956) followed embryo
development and the state of food materials
in the female gametophytes of Picea and Plnus
and Slmola (1974) made an ultrastructural study
of the embryo and female gametophytes in dry and
germinating seeds of P^. sylvestris. Many more
complete developmental and physiological studies
are needed.

SUMMARY

Cone and seed biology of conifers is a broad
subject, the species are many but the
researchers are few. Cones and seeds are

25

essential for reforestation but many factors
often prevent their production. These factors
are generally poorly understood. Cone and seed
production can be significantly increased for
many species by cone induction or enhancement
treatments and improved pollination techniques.
Many factors are difficult to control but
warrant further study. These include factors
which prevent fertilization, cause ovule and
embryo retardation and abortion and cause the
abortion of cones. Much of the technology is
now available to significantly increase cone and
seed production in many forest trees. However,
further research is required to refine the
technology and provide basic information about
conifer reproductive development and
physiology.

REFERENCES

Allen, G.S. Origin and development of the ovule
in Douglas-fir. For. Sci. 9: 386-393; 1963.

Allen, G.S.; Owens, J.N. The life history of
Douglas-fir. Env. Can. For. Serv., Ottawa.
1972. 139 pp.

Barnes, B.V. Effects of thinning and

fertilizing on production of western white
pine seed. U.S.D.A. For. Serv. Res. Pap. No.
INT-58. 1969.

Barnes, B.V.; Bingham, R.T. Flower induction
and stimulation in western white pine.
U.S.D.A. For. Serv. Res. Pap. No. INT-2,
1-10. 1963.

Boyer, W.D.; Woods, F.W. Date of pollen
shedding by longleaf-pine advanced by
increased temperatures at strobili. For.
Sci. 19: 315-318; 1973.

Bramlett, D.L. Cone crop development records
for six years in shortleaf pine. For. Sci.
18: 31-33: 1972.

Brazeau, M.; Veilleux, J.M. Bibliographie
annote'e sur les effets de la fertilization
sur la production de cones et de semences.
Quebec, Ministere des Terres et Forets,
Service de la Recherche, Me'moire 25. 1976.

Brown, I.R. Premature cone loss in grafted
clones of Scots pine. In: Sexual
reproduction of forest trees. Vol. I.
Finnish Forest Res. Inst., Helsinki. 1970.
15 pp.

Cecich, R.A. The histochemistry and

ultrastructure of jack pine microsporangia
during the winter. Amer. J. Bot. 71:
851-864; 1984.

Ching, T.M.; Ching, K.K. Physical and

physiological changes in maturing Douglas-fir
cones and seed. For. Sci. 8: 21-31; 1962.

Coffen, D.O; Bordelon, M.A. Stem breakage
effect on cone and pollen production in
Pinus monticola (Dougl.). U.S.D.A. For. Ser.
Res. Note INT-312. 1981.

Colangeli, A.; Owens, J.N. Pollination

mechanism in western hemlock. In: Genetic
Gain: Incorporation and Implications for
Forestry. Proc. West. For. Gen. Ass. 1984
Annual Meeting. 1984. 20 pp.

Copes, D.L. Efffect of annual leader pruning on
cone production and crown development of
grafted Douglas-fir. Silvae Genet. 22:
167-173; 1973.

Curtis, J.D.; Popham, R.A. The developmental
anatomy of long-branch terminal buds of
Pinus banksiana. Amer. J. Bot. 59: 194-202;
1972.

Dallimore, W.; Jackson, A.B. A handbook of
Coniferae and Ginkgoaceae. Revised by
S.G. Harrison. St. Martin's Press, N.Y.
1966. 729 pp.

Daubenmire, R. A seven-year study of cone
production as related to xylem layers and
temperature in Pinus ponderosa . Amer. Mid.
Nat. 64: 187-193; 1960.

Doak, C.C. Evolution of foliar types, dwarf

shoots, and cone scales of Pinus . 111. Biol.
Monogr. 13: 1-106; 1935.

Dogra, P.D. Pollination mechanisms in

gymnosperms. In: Advances in Palynology.
Nat. Bot. Gard. Lucknow. Ann. Rep. 1964.
1964. 142-175 pp.

Dogra, P.D. Seed sterility and disturbances in
embryogeny in conifers with particular
reference to seed testing and tree breeding
in Pinaceae. Stud. For. Suec. 45; 1967. 97
pp.

Doyle, J. Developmental lines in

pollination mechanisms in the Coniferales.
Sci. Proc. Roy. Dublin Soc. 24: 43-62; 1945.

Duff, G.H. and N.J. Nolan. Growth and

morphogenesis in the Canadian forest species.
III. The time scale of morphogenesis at the
stem apex of Pinus resinosa Ait. Can. J.
Bot. 36_: 687-706; 1958.

Dunberg, A.R. Flower induction in Norway
spruce. IURFO Norway Spruce Meeting
S. 2. 03. 11 - S. 2. 02. 11, Bucharest 1979. 1979.
139-157 pp.

Durzan, D.J.; Campbell, R.A. Red light inhibits
female cone production in white spruce trees.
In: F. Bonner (ed.) Proc. Symp. Flower and
seed development in trees. IUFR0. 1978.
Miss. State Univ. South For. Exp. Sta. U.S.
For. Serv. 1979. 57-62 pp.

26

Ebell, L.F. Girdling: its effect on

carbohydrate status and on reproductive bud
and cone development of Douglas-fir. Can. J.
Bot. 49: 453-466; 1971.

Ebell, L.F. Cone induction response of

Douglas-fir to form of nitrogen fertilizer
and time of treatment. Can. J. For. Res. 2:
317-326; 1972a.

Forbes, D.C. Loss of short leaf pine flowers
under natural conditions - a probable
explanation for high losses in controlled
pollinations. Tree Planters' Notes 22: 30;
1971.

Fowells, H.A. (compiler). Silvics of forest
trees of the United States. U.S.D.A. For.
Serv. Agric. Handbook. No. 271. 1965.

Ebell, L.F. Cone-production and stem-growth

response of Douglas-fir to rate and frequency
of nitrogen fertilization. Can. J. For. Res.
2: 327-338; 1972b.

Ebell, L.F; McMullan, E.E. Nitrogenous

substances associated with differential cone
production responses of Douglas-fir to
ammonium and nitrate fertilization. Can. J.
Bot. 48: 2169-2177; 1970.

Eckert, R.L.; Eckert, B.S. A procedure for
study of immunochemical properties of
megagaraetophy te tissue of Pinus strobus L.
Can. J. For. Res. 14: 142-145; 1984.

Eis, S. Cone crops of white and black spruce
are predictable. For. Chron. 43: 247-252;
1967.

Eis, S. Cone production of Douglas-fir and
grand fir and its climatic requirements.
Can. J. For. Res. 3: 61-70; 1973.

Eis, S.; Craigdallie, D. Reproduction of
conifers. A handbook for cone crops
assessment. Env. Can., Can. For. Serv., Pac.
For. Res. Gen., Victoria, B.C. B.C.-X-219;
1981.

Ekberg, I.; Ericksson, G. ; Kartel, N. ;

Sulikova, Z. The meiotic development in male
aspen. Stud. For. Suec. 58: 3-16; 1967.

Eriksson, G. Meiosis and pollen formation in
Larix. Roy. Coll. For., Stockholm. 1968a.
18 pp.

Eriksson, G. Temperature response of pollen
mother cells in Larix and its importance for
pollen formation. Stud. For. Suec. 63: 1-31;
1968b.

Eriksson, G.; Ekberg, I.; Jonsson, A. Meiotic
investigations in pollen mother cells of
Norway spruce cultivated in a plastic
greenhouse. Hereditas 66: 1-20; 1970a.

Eriksson, G; Ekberg, I.; Jonsson, A. Further
studies on meiosis and pollen formation.
Stud. For. Suec. 87: 1-65; 1970b.

Favre-Duchartre , M. Contribution a" l'e'tude de
la reproduction chez Ginkgo biloba. Rev.
Cytol. Biol. V£g. 17: 1-218; 1956.

Favre-Duchartre, M. Ginkgo, an oviparous plant.
Phytomorphology 8: 377-390; 1958.

Francini, E. Ecologia comparata di

Pinus halepensis 111., Pinus pinaster Sol. e
Pinus plnea L. sulla base del compor tamento
del gametofito femminile. Acad. Italiana
Sci. For. 7: 107-172; 1958.

Fraser, D.A. The relation of environmental
factors to flowering in spruce. In:
K.V. Thimann (ed.) The Physiology of Forest
Trees. Ronald Press Co., N.Y. 1958.
629-642 pp.

Fraser, D.A. Apical and radial growth of white
spruce (Picea glauca (Moench) Voss) at Chalk
River, Ontario, Canada. Can. J. Bot. 40:
659-668; 1962.

Fraser, D.A. Vegetative and reproductive growth
of black spruce (Picea mariana (Mill.) BSP.)
at Chalk River, Ontario, Canada. Can. J.
Bot. 44: 567-580; 1966.

Gifford, E.M. Jr.; Mirov, N.T. Initiation and
ontogeny of the ovulate strobilus in
Ponderosa pine. For. Sci. 6: 19-25; 1960.

Greenwood, M.S. Reproductive development in
loblolly pine. I. The early development of
male and female strobili in relation to the
long shoot growth behaviour. Amer. J. Bot.
67: 1414-1422; 1980.

Hagraan, M. Incompatibility of forest trees.
Proc. R. Soc. Lond. B. 188: 313-326; 1975.

Hagman, M.; Mikkola, L. Observations on cross-,
self-, and interspecific pollinations in
Pinus peuce Griseb. Silvae Genet. 12:
73-79; 1963.

Hakansson, A. Seed development of Picea abies
and Pinus sllvestrls. Medd. Statens
Skogoforskningsinst. (Swed.) 46: 1-23; 1956.

Hall, J. P. Microsporogenesis in Larix lar lclna .
Can. J. Bot. 60: 797-805; 1982.

Hard, J.S. Frost damage to red pine conelets.
U.S.D.A. For. Serv. Res. Note. LS-5. 1963.

Harrison, D.L.S.; Owens, J.N. Bud development
in Picea engelmanni 1 . I. Vegetative bud
development, differentiation, and early
development of reproductive buds. Can. J.
Bot. 61: 2291-2301; 1983.

27

Heitmuller, H.H.; Melchior, G.H. Uber die
blahf OTdernde Wirkung des Wurzelschnitts ,
des Zweigkrtfmmens und der Strangulation au
japanischer LaTche (Larix leptolepis [Sieb.
et Zucc] Gord.). Silvae Genet. 9: 65-72:
1960.

Ho, R.H. Pollination mechanism and seed

production potential in Douglas-fir. For.
Sci. 26: 522-528; 1980.

Hoist, M.J. The stimulation of flowering in

spruce and pine. In: E.K. Morgenstern (ed.)
Proc. 12th Meeting Comm. For. Tree Breeding
Can. Vol. II. Laval, Univ., F.Q. 1970. Can.
For. Serv., Dept. Fisheries and Forestry,
Ottawa. 1971. 125-128 pp.

Hutchinson, J.G.; Bramlett, D.L. Frost damage
to short leaf pine flowers. J. For. 62: 343;
1964.

Jackson, D.I.; Sweet, G.B. Flower initition in
temperate woody plants. A review based
largely on the literature of conifers and
deciduous fruit trees. Horticult. Abstr. 42:
9-24; 1972.

Johansen, D.A. Plant embryology. Chronica
Botanica Co., Waltham, MA. 1950.

Kato, Y.; Fukuhara, N.; Kobayashi, R.

Stimulation of flower bud differentiation of
conifers by gibberellins . J. Jap. For. Soc.
41: 309-311; 1959.

Kotter, W. Normale und anomale Fruchtbildung
bei Juniperus communis. Inagural Diss. Univ.
Hamburg, 1-75 ~pp~. Cited in: H. Singh, 1978.
Embryology of Gymnosperms . Gebrffder
Borntraeger Berlin. 1931. 248 pp.

Krugman, S.L. Freezing spring temperatures

damage knobcone pine conelets. U.S. D.A. For.
Serv. Res. Pap. PSW-37. 1966.

Kupila-Ahvenniemi, S.; Pihakaski, S.;

Pihakaski, K. Wintertime changes in the
ultrastructure and metabolism of the
microsporanglate strobili of Scotch pine.
Planta 144: 19-29; 1978.

Lanner, R.M.; Van Den Berg, D.A. The vegetative
buds and shoots of lodgepole pine. In:
Management of lodgepole pine ecosystems.
Vol. 1. Proc. Pullman Wash. 68-85; 1975.

Lee, K.J. Factors affecting cone initiation in
pines: a review. Research Rep. Inst, of
For. Genet. No. 15: 45-85; 1979.

Lill, B.S. Development of the female cone of
Pinus radiata D. Don. Vol. I. Ph.D. Diss.,
Univ. Canterbury, Christchurch, N.Z. 1974.

Lill, B.S.; Sweet, G.B. Pollination in

Pinus radiata. N.Z.J. For. Sci. 7: 21-34;
1977.

Longman, K.A. Factors affecting flower

initiation in certain conifers. Linn. Sco.
Lond. Proc. 172: 124-127; 1961.

Lowry, W.P. Apparent meterologlcal requirements
for abundant cone crop in Douglas-fir. For.
Sci. 12: 185-192; 1966.

McWilliam, J.R. The role of the micropyle in
the pollination of Pinus. Bot. Gaz. 120:
109-117; 1958.

Maguire, W.P. Are ponderosa pine crops

predictable? J. For. 54: 778-779; 1956.

Matthews, J.D. Factors affecting the production
of seed by forest trees. For. Abs. 24(1):
i-xiii; 1963.

Melchior, G.H. Girdling experiments to increase
flowering ability in Japanese larch
L_. leptolepis (Sieb. et Zucc.) Gord. and
European larch, L. decidua
Mill. Silvae Genet. 9: 105-111; 1960.

Melchior, G.H. The induction of flowering in
Japanese larch graftings. Silvae Genet. 10:
20-27; 1961a.

Melchior, G.H. Experiments with girdling of
grafts of European and Japanese larch
(Larix decidua and Larix leptolepis) . Silvae
Genet. 10: 97-124; 1961b.

Melchior, G.H. Experiments of flower

stimulation in Douglas-fir grafts. Silvae
Genet. 17: 26-31; 1968.

Mergen, F.; Burley, J.; Furnival, G.M. Embryo
and seedling development in Picea glauca
(Moench) Voss after self-, cross-, and wind
pollination. Silvae Genet. 14: 188-194;
1965.

Orr-Ewlng, A.L. A cytological study of the
effects of self-pollination on
Pseudotsuga menziesii (Mirb.) Franco. Silvae
Genet. 6: 179-185; 1957.

Ottley, A.M. The development of the
gametophytes and fertilization in
Juniperus communis and Juniperus virginiana .
Bot. Gaz. 48: 31-46; 1909.

Owens, J.N. The relative importance of

initiation and early development on cone
production in Douglas-fir. Can. J. Bot. 47:
1039-1049; 1969.

Owens, J.N. Reproductive patterns in conifers.
In: Sexual reproduction of conifers.
Academy of Sciences of the USSR. Siberian
branch of the Biological Institute. Forest
and Wood Institute. ALL-Union symposium.
April 1973. 1973.

Owens, J.N. Reproductive biology of conifers.
In: B.P. Dancik and K.0. Higgenbothan (eds.)
Proc. 6th North Amer. For. Biol. Workshop.,
Univ. Alta. 1980. 77-92 pp.

28

Owens, J.N. Pollen management through

phenologlcal control of development. In:
D.F.W. Pollard, D.G. Edwards and C.W. Yeatman
(eds.) Proc. 18th Meeting Can. Tree Improv.
Assoc. Pt. 2, Duncan, B.C. Aug. 17-20, 1981.
Can. For. Serv., Env. Can. for CTIA/ACAA,
Ottawa. 1982.

Owens, J.N. Bud development in mountain hemlock
(Tsuga mertensiana) . II. Cone-bud
differentiation and predormancy development.
Can. J. Bot. 62: 484-494; 1984a.

Owens, J.N. Bud development in grand fir
(Abies grandls) . Can. J. For. Res. 14:
575-588; 1984b.

Owens, J.N.; Blake, M.D. Pollen morphology and
development of the pollination mechanisms in
Tsuga heterophylla and T. mertensiana. Can.
J. Bot. 61: 3041-3048; 1983 .

Owens, J.N.; Blake, M.D. The pollination

mechanism of Sitka spruce (Picea sitchensis) .
Can. J. Bot. 62: 1136-1148; 1984.

Owens, J.N.; Molder, M. Pollen development in
Douglas-fir (Pseudotsuga menzlesil). Can. J.
Bot. 49: 1263-1266; 1971a.

Owens, J.N.; Molder, M. Meiosis in conifers:
prolonged pachytene and diffuse diplotene
stages. Can. J. Bot. 49: 2061-2064; 1971b.

Owens, J.N.; Molder, M. Bud development in
western hemlock. II. Initiation and early
development of pollen cones and seed cones.
Can. J. Bot. 52: 283-294; 1974a.

Owens, J.N.; Molder, M. Yellow cedar cone

production. B.C. For. Serv. Res. Notes E.P.
730 No. 68. 1974b. 4 pp.

Owens, J.N.; Molder, M. Development of
long-shoot terminal buds of
Pinus contorta spp. contorta. In:
D.M. Baumgartner (ed.) Management of
lodgepole pine ecosystems symp. proc. Wash.
State Univ. Coop Ext. Serv., Pullman. 1975a.
85-104 pp.

Owens, J.N.; Molder, M. Sexual reproduction of
mountain hemlock (Tsuga mertensiana) . Can.
J. Bot. 53: 1811-1826; 1975b.

Owens, J.N.; Molder, M. Bud development in

Sitka spruce. II. Cone differentiation and
early development. Can. J. Bot. 54:
766-779; 1976.

Owens, J.N.; Molder, M. Vegetative bud

development and cone differentiation In Abies
amabilis. Can. J. Bot. 55: 992-1008; 1977a.

Owens, J.N.; Molder, M. Development of

long-shoot terminal buds of western white
pine (Pinus monticola). Can. J. Bot. 55:
1308-1321; 1977b.

Owens, J.N.; Molder, M. Seed-cone

differentiation and sexual reproduction in
western white pine (Pinus monticola). Can.
J. Bot. 55: 2574-2590; 19771T

Owens, J.N.; Molder, M. Sexual reproduction of
Abies amabilis. Can. J. Bot. 55: 2653-2667;
1977d.

Owens, J.N.; Molder, M. Bud development in

Picea glauca. II. Cone differentiation and
early development. Can. J. Bot. 55:
2746-2760; 1977e.

Owens, J.N.; Molder, M. Cone induction in

yellow cypress (Chamaecyparis nootkatensls)
by gibberellin A^ and the subsequent
development of seeds within the induced
cones. Can. J. For. Res. 7: 605-613; 1977 f .

Owens, J.N.; Molder, M. The times and patterns
of cone differentiation in western conifers.
In: 3rd World Consultation on Forest Tree
Breeding. Canberra, Aust. 1977g. 6 pp.

Owens, J.N.; Molder, M. Sexual reproduction of
white spruce (Picea glauca) . Can. J. Bot.
57: 152-169; 1979a.

Owens, J.N.; Molder, M. Bud development in
Larix occidentalis . II. Cone
differentiation and early development. Can.
J. Bot. 57: 1557-1572; 1979b.

Owens, J.N.; Molder, M. Sexual reproduction of
Larix occidentalis. Can. J. Bot. 57:
2673-2690; 1979c.

Owens, J.N.; Molder, M. The times and patterns
of cone differentiation in western North
American conifer. In: F. Bonner (ed.) Proc.
Symp. Flowering and seed development in
trees. IUFRO. 1978. Miss. State Univ.
South. For. Exp. Sta. U.S. For. Serv. 1979d.
25-32 pp.

Owens, J.N.; Molder, M. The reproductive cycles
of western red cedar and yellow cedar. B.C.
Min. For. Info. Serv. Br. Victoria. 1984a.
28 pp.

Owens, J.N.; Molder, M. The reproductive cycle
of lodgepole pine. B.C. Min. For. Info.
Serv. Br. Victoria. 1984b. 29 pp.

Owens, J.N.; Molder, M. The reproductive cycle
of interior spruce. B.C. Min. For. Info.
Serv. Br. Victoria. 1984c. 30 pp.

Owens, J.N.; Molder, M. The reproductive cycles

of western and mountain hemlock. B.C. Min.

For. Info. Serv. Br. Victoria. 1984d. 34
pp.

Owens, J.N.; Molder, M. The reproductive cycles
of the true firs. B.C. Min. For. Info. Serv.
Br. Victoria (in press); 1985.

29

Owens, J.N.; Pharis, R.P. Initiation and
ontogeny of the microsporangiate cone in
Cupressus arlzonica in response to
gibberellin. Amer. J. Bot. 54: 1260-1272;
1967.

Owens, J.N.; Pharis, R.P. Initiation and
development of western red cedar cones in
response to gibberellin induction and under
natural conditions. Can. J. Bot. 49:
1165-1175; 1971.

Owens, J.N.; Simpson, S.J. Further observations

on the pollination mechanism and seed

production of Douglas-fir. Can. J. For. Res.
12: 431-434; 1982.

Owens, J.N.; Singh, H. Vegetative bud

development and the time and method of cone
initiation in subalpine fir
(Abies lasiocarpa) . Can. J. Bot. 60:
2249-2262; 1982.

Owens, J.N.; Molder, M.; Langer, H. Bud
development in Picea glauca. I. Annual
growth cycle of vegetative buds and shoot
elongation as they relate to date and
temperature sums. Can. J. Bot. 55:
2728-2745; 1977.

Owens, J.N.; Simpson, S.; Molder, M. The
pollination mechanism in yellow cypress
( Chamaecyparis nootkatensis) . Can. J. For.
Res. 10: 564-572; 1980.

Owens, J.N.; Simpson, S.J.; Molder, M. The

pollination mechanism and the optimal time of
pollination in Douglas-fir

(Pseudotsuga menziesii) . Can. J. For. Res.
11: 36-50; 1981a.

Owens, J.N.; Simpson, S.J.; Molder, M. Sexual
reproduction of Pinus contorta. I. Pollen
development, the pollination mechanism, and
early ovule development. Can. J. Bot. 59:
1828-1843; 1981b.

Owens, J.N.; Simpson, S.J.; Molder, M. Sexual
reproduction of Pinus contorta. II.
Postdormancy ovule, embryo, and seed
development. Can. J. Bot. 60: 2071-2083;
1982.

Owens, J.N.; Webber, J.E.; Ross, S.D.;

Pharis, R.P. Interaction between gibberellin
A^^ and root-pruning on the reproductive and
vegetative processes in Douglas-fir. III.
Effects on anatomy of shoot elongation and
terminal bud development. Can. J. For. Res.
15: 354-364; 1985.

Owston, P.W. The shoot apex in eastern white
pine: its structure, seasonal development
and variation within the crown. Can. J. Bot.
47: 1181-1188; 1969.

Pettitt, J.M. Detection in primitive

gymnosperms of proteins and glycoproteins of
possible significance in reproduction.
Nature 266: 530-532; 1977a.

Pettitt, J.M. The megaspore wall in
gymnosperms: ultrastructure in some
zooidogamous forms. Proc. R. Soc. Lond. B.
195: 497-515; 1977b.

Pettitt, J.M. Precipitation reactions occur
between components of the ovule tissue in
primitive gymnosperms. An. Bot. 44: 369-371;
1979.

Pettitt, J.M. Ultrastructural and

immunocytochemical demonstration of
gametophytic proteins in the pollen tube wall
of the primitive gymnosperm Cycas . J. Cell
Sci. 57: 189-213; 1982.

Pharis, R.P.; Morf; W. Experiments on the

precocious flowering of western red cedar and
four species of Cupressus with gibberellins
A3 and A4/7 mixture. Can. J. Bot. 45:
1519-1524; 1967.

Powell, G.R. Initiation and development of

lateral buds in Abies balsamea . Can. J. For.
Res. 4: 458-469; 1974.

Powell, G.R. Patterns of development in
Abies balsamea crowns and effects of
megastrobilus production on shoots and buds.
Can. J. For. Res. 7: 498-509; 1977a.

Powell, G.R. Bienniel strobilus production in
balsam fir: a review of its morphogenesis
and a discussion of its apparent
physiological basis. Can. J. For. Res. 7:
547-555; 1977b.

Powell, G.R.; Tosh, K.J.; Remphrey, W.R.

Occurrence and distribution of cones borne
laterally on long shoots of Larix laricina.
Can. J. Bot. 62: 771-777; 1984.

Puritch, G.S. Cone production in conifers. A
review of the literature and evaluation of
research needs; with an economic analysis by
A.H. Vyse. Pacific For. Res. Cen., Can. For.
Serv., Victoria, B.C. Info. Rep. BC-X-65.
1972. 94 pp.

Rehfeldt, G.E.; Stage, A.R.; Bingham, R.T.

Stroblli development in western white pine:
periodicity, prediction and association with
weather. For. Sci. 17: 454-461; 1971.

Reukema, D.L. Seed production of Douglas-fir
increased by thinning. Res. Note Pacific NW
For. Range Exp. Sta. No. 210. 1961.

Ritchie, G.A. Phenology and ontogeny of the
reproductive and primary vegetative
structures of Abies amabilis and
Abies procera. M.Sc. Thesis, Univ.
Washington, Seattle, WA., U.S.A. 1966.

Ross, S.D.; Webber, J.E.; Pharis, R.P.;

Owens, J.N. Interaction between gibberellin
A4/7 and root pruning on the reproductive and
vegetative process in Douglas-fir. I.
Effects on flowering. Can. J. For. Res. 15:
341-347; 1985.

30

Roy Chowdhury, C. The embrogeny of conifers: A
review. Phy tomorphology 12: 313-338; 1962.

Sacher, J. A. Structure and seasonal activity of
the shoot apices of Plnus lambertlana and
Pinus ponderosa . Amer. J. Bot. 41: 749-759;
1954 .

Sarvas, R. Investigations on the flowering and
seed crop of Pinus sylvestrls. Inst. For.
Fenn. 53: 1-198; 1962.

Sarvas, R. The annual period of development of
forest trees. Tiedeaketaraia . Sitzungsber.
Finn. Acad. Wlss. 1965: 211-231; 1965.

Sarvas, R. Investigations on the flowering and
seed crop of Picea abies. Comm. Inst. For.
Fenn. 67: 84 ppT 1968.

Schopraeyer, C.S. Seeds of woody plants in the
United States. U.S.D.A. For. Serv. Handbook.
No. 450. 1974. 883 pp.

Schubert, G.H. Effect of fertilizer on cone

production of sugar pine. U.S.D.A. For. Ser.
C.F.R.E.S. For. Res. Note NO. 116. 1956. 4
pp.

Simola, L.K. The ultrastructure of dry and
germinating seeds of Pinus sylvestrls L.
Acta Bot. Finn. 103: 1-31; 1974.

Singh, H. Embryology of Gymnosperms.

Gebru'der borntraeger, Berlin. 1978. 302
pp.

Singh, H.; Owens, J.N. Sexual reproduction of
Engelraann spruce (Picea engelmannii) . Can.
J. Bot. 59: 793-810; 1981a.

Singh, H. ; Owens, J.N. Sexual reproduction in
subalplne fir (Abies laslocarpa) . Can. J.
Bot. 59: 2650-2666; 1981b.

Singh, H.; Owens, J.N. Sexual reproduction in
grand fir (Abies grandls). Can. J. Bot. 60:
2197-2214; 1982.

Singh, H.; Owens, J.N.; Dietrich, H.F.
Ultrastructure and development of the
ralcrosporangiura of Pseudotsuga menzlesi t
(Mirb.) Franco. I. Dormant stages. Amer.
J. Bot. 70: 1272-1280; 1983.

Sorensen, F.C. The roles of polyembryony and
embryo viability in the genetic system of
conifers. Evolution 36: 725-733; 1982.

Stephens, G.R. Jr. Flower induction In woody
plants. Ph.D. Thesis. Forestry, Yale Univ.
1961. 75 pp.

Stephens, G.R. Jr. Stimulation of flowering in
eastern white pine. For. Sci. 10: 28-34;
1964.

Sucoff, E. Timing and rate of bud formation in
Plnus resinosa. Can. J. Bot. 49: 1821-1832;
1971.

Sweet, G.B. Shedding of reproductive structures
in forest trees. In: T.T. Kozlowskl (ed.)
Shedding of plant parts. Academic Press,
New York. 1973. 341-382 pp.

Sweet, G.B. A physiological study of seed cone
production in Pinus radlata . N.Z.J. For. Sc.
9: 20-33; 1979.

Sweet, G.B.; Thulin, I.J. The abortion of

conelets of Plnus radiata. N.Z.J. For. 14:
59-67; 1969.

Takao, A. Histocheralcal studies on

embryogenesis of Pinus thunbergil. Pari.
Bot. Mag. Tokyo 73: 379-388; 1960.

Van Den Berg, D.A.; Lanner, R.M. Bud

development in lodgepole pine. For. Sci. 17:
479-486; 1971.

Van Vredenburch, C.L.H.; La Bastide, J.G.A. The
influence of meterologlcal factors on the
cone crop of Douglas-fir In the Netherlands.
Silvae Genet. 18: 182-186; 1969.

Villar, M.; Knox, R.B.; Dumas, C. Effective
pollination period and nature of
pollen-collecting apparatus in the
Gymnosperm, Larix leptolepis. Ann. Bot. 53:
279-284; 1984^

Wheeler, N.C.; Ying, C.C.; Murphy, J.C. Effect
of accelerating growth on flowering in
lodgepole pine seedlings and grafts. Can. J.
For. Res. 12: 533-537; 1982.

White, T.L.; Knopp, J. A. Conelet abortion and
ATP levels in long leaf pine. Can. J. Bot. 4
56: 680-685; 1978.

Willson, M.F.; Burley, N. Mate choice in

plants. Princeton Univ. Press, Princeton,
N.J. 1983. 251 pp.

Wlnton, L.L. Phenology of normal and forced
microsporogenesis in white and black spruce
and their P, hybrid. School of For. Univ.
Minn. St. Paul, Minn. For. Notes 153. 1964.

Yanagihara, T.; Tochlaki, K . ; Aral, K. On the
relation between the harvest of Japanese
larch seed and meteorological conditions. J.
Jap. For. Soc. 42: 347-351; 1960.

Yokoyaraa, T.; Askawa, S. Effect of photoperiod
on female stroblll formation in
Larix leptolepis (Sieb. et Zucc.) Gordon. J.
Jap. For. Soc. 55: 388-393; 1973.

Young, E.; Hanover, J.W. Accelerating maturity
in Picea seedlings. Acta Hort. 56: 105-114;
1976.

31

5s

IDAHO WESTERN REDCEDAR (THUJA PL I CAT A DONN.) CONEBEARING
STATUS IN AN UNUSUALLY DRY YEAR^
William R^Gall and Allen S.j^Rowley

ABSTRACT: Twenty-six 1/5 acre (0.0809 hectare)
plots were measured in western redcedar (Thuja
plicata Donn.) stands in northern Idaho in 1979.
Each plot represented one stand. Half of the
plots contained western redcedar which were
bearing cones of the current year, or cones that
matured in 1978. The western redcedar stands
bearing cones had a mean cedar d.b.h. of
11.4 inches (29 cm) versus 13 inches (33 cm) for
nonbearing stands. Conebearing stands contained
less basal area for the stand and western redcedar
and more basal area of grand fir (Abies grandis) ,
and western hemlock (Tsuga heterophylla) . Cone-
bearing stands had a mean elevation of 3120 feet
(1125 m) , and contained reproduction consisting of
some combination of grand fir, western redcedar,
and western hemlock; nonbearing stands were lower
in elevation and had reproduction of western
redcedar only and occasional Douglas-fir
(Pseudotsuga menziesii) .

Total height of two dominant or codominant cedars
had a correlation of 0.694 with d.b.h. in conebear-
ing stands; a larger value was found in nonbearing
stands. The ratio of total height/d.b.h. had a
correlation of 0.751 with the number of stems/ha
and 0.751 with the number of cedar stems/ha in
nonbearing stands. No significant correlation
with these variables was found for conebearing
stands .

INTRODUCTION

for a tree improvement program and to what extent
the program would be worthwhile.

Currently little is known about the reproductive
biology and natural variation of western redcedar
in Idaho. Owens and Pharis (1971) described the
phenology of western redcedar pollen and ovulate
strobilus initiation, meiosis, and pollination in
coastal British Columbia. Ebell and Schmidt
(1964) reported on the effect of climatic condi-
tions on western redcedar pollen release on
Vancouver Island.

MATERIALS AND METHODS
Plant Materials

Twenty-six 1/5 acre (0.0809 hectare) plots were
laid out in 1979 within the natural distribution
of western redcedar in Idaho. Half of the plots
contained western redcedar which were bearing
cones that matured in 1978 or would mature in
1979. Representative stands were chosen on
relatively dry sites or moist, well-drained sites;
cedar bogs were avoided. Stands with a closed
canopy and of approximately 50 percent cedar and a
mean d.b.h. of 12 inches (30 cm) were chosen.
Diameter at breast height was measured for all
trees on the plot with a diameter of 3 inches
(7.6 cm) or larger. Basal area was calculated
accordingly for all the trees. No increment cores
were taken to provide an estimate of age.

This study originated to obtain preliminary
information about variation in the natural western
redcedar (Thuj a plicata Donn.) stands, including
information about the reproductive biology of
western redcedar in Idaho. Knowledge of this kind
is necessary for planning a forest tree improvement
program in any species or for deciding the need

Paper presented at the Conifer Tree Seed in the
Intermountain West Symposium, Missoula, MT, August
5-6, 1985.

William R. Gall is a former Forest Tree
Physiologist/Geneticist, University of Idaho,
Moscow, ID. Allen S. Rowley is a Forestry
Technician, Nezperce National Forest, Forest
Service, U.S. Department of Agriculture,
Grangeville, ID.

The total height of two dominant or codominant
cedar trees was recorded. The western redcedar
trees were scrutinized for cones with an 8 x 26
pocket monocular.

All trees between 7.6 and 15.0 cm d.b.h. were
considered to be reproduction. If any tree
species or a combination of any three tree species
was represented by 10 or more trees in that
diameter class in a plot, that species or combina-
tion of species was considered to be reproducing.

Environmental Variables

Elevation, latitude, and longitude of each stand
were obtained from either a 7 1/2' or 15' topo-
graphic quadrangle map. Aspect of each stand was
observed with a hand held compass. Weather data
were obtained from the National Climatic Data
Center (1977, 1978, 1979).

32

Statistical Analysis

It was planned that data would be analyzed
according to this analysis of variance:

D

Source of Variation

d.f.

2

Conebearing Status

1

13

Stands/Conebearing Status

24

t

Trees/Stands/Conebearing Status

1874

1900

Total Variation

1899

However, when analysis was begun it was observed
that means for the two groups were sometimes very
similar, and variation among stands within cone-
bearing status sometimes obscured variation
between the two kinds of stands. The exceptions
were: percent cedar by number of stems and by
basal area, total number of stems, number of cedar
stems, 2-tree total height, 2-tree d.b.h. and
2-tree total height/d.b.h. A two-tailed F test
(Snedecor and Cochran 1980) showed that the
variance estimates for cedar basal area, plot
basal area, cedar mean d.b.h., plot mean d.b.h.,
grand fir (Abies grandis) basal area, and western
hemlock (Tsuga heterophylla) basal area were
heterogenous. Therefore, it would not be appro-
priate to combine such variables into a single
analysis as above. Instead, these variables were
analyzed according to the following analysis of
variance by conebearing status:

n Source of Variation d.f.

13 Stands 12

t Trees/Stands 13(t-l)
Total Variation

Percent cedar by number of stems and by basal area
were tested for skewness and kurtosis and found to
be normally distributed. These two variables and
total number of stems/acre and number of cedar
stems/acre were analyzed according to this
analysis of variance:

n Source of Variation d.f.

2 Conebearing Status 1

13 Stands/Conebearing 24

26 Total Variation

Correlations were computed by conebearing status
among cedar basal area, plot basal area, cedar
d.b.h., plot d.b.h., number of cedar stems, total
number of stems, 2-tree total height, 2-tree
d.b.h., and 2-tree total height/d.b.h. Elevation
was analyzed by a 2-way Chi-square test.

RESULTS

Conebearing status was found not to be independent
of high and low elevation by a Chi-square test
(table 1). Nonbearing stands occurred mostly at
lower elevations below 3500 feet (1067 ra) .
Conebearing stands were more variable but occurred
most often at higher elevations. The stands were
located from 46° 9' N to 48° 50' N and from 115°
37' W to 117° 2' W. No significant effect of the
stands' location or aspect was found.

Table I, — Number of stands in eight elevation
intervals by conebearing status

Elevation

Number

of

Number

Class (m)

Conebearing

Stands

Nonbearing

LOW

457-609

0

1

610-762

0

1

762-914

3

3

914-1066

3

6

HIGH

1067-1219

3

0

1219-1371

1

1

1372-1524

1

0

1524-1676

2

1

13

13

Chi-square = 4.25
d.f. = 1

indicates the probability of a greater Chi-
square is less than 5 percent.

The field measurements define a difference in
stand composition between stands bearing cones and
those without cones. Table 2 displays the means
of several variables for the two types of stands.
The conebearing stands had less basal area for the
whole stand and less western redcedar basal area.
Conebearing stands did have a higher amount of
basal area for grand fir, which occurred in
21 stands, 10 bearing redcedar cones. Also as
seen in table 2, more basal area of western
hemlock was found in conebearing stands where it
existed in 12 stands, six bearing cones. Notice
that the means of the two dominant tree heights
were the only variables at all similar.

Table 2. — Means for selected variables by cone-
bearing status

Variable Conebearing Nonbearing

Redcedar Basal area (m /ha) 5.16 9.99

Grand fir Basal area 3.10 1.42

Western hemlock Basal area 2.32 0.55

Total plot Basal area 10.49 13.80

Plot d.b.h. (cm) 25.9 33.0

Redcedar d.b.h. 29.0 33.0
Total height for dominant

redcedar trees (m) 25.9 26.5

Stems/ha 865 919

Redcedar stems/ha 450 526

Counts of stems between 7.6 and 13.0 cm (3.0 and
5.9 inches) were used to quantify reproduction,
although in shade-tolerant species d.b.h. may not
be well correlated with age. Any species or
combination of any three species having 10 or more
stems in this d.b.h. class was arbitrarily
considered to be reproducing on that plot. In
some stands no stems of any species were present
in this size class in sufficient numbers to
qualify as reproduction. In conebearing stands
reproduction consisted of cedar, grand fir, or
western hemlock, singly or in a combination. In
nonbearing stands reproduction consisted of cedar
alone in five stands, along with Douglas-fir

33

(Pseudotsuga menziesii) or western larch (Larix
occidentalis) in two stands.

Table 3. — Tree species of reproduction size (10
stems or more between 7.6 and 15.0 cm
d.b.h.) by conebearing status of stands

Conebearing Stands

Nonbearing Stands

S tand

Spec ie s

S tand

Species

INO .

Reproducing

wo .

Reproducing

9

L ■ . ! 1 U 11L

7

10

cedar

8

cedar

11

cedar and hemlock

16

12

cedar and hemlock

17

cedar

13

cedar, hemlock

18

and grand fir

14

cedar and hemlock

19

cedar

15

20

cedar

22

grand fir

21

cedar and

Douglas-fir

23

cedar and grand fir

24

cedar and larch

28

25

29

cedar

26

30

27

cedar

32

cedar and grand fir

31

Variance estimates among conebearing and nonbearing
stands were found by a two-tailed F test to be
heterogenous for cedar basal area, plot basal area,
cedar mean d.b.h., plot mean d.b.h., western
hemlock basal area, and grand fir basal area
(table 4). The variance estimates were larger in
conebearing stands only for grand fir and western
hemlock basal area.

Table 4. — Two-tailed F values for selected vari-
ables by conebearing status

Variable

Conebearing

Nonbearing

Plot mean d.b.h.

C1)

7. 249**

Plot basal area

17.007**

Cedar mean d.b.h.

3.667*

Cedar basal area

25.165**

Stems/ acre

2.098 n.s.

Cedar Stems/acre

2.236 n.s.

Western hemlock

basal area

8.589**

Grand fir basal area

4.442**

^ach F ratio is listed under the type of
stand with the largest variance estimate.

**Indicates the effect was significant at P<1%.

*Indicates the effect was significant at P<5%.

n.s. Indicates the effect was not significant
at P<5%.

Correlation coefficients among variables were
calculated for conebearing stands and are
presented in table 5. For conebearing stands the
ratio two-tree mean total height/d . b . h . was not
significantly correlated with any other variable.
In nonbearing stands the two-tree mean total
height/d. b.h. ratio was only significantly
correlated (P<1%) with the number of stems/acre

and number of cedar stems/acre. Two-tree mean
d.b.h. was significantly correlated with plot mean
d.b.h. and cedar d.b.h. (P<1%) and with plot basal
area, cedar basal area and number of stems/acre
(P<5%) .

Variance estimates for the ratio total height/
d.b.h. between conebearing and nonbearing stands
were not heterogeneous by a two-tailed F test
(P<5%) , and the effect of conebearing status was
not significant in the analysis of variance.

Relative amounts of among-stand and within-stand
phenotypic variation are given in table 5 for the
total height/d. b.h. ratio by conebearing status.
Among-stand variation was larger than within-stand
variation in both conebearing categories. These
figures are based on a separate analysis of
variance computed by conebearing status for the
ratio of total height to d.b.h.

DISCUSSION

Conebearing stands were more common at higher
elevation. Habeck (1979) reported that climax
western redcedar stands in the Selway-Bitterroot
Wilderness in Idaho were reproducing at higher
elevations but not at lower elevations. In spite
of the fact that Fovells (1965) called western
redcedar a prolific seeder, Gashwiler (1969) said
that seedfall of western redcedar in west central
Oregon over a period of 12 years was erratic. As
a possible explanation for these observations
Ebell and Schmidt (1964) report a strong climatic
effect on pollen production in western redcedar.
For the cones observed in this study weather or
elevational moisture differences could have been a
factor .

The winter of 1977-78 was drier than normal. The
months of January through April were 25 percent
below normal for precipitation recorded at weather
stations in Idaho. This time period is also a key
time for western redcedar reproduction as meiosis
followed by pollination is occurring (Owens and
Pharis 1971). The cones we observed in 1979 were
undergoing meiosis and pollination during this dry
period. During a dry winter in 1938-39 in Illinois
Martin (1952) found that the sporophylls and
microsporangia of northern white cedar (Thuja
occidentalis L.) became desiccated. The
desiccation accounted for a low rate of survival of
staminate cones. Similarly the dry winter of
1977-78 could have produced the same effect in the
stands we examined.

The fact that the correlation of d.b.h. and total
height is different and smaller for means of cone-
bearing than nonbearing stands suggests the two
variables can be manipulated somewhat independ-
ently. Wright (1975) reported that measurement of
Scotch pine (Pinus sylvestris L.) plantations in
Michigan after 12 growing seasons in the field
showed that there are general genes for growth
rate, as well as specific genes for height and
diamater growth. If improvement is to be made in
the ratio total height/d .b . h . , it is necessary to
know if this is true for western redcedar. If so
d.b.h. and height can be manipulated independently.

34

Table 5. — Correlation among plot variables for conebearing stands, 11 degrees of freedom

2-tree

mean

d.b.h.

2-tree
mean
d.b.h. /
height

plot
mean

cedar
mean

d.b.h. d.b.h.

plot cedar total cedar
basal basal stems/ stems/
area area acre acre

2-tree mean
height

0.694
**

0.485
n . s .

0.375
n.s.

0.526
n.s.

0.2 39
n.s.

0.261
n.s.

-0.376
n.s.

-0.356
n.s.

2-tree mean
d.b.h.

-0.282
n.s.

0.702
**

0.825
**

0.587
*

0.623
*

-0.554

-0.301
n.s.

2-tree mean
height/d.b.h.

plot mean

-0.378 -0.317

0.829
**

-0.341
n.s.

0.784
**

-0. 362
n.s.

0.314
n.s.

0.200
n.s.

-0.866
**

-0.059
n.s.

-0.596
*

cedar mean
d.b.h.

0.526
n.s.

0.454
n.s.

-0.860
**

-0.627
*

plot basal area

0.528
n.s.

-0.450
n.s.

-0.240
n.s.

cedar basal area

-0.027
n.s.

0.330
n.s.

total stems/acre

0.852
**

* Indicates the correlation was significant at P<5%.
** Indicates the correlation was significant at P<1%.
n.s. Indicates the correlation was not significant at P<5%.

Table 6. — Relative amounts of phenotypic variation
in ratio of total height to d.b.h. by
conebearing status

Among-stand

Within-stand

variation as 1

variation as % of

of total pheno-

total phenotypic

typic variation

variation

Nonbearing

85.4

14.6

stands

Conebearing

75.6

24.4

stands

The correlation of total number of stems and
number of cedar stems with the ratio total
height/d.b.h. in nonbearing stands may indicate an
environmental effect of stand dynamics on that
ratio. The lack of such correlation in
conebearing stands suggests that the number of
stems cannot explain all the variation in that
ratio. More information is needed to separate the
environmental factors out of the height to d.b.h.
ratio.

Western redcedar conebearing in closed stands in
Idaho is not as common, especially in a dry year,

as it may be in coastal forests. This variability
in cone production may mean that vegetative
propagation of materials obtained from ortets in
natural stands is a better way of sampling genetic
variation than collecting cones. If it is desired
to sample genetic variation by collecting cones, it
may be useful to be able to discriminate between
stands that will be likely to be bearing cones and
those that will not. The sample size used in the
present research was not large enough to permit
development of a predictive model. Use of the
means reported here, however, may permit a
reduction in time and travel required in locating
stands bearing western redcedar cones. Further
research on conebearing status of western redcedar
stands would increase the sample size and permit
estimation of differences between years. It might
also permit development of a predictive model which
would reduce the tine and travel involved in
finding natural western redcedar stands which are
bearing cones.

If there are wet-r.1 te-adapted and dry-site-adapted
genotypes of western redcedar this knowledge could
be utilized in reforestation efforts. Perhaps it
would be economical to plant western redcedar on
drier sites than where it is climax. However, the
present research cannot answer whether there are
dry-site-adapted and wet-site-adapted genotypes of

35

western redcedar. Further research using
replicated field plantations of progenies from
trees on both kinds of site could be used to
estimate genetic and environmental components of
the difference between cedar producing cones on
moist sites and cedar not producing cones on dry
sites. Chemical and biochemical estimates of
variability might also provide estimates of
genotypic and environmental effects.

The outstanding characteristic of western redcedar
is its conical shape (Anderson 1961). While a
conical shope may be ideal for utility poles, it
may not be the best stem form for other products.
If the stem form of western redcedar could be
improved, the yield of other products such as
lumber might be improved. Estimates of the effect
of environmental factors on the height-to-diameter
ratio can be made by further research in natural
stands. A possible association with age or
buttressing should be investigated, and the effect
of slope position and elevation should be tested.
The present research has shown that number of stems
per hectare and number of cedar stems per hectare
were correlated with the height-to-diameter ratio
in nonbearing but not in conebearing stands.

To estimate the extent to which breeding progress
can be made in changing the height-to-diameter
ratio, replicated plantations should be established
containing progenies of trees from a number of
stands. The number of trees per stand whose
progenies are represented should be fairly high.

Knowledge of the site requirements for western
redcedar cone production and of climatic factors
affecting cone production may permit seed orchard
establishment on sites where seed production can be
maximized and outside pollen can be minimized.
Some research needs to be done to clarify whether
western redcedar seed orchards can be established
on a dry site, so that irrigation can supply
necessary soil moisture during each month and the
seed orchard can be established some distance away
from sources of western redcedar pollen. In some
species, a slight drought at or before the period
of cone initiation promotes the induction of large
numbers of cones. Research is needed to test the
effect of a slight drought on western redcedar cone
initiation. Results of the present research
provide some information which has a bearing on
western redcedar seed orchard management and cone
production. The possibility of pollen desiccation
and conditions under which it may occur should
also be investigated.

Research on phenology of western redcedar flower
induction, flowering, and pollination in Idaho is
needed to at least partially explain the results
obtained in the present research, to indicate which
months and what kind of weather are critical and as
preparation in case any controlled pollination
experiments in an outdoor breeding arboretum need
to be done.

CONCLUSIONS

It was not possible in the present research to
separate environmental effects from genetic ones.
The objective of the present research was to obtain
preliminary information about variation in and
among natural western redcedar (Thuja plicata
Donn.) stands, including information about the
reproductive biology of western redcedar in Idaho.

Twenty-six stands were selected having an average
composition of approximately 50 percent cedar by
number of stems or by basal area and a mean d.b.h.
of approximately 12 inches (30 cm). Half the
stands were bearing western redcedar cones; all
but one were bearing the previous year cones. In
general, stands selected had closed canopies and
no standing water on the site.

The ratio of total height to d.b.h. of two
dominant or codominant cedars did not differ
significantly between the two kinds of stands.
The ratio of total height/d .b .h. was significantly
correlated with number of stems/ha and with number
of cedar stems/ha in nonbearing stands indicating
a possible stand influence. Total height of two
dominant or codominant cedars had a correlation of
0.858 with d.b.h. of the same trees in nonbearing
stands, but only 0.694 in conebearing stands.
Both correlations were significant at the 1
percent level. These correlations, being
different, suggest a possibility of d.b.h. and
height having a degree of independence.

Conebearing stands had a mean elevation of
1125 m. Conebearing stands had higher basal areas
of grand fir and/or western hemlock. Reproduction
in conebearing stands was more likely to be a
combination of grand fir, western hemlock, and
cedar .

Conebearing stands tended to be on high-elevation,
moist sites, and nonbearing stands tended to be on
drier sites at lower elevation. There is some
variation in the total height to d.b.h. ratio.
More research will be needed to estimate
environmental and genetic components of the
variation in ratio of total height to d.b.h. and
to estimate environmental and genetic components
of the differences between conebearing and
nonbearing stands, particularly with regard to the
difference between stands bearing cones on moist
sites and stands not bearing cones on drier sites.
Improvement in the conical shape of cedar is a
potential. If there are dry-site-adapted
genotypes of cedar, they could be utilized on a
wider range of sites than those on which cedar is
climax. However the present research is not able
to separate environmental effects from genetic
ones, which is information needed to determine any
wet-site or dry-site adaptations.

ACKNOWLEDGMENT

This research was supported by a grant from the
Stillinger Fund at the University of Idaho, and by
a grant from the Program for Forest Utilization
Research of the State of Idaho.

36

REFERENCES

Anderson, I. V. Western redcedar poles from British
Columbia. Journal of Forestry. 59:14-16; 1961.

Ebell, L . F.; Schmidt, R. L. Meteorological
factors affecting conifer pollen dispersal on
Vancouver Island. Canadian Dept. of Forestry
Pub. No. 1036; 1964. 28 p.

Fowells, H. A. Silvics of forest trees of the
United States. Agriculture Handbook 271.
Washington, DC: U.S. Department of Agriculture,
Forest Service; 1965. 762 p.

Gashwiler, J. S. Seed fall of three conifers in
west-central Oregon. Forest Science. 15:290-295;
1969.

Haberk, J. R. A study of climax western redcedar
(Thuja plicata Donn.) forest communities in the
Selway-Bitterroot Wilderness in Idaho. Northwest
Science. 52:67-76; 1978.

Martin, P. C. A morphological comparison of Biota
and Thuja. Pennsylvania Academy of Science;
65-112; 1952.

National Climatic Data Center. Climatological data:
annual summary 82(13) Idaho. Washington, DC: U.S.
Department of Commerce, National Oceanographic
and Atmospheric Administration; 1979.

Owens, J. N.; Pharis, R. P. Initiation and

development of western redcedar cones in response
to gibberellin induction and under natural
conditions. Canadian Journal of Botany.
49 : 1 165-1 175 ; 1971.

Snedecor, G. W. ; Cochran, W. G. Statistical

methods. 7th ed. Ames, IA: Iowa State Univeristy;
1980. 507 p.

Wright, J. W. Genetic variation in the

height-diameter ratio in scotch pine. Proceedings
of the 12th Lake States Forest Tree Improvement
Conference; 1975:113-121.

37

EFFECTS OF COLD STRATIFICATION AND SEED COAT STERILIZATION
TREATMENTS ON PINYON INUS EPULIS) GERMINATION,,
Gerald J.jGottfried and L. J.j^Heidmann

ABSTRACT: One experiment compared germination
capacity and energy of pinyon (Pinus edulis
Engelm.) after 30- and 60-day cold stratification.
Seed from five Arizona seed sources were compared.
Both treatments did not affect germination
capacity, but increased speed of germination. A
second study evaluated the effects of six hydrogen
peroxide treatments designed to improve
germination and suppress mold growth during
prolonged tests. Results were variable. Except
for one seedlot, treatment did not improve
germination capacity or energy, nor did it
suppress mold. The most concentrated treatment
tended to suppress germination.

INTRODUCTION

Pinyon-juniper woodlands occupy approximately
13.4 million hectares (33 million acres) in the
southwestern United States. The type covers
17 percent of Arizona, mostly in the northern
half, and 26 percent of New Mexico (West and
others 1975) where it occurs statewide, except in
the southeastern and south-central areas
(Springfield 1976) . The woodlands consist of
pinyon (Pinus edulis Engelm.) and one or more of
the following junipers: Utah juniper ( Juniperus
osteosperma (Torr.) Little); one-seed juniper (J.
monosperma (Engelm.) Sarg.); alligator juniper (J.
deppeana Steud.), and Rocky Mountain juniper (J.
scopulorum Sarg.). The type is found between
1 371 and 2 438 m (4,500 and 8,000 ft) elevation;
precipitation can vary from 300 to 550 mm (12 to
22 inches) . It grows on a wide variety of soils
and parent materials (Springfield 1976) .

Management objectives for pinyon-juniper woodlands
in the Southwest appear to have recently shifted
towards more intensive production of wood
products. Better knowledge of woodland ecology is
needed to develop suitable forestry practices for
these areas. Information is needed about the
germination requirements of pinyon and junipers
with respect to climatic factors, such as light
and moisture, and physiological factors, such as
seed dormancy.

Paper presented at the Conifer Tree Seed in the
Inland Mountain West Symposium, August 5-6, 1985,
Missoula, MT.

Gerald J. Gottfried is Research Forester, and
L. J. Heidmann is Research Plant Physiologist,
Rocky Mountain Forest and Range Experiment
Station, Forest Service, U.S. Department of

Agriculture, Flagstaff, AZ.

The current studies concentrate on dormancy in
pinyon seed. This species is of particular
interest because it occurs throughout the woodland
zone, and is economically important for fuelwood,
especially in New Mexico, and for edible nuts.
Little is known about seed dormancy of pinyon
under natural conditions, or how storage will
affect dormancy. Some pine species show little
evidence of dormancy, while in others, up to
3 years may elapse before germination occurs
(Krugman and Jenkinson 1974) . Cold stratification
is a common technique for breaking embryo
dormancy, but its application to pinyon is poorly
documented according to Krugman and Jenkinson
(1974) .

Understanding seed dormancy is important in
research involving the effects of various
environmental factors on pinyon germination and
establishment. This knowledge is critical to
successful growth of planting stock or for
regeneration by artificial seeding. Sowing
untreated seed during dormancy would lengthen the
germination period, and expose seed to adverse
conditions for longer periods. Seed from
different locations might exhibit different
degrees of dormancy and might respond differently
to stratification treatments.

To solve the problem of mold in long-term
germination studies, Riffle and Springfield (1968)
used a 30-minute soak with 30-percent hydrogen
peroxide alone, or in combination with a water
soak or wash, to prevent the development of
seedborne fungi. They also found that pinyon seed
treated with peroxide, alone or in combination
with water, germinated twice as fast as untreated
seed during the first 8 days of a 28-day test
period. According to Kramer and Kozlowski (1979),
hydrogen peroxide stimulates seed respiration,
which is essential for the early phases of
germination.

The two studies reported here were designed to
evaluate the effects of cold stratification and
various hydrogen peroxide treatments on pinyon
germination capacity and germination energy.
Germination capacity is the percent of filled
seeds which germinate during a period of time
ending when germination is essentially complete.
Germination energy is defined as the number of
days necessary to achieve 50 percent germination
of filled seeds. The hydrogen peroxide treatments
also were evaluated for their ability to suppress
mold contamination of seed during testing.

38

METHODS

Cold Stratification Experiment

Seed Sources

Pinyon seed was collected from five locations in
central Arizona (fig. 1). Trees from the Prescott
area contained both 1- and 2-needle fascicles, and
have been classified as either P. edul is var.
f a 1 1 a x or as a P. edul is x P. monophyl la hybrid
(Lanner 1 9 7 A ) . Seed from Deadman Flat was
collected in 1979, and from the other sources in
the fall of 1983. The seed was stored at 3° C
(37° F) . Site characteristics of the collection
areas are given in table 1 ; elevations were taken
from U.S. Geological Survey topographic maps.
Mean annual precipitation values are for the
nearest weather station (Sellers and others 1985) ,
except for Cedar Ranch, which is not near any
station .

Distribution of junipers and
pinyon in Arizona

Figure L. — Map of the pinyon-juniper woodlands in
Arizona (Arnold and others 1964) and the location
of the seedlots: (1) Prescott, (2) Deadman Flat,
(3) Cedar Ranch, (4) Tusayan, and (5) North
Kaibab.

Table 1. — Characteristics of the five pinyon
provenances

Location Elevation Pinyon height Mean annual

in stand precipitation

- - meters

mm

Prescott

1,689

7

3

480

Deadman Flat

2,054

4

8

408

Cedar Ranch

2, 195

12

7

Tusayan

2,091

15

1

385

N. Kaibab

2,073

10

6

508

The experiment compared 30- and 60-day
cold-stratification treatments with a control
treatment. Seeds from each seedlot were rinsed in
distilled water, then were wrapped in moistened
paper towels. Wrapped seeds were put into a
double plastic bag, loosely sealed, and placed in
a refrigerator at 3° C (37° F) . Towelling for the
60-day treatment was changed after the first
30 days. The first lot was started in March and
the second in April 1984.

After stratification, 25 seeds from each of the
three treatments in each seedlot were placed on
glass-fiber filter paper in 10-centimeter
disposable petri dishes. There were five
subsamples of each seedlot-treatment combination,
for a total of 75 dishes. Dishes were randomized
on each of five shelves, which contained one
subsample . Seeds were germinated in a Hoffman
germinator programmed for 12 hours of light at
23.9° C (75° F) and 12 hours of dark at 20° C
(68° F) . Daytime temperature was 6.1° C (11° F) ,
lower than recommended by the International Seed
Testing Association (Krugman and Jenkinson 1974) ;
however, preliminary tests indicated better
germination at 23.9° C (75° F) than at 30° C
(86° F) . Although Krugman and Jenkinson (1974)
recommended a schedule of 8 hours of light and
16 hours of darkness, the 12-hour interval was
used, because it may better approximate conditions
during the germination period, which was assumed
to be summer. However, it is not known whether P.
edul is germinates in the spring, summer, or during
both seasons under natural conditions. The trials
were not replicated in different germinators or
for other time periods.

Dishes were monitored daily and were watered as
necessary. Germination was recorded when the
radicle equaled the length of the seed.
Germinated seeds were discarded. Counts were made
for 33 days until germination had essentially
ended. Ungerminated seeds were cut open to
determine if they were filled.

Hydrogen Peroxide Experiment

The hydrogen peroxide trial consisted of seven
treatments as follows:

Treatment Number
I

2
3
4
5
6
7

Treatment Description
Control

3% H202 2i-minute soak
5-minute soak
10-minute soak

30% H202 2i-minute soak
5-minute soak
10-minute soak

Seeds were stirred continuously during the soaking
period. Four seedlots were treated; seed from
Deadman Flat was not included because of
insufficient seed. Five subsamples were set up
for each treatment combination. Twenty-five seeds
were placed on glass-fiber filter paper in each
10-centimeter petri dish; there were 140 dishes
with 28 dishes on each of five shelves. The
experiment was started in January 1985.

39

The same germination and counting procedures used
in the stratification study were followed. Counts
were made for an average of 44 days. Shelf order
in the germinator was rotated to avoid possible
germination differences caused by microclimatic
variations in the equipment. In addition, the day
when mold first appeared in a dish was noted.
Experience indicates that most seed becomes
contaminated soon after mold is first detected.

Treatment differences in both studies were
compared by one-way analysis of variance using
pooled seed source as a blocking factor. Pairwise
differences were calculated using Fisher's
F-protected least significant difference test.
One-way analysis of variance procedures were used
to test for treatment differences within seedlots,
and Duncan's multiple-range test was used to
isolate any differences. Significance was
indicated by P values at or below the 5 percent
level. Germination capacity was analyzed after
arc sin transformation of these data to improve
variance homogeneity. Although there appeared to
be differences among seedlots, they could not be
evaluated statistically, because by combining the
seed, spatial variability within each seed source
area could not be determined. All analyses within
seedlots are based on estimates of subsampling
variability, not spatial variability within a seed
source . A t-test was used to compare germination
differences within control treatments over time.

RESULTS

Cold Stratification

Germination capacity. — The germination capacity of
filled seeds was not significantly different among
treatments. The averages, with standard
deviations, ranged from 92.26 1 0.05 percent for
the control to 88.20 ± 0.08 percent for 30-day
stratification. The range among seedlots was also
small (14 percent), varying from 98 percent on the
North Kaibab to 84 percent at Deadman Flat.

Germination energy. — There were differences
between treatments in the number of days to reach
50 percent germination, with germination
significantly faster for cold stratification
treatments than for the control (table 2) . Seed
from the Prescott and North Kaibab showed the
greatest response to cold stratification, with 50

and 82 percent decreases, respectively, in time
necessary for 50 percent germination. Sixty days
of cold stratification produced more rapid
germination than the 30-day treatment or the
control for Deadman Flat, Cedar Ranch, and Tusayan
seed. However, combining the data for the five
seedlots tended to reduce the difference between
stratification treatments, although both were
still different than the control.

The control treatments. — There appeared to be no
differences in germination capacity among
seedlots. Differences were apparent in
germination energy. The Cedar Ranch and North
Kaibab seed germinated in an average of 10.6 days,
while the other three seedlots germinated in an
average of 17.1 days.

Hydrogen Peroxide Treatments

Germination capacity. — Analysis of the hydrogen
peroxide treatments indicated differences for the
pooled seedlots (table 3) . Seeds receiving the
30-percent hydrogen peroxide soak for 10 minutes
germinated least (52.4 ± 25.3 percent), and
significantly less than seed receiving the
3-percent solution for 5 or 10 minutes or the
30-percent solution for 2.5 minutes. These three
treatments averaged 83.3 percent. The 30-percent
solution for 10 minutes was similar to the
control, the 3-percent for 2.5 minutes, and the
30-percent for 5-minute treatments.

Some treatment differences also were noted within
three of the four seedlots (table 3) . Only seeds
from Cedar Ranch showed no significant differences
among treatments. Germination percentages for
seeds receiving the 30-percent hydrogen peroxide
soak for 10 minutes were consistently lower than
any other treatment within seedlots. Even the
Cedar Ranch ecotype showed a slight drop of
10 percent in germination compared to the control.
A drop in germination of 38 percent was noted for
North Kaibab seed. The Tusayan control did not
germinate well; however, the difference between
the 30-percent peroxide for 10-minute treatment
and the best Tusayan germination (30-percent
peroxide for 2.5 minutes), which was helped by
treatment, was about 46 percent.

Two of the seedlots also showed significant
germination differences for the 30-percent

Table 2. — Germination energy expressed as the average number of days (with standard deviation) to achieve
50-percent germination of filled seed after cold stratification

Seedlot Prescott Deadman Cedar Ranch Tusayan N. Kaibab Total

treatment Flat

Control 16.2+2.8 a1 17.6+1.5 a 10.2+0.8 a 17.6+0.9 a 11.0±0.7 a 14.5+3.6 a

30-day 8.213.0 b 11.6 + 2.4 b 6.2+0.8 b 12. Oil. 6 b 1.8+0.4 b 8.0+4.2 b
stratification

60-day 5.6+2.1 b 9.6+0.9 c 4.4+1.1 c 8.412.7 c 1.8+0.4 b 6.013.1 b
stratification

Means within each column followed by different letters are significantly different at the 5 percent level.

40

Table 3. — Average germination capacity (percent with standard deviation) for filled seed after hydrogen
peroxide treatment

Prescott Cedar Tusayan N. Kaibab Combined

Ranch

Control
3Z x 2.5 min
3Z x 5 min
3% x 10 min
30Z x 2.5 min
30% x 5 min
30Z x 10 min

83. 5± 3.7 be

87.5+10. A c

88. 71 6.7 c

87. 5± 3.3 c

83. 5± 6.8 c

69.2116.7 ab

54.2+15.2 a

94. 4 + 3.6 a
90.3+ 3.7 a
90.4+10.4 a
92.8+ 6.6 a
94.4+ 3.6 a
94.4+ 4.5 a
84.7+10.6 a

26.7+14.3 a
30.3+11.7 a
65.1+ 9.6 b
61.6+17.6 b
69.6+ 8.3 b
64.8+10.4 b
23.2+ 8.2 a

85.0+ 3.9 c
89.9+ 6.4 c
87.9+ 6.3 c
86.3+ 5.4 c
91.9+ 3.0 c
70.8+ 7.6 b
47.4+16.7 a

72.4+30.8 ab
74.5+29.5 ab
83.0+12.0 b
82.0+13.9 b
84.8+11.2 b
74.8+13.3 ab
52.4+25.3 a

Means within each column followed by different letters are significantly different at the 5 percent
level .

5-minute treatment (table 3). This treatment for
Prescott (69 percent germination) was
statistically similar to the control and to the
30-percent 10-minute soak, but less than for the
other treatments, which averaged 87 percent. The
30-percent 5-minute soak for North Kaibab
(71 percent) was higher than the 10-minute soak in
30-percent peroxide (47 percent) but lower than
the other five treatments, which averaged
88 percent. However, the 30-percent 5-minute
treatment produced relatively good germination at
Tusayan .

Comparing data from the controls of the
stratification and the peroxide tests indicated
that Tusayan germination capacity decreased from
May 1984 to January 1985. The control averaged
87 percent germination for the cold stratification
test, and 27 percent for the hydrogen peroxide
test. A comparison of control germination
capacities for the other seedlots also showed
decreased germination:

1984 1985

Prescott 87.8 83.5

Cedar Ranch 97.5 94.4

North Kaibab 97.4 85.0

T-tests, however, indicated that only the North
Kaibab seed germination capacity had also dropped
significantly over the 8 months.

Germination energy. — The analysis of variance for
the combined data also indicated significant
differences among treatments (table 4); but the
differences between treatments could not be
demonstrated by the Fisher's F-protected test. It
appears that seeds which received the 30-percent
10-minute soak took longer to reach 50 percent
germination than it took seeds receiving the other
treatments .

Comparisons within seedlots generally indicated
that the 30-percent 10-minute treatments retarded
germination energy, while the other treatments had
no effect. Tusayan was an exception, however,
exhibiting the same trend as germination capacity.
The control group did very poorly, while some
peroxide treatments stimulated germination. The
control group never achieved 50 percent
germination .

Mold . — The various peroxide treatments did not
prevent or delay the contamination of seed by
various seed or airborne fungi. The only
significant delay — about 7 days compared to the
control — was for the 30-percent 5-minute treatment
of Prescott seed. There also appeared to be
differences among seedlots. Cedar Ranch seed
became contaminated after 23 days, while Prescott
seed became contaminated in 4 days. Contamination
was noted on the Prescott seed during the
stratification period in the first trial. Tusayan
and North Kaibab were contaminated within 10 days.

Table 4. — Germination energy expressed as the average number of days (with standard deviation) to achieve
50-percent germination of filled seed after hydrogen peroxide treatment

Prescott Cedar Tusayan N. Kaibab Combined

Ranch

Control
3% x 2.5 min
3% x 5 min
3% x 10 min
30% x 2.5 min
30% x 5 min
30% x 10 min

12.8+ 1 . I a
12.2+ 0.4 a
12.0+ 0.7 a
12.0+ 1.2 a
L2.4+ 1.5 a
20.6+15.4 a
33.0+17.0 b

10.2+0.2 a

10.2+0.2 a

10.4+0.2 a

10.2+0.2 a

10.4+0.2 a

10.4+0.2 a

11.6±0.4 b

44.41 2.2 a
44.41 2.2 a
24.41 5.5 b
31.6111.7 b
26.41 5.1 b
29.01 5.7 b
44.4+ 2.2 a

10.2+ 0.8 a

10.41 0.5 a

11.2+ 1.1 a

11.41 1.3 a

11.0+ 0.7 a

14.21 3.0 a

31 .2114. 1 b

19.4116.7 a
19.3+16.8 a
14.51 6.6 a
16.3110.2 a
15.01 7.6 a
19.31 8.1 a
30.0H3.6 a

Number within a column followed by the same letter is not significantly different at the 5 percent level.

41

A sample of contaminated seeds was inspected by
Dr. J. States, Biology Department, Northern
Arizona University. He identified: Penicillium
(2 species), yeast (3 species), Alternar ia ,
Cladosporium, Bispora , Gliocladium, and Rhizopus .
Dr. States indicates that the molds will attack
and kill germinating seedlings under high moisture
conditions. No effort was made to relate genera
to specific hydrogen peroxide treatments.

DISCUSSION AND CONCLUSIONS
Cold Stratification

Although the germination capacity of filled seeds
was not significantly different among treatments,
cold stratification affected the germination
energy of pinyon. Both the 30-day and 60-day
treatments germinated faster than the control.
Some seedlots did better after 60 days than after
30 days. Results over all seedlots, however,
indicated no differences between stratification
periods. While it is difficult to make
recommendations after testing five seedlots, the
more rapid germination of pinyon seed after cold
stratification was consistent. It appears that a
30-day treatment should be sufficient for
laboratory or greenhouse purposes where ideal
environmental conditions are uniformly maintained.
A 60-day stratification period may be more
beneficial for field sowing because suboptimal
conditions often occur. For example, Allen (1960)
found more rapid germination of Douglas-fir
(Pseudotsuga menziesii Mirb . Franco) seed
following long stratification periods of up to
150 days when spring temperatures were low. Local
seed should be tested to evaluate the effects of
cold stratification, especially if large seeding
operations are being considered.

A comparison of germination energy for the control
treatments suggested differences among seedlots.
The Cedar Ranch and North Kaibab seed germinated
in an average of 10.6 days, 6.5 days earlier than
the average for the other three seedlots.

Some combination of elevation and latitude may be
responsible for the difference between the two
groups. Either factor alone did not seem to
influence this difference, because Cedar Ranch is
south of Tusayan, while Tusayan and Deadman Flat
are at a similar elevation as the North Kaibab.
The growing season most likely starts later, and
is shorter on more northern sites or above 2 134 m
(7,000 ft). The ability to germinate quickly once
proper climatic conditions are reached would give
higher elevation or northern seedlings more time
to become established before the onset of severe
conditions, especially drought.

Hydrogen Peroxide Treatments

Although most treatments did not affect
germination capacity, relative to the controls
(table 3) , germination capacity generally appeared
lowest for seed receiving the 30-percent 10-minute
treatment. Riffle and Springfield (1968), working
with pinyon seed from Santa Fe , New Mexico, found

that a 30-minute soak with 30-percent hydrogen
peroxide treatment produced a slight but
nonsignificant 18-percent increase in germination.
They reported that combining water treatments with
the hydrogen peroxide treatment improved
germination relative to their control; but the
results from all hydrogen peroxide treatments were
similar .

Germination energy was generally depressed, or in
the case of Tusayan, not improved by the
30-percent 10-minute treatment (table 4). Riffle
and Springfield (1968), in contrast, indicated
that all of their hydrogen peroxide treatments
improved germination.

Ecotypic variability could account for the
different results encountered in the seedlots, and
for the differences with Riffle and Springfield
(1968). They only tested one seed source and
recognized the fact that other seedlots could
react differently. The conclusions from this
study also would have been different if the tests
had been conducted on either Cedar Ranch or
Tusayan seed alone.

Ecotypic differences also may have been a factor
causing the decrease in germination capacity of
the two northern seedlots (Tusayan and North
Kaibab) while in storage. All attempts were made
to handle all seed similarly. Meeuwig and Bassett
(1983) reported that pinyon seed lost viability
rapidly after 1 year; however, this may be under
natural conditions. Other investigators indicated
difficulties maintaining viability of singleleaf
pinyon (P. monophylla) seeds under storage (Budy
1985). However, the Deadman Flat seed used in the
stratification test had a germination capacity of
84 percent after 4 1 years in storage. This seed
was stored in a closed glass jar, while the 1983
seeds were stored in paper bags placed in loosely
closed plastic bags. While this may account for
better germination of Deadman Flat seed, it does
not explain the differences with Cedar Ranch or
Prescott seed.

The six hydrogen peroxide treatments did not
prevent or delay contamination of seed by mold.
Riffle and Springfield (1968) found that their
treatments prevented contamination. Differences
in techniques, and the longer period of treatment,
may account for the different results. Dishes in
this study could have been recontaminated by
airborne spores. However, the hydrogen peroxide
did not reduce the growth of seed fungi initially
present or of new contamination by airborne fungi.
Riffle and Springfield (1968) also identified
Penicillium , Alternaria , and Rhozopus plus
Tr ichoderma , Cephalosporium , Aspergillus , and
Fusarium . Differences in the number of days
before contamination was noted and varied by
seedlot, again indicating possible ecotypic
variation .

42

REFERENCES

Allen, G. S. Factors affecting the viability and
germination behavior of coniferous seed. IV.
Stratification period and incubation
temperature, Pseudotsuga menziesii (Mirb.)
Franco. For. Chron. 36: 18-29; 1960.

Arnold, Joseph F.; Jameson, Donald A.; Reid,

Elbert H. The pinyon- juniper type of Arizona:
Effects of grazing, fire, and tree control.
Prod. Rep. 84. U.S. Department of Agriculture,
Forest Service; 1964. 28 p.

Budy, J. D. Personal communication. Reno, NV:

University of Nevada, Assistant Professor; 1985
April.

Kramer, Paul J.; Kozlowski, Theodore T. Physiology
of woody plants. New York, NY: Academic Press;
1979. 811 p.

Krugman, Stanley L.; Jenkinson, James L. Pinus L.
In: Schopmeyer, C. S., tech. coord. Seeds of
woody plants in the United States. Agric .
Handb. 450, 883 p. Washington, DC: U.S.
Department of Agriculture, Forest Service;
1974. p. 598-638.

Lanner, Ronald M. Natural hybridization between
Pinus edulis and Pinus monophylla in the
American Southwest. Silvae Genetica. 23:
108-116; 1974.

Meeuwig, Richard 0.; Bassett, Richard L.

Pinyon-juniper . In: Burns, R. M., tech. comp.
Silvicultural systems for the major forest
types of the United States. Agric. Handb. 445,
191 p. Washington, DC: U.S. Department of
Agriculture, Forest Service; 1983. p. 84-86.

Riffle, J. W. ; Springfield, H. W. Hydrogen
peroxide increases germination and reduces
microflora on seed of several southwestern
woody species. For. Sci. 14: 96-101; 1968.

Sellers, William D.; Hill, Richard H . ;

Sanderson-Rae , Margaret. Arizona climate.
Tucson, AZ: University of Arizona Press; 1985.
143 p.

Springfield, H. W. Characteristics and management
of southwestern pinyon-juniper ranges: The
status of our knowledge. Res. Pap. RM-160. Fort
Collins, CO: U.S. Department of Agriculture,
Forest Service, Rocky Mountain Forest and Range
Experiment Station; 1976. 32 p.

West, Neil E . ; Rea, Kenneth H.; Tausch, Robin J.
Basic synecological relationships in
juniper-pinyon woodlands. In: The
pinyon-juniper ecosystem: A symposium. Logan,
UT: Utah State University; 1975: 41-53.

.3

ACETONE IS UNRELIABLE AS A SOLVENT FOR INTRODUCING GROWTH REGULATORS INTO

SEEDS OF SOUTHWESTERN PONDEROSA PINE £PINUS PONDEROSA VAR. SCOPULORUM) n

L . J . ^Heidmann

ABSTRACT: Acetone is reported to be a reliable
solvent for introducing plant growth regulators
into various seeds without affecting germination.
Several experiments in which southwestern
ponderosa pine seeds were soaked in acetone for
periods of 1 to 24 hours, then germinated under
various temperature regimes, gave negative results
even at soaking periods as short as 1 hour.
Acetone is not an appropriate solvent for intro-
ducing plant growth regulators into southwestern
ponderosa pine seeds.

INTRODUCTION

Germination of southwestern ponderosa pine (Pinus
ponderosa var. scopulorum) is temperature-
dependent (Pearson 1950; Larson 1961). Under
ideal temperatures in the laboratory, 20° to 25° C
(68° to 77° F) , 50 percent of seeds germinate in a
few days (Heidmann 1981). In the field, during
the germination period, diurnal fluctuations in
temperature of 4.4° to 26.7° C (40° to 88° F) are
common. Under these conditions, germination may
take several weeks. Cytokinin and abscisic acid
(ABA) levels appear to be related to temperature
changes. An increase in the level of ABA and a
decrease in cytokinin have been shown during
temperature stress, and are correlated with a
reduction in shoot growth (Itai and others 1973) .
Lettuce seeds, in the dark at 35° C (95° F) , were
not released from dormancy by kinetin or
gibberellic acid (GA^) ; however, when GA^ was
supplied with kinetin, germination occurred (Khan
1975) .

Dormancy in plants and seeds is likely under the
control of naturally occurring hormones.
According to Khan (1975), gibberellins (GAs) ,
inhibitors such as ABA, and cytokinins have
primary, preventive, and permissive roles in
control of seed germination. Gibberellic acid
will promote germination in the absence of ABA;
but when ABA is present, GA^ will only promote
germination in the presence of cytokinin.

Paper presented at the Conifer Tree Seed in the

Inland Mountain West Symposium, August 5-6, 1985,
Missoula, MT.

L. J. Heidmann is Principal Research Plant
Physiologist, Rocky Mountain Forest and Range
Experiment Station, Forest Service, U.S.
Department of Agriculture, Flagstaff, AZ.

Heidmann (1981) showed that GA_ and several other
compounds significantly reduced the time required
for 50 percent germination of southwestern pondero
pine seeds under fluctuating temperatures similar
to those encountered in the field. The compounds
were applied as an aerated soak. Aeration in
water alone significantly reduced the time to
50 percent germination, presumably because of the
effect of oxygen. Although stimulation of
germination by oxygen is a desirable trait, it
masks the effect of the compounds being studied.
Therefore, to eliminate the confounding effect of
oxygen and to isolate the effect of the growth
regulators under study, another method of
introducing regulators into the seeds is required.

Tao and Khan (1974) showed that GA^ and indole-
3-acetic acid (IAA) can be introduced into the
embryos of various dry seeds with acetone.
Acetone itself did not adversely affect
germination .

In 1983 and 1984, a series of experiments was
conducted to determine the effect of acetone on
the germination of southwestern ponderosa pine
seeds. If acetone did not modify germination
patterns, then it was to be used to introduce GA^,
ABA and N-6-benzyladenine (a cytokinin) , singly or
in combination, into pine seeds.

METHODS

Four experiments were conducted. The first three
experiments were conducted in a similar manner.
Seeds were collected on the Coconino National
Forest, approximately 40 km (25 mi) north of
Flagstaff, AZ, at an elevation of about 2 440 m
(8,000 ft). Seeds were collected in 1979 and were
stored at -17° C (2° F) until use. Viability in
1980 was 98 percent. Seeds were soaked in 100 ml
of acetone for 0, 4, 8, 16, and 24 hours. After
soaking, seeds were rinsed several times in
distilled water, then were placed on Whatman glass
fibre filter paper in 10-cm (3.9-inch) disposable
petri dishes. International Seed Testing
Association guidelines (1976) suggest using four
replications of 100 seed for conducting
germination experiments. However, because of a
scarcity of seed, six replications of 50 seeds
were used in the first three experiments, and five
replications were used in the fourth. Treatments
were randomized within replications, then dishes
were placed on a single shelf in a Hoffman seed
germinator set, for 16-hour days and 8-hour
nights. In experiment 1, day temperature was

44

maintained at 27° C (80° F) and night temperature
at 7° C (45° F) . For experiment 2, temperatures
were 24° C (75° F) for days and 4.4° C (40° F) for
nights. Temperatures for experiment 3 were 24° C
(75° F) days and 18° C (65° F) nights. Temperatures
chosen were those likely to be used in conducting
germination tests with plant growth regulators.
Results from tests conducted at optimum temperatures
were compared to those from experiments using
temperatures similar to those encountered in the
field during the germination period.

Dishes were watered as necessary with distilled
water. Care was taken not to overwater, because
germination percentages of southwestern ponderosa
pine seeds are lowered considerably by excess
water (Rietveld 1985). Germination counts were
made daily. Seeds with a radicle of 5 mm
(0.2 inch) in length were recorded as germinated,
then were discarded.

Experiment A was conducted with seeds collected on
the North Kaibab Ranger District of the Kaibab
National Forest, in 1962, at an elevation of
approximately 2 440 m (8,000 ft). Viability of
the seed was 90 percent in 1980. The experiment
was conducted in the same manner as the first
three experiments, except that seeds were soaked
in acetone for 0, 1, 2, and 3 hours. For this
experiment, a Perclval germinator was used. Day
length was 16 hours at 24° C (75° F) , and night
length was 8 hours at 18° C (65° F) .

Treatment means for all experiments were analyzed
using Dunnett's T3 multiple comparison procedures
for heterogeneous variance (P=0.10). Individual
T-tests between treatments were also run (P=0.01
and 0.05) .

RESULTS AND CONCLUSIONS

Results from the four experiments were inconsistent
with wide variability. In the first three
experiments, soaks in acetone of 8 hours or more
generally reduced germination (table I, P=0.1).
In experiments 1 and 3, a 24-hour soak almost
completely repressed germination; in the second
experiment, this treatment was not significantly
different from the control. Although a 4-hour
soak is not significantly lower than controls, the
germination is as much as 30 percent less
(table 1).

Responses to germination temperatures were equally
unpredictable. The first two experiments utilized
minimum temperatures likely to be encountered in
the field during the summer germination period. A
temperature regime of 27° C (80° F) days and 7° C
(45° F) nights, resulted in the highest germination
percentage for the control (84 percent) , while
germination for the 24-hour soak was only
4 percent. Dropping the maximum and minimum
temperatures 2.8° C (5° F) reduced control
germination 22 percent, but resulted in a
40 percent increase in germination for seeds
soaked 24 hours. In the third experiment,
conducted under what are considered optimum
temperatures, there was virtually no germination
for the 16- and 24-hour soaking periods.

Information from the first three experiments
seemed to indicate that a shorter soaking time
might be desirable. This is refuted by results
from experiment 4 which show that a 1-hour soak
significantly reduced germination. Individual
T-tests between the control and the other three
treatments showed that each of the soak treatments

Table L. — Effect of various acetone treatments on germination of southwestern ponderosa pine seeds

Experiment Number

Soak Time 12 3 4
(hours)

i i i I

Germ Std. Germ Std. Germ Std. Soak Time Germ Std.

%

Dt v .

%

Dev.

%

Dev.

(hours )

Z

Dev.

0

84

2

33a

3

a

7.53

62

00a2

3

a

15.65

79

67a2

3

a

4.46

0

77.60a2

3

a

3.85

4

69

67a

ab

27.75

58

00a

a

9.63

48

33a

a

25.91

1

46.80b

b

20.62

8

21

00b

be

23.58

14

33c

b

12.42

8

67b

b

5.89

2

52.40a

b

22.60

16

14

67b

c

15.68

19

33br

b

23.55

0

00c

c

0.00

3

39.60a

b

31 .09

24

4

00b

c

6.20

44

33ab

a

5.57

0

33c

c

0.82

Mean of 6 replications of 50 seeds.

Means within each column followed by the same letter are not statistically different at the 0.1 level
according to Dunnett's T3 multiple comparison procedure.
3

Means within each column followed by the same letter are not statistically different at the 0.01 level
for experiments 1, 2, and 3 or at the 0.05 level for experiment 4 according to individual T-tests between
pairs of treatments.
4

Mean of 5 replications of 50 seeds.

45

reduced germination significantly (P=0.05). The
discrepancy between Dunnett's multiple comparison
procedure and individual T-tests is because
Dunnett's method is based on an overall
experiment-wise Type I error, and individual
T-tests use a comparison-wise Type I error (King
1985) .

These results confirm the erratic behavior of
southwestern ponderosa pine seeds when treated
with various materials. Although seeds can be
treated with strong oxidizing agents, such as
hydrogen peroxide, without affecting germination
(Heidmann 1981), other treatments, such as a
15-second soak in sodium hypochlorite (Clorox) ,
significantly repressed germination (Ronco 1985).
It appears that acetone adversely affects the
germination of ponderosa pine seeds. Soaking
periods as short as 1 hour significantly repressed
germination. This is contrary to findings by Tao
and Khan (1974) that indicate various seeds can be
soaked in acetone for as long as 28 hours without
affecting germination. Milborrow (1963) soaked
several species of seeds in acetone for periods as
long as 3 months without deleterious effects. The
primary conclusion to be drawn from these
experiments is that acetone is unreliable as a
solvent for introducing materials into the embryos
of ponderosa pine seeds. In experiments currently
in progress, growth regulators are being
introduced to the embryos under vacuum.

Milborrow, B.V. Penetration of seeds by acetone

solutes. Nature. 199: 716-717; 1963.

Pearson, G.A. Management of ponderosa pine in the
Southwest. Monogr . 6. Washington DC: U.S.
Department of Agriculture; 1950. 218 p.

Rietveld, W.J. Personal communication. Rhinelander
WI: U.S. Department of Agriculture, Forest
Service, Rocky Mountain Forest and Range
Experiment Station; 1985 March 4.

Ronco, Frank. Personal communication. Flagstaff,
AZ: U.S. Department of Agriculture, Forest
Service, Rocky Mountain Forest and Range
Experiment Station; 1985 January 20.

Tao, Kar-Ling; Khan, Anwar A. Penetration of dry
seeds with chemical applied in acetone. Plant
Physiol. 54: 956-958; 1974.

REFERENCES

Heidmann, L. J. Overcoming temperature-dependent
dormancy of southwestern ponderosa pine seed.
Res. Note RM-406. Fort Collins, CO: U.S.
Department of Agriculture, Forest Service,
Rocky Mountain Forest and Range Experiment
Station; 1981. 4 p.

International Seed Testing Association.

International rules for seed testing. Seed
Sci. Technol. 4(1): 3-180; 1976.

Itai, C; Benzioni, A.; Ordin, L. Correlative

changes in endogenous hormone levels and shoot
growth induced by short heat treatments to the
root. Physiol. Plant. 29: 355-360; 1973.

Khan, Anwar A. Primary, preventive, and permissive
roles of hormones in plant systems. Bot. Rev.
41(4): 391-420; 1975.

King, Rudy. Personal communication. Fort Collins,
CO: U.S. Department of Agriculture, Forest
Service, Rocky Mountain Forest and Range
Experiment Station; 1985 February 4.

Larson, M. M. Seed size, germination date, and

survival relationships of ponderosa pine in the
Southwest. Res. Note RM-66. Fort Collins, CO:
U.S. Department of Agriculture, Forest Service,
Rocky Mountain Forest and Range Experiment
Station; 1961. 4 p.

46

4

ARE YOU GETTING PONDEROSA PINE AND DOUGLAS-FIR
CONE CROPS AT HIGH ELEVATIONS?
Glenn L. iJacobsen

ABSTRACT: Research documentation and personal
observations of cone crops at the upper elevation
ranges of old-growth ponderosa pine and Douglas-
fir in Central Idaho indicate a longer cone
crop interval than the normal interval of A to 5
years for ponderosa pine. It appears that cone
croDS for ponderosa pine at 5,500- to 6,000-foot
elevations and Douglas-fir at 6,000- to 6,500-
foot elevations must coincide with a heavy to
extremely heavy cone crop at lower elevations
before we can obtain a collectible cone crop.

INTRODUCTION

Review of research literature for the Idaho area,
Lucky Peak Nursery reports from 1965 to 1984,
and observations made on the Payette and Boise
National Forests in Central Idaho indicate a
wide range of variability of cone crops,
especially with ponderosa pine and Douglas-fir.

In Central Idaho, ponderosa nine on commercial
forest land occurs between the elevations of
3,000 to 6,500 feet above sea level. Douglas-fir
occurs between the elevations of 3,500 to 7,000
feet. The vast majority of these species are in
old growth stands that are in a static condition.
Cone crop records generally indicate the lower
the elevation the heavier the cone crop. Cone
crops of ponderosa pine at the 5,500- to 6,000-
foot elevation and Douglas-fir at the 6,000- to
6,500-foot elevation have been very sparse.

The trend appears to be that whenever conditions
are right for a better than average ponderosa
pine cone crop, we also have at least an average
Douglas- fir cone crop. This may vary by Ranger
District and areas within Districts.

RESEARCH

Heavy ponderosa pine crops were documented
(Curtis and Foiles 1961) in Central Idaho in the
following years:

1936 - Heavy cone cron

1940 - Heavy cone crop

1941 - Cool moist summer — widesnread estab-

lishment of seedlings

1958 - Extremely heavy cone crop

1959 - Below normal summer rainfall — poor

establishment of seedlings

1962 - Average crop - less than 1/2 size of

1958 crop

1963 - Cool, moist summer months — unusually

successful natural regeneration

No research data are available on cone crops from
1964 to the present time for Central Idaho.

CONE COLLECTION RECORDS

Cone collection records for the Payette National
Forest indicate collectible cone crops in the
following years, although most crops were light.
Heavy nonderosa pine crops occurred in 1958,
1962, and 1971. Collectible crops are defined as
one bushel of cones per tree.

Paper presented at the Conifer Tree Seed in the
Inland Mountain West Symposium, Missoula, MT,
August 5-6, 1985.

Glenn L. Jacobsen is Forest Silviculturist,
Payette National Forest, Forest Service,
U.S. Department of Agriculture, McCall, ID.

Year

Species

Elevation rang

1958

Ponderosa pine

(fe
Unknown

et)

1962

Ponderosa pine

3,000 -

4,000

1963

Ponderosa pine

3,000 -

4,000

1964

Ponderosa pine
Douglas-f ir

3,000 -
4,000 -

5,000
6,000

1965

Ponderosa pine
Douglas-f ir

3,000 -
4,000 -

4,000
5,000

1971

Ponderosa pine
Douglas-fir

4,000 -
5,000 -

6,000
6,000

1974

Ponderosa pine
Douglas-fir

4,000 -
5,000 -

5,000
6,000

1975

Ponderosa pine

4,000

1978

Ponderosa pine
Douglas-f ir

3,000 -
4,000 -

5,400
6,u00

1980

Douglas-f ir

4,500 -

5,500

1982

Couglas-f ir
Ponderosa pine

4,500 -
4,000 -

6,000
5,500

47

Note the gap in cone collection between 1965 and
1971. Cone crops, if any, were light and the
Forest did not believe there was a need to
collect additional seed to supplement the seed
inventory during this period. The seed inventory
was considered adequate for the artificial regen-
eration need.

OBSERVATIONS

On the Payette National Forest, ponderosa pine-
Douglas-fir appears to have the following cone
crop intervals:

Elevation Range Collectible Cone Crop Interval

3,000 - 4,000 2 to 3 years
4,000 - 5,000 3 to 4 years

5,000 - 6,500 Coincides with an above average
crop at the lower elevations

For ponderosa pine at 5,500 to 6,000 feet and
Douglas-fir at 6,000 to 6,500 feet, it appears
the cone crop must coincide with a heavy to
extremely heavy cone crop at the lower elevations
before we get a collectible crop at the higher
elevation range of these species.

The last heavy crop in ponderosa pine was 1971.
Prior to that a heavy crop occurred in 1958, a
13-year interval.

MANAGEMENT IMPLICATIONS

Cone croDS are not uniform in Central Idaho,
with variations occurring primarily due to
elevation range, tree age and vigor, and soil
type .

Schmidt and Shearer (1971) found that 24 out of
25 ponderosa Dine seed that reach maturity are
eaten by small forest animals. This information
was determined over a 6-year period with various
cone crons. This, combined with insects feeding
on cones and seeds, only leaves seed available
for natural regeneration during average or
better cone crop years.

The silvicultural and management implications of
the above observations and trends are:

1. We have less ODDortunities for regeneration,
both natural and artificial, at the higher
elevation of a species range due to longer cone
crop intervals.

2 . If a cone crop occurs at the higher eleva-
tion of a species range, we must not miss the
opportunity to collect cones or coordinate site
preparation with the cron to achieve natural
regeneration .

SUMMARY

The average ponderosa pine cone crop interval
documented currently is 4 to 5 years (Barrett
1979) .

TRENDS

With the elevation range of ponderosa pine and
Douglas-fir, the trend appears to be: At the
lower elevation of a species range, cone crons
are 3 to 4 times more frequent than at the
higher elevation range.

This trend is supported by the reforestation
personnel who have tenure on Districts of the
Payette National Forest and have observed cone
crops throughout the years . Tenure varies from
10 to 25 years depending on the respective
District. Due to the lack of high-elevation seed
in the Payette's seed inventory, emphasis has
been placed on trying to collect high-elevation
seed if cones are available. This trend is not
documented by research to my knowledge, but many
forests are reporting difficulty in obtaining
high-elevation seed for different species due to
infrequent cone crops. Research (USDA Forest
Service 1965) has documented in Central Idaho
that mature and overmature ponderosa pine growing
at 5,500 feet produce lower quality seed than
similar trees at 4,000 feet.

Observation and trends in Central Idaho indicate
ponderosa pine and Douglas-fir cone crops at
the higher elevation range of these species are
infrequent. This adds considerable pressure to
the timber manager to collect cones at these
elevations when a collectible crop does occur,
plus accomplish site preparation for natural
regeneration with a cone crop.

Managers in areas other than Central Idaho need
to ask themselves if they are getting ponderosa
and Douglas-fir cone crops at higher elevation
ranges. If not, they may want to place more
management emphasis on cone collection during a
cone crop or site preparation for natural
regeneration. Douglas-fir and ponderosa nine
may not be the only species with cone crop
frequency problems in the higher elevation
ranges .

REFERENCES

Barrett, James W. Silviculture of ponderosa pine
in the Pacific Northwest: the state of our
knowledge. General Technical Report PNW-97.
Portland, OR: U.S. Department of Agriculture,
Forest Service, Pacific Northwest Forest and
Range Experiment Station; 1979. 106 p.

Curtis, James D.; Foiles, Marvin W. Ponderosa
pine seed dissemination into group clear-
cuttings. Journal of Forestry. 59(10):
766-767: 1961.

48

Schmidt, Wyman C; Shearer, Raymond C. Ponderosa
pine seed — for animals or trees? Research
Paper INT-112. Ogden, UT: U.S. Department of
Agriculture, Forest Service, Intermountain
Forest and Range Experiment Station: 1971.
14 p.

USDA Forest Service. Silvics of forest trees of
the United States. Agricultural Handbook 271.
Washington, DC: U.S. Department of Agriculture,
Forest Service; 1965. 762 p.

49

SEED-DISPERSAL CHARACTERISTICS OF CONIFERS
IN THE INLAND MOUNTAIN JEST,

Ward W. ^McCaughey , Wyman C.|_Schmidt, and Raymond C.^Shearer

•Is

ABSTRACT: This paper summarizes seed-dispersal
characteristics and factors affecting dispersal
of the major conifers found throughout the
Inland Mountain West. Seed dispersal of these
conifers is influenced by a number of physical,
climatic, and biotic factors such as seed and
wing size, height of cone-bearing trees,
distance from seed source, physiographic
position, and wind patterns. Birds and
mammals also aid in the dispersal of larger
seeds, particularly those that are wingless.

This paper aims at consolidating published and
unpublished seed-dispersal information for
23 major conifer species found throughout the
Inland Mountain West — that geographic area
between the east slopes of the Rockies and
east slopes of the Cascades and Sierras in the
United States and Canada. Because seed and
wing size, height of cone-bearing trees,
distance from seed source, physiographic
position, and wind patterns also influence
seed dispersal, they are also included in this
discussion.

INTRODUCTION

Seed-dispersal information can tell us where
seed will be dispersed, how much seed can be
expected, and when it can be expected. This
knowledge, and an understanding of seed
germination and seedling survival, provides
much of the basic information needed to make
valid decisions about regenerating an area.
For example, the seed-dispersal characteristics
of a species may indicate that adequate seed
can be expected throughout an area, and that
natural regeneration has a high probability of
success .

Conversely, dispersal characteristics may
indicate that much of the area is far too
distant from a seed source and will have to be
artificially regenerated.

Because natural regeneration is often very
desirable and practical for many species,
seed-dispersal characteristics should be
considered in planning regeneration cuttings.
This consideration can influence the choice of
cutting system such as seed tree, shelterwood,
or clearcuttings , the layout and timing of
cutting, and site preparations.

Paper presented at the Conifer Tree Seed in
the Inland Mountain West Symposium, Missoula,
MT, August 5-6, 1985.

Ward W. McCaughey is Forester, Wyman C.
Schmidt is Research Silviculturist and Project
Leader, and Raymond C. Shearer is Research
Silviculturist, "Silviculture of Subalpine
Forest Ecosystems" Research Unit, Intermountain
Research Station, Forest Service, U.S. Depart-
ment of Agriculture. McCaughey and Schmidt
are stationed at Bozeman, MT and Shearer at
Missoula, MT.

SEED DISPERSAL CHARACTERISTICS

Throughout forests of the Inland Mountain West
the typical shape of the seed-dispersal
pattern for wind-dispersed seeds is a negative
exponential curve. Quantity of dispersed seed
decreases rapidly as distance from the windward
seed source increases and then remains at a
low level. The shape of the dispersal curve
is much the same at the leeward side of
openings, but total distance of dispersal and
amount of seed at a given distance from seed
source is less than that from the windward
edge. Thus, there is a U-shaped distribution
of seed dispersal across openings. Some
exceptions to this dispersal pattern occur for
wingless seeds of juniper (Juniperus spp.) and
pinyon (Pinus spp.), which are dispersed by
other modes such as birds, mammals, and
gravity.

To provide a basis for comparing dispersal
characteristics of nine species, we developed
dispersal curves that best fit the variety of
available data (fig. 1). Using number of
seeds dispersed at the windward timber edge as
a species base value, we related the number of
seeds dispersed to different distances from
the source as a percent of the amount at the
timber edge. For species comparison purposes,
we used the term "dispersal efficiency,"
defined as the ratio of number of seeds at the
source to the quantity dispersed at a given
distance from the source.

A natural logarithm model was used to develop
curves for each of the nine species; therefore,
curve shapes were similar. However, there
were substantial differences in the percent of

50

FIR

0 20 40 60 80 100 120

DISTANCE FROM SEED SOURCE, M

Figure L. — A comparison of the seed dispersal characteristics of nine conifers of the Inland Mountain
West. This relates the number of seeds found at the seed source to the numbers found at increasing
distances from the source. Percent values are used to arrive at a common base for the different
species.

seeds dispersed to increasing distances. For
example, western hemlock (Tsuga heterophylla
[Raf.] Sarg.) disperses seeds very efficiently
but lodgepole pine (Pinus contovta var.
latifolia Engelm.) does rather poorly. At
50 m (164 ft), hemlock dispersal was 70 percent
while lodgepole pine was only 20 percent of
that at the seed source.

It is also apparent that species of the same
genus do not necessarily exhibit similar seed
dispersal efficiencies. For example, white
spruce (Picea glauca [Moench] Voss) , Engelmann
spruce (Picea engelmannii Parry ex Engelm.),
and black spruce (Picea mariana [Mill.]
B.S.P.) range from good to poor in seed
dispersal efficiency. The percentage of white
spruce seeds reaching 50 ro (164 ft) is more
than double that of black spruce. Explanations
for differences in dispersal efficiencies of
these and other species are speculative
because definitive data are in short supply.

Mathematical models have been developed for
some of the western conifers to illustrate
seed dispersal. These models usually fit the
data well but offer little explanation of why
each species behaves as it does.

Some species, such as Engelmann spruce, have
several mathematical models of seed dispersal
developed from data collected in different
areas within their geographic range (fig. 2).
Here, in spite of different independent
variables, the dispersal curve shapes are
similar. Absolute quantities of dispersed
seeds are similar when you consider that one
model (Roe 1967) was developed from a bumper
crop while the other two models represent
several years of seed crop data.

Seed-dispersal patterns of winged-seeded
species are influenced by seed and wing
length, height and physiographic location of
seed source trees, quantity of seed-bearing

51

DISTANCE FROM SEED SOURCE, M

Figure 2. — A comparison of seed dispersal curves from three studies of Engelmann spruce. All three
models were calibrated to 1 million seeds per hectare at seed source.

trees, yearly variation of seed production,
and time of dispersal. Sharpe and Fields
(1982) developed a mathematical model simu-
lating dispersal of winged seeds. Wind speed
at the seed source, height of source, and
humidity greatly influenced dispersal.
Although there are no data to support our
claim, small seeds with large wings probably
disperse farther than larger seeds with small
wings. Seeds likely disperse farther from
tall than from short trees and high winds
disseminate seeds greater distances than light
winds (Isaac 1930). Prevailing winds usually
determine the direction of seed dispersal.
Rising thermal winds can also disperse seeds
uphill in mountainous terrain at mid to lower
elevations (Alexander and others 1984; Shearer
1980) .

The amount of seed produced is related to the
number and character of seed-producing trees.
In the Intermountain region the quantity of
seed produced was proportional to the square
meters of basal area per hectare of Engelmann
spruce trees 25.4 cm (10 in) and larger in
diameter (Roe 1967; McCaughey and Schmidt in
preparation) . Each species differs in the
frequency of heavy, moderate, and light seed
crops. The period between heavy seed crops
for various species ranges from about 3 to
10 years. Sound seed production increases
linearly with total seedfall, although there
is considerable variability between years and
locations (Alexander and others 1982).

The following sections present brief descrip-
tions of seed dispersal for each of 23 coni-
fers found within the Inland Mountain West.
Table 1 presents descriptions of seed charac-
teristics and cone phenology.

52

Table 1. — Seed characteristics and tree phenology of some conifers of the Inland Mountain West

Seed characteristics1 Phenology

Average

Primary

number of

Cone

seed

Seed

Wing

seeds per

ripening

dispersal

Species

length

length

kilogram

dates

dates

mm

mm

Thousands

Abies

Grand fir

9.5

19.0

40.6

Aug.

Aug. -Sept.

Subalpine fir

-

-

76.7

Aug.

Sept . -Oct .

White fir

12.7

-

24.5

Sept. -Oct.

Sept. -Oct.

Juniperus

Rocky Mountain juniper

36.4

none

59.7

Sept .-Dec.

Oct.14

Western juniper

-

none

27.1

Sept .

Extended1*

Larix

Subalpine larch

-

-

313. 1

Aug. -Sept .

Sept.

Tamarack

3.2

6.4

701. 1

Aug . -Sept .

Sept . -Spring

Western larch

6.4

12.7

302.0

Aug.

Sept. -Oct.

Picea

Black spruce

3.2

6.4-9.5

890. 1

Sept.

Oct. 5

Blue spruce

-

-

233.4

Fall

Fall-Winter

Engelmann spruce

3.2

12.7

297.6

Aug. -Sept .

Sept . -Oct .

White spruce

3.2

6.4-9.5

498.2

Aug.

Sept.

Pinu8

Jack pine

2.1

8.5

288.8

Sept .

Sept.6

Limber pine

10.8

Aug. -Sept.

Sept . -Oct .

Lodgepole pine

4.2

12.7

207.2

Aug. -Sept.

Sept. -Oct. 6

Pinyon

4.2

Aug. -Sept.

Sept. -Oct.

Ponderosa pine

6.4

25.4

26.5

Aug . -Sept .

Aug. -Sept .

Western white pine

8.5

25.4

59.5

Aug.

Aug. -Sept .

Whitebark pine

5.7

Aug . -Sept .

Oct.7

Pseudotsuga

Douglas-fir (interior)

85.5

July-Aug .

Aug. -Sept .

Thuja

Western redcedar

3.2

3.2

912.7

Aug. -Sept .

Aug. -Sept.

Tsuga

Mountain hemlock

3.2

12.7

251.5

Aug. -Oct .

Sept . -Oct .

Western hemlock

1.6

573.2

Aug. -Oct .

Sept . -Winter

'information from: USDA Forest Service 1974; Harlow and others 1979.

information from: USDA Forest Service 1965; USDA Forest Service 1974; Schmidt and Lotan 1980.
3Cone is 6.4 mm in diameter and has one or two, rarely three or four seeds.
''Cones may persist on trees for 2 to 3 years.

5Black spruce retains its cones in a semiserotinous state for several years providing a continuous
source of seed in stands over 40 years old.

6Many serotinous cones remain closed for several months or years.
7Seeds are dispersed when the detached cone disintegrates.

5 3

Abies

Grand fir. — Wind disperses grand fir (Abies
grandis [Dougl. ex D. Don] Lindl.) seed in
early fall as cones disintegrate and fall from
the tree. The seeds are large, and as a
result dispersal distances are relatively
short. Haig and others (1941) report that fir
seed disperses up to 122 m (400 ft) from the
source, but 46 to 61 m (150 to 200 ft) was the
average dispersal distance. On steep topog-
raphy of a western Montana site, grand fir
seed dispersed upslope nearly 244 m (800 ft)
from the nearest source (Shearer 1985).

Grand fir seed dispersal follows the same
patterns as the species shown in figure 1.
Eight-year results of seed production studies
in Idaho showed that grand fir produced the
least amount of seed of the species associated
with western white pine (USDA Forest Service
in press) .

Subalpine fir. — Subalpine fir (Abies lasiocarpa
[Hook.] Nutt.) seed is wind-disseminated.
Prevailing winds have been found to carry
seeds up to 80 m (264 ft) from the windward
edge of openings in Colorado (Noble and Ronco
1978). In Montana, Shearer (1980) found that
thermal upslope winds dispersed fir seed
uphill into clearcuts.

Seed dispersal of subalpine fir follows the
typical wind-dispersal pattern (fig. 1). In
clearcut openings in Colorado, quantity of
seed declined rapidly to about 30 m (100 ft),
and then remained at low levels to 80 m
(264 ft) from the timber edge (Noble and Ronco
1978) . Dispersed seed quantities increase at
about 10 m (33 ft) from the leeward edge of
the opening, giving a U-shaped distribution
profile of dispersal across the openings.

Nobel and Ronco (1978) reported that the
velocity of prevailing winds, amount of seed
produced, and distance from the source influ-
enced subalpine fir seed dispersal into
openings in Colorado. In the mountains of
Montana, peak dispersals were usually asso-
ciated with upslope winds, temperatures
greater than average, and humidity levels
lower than average (Shearer 1980) .

White fir. — White fir (Abies concolor [Gord. &
Glend.] Lindl. ex Hildebr.) seed is wind-
disseminated as cones disintegrate on the tree
in September and October (USDA Forest Service
in press). Some seed disperses at least 114 m
(375 ft) from the source into openings
(Franklin and Smith 1974b) . Because fir seed
wings are short and broad in relation to seed
size, dispersal distances are thought to be
small (Gordon 1970; Siggins 1933).

In Oregon, the quantity of white fir seed
decreased rapidly from the windward source out
to 38 m (125 ft) and then leveled off.
Dispersed seed quantities averaged about 25
and 10 percent at 38 m (125 ft) and 114 m
(375 ft), respectively, of amounts at the
windward source (Franklin and Smith 1974b).

Some white fir seed is dispersed by animals
such as the Douglas1 squirrel (Tamiasciurus
douglasii spp.), which cuts and caches fir
cones before disintegration (Fowells and
Schubert 1956).

Distance was the only measured factor that
affected white fir seed dispersal. The
quantity dispersed was significantly correlated
with distance from the stand edge (Franklin
and Smith 1974b). Although only 20 to 50 per-
cent of fir seeds are sound, even in good
production years, no significant differences
in dispersal distance's of sound and empty
seeds have been detected (USDA Forest Service
1965; Franklin and Smith 1974b).

Juniperus

Rocky Mountain juniper. — The fruits of Rocky
Mountain juniper (Juniperus scopulorum Sarg.)
are small, indehiscent strobili called "ber-
ries" that contain one to four, rarely as many
as 12, brownish seeds (USDA Forest Service
1974). The berries are heavy and normally
fall close to the parent tree accounting for
the slow expansion of juniper forests
(Burkhardt and Tisdale 1969). Most of the
fruit remains on the tree until late spring.
However, after fall ripening, birds and
animals have ample time to use it as a food
supply. Seed quickly passes through the
digestive tracts of birds and animals with
little effect on germination capability.
Randies (1949) reported that turkeys help
disseminate juniper seeds in the Southwest as
do bighorn sheep, chipmunks, foxes, and some
small mammals. Dispersal of juniper seed
depends upon the movement patterns of animals
that use the berries as a food source (Randies
1949). The Bohemian waxwing (Bombycilla
garrulus [Linnaeus]) can eat more than 900
seeds a day and spread them over a wide area
(Phillips 1910).

Western juniper. — The fruits of western
juniper (Juniperus occidentalis Hook.) are
small "berries" that contain one to four,
rarely as many as 12, brownish seeds. Disper-
sal occurs when the berries fall off the
trees, but usually birds or other animals eat
the fruit and disperse the seeds via their
excrement (USDA Forest Service 1974). Seed-
lings are found along fence rows, where they
are closely spaced as in a hedge, indicating

54

that seeds were bird-dispersed. Groups of
juniper seedlings are also found beneath other
trees to which birds apparently carried
juniper seed (USDA Forest Service 1965).
Specific information is limited on the types
of animals that disperse western juniper seed
and how far these animals carry the seed.

Larix

Subalpine larch. — Little is known about seed-
dispersal distances of subalpine larch (Larix
lyallii Pari.). The relatively lightweight,
winged seeds fall from the cones in September
and are wind-disseminated (USDA Forest Service
1974). Good seed crops occur about once every
10 years. Therefore, seedling establishment
is sporadic and limited (USDA Forest Service
in press) .

Tamarack. — Little information is available on
the seed-dispersal characteristics of tamarack
(Larix larioina [Du Roi] K. Koch). Duncan
(1954) reports large numbers of tamarack
seedlings located within 1 tree height, a
moderate number within 2 tree heights from the
seed source, but only scattered seedlings at
greater distances. This short dispersal
distance is probably explained by the fact
that mature seed-producing trees average only
15 to 25 m (50 to 80 ft) tall. Because
tamarack seed is winged and wind-dispersed, we
assume it follows the same distributional
pattern as other wind-dispersed species
(fig. 1). Characteristically, tamarack grows
in low flat areas, making topographic in-
fluences, so common with other western coni-
fers, of lesser consequence. Tamarack seeds
are also dispersed by red squirrels
(Tamiasciurus hudsonicus [Erxleben]), which
cut cone-bearing branchlets and cache the
cones (Duncan 1954).

Western larch. — Western larch (Larix
occidentaHs Nutt.) seed is small, enabling it
to wind-disperse long distances from the
source (USDA Forest Service 1965). Numerous
studies in Montana have indicated that larch
seed disperses at least 250 m (800 ft)(Boe
1953; Shearer 1959; Schmidt and others 1976;
Shearer 1985).

The amount of sound larch seed dispersing from
the source into clearcuts decreases rapidly
out to 122 m (400 ft), and then remains at a
low level. Seed crops vary substantially by
year, which affects the dispersal pattern. In
poor seed years, dispersal beyond 80 m (264 ft)
from the timber edge is usually not detected
with normal sampling procedures.

Thermal upslope winds in mountain terrain have
a direct effect on larch seed dispersal in
cuttings on mid to lower elevation slopes in
early fall. Cones on upper elevation slopes
mature 2 to 4 weeks later than at mid to lower
elevations (Fiedler 1976). Therefore, seed is
not released during the early fall when
thermal upslope winds are prevalent. As a
result a high proportion of larch seed on
upper slopes is dispersed by storm fronts in
mid to late fall (Shearer 1985).

Picea

Black spruce. — The maximum dispersal distance
of black spruce (Picea mariana [Mill.] B.S.P.)
seed is about 100 m (328 ft) from the source
(LeBarron 1939; Payandeh and Haavisto 1982).
LeBarron (1948) reported that although seed
disperses more than 91 m (300 ft) from a
source, the number of dispersed seed at 30 m
(100 ft) is only 6 percent of the amount of
seed falling in the uncut timber.

Dispersal is also affected by windspeed and
direction, time of release, and physiographic
location, but distance is the only factor that
has been quantified. Payandeh and Haavisto
(1982) showed that the two variables, stripcut
width and distance from the windward stand
edge, were highly correlated with the quantity
of spruce seed dispersed across a clearcut.

Blue spruce. — We found no specific studies
that described seed dispersal of blue spruce
(Picea pungens Engelm.). Shepperd (1985)
speculated that because blue spruce seed is
slightly larger than Engelmann spruce, disper-
sal distances are less. The effective seeding
distance, of blue spruce, to obtain adequate
natural regeneration, probably is about three
to four times the height of the seed-bearing
trees .

Engelmann spruce. — Wind, the main disseminator
of Engelmann spruce (Picea engelmannii Parry
ex Engelm.) seeds, carries seeds to distances
of 244 m (800 ft) from the uncut timber
bordering the windward edges of clearcuts
(USDA Forest Service in press). In the
Intermountain area, wind commonly disseminates
spruce seeds to distances of 201 m (660 ft)
from the source (Squillace 1954; Roe 1967;
McCaughey and Schmidt in preparation). These
distances were similar to distances reported
for clearcuts in Colorado (Alexander and
Edminster 1983). Seeds have also been observed
skidding great distances over a glazed snow
surface .

5 5

General patterns of spruce seed dispersal
across clearcuts are shown in figure 2. After
an initial rapid decline, the quantity dis-
persed levels off or gradually declines beyond
100 m (348 ft) from the seed source (Noble and
Ronco 1978; Roe 1967). The quantity of
dispersed seed increases again at about 10 m
(33 ft) from the leeward edge, producing a
U-shaped dispersal pattern across openings
(Noble and Ronco 1978). Alexander and
Edminster (1983) reported that, in Colorado,
nearly 40 percent of the spruce seedfall under
the uncut windward stand dispersed about 30 m
(100 ft), and 10 percent dispersed 91 m
(300 ft) from the source. Similar results
were found in the Intermountain region where
dispersed seed quantities started leveling off
at about 120 m (394 ft) from the source
(McCaughey and Schmidt in preparation).

Distance from seed source had the greatest
impact on the dispersal patterns of Engelmann
spruce (fig. 2) (Roe 1967; Noble and Ronco
1978; Alexander and Edminster 1983; McCaughey
and Schmidt in preparation). Also, studies in
the Intermountain area show that quantity of
seeds dispersing into clearcuts is strongly
correlated with basal area of mature spruce in
the adjacent uncut timber. As stand basal
area of mature spruce 25 cm (10 in) and larger
increased, quantity of dispersed seed increased
(Roe 1967; McCaughey and Schmidt in prepara-
tion). Colorado studies also show a strong
positive correlation of seedfall under the
uncut stand with quantity dispersed into
clearcuts (Alexander and Edminster 1983; Noble
and Ronco 1978) .

White spruce. — White spruce (Picea glauca
[Moench] Voss) has very small seeds, which
wind disperse in the air and over snow. Early
studies indicated 101 m (330 ft) was the
greatest distance seeds disseminated via air,
but with sufficient wind much greater distances
would be expected (Rowe 1955). Late-falling
seeds were blown considerable distances over
crusted snow (Rowe 1955). Nearly 88 percent
of seeds fall in September, the first month of
dispersal, with the remaining seeds dispersing
during the winter and spring (Crossley 1955).
About 20 percent of the total seeds dispersing
during the winter were trapped beyond 100 m
(328 ft), and 9 percent beyond 200 m (656 ft)
from the source. The presence of seedlings
indicated that spruce seed dispersed at least
300 m (984 ft) from the stand edge in strip
and clearcuttings (Dobbs 1976).

Quantity of dispersed spruce seed decreases
rapidly from timber edge to about 140 m
(460 ft) into clearcuts, and then levels off
for a considerable distance from the source.
This pattern suggests that significant quanti-
ties of seed are released in high winds (Dobbs
1976).

Pinus

Jack pine. — Jack pine (Pinus banksiana Lamb.)
has two seed dispersal modes because its cones
are serotinous and nonserotinous. Although
seeds are winged and small, about 2.1 mm
(0.33 in) long, they have a relatively short
dispersal distance. We found no studies that
described seed dispersal from nonserotinous
cones. However, indirect evidence of dispersal
through seedling establishment suggests the
effective dispersal range is about 34 to 40 m
(110 to 130 ft) — 2 tree heights — but, the
number of established seedlings more than
1 tree height from the seed source is low
(USDA Forest Service 1939). Seed dispersal
distances from serotinous cones are very short
because cones must be near the ground surface
before ambient air temperatures are high
enough to melt the resin bond. Cameron (1953)
reported that the bonding resin on serotinous
cones melts at about 50 °C (122 °F).

Limber pine. — Limber pine (Pinus flexilis
James) seeds are very large and have rudimen-
tary wings or are wingless (Harlow and others
1979). Because of their large size, they are
seldom dispersed by wind. The primary disper-
ser in northern Utah, Wyoming, and Arizona
(and presumably throughout the range of limber
pine) is the Clark's nutcracker (Nucifraga
oolumbiana [Wilson] ) (Tomback and Kramer 1980;
Benkman and others 1984). Nutcrackers carry
seeds to caches under forest litter. Some of
these cache sites are favorable to germination
(Tomback 1981). Red squirrels cache cones
under deep litter piles where seeds are
unlikely to germinate (Benkman and others
1984). Presumably, birds can disperse these
seeds great distances, but animal, wind, and
gravity dispersal likely provide limited
distribution.

Lodgepole pine. — Lodgepole pine (Pinus contorta
var. latifolia Engelm. ) has the smallest seeds
of any pine except jack pine. Yet, dispersal
distances are far less than such associates as
western larch and Engelmann spruce. Seeds
from nonserotinous cones, sufficient to
restock cutover areas, seldom disperse over
61 m (200 ft) from the source (Boe 1956; Perry
and Lotan 1977a). However, in western Montana,
winds dispersed seeds uphill nearly 244 m
(800 ft) from the downhill side of clearcuts
(Shearer 1985). Seeds have been observed
skidding great distances over a glazed snow
surface .

Dispersal distances for seeds from serotinous
cones depend upon the distance cones are
scattered during logging. Serotinous cones
need temperatures near 60 °C (140 °F) to break
resin bonds. Fire will easily melt the resin
bond, but cones must be within 30 cm (12 in)

56

of Che ground before ambient air temperatures
are high enough to open the cones (Perry and
Lotan 1977b; Lotan and Perry 1983). Eighty-
three percent of serotinous cones on south
slopes and 40 percent on north slopes open
when they are less than 30 cm (12 in) above
the ground (Lotan 1964).

Quantity of lodgepole seed decreases rapidly
as distance from the source increases — at 20 m
(66 ft) into clearcuts seedfall varies from
about 10 to 30 percent of that at timber edge
(Boe 1956; Dahms and Barrett 1975; Lotan and
Perry 1983). At 20 m (66 m) from the windward
side (fig. 1) the seedfall is about 45 percent
of that at stand edge. A U-shaped distribution
occurs between stand edges across clearcuts
(Lotan and Perry 1983).

Wind disperses lodgepole seeds from nonseroti-
nous cones, but seed from serotinous cones is
usually dispersed by scattering conebearing
slash throughout logging areas. Seed disper-
sal distances are likely to be greater on
south than on north slopes because of higher
velocities of thermal upslope winds on warmer
south-facing slopes (Fiedler 1974).

Pinyon . — Seed dispersal characteristics are
similar for pinyon (Pinus edulis Engelm.) and
singleleaf pinyon {Pinus monophylla Torr. and
Frem.), therefore, they are presented together.

The large seeds of pinyon are dispersed by
gravity or animals since the seed wing is
easily detached, and of no practical use in
dispersal (Phillips 1909). Clark's nutcrackers
cache seeds long distances from the source, 1
to 3 cm (.4 to 1.2 in) below the soil surface
in open areas (Vander Wall and Balda 1977;
Lanner and Vander Wall 1980). Some cached
seeds, not utilized by nutcrackers during the
winter months, germinate and thus expand
pinyon distribution. Everett (1985) speculates
that rodents disperse seed locally under, or
short distances from, existing pinyon stands.
Preliminary results indicate that nearly
87 percent of seedlings are located under or
near fully stocked stands. Rasmussen (1941)
reported that woodrats (Neotoma spp.), mice
(Peromyscus spp.), and chipmunks (Eutamias
minimus spp.) cached pinyon seeds, but whether
caches were on sites favorable for germination
was unknown.

Ponderosa pine. — Seed dispersal information
presented here is for ponderosa pine {Pinus
ponderosa Dougl. ex Laws.), since little
information is available for Arizona pine
(Pinus ponderosa var. arizonica [Engelm.]
Shaw) and Rocky Mountain ponderosa pine (Pinus
ponderosa var. scopulorum Engelm.).

Ponderosa pine seeds are fairly large, and as
a result do not disseminate great distances
(Barrett 1966). Theoretical calculations of
seed flight indicated that a 30-m (100-ft)
tree exposed to 32-km/h (20-mi/h) winds would
disperse seeds up to about 180 m (594 ft)
(Siggins 1933). Data indicate that most seeds
fall within 20 to 40 a (60 to 132 ft) from
seed-producing trees (Curtis and Foiles 1961;
Barrett 1966). However, a minor amount of
seed will disperse up to 160 m (528 ft) in
central Oregon and 244 m (800 ft) in western
Montana (Barrett 1966; Shearer 1985). In
Idaho, 82 percent of the seed dispersed within
152 m (500 ft) of the timber edge was confined
to 30 m (100 ft) of the edge. Some of the
seed falls to the ground in cones (USDA Forest
Service 1940). Seed is released when high
ground temperatures dry out the cones causing
scales to open (Curtis and Foiles 1961).

Western white pine. — Western white pine (Pinus
montiaola Dougl. ex D. Don) seed is mainly
wind dispersed, although squirrels, mice, and
birds disseminate some seed. Clumps of
seedlings have been found where seeds cached
by mice have germinated (USDA Forest Service
1965) . Seeds are large and do not disperse
great distances. Haig and others (1941)
reported that seed sufficient for adequate
reproduction seldom reaches more than 122 m
(400 ft) from the source. Isaac (1930)
demonstrated that when white pine seed was
dropped from a height of 61 m (200 ft), in a
21 km/h (13 mi/h) wind, it dispersed 792 m
(2,600 ft) from the point of release. Shearer
(1985) reported that white pine seeds dispersed
at least 244 m (800 ft) uphill from the source
at the bottom edge of steep mountain clearcuts
in Montana.

Whitebark pine. — Little information is avail-
able on the seed-dispersal characteristics of
whitebark pine (Pinus albicaulis Engelm.).
Its large seeds are wingless and disseminate
after the cones detach from the tree and
disintegrate (USDA Forest Service 1974). The
primary disseminators are animals, especially
the Clark's nutcracker. A single nutcracker
eats up to 32,000 seeds per year. These birds
store whitebark seeds in small caches on the
ground, some on microsites favorable for
germination. This dispersal by nutcrackers
helps maintain the "pioneering" status of
whitebark pine (Tomback 1981).

Squirrels also collect and cache whitebark
pine cones and seeds, often carrying them
great distances from the source. Bears also
feed on the seeds, especially grizzly bears
(Ursus arotos [Linnaeus]), which depend
heavily on these seeds as an important source
of energy, particularly in the Yellowstone

57

ecosystem. Some seeds pass through the bear's
digestive system intact; therefore, bears may
serve as a minor mode of dispersal.

Pseudotsuga

Rocky Mountain Douglas-fir. — Rocky Mountain
(interior) Douglas-fir (Pseudotsuga menziesii
var. glauca [Beissn.] Franco) has moderate-size
seeds and seed wings and is wind-disseminated.
Boe (1953) reports that large quantities of
seeds dispersed up to 80 m (264 ft) from the
source in clearcuts on the Coram Experimental
Forest in northwestern Montana. Seed also
disperses into clearcuts on steep topography
up to 244 m (800 ft) uphill from the source
(Shearer 1985). Reproduction indicates a
seed-dispersal radius of about 91 to 183 m
(300 to 600 ft) around open-grown trees on
level land (Frothingham 1909).

Douglas-fir has a seed-dispersal pattern
similar to other species (fig. 1). Quantity
of dispersed seed decreases rapidly out to
80 m (264 ft) and remains at low levels from
80 to 241 m (264 to 792 ft) from the source
(Boe 1953). Shearer (1985) reported that
early-ripening seeds were dispersed by upslope
thermal winds and late-ripening seeds by winds
of unstable air masses from storm fronts.

Thuja

Western redcedar. — Western redcedar (Thuja
plicata Donn ex D. Don) seed is wind-
disseminated, but because of a small wing
surface, its rate of fall is fast and flight
distance is short (Siggins 1933). Isaac
(1930) reports that seed did not disperse more
than 122 m (400 ft) when released at 46 m
(150 ft) above the ground. However, in
British Columbia, an examination of regenera-
tion indicated that seed dispersed at least
201 m (660 ft) from the source (Clark 1970).

The pattern of redcedar seed dispersal is
apparently similar to patterns shown for other
conifer species (fig. 1). Nearly 80 percent
of filled seeds in clearcuts are found within
15 to 30 m (50 to 100 ft) from the source.
(Gashwiler 1969). Seeds, dispersed between 61
to 76 m (200 to 250 ft) from the source,
accounted for 17 percent of the total seed
count in clearcuts with only 4 percent dis-
persing 107 to 122 m (350 to 400 ft).

Tsuga

Mountain hemlock. — Seeds of mountain hemlock
(Tsuga mertensiana [Bong.] Carr.) are small,
and have a large wing, making them well-suited
for wind dispersal. Mountain hemlock is a
prolific seed producer and bears seed nearly

every year (USDA Forest Service 1965) . Its
seed disperses at least 114 m (375 ft) into
clearcuts from the source (Franklin and Smith
1974a) .

The dispersal pattern of mountain hemlock is
very similar to Engelmann spruce (fig. 1).
Quantity of hemlock seed decreases rapidly
from the source out to about 38 m (125 ft),
and then levels off at a low level or decreases
slowly from 38 to 114 m (125 to 375 ft)
(Franklin and Smith 1974a). The amount of
seed dispersed at 120 m (394 ft) from the
source is less than 10 percent of that at the
source (f ig . 1 ) .

Mature seed-producing trees are relatively
short, usually 15 to 23 m (50 to 75 ft). Were
it not for this, seed-dispersal distances
would likely be much greater. The number of
seed-bearing trees in adjacent stands also
influences the amounts of seed dispersed into
clearcuts (Franklin and Smith 1974a).

Western hemlock. — Western hemlock (Tsuga
heterophylla [Raf.] Sarg.) seeds are light and
are dispersed great distances by wind. During
bumper crop years, enough seed falls within
100 m (330 ft) from the source to provide
optimum seedling establishment (Clark 1970).
To produce sufficient seed for adequate
stocking 201 m (660 ft) from the source, at
least two bumper seed crops are usually
required .

Considerable amounts of hemlock seed have been
observed dispersing up to 610 m (2,000 ft)
from the source. In a controlled test,
hemlock seeds released at a height of 61 m
(200 ft) in a 20.1 km/h (12.5 mi/h) wind dis-
persed as far as 1 158 m (3,800 ft) (Isaac
1930). In our comparisons of dispersal curves
(fig. 1), western hemlock stood out as the
most efficient of the western conifers. For
example, the number of seeds dispersed to
120 m (394 ft) was nearly half of the number
dispersed to the stand edge. This was about
double that of any other associated species.

Hemlock seed disseminates from cones throughout
fall and early winter months. During the
winter, winds sometimes disperse seeds across
crusted snow, depositing large amounts in
small depressions (Harris 1969).

DISCUSSION

Managers depending on natural regeneration in
their silvicultural practices should consider
seed dispersal characteristics in their plans.
This paper synthesizes information available
for our western conifers. Of this information,
the following factors stand out as being very
important in describing seed dispersal:

58

1 . Wind is the primary dispersing mechanism
for seeds of western conifers.

2. The shapes of seed dispersal curves are
very similar for most western conifers,
but species vary substantially in dis-
persal distances.

3. Of the tree, stand, and site character-
istics that largely determine the disper-
sal patterns of western conifers, several
stand out:

• seed aerodynamics

• tree height

• time of dispersal

• character of the seed-producing stand
(size and number of each species)

• annual variability of seed production

• physiographic position of seed source
as related to wind character.

A. Thermal winds are very important for
early season dispersal of seeds; pre-
vailing winds and storm fronts are very
important for late season dispersal.

5. Most seeds are dispersed through the air
in the fall, but seeds released in late
fall and early winter may also be blown
over crusted snow.

6. Birds and mammals are the primary disper-
sal agents for wingless seeds and can
carry them for substantial distances.

REFERENCES

Alexander, Robert R. ; Edminster, Carleton B.
Engelmann spruce seed dispersal in the
Central Rocky Mountains. Research Note
RM-A2A. Fort Collins, CO: U.S. Department
of Agriculture, Forest Service, Rocky
Mountain Forest and Range Experiment
Station; 1983. A p.

Alexander, Robert R. ; Shearer, Raymond C.J

Shepperd, Wayne D. Silvical characteristics
of subalpine fir. General Technical Report
RM-115. Fort Collins, CO: U.S. Department
of Agriculture, Forest Service, Rocky
Mountain Forest and Range Experiment
Station; 198A. 29 p.

Alexander, Robert R. ; Watkins, Ross K . ;
Edminster, Carleton B. Engelmann spruce
seed production on the Fraser Experimental
Forest, Colorado. Research Note RM-A19.
Fort Collins, CO: U.S. Department of
Agriculture, Forest Service, Rocky Mountain
Forest and Range Experiment Station; 1982.
6 p.

Barrett, James W. A record of ponderosa pine
seed flight. Research Note PNW-38. Portland,
OR: U.S. Department of Agriculture, Forest
Service, Pacific Northwest Forest and Range
Experiment Station; 1966. 5 p.

Benkman, Craig W. ; Balda, Russell P.; Smith,
Christopher C. Adaptations for seed disper-
sal and the compromises due to seed preda-
tion in limber pine. Ecology. 65(2):
632-6A2; 198A.

Boe, Kenneth H. Western larch and Douglas-fir
seed dispersal into clearcuttings . Research
Note INT-129. Ogden, UT: U.S. Department of
Agriculture, Forest Service, Intermountain
Forest and Range Experiment Station; 1953.
3 p.

Boe, Kenneth N. Regeneration and slash disper-
sal in lodgepole pine clear cuttings.
Northwest Science. 30(2): 1-11; 1956.

Burkhardt, Wayne J.; Tisdale, E. W. Nature and
successional status of western juniper
vegetation in Idaho. Journal of Range
Management. 22: 26A-270; 1969.

Cameron, H. Melting point of the bonding
material in lodgepole pine and jack pine
cones. Silvicultural Leaflet 83. Ottawa,
ON: Canadian Forestry Branch, Division of
Forestry Research; 1953. 3 p.

Clark, M. B. Seed production of hemlock and
cedar in the interior wet belt region of
British Columbia related to dispersal and
regeneration. Research Note No. 51.
Victoria, BC: British Columbia Forest
Service; 1970. 11 p.

Crossley, D. I. Survival of white spruce

reproduction resulting from various methods
of forest soil scarification. Technical
Note 10. Ottawa: Canadian Department of
Northern Affairs and Natural Resources,
Forestry Research Division; 1955. 9 p.

Curtis, James D. ; Foiles, Marvin W. Ponderosa
pine seed dissemination into group clear-
cuttings. Journal of Forestry. 59: 766-767;
1961.

Dahms, Walter G. ; Barrett, James W. Seed

production of central Oregon ponderosa and
lodgepole pine. Research Paper PNW-191.
Portland, OR: U.S. Department of Agricul-
ture, Forest Service, Pacific Northwest
Forest and Range Experiment Station; 1975.
13 P.

59

Dobbs, R. C. White spruce seed dispersal in
central British Columbia. Forestry
Chronicle. 52(5): 225-228; 1976.

Duncan, Donald P. A study of some of the

factors affecting the natural regeneration
of tamarack (Larix lariaina) in Minnesota.
Ecology. 35: 498-521; 1954.

Everett, Richard L. Telephone interview on
pinyon seed dispersal. Research Range
Scientist, Intermountain Research Station,
Forest Service, U.S. Department of Agricul-
ture, Reno, NV; June 1985.

Fiedler, Carl E. Atmospheric conditions

surrounding a seedfall in western Montana.
Missoula, MT: University of Montana; 1974.
46 p. Master of Science Thesis.

Fiedler, Carl E. Conifer establishment follow-
ing prescribed broadcast burning in larch-
fir forests — a research summary report.
Missoula, MT: University of Montana; 1976.
70 p.

Fowells, H. A.; Schubert, G. H. Seed crops of
forest trees in the pine region of
California. Technical Bulletin 1150.
Washington, DC: U.S. Department of Agricul-
ture; 1956. 48 p.

Franklin, Jerry F. ; Smith, Clark E. Seeding
habits of upper-slope tree species. II.
Dispersal of a mountain hemlock seedcrop on
a clearcut. Research Note PNW-214. Portland,
OR: U.S. Department of Agriculture, Forest
Service, Pacific Northwest Forest and Range
Experiment Station; 1974a. 9 p.

Franklin, Jerry F. ; Smith, Clark E. Seeding
habits of upper-slope tree species. III.
Dispersal of white and shasta red fir seeds
on a clearcut. Research Note PNW-215.
Portland, OR: U.S. Department of Agricul-
ture, Forest Service, Pacific Northwest
Forest and Range Experiment Station; 1974b.
9 P.

Frothingham, E. H. Douglas fir: a study of the
Pacific Coast and Rocky Mountain forms.
Circular 150. Washington, DC: U.S. Depart-
ment of Agriculture, Forest Service; 1909.
38 p.

Gashwiler, Jay S. Seed fall of three conifers
in west-central Oregon. Forest Science.
15(3): 290-295; 1969.

Gordon, Donald T. Natural regeneration of
white and red fir . . . influence of
several factors. Research Paper PSW-58.
Berkeley, CA: U.S. Department of Agricul-
ture, Forest Service, Pacific Southwest
Forest and Range Experiment Station; 1970.
32 p.

Haig, Irvine T.; Davis, Kenneth P.; Weidman,
Robert W. Natural regeneration in the
western white pine type. Technical Bulletin
767. Washington DC: U.S. Department of
Agriculture, Forest Service; 1941. 99 p.

Harlow, William M. ; Harrar, Ellwood S.; White,
Fred M. Textbook of Dendrology. 6th ed. New
York; McGraw-Hill; 1979. 510 p.

Harris, A. S. Ripening and dispersal of a
bumper western hemlock-sitka spruce seed
crop in southwest Alaska. Research Note
PNW-105. Portland, OR: U.S. Department of
Agriculture, Forest Service, Pacific
Northwest Forest and Range Experiment
Station; 1969. 11 p.

Isaac, L. A. Seed flight in the Douglas-fir
region. Journal of Forestry. 28: 492-499;

1930.

Lanner, Ronald M. ; Vander Wall, Stephen B.
Dispersal of limber pine seed by Clark's
nutcracker. Journal of Forestry. 78(10):
637-639; 1980.

LeBarron, Russel K. The role of forest fires
in the reproduction of black spruce.
Minnesota Academy of Science Proceedings.
7: 10-14; 1939.

LeBarron, Russel K. Silvicultural management
of black spruce in Minnesota. Circular 791.
Washington, DC: U.S. Department of Agricul-
ture; 1948. 60 p.

Lotan, James E. Regeneration of lodgepole
pine: a study of slash disposal and cone
opening. Research Note INT-16. Ogden, UT:
U.S. Department of Agriculture, Forest
Service, Intermountain Forest and Range
Experiment Station; 1964. 4 p.

Lotan, James E. ; Perry, David A. Ecology and
regeneration of lodgepole pine. Agriculture
Handbook 606. Washington, DC: U.S. Depart-
ment of Agriculture, Forest Service; 1983.
51 p.

McCaughey, Ward W. ; Schmidt, Wyman C. Seed
dispersal of Engelmann spruce in the
Intermountain West. U.S. Department of
Agriculture, Forest Service, Intermountain
Forest and Range Experiment Station,
Bozeman, MT. Manuscript in preparation.

Noble, Daniel L. ; Ronco, Frank Jr. Seedfall
and establishment of Engelmann spruce and
subalpine fir in clearcut openings in
Colorado. Research Paper RM-200. Fort
Collins, CO: U.S. Department of Agriculture,
Forest Service, Rocky Mountain Forest and
Range Experiment Station; 1978. 12 p.

60

Payandeh, B. ; Haavisto, V. F. Prediction

equations for black spruce seed production
and dispersal in northern Ontario. Forestry
Chronicle. 58:96-99; 1982.

Perry, David A.; Lotan, James E. Regeneration
on early growth strip clearcuts in the
lodgepole pine/bitterbrush habitat type.
Research Note INT-238. Ogden, UT: U.S.
Department of Agriculture, Forest Service,
Intermountain Forest and Range Experiment
Station; 1977a. 7 p.

Perry, David A.; Lotan, James E. Opening

temperatures in serotinous cones of lodge-
pole pine. Research Note INT-228. Ogden,
UT: U.S. Department of Agriculture, Forest
Service, Intermountain Forest and Range
Experiment Station; 1977b. 6 p.

Phillips, Frank J. A study of pinyon pine.
Botanical Gazette. 48: 216-223; 1909.

Phillips, Frank J. The dissemination of

junipers by birds. Forestry Quarterly. 8:
60-73; 1910.

Randies, Quincy. Pinyon-juniper in the South-
west. In: U.S. Department of Agriculture
Yearbook 1949. Washington, DC: U.S. Depart-
ment of Agriculture; 1949: 342-347.

Rasmussen, Irvin D. Biotic communities of
Kaibab Plateau, Arizona. Ecological Mono-
graphs. 11(3): 229-275; 1941.

Roe, Arthur L. Cutting practices in Montana
larch-Douglas-fir. Northwest Science. 29:
23-34; 1955.

Roe, Arthur L. Seed dispersal in a bumper
spruce seed year. Research Paper INT-39.
Ogden, UT: U.S. Department of Agriculture,
Forest Service, Intermountain Forest and
Range Experiment Station; 1967. 10 p.

Rowe, J. S. Factors influencing white spruce
reproduction in Manitoba and Saskatchewan.
Technical Note 3. Ottawa: Canadian Depart-
ment of Northern Affairs and Natural
Resources, Forestry Research Division;
1955. 27 p.

Schmidt, Wyman C; Shearer, Raymond C. ; Roe,
Arthur L. Ecology and silviculture of
western larch forests. Technical Bulletin
No. 1520. Washington, DC: U.S. Department
of Agriculture; 1976. 96 p.

Schmidt, Wyman C; Lotan, James E. Phenology
of common forest flora of the Northern
Rockies — 1928 to 1937. Research Paper
INT-259. Ogden, UT: U.S. Department of
Agriculture, Forest Service, Intermountain
Forest and Range Experiment Station; 1980.
20 p.

Sharpe, David M. ; Fields, David E. Integrating
the effects of climate and seedfall velo-
cities on seed dispersal by wind: a model
and application. Ecological Modeling. 17:
297-310; 1982.

Shearer, Raymond C. Regeneration establishment
in response to harvesting and residue
management in a western larch-Douglas-fir
forest. In: Environmental consequences of
timber harvesting in the Rocky Mountain
coniferous forests: symposium proceedings;
1979 September 11-13; Missoula, MT. General
Technical Report INT-90. Ogden, UT: U.S.
Department of Agriculture, Forest Service,
Intermountain Forest and Range Experiment
Station; 1980: 249-269.

Shearer, Raymond C. Western larch seed disper-
sal over clearcut blocks in northwestern
Montana. Montana Academy of Sciences
Proceedings. 19: 130-134; 1959.

Shearer, Raymond C. Establishment of conifers
following broadcast burning on clearcuts in
western Montana. Missoula, MT: University
of Montana; 1985. 115 p. PhD Dissertation.

Shepperd, Wayne D. Telephone interview on blue
spruce seed dispersal. Research Silvicul-
turist, Rocky Mountain Forest and Range
Experiment Station, Forest Service, U.S.
Department of Agriculture, Fort Collins,
CO; February 1985.

Siggins, H. W. Distribution and rate of fall
of conifer seeds. Journal of Agricultural
Research. 47: 119-128; 1933.

Squillace, A. E. Engelmann spruce seed disper-
sal into a clearcut area. Research Note
INT-11. Ogden, UT: U.S. Department of
Agriculture, Forest Service, Intermountain
Forest and Range Experiment Station; 1954.
4 p.

Toroback, D. F. Notes on cones and vertebrate-
mediated seed dispersal of Pinus albicaulis
(Pinaceae). Madrono. 28(2): 91-94. 1981.

Tomback, D. F. ; Kramer, K. A. Limber pine seed
harvest by Clark's Nutcracker in the Sierra
Nevada: timing and foraging behavior.
Condor. 82: 467-468. 1980.

U.S. Department of Agriculture, Forest Service.
Jack pine seed habits. Technical Note 149.
St. Paul, MN: U.S. Department of Agricul-
ture, Forest Service, Lake States Forest
Experiment Station; 1939. 1 p.

U.S. Department of Agriculture Forest Service.
Annual report. Ogden, UT: Intermountain
Forest and Range Experiment Station; 1940.

61

U.S. Department of Agriculture, Forest Service.
Silvics of forest trees of the United
States. Agriculture Handbook 271.
Washington, DC: U.S. Department of Agricul-
ture, Forest Service, 1965. 762 p.

U.S. Department of Agriculture, Forest Service.
Seeds of woody plants in the United States.
Agriculture Handbook. 450. Washington, DC:
U.S. Department of Agriculture, Forest
Service; 1974. 883 p.

U.S. Department of Agriculture, Forest Service.
Silvics of native and naturalized trees of
the United States and Puerto Rico. Agricul-
ture Handbook. Washington, DC: U.S. Depart-
ment of Agriculture, Forest Service; [In
press] .

Vander Wall, Stephen B.; Balda, Russell P.

Coadaptations of the Clark's nutcracker and
the piflon pine for efficient seed harvest
and dispersal. Ecological Monographs. 47:
89-111; 1977.

62

L
>s 1

CONE PRODUCTION ON DOUGLAS-FIR AND WESTERN LARCH IN MONTANA.

Raymond C. ^Shearer

ABSTRACT: This study determined the number of
cones that matured on Douglas-fir and western
larch growing on twenty 0.1-ha (0.25-acre) plots
from 1980 through 1983 at each of four locations
near Missoula, MT. A good cone crop in 1980 pro-
duced 80 percent of all cones counted during the
4-year study. In 1980, cones were produced on
45 percent of the Douglas-fir and 38 percent of
the larch in the 10- to 15-cm (4- to 6-inch)
diameter class while 85 percent were produced by
both species in the 30- to 36-cm (12- to 14-inch)
diameter class. The average number of cones per
tree in 1980 increased 10 times for Douglas-fir
and 27 times for larch as the diameter classes
increased from 10 to 15 cm (4 to 6 inches) to 30
to 36 cm (12 to 14 inches). In years of fair or
poor cone production the average number of cones
per tree was about seven times greater for
Douglas-fir and 15 times greater for western
larch in the 30- to 36-cm (12- to 14-inch) diame-
ter class than in the 10- to 15-cm (4- to 6-inch)
diameter class. More than half of the Douglas-
fir and larch in the 10- to 15-cm (4- to 6-inch)
diameter class failed to produce any cones during
the study; only 7 percent of the trees in the 30-
to 36-cm (12- to 14-inch) diameter class failed
to produce cones.

INTRODUCTION

Cone production in natural forest stands has been
studied at several locations in the Inland Moun-
tain West (Fowells and Schubert 1956; Franklin
and others 1974; Alexander and Noble 1976). Cone
production of conifers native to Montana was
studied by Boe (1954). He classified Rocky Moun-
tain Douglas-fir (Pseudotsuga menziesii var.
glauca [Beissn.] Franco) west of the Continental
Divide as a prolific seeder and western larch
(Larix occidentalis Nutt.) as a good seeder.

was to determine for Douglas-fir and larch at
four study sites in western Montana from 1980
through 1983: (1) the influence of insect lar-
vae, particularly western spruce budworm
(Chor istoneura occidentalis Freeman), on cone
and seed losses (Hedlin and others 1980) and (2)
the number of cones produced each year. Shearer
(1984) published information on the cone and
seed reduction caused by insect feeding. This
paper reports the number of cones that were pro-
duced on the Douglas-fir and larch growing with-
in each of 80 plots.

STUDY SITES

In 1980, four study sites were selected near
Missoula in western Montana. Two were in the
warm and dry Douglas-fir forest series and two
in the cooler and moister subalpine fir (Abies
lasiocarpa [Hook.] Nutt.) forest series (Pfister
and others 1977) :

Forest series

Douglas-fir

Distance and direction of
sites from Missoula, MT

Ashby Creek (Ashby)
29 km (18 miles) east

Blue Mountain Road (Blue)
8 km (5 miles) southwest

Subalpine fir West Fork Lolo Creek (Lolo)

35 km (22 miles) southwest

Spring Creek (Spring)

60 km (36 miles) northeast

The study sites are identified hereafter by the
names in parentheses.

Cone production either on individual trees or
within natural stands usually varies greatly by
year. Because Douglas-fir and larch cones de-
velop throughout the crowns, trees of greater
diameter and crown volume usually produce more
cones (Fowells 1965). The purpose of this study

Paper presented at the Conifer Tree Seed in the
Inland Mountain West Symposium, Missoula, MT,
August 5-6, 1985.

Raymond C. Shearer is a Research Silviculturist ,
Intermountain Research Station, Forest Service,
U.S. Department of Agriculture, Missoula, MT.

Ashby is a second-growth stand averaging 50
years old, composed of 35 percent Douglas-fir,
24 percent western larch, 20 percent Rocky
Mountain ponderosa pine ( Pinus ponderosa var.
scopulorum Engelm.), and 21 percent lodgepole
pine (P. contorta var. latifolia Engelm.).
Blue, a partially cut stand averaging 80 years
old, is composed of 67 percent Douglas-fir, 29
percent western larch, and 4 percent ponderosa
pine. Lolo, a partially cut stand averaging 70
years of age, is composed of 15 percent Douglas-
fir, 35 percent western larch, 48 percent
lodgepole pine, and 2 percent ponderosa pine,
Engelmann spruce (Picea engelmannii Parry ex
Engelm.), and subalpine fir. Spring is a more
open partially cut stand with trees ranging from
60 to 175 years of age. Composition of the

h3

cone-bearing trees is 30 percent Douglas-fir, 42
percent western larch, 13 percent subalpine fir,
8 percent lodgepole pine, 6 percent Engelmann
spruce, and 1 percent ponderosa pine.

At each of these four study sites, twenty 0.1-ha
(0.25-acre) circular plots were randomly estab-
lished. All trees 8 cm (3 inches) d.b.h. and
larger were numbered. Trees were grouped into
5-cm (2-inch) diameter classes for this paper:
10 cm [10.2-15.0] (4-inch) [4.0-5.9], 15 cm
[15.2-20.1] (6-inch) [6.0-7.9], for example.
Broader groupings were also used; for example,
10-15 cm (10.2-20.1 cm) [4-6 inch (4.0-7.9 inch)]
and 20-25 cm (20.2-30.2 cm) [8-10 inch (8.0-11.9
inch) ] .

The number of trees per acre was greatest at
Blue, followed by Lolo, Ashby, and Spring
(table 1). Each location had both Douglas-fir
and larch in the 10- through 36-cm (4- through
14-inch) diameter classes. Douglas-fir and west-
ern larch were most abundant in the 15-cm (6-
inch) diameter class at each location. Only
occasional larger diameter trees grew within the
plots .

Table 1. — Average number of Douglas-fir (PSME) ,
western larch (LA0C) , and other trees
per acre by diameter class, forest
series, and study area in western
Montana, 1980. Basis: twenty 0.1-ha
(0.25-acre) plots at each location

Diameter

class PSME LAOC Other PSME LA0C Other

Pseudotsuga menziesii Forest Series

Ashby Creek Blue Mountain

4

7.

4

3.8

0

.8

7.6

6.

4

6

19.

8

13.2

18

.6

50.8

20.

4

3

.0

8

10.

6

. . 8.4

14.0

40.6

16.

2

1.

.8

10

6.

2

5.0

10

.2

23.4

9.

2

1,

.4

12

3.

6

2.2

7

.4

6.6

4.

2

.6

14

8

1.0

4

.0

1.2

6

.2

16

4

1

.4

.2

2

.4

18

2

1

.0

20

.4

.2

Sum

49.

0

33.6

57

.8

130.4

57.

2

7

.6

Abies .

lasiocarp

a '.

Forest Series

Lolo Creek

Spring

Creek

4

1.

0

1.0

1

.4

7.2

3.

2

6

.6

6

11.

4

28.0

35

.2

7.6

15.

4

7

.8

8

5.

0

17.6

27

.4

3.8

9.

8

3

.6

10

3.

4

6.4

13

.8

3.6

1.

6

3

.6

12

2.

,6

3.4

4

.8

.6

3.

.2

.6

14

4

1.0

.2

.6

.4

.2

16

,2

.6

18

2

.6

22

.4

30

.4

Sum

24.

.2

58.8

82

.8

24.0

33.

.6

22

.4

The Blue study area had 1.1 to 2.8 times more
total basal area than the other three locations;
Spring had only 0.4 as much as the other sites
(table 2). Blue had from 2.7 to 6.0 times more
Douglas-fir basal area and from 0.9 to 1.8 times
more larch than the other stands.

Table 2. — Basal area (ft2/acre) of Douglas-fir

(PSME), western larch (LAOC), and other
trees by forest series and study area
in western Montana, 1980. Basis:
twenty 0.1-ha (0.25-acre) plots at
each location

Forest series Plot PSME LAOC Other Total

Douglas-fir

Ashby

20.

.1

13.

9

34.

.1

68.

.1

Blue

54,

,7

24.

,2

4,

,4

83.

.2

Subalpine fir

Lolo

11.

.0

27.

8

34.

.5

73.

.3

Spring

9.

. 1

13.

1

7,

.5

29.

,7

CONE PRODUCTION ESTIMATES

Douglas-fir and western larch cone production
was estimated in 1980, 1981, 1982, and 1983 on
the 80 plots by counting the fully elongated
cones between late July and early September. No
attempt was made to judge the amount of cone
mortality from bud burst to the time of examina-
tion. Cone counts on trees of other coniferous
species growing on the plots were not made.

In 1980, counts were made from a truck-mounted
hydraulic bucket. In 1981, 1982, and 1983,
cones were counted on each tree from points
where the full length of the tree could be
scanned with binoculars. Trees that produced no
or few cones (usually less than 10 cones) , were
carefully examined to be sure no cones were
missed. This meant observing these trees from
two or more locations so the entire crown could
be scanned. Trees with greater numbers of cones
were examined from only one location.

RESULTS

The average total number of mature cones per
tree counted from 1980 through 1983 varied by
species and by study location (table 3) .
Douglas-fir averaged greater total 4-year cone
production than western larch at Ashby, Lolo,
and Spring. These numbers were substantially
lower than the potential because of insect-
caused mortality in the early development of the
conelets (Shearer 1984) . Cone production of
both species was greatly influenced by year and
diameter class.

64

Table 3. — Average number of Douglas-fir and western larch cones per tree and number of trees sampled per
year by study area, western Montana

Douglas-fir

Western larch

Average and

Average and

Year

Study area

standard deviation

T rees

standard deviation

Trees

—————— Number — —

Numbe r

1980

Ashby

1 1 ft _i_ 1 1 o

219 ± 318

7 ft £

206

147 + 274

1 7 /

1 34

Blue

46 i 141

497

107 ± 238

192

Lolo

187 + 300

61

74 i 271

132

Spring

7 Q 7 * /.TO

Do

1 CQ + 7 /. ft

1 1 J

1981

Ashby

< 1 + < 1

7 / 1

24 1

7 .4. 1 C

1 z 16

16 /

Blue

0 t 0

600

1 ± 2

275

Lolo

<1 ± 3

120

3 ± 12

291

Spring

< 1 ± < 1

1 AT

1U/

1 <4> IT
J - 11

lb J

1982

Ashby

8 ± 23

Zw

9 ± 19

1 C ~i

lo /

DJ.UG

f\L * QQ

1 1 + ^9

J 1 X D L

7 7 1

Lolo

101 i 82

82

19 ! 51

265

Spring

20 ♦ 119

119

3+11

168

1983

Ashby

<1 t 1

144

21 ± 47

115

Blue

<1 i <1

621

10 ± 29

283

Lolo

<1 t <1

64

3 ± 10

206

Spring

9 ± 25

119

10 ± 35

167

Production by Year

About 80 percent of all Douglas-fir and western
larch cones counted during the 4 years of this
study were produced in 1980. Twenty percent of
the Douglas-fir cones were counted in 1982, and
less than 1 percent in 1981 and 1983. Two, 11,
and 7 percent of the larch cones were counted in
1981, 1982, and 1983.

The average number of cones per tree also varied
each year by location (table 3). Greatest aver-
age cone maturity for the 4-year period occurred
at Spring, where average basal area and crown
competition were least. Lowest cone production
occurred at Blue, where the highest basal area
was measured. Although most of the Douglas-fir
cones were produced in 1980, more cones matured
at Blue in 1982 than in 1980 (table 3). This
difference would have been even greater if
insect-caused mortality of conelets in 1982 was
kept at the same level as in 1980 (Shearer 1984).
The low Douglas-fir cone counts in 1983 at Ashby
and Blue resulted from high conelet mortality
caused by insects.

Production by Diameter Class

More Douglas-fir and larch cones were produced on
larger diameter trees, probably because of great-
er crown volume. In 1980, cone production at
least tripled in each larger diameter class
(table 4). This trend was similar at all loca-
tions, although the average number of cones
varied from lower counts at Blue to higher counts
at Spring. The number of Douglas-fir cones on

each tree ranged from 0 to 698 in the 10- to
15-cm (4- to 6-inch) (small) diameter class, 44
to 1,065 in the 20- to 25-cm (8- to 10-inch)
(mid) diameter class, and 81 to 1,482 in the 30-
to 36-cm (12- to 14-inch) (large) diameter
class. The number of western larch cones on
each tree ranged from 0 to 750 within the small
diameter class, 0 to 1,200 within the mid diame-
ter class, and 0 to 2,450 within the large diam-
eter class.

Cone production in 1981, 1982, and 1983 was much
lower than in 1980. Nevertheless, the average
number of Douglas-fir and western larch cones
per tree increased in each larger diameter class
except in 1981 when so few Douglas-fir cones
were produced that the trend was not evident
(table 4). One Douglas-fir at Lolo had 30
cones — the greatest number found in the four
study areas. Most western larch also failed to
produce cones in 1981 and the maximum number of
cones was 85, 130, and 55 in the small, mid, and
large diameter classes.

The number of Douglas-fir cones in 1982 ranged
from 0 to 360 within the small, 0 to 550 within
the mid, and 0 to 400 within the large diameter
classes; larch cones ranged from 0 to 150 within
the small, 0 to 250 within the mid, and 0 to 300
within the large diameter classes.

In 1983, cone production on Douglas-fir ranged
from 0 to 66 within the small, 0 to 104 within
the mid, and 0 to 176 within the large diameter
classes. Larch cone production in 1983 ranged
from 0 to 57 within the small, 0 to 117 within

65

Table 4. — Average number of Douglas-fir and western larch cones per tree and number of trees sampled
each year by diameter class for four locations, western Montana

Douglas-fir Western larch

Average and Average and

Year Diameter class standard deviation Trees standard deviation Trees

Inch ------ Number ------ ------ Number ------

1980

4- 6

41 ± 114

377

21 +

79

261

8-10

132 ± 232

366

116 ±

210

247

12-14

434 + 519

63

575 ±

536

55

1981

4- 6

<1 ± 1

509

1 ±

5

444

8-10

<1 ± <1

464

4 ±

13

361

12-14

0 ± 0

77

6 ±

12

78

1982

4- 6

17 ± 49

490

5 ±

16

434

8-10

72 ± 99

450

• 26 ±

47

348

12-14

122 ± 139

72

48 ±

78

76

1983

4- 6

<1 ± 4

434

1 ±

5

376

8-10

1 ± 7

426

9 ±

17

319

12-14

6 ± 26

70

56 ±

82

69

the mid, and 0 to 272 within the large diameter
classes .

The percentage of trees that produced cones
increased with size of cone crop:

Size of

cone crop Douglas-fir Western larch

1 (largest)

2

49 (1980)
43 (1982)

48 (1980)
24 (1983)

Douglas-fir cone crop of 1982 matured cones on
22, 60, and 75 percent in the small, mid, and
large diameter classes. The larch cone crops of
1982 and 1983 produced cones on an average of
12, 27, and 58 percent in the small, mid, and
large diameter classes.

Table 5. — Percent of Douglas-fir and western
larch in four western Montana study
areas that had 0, 1-99, and more than
99 cones per tree each year by
diameter class

3 2 (1983) 22 (1982)

4 (smallest) <1 (1981) 13 (1981)

In 1981, of 1,050 Douglas-fir trees examined,
only seven produced cones. In 1983, 2, <1, and 3
percent of the trees at Ashby, Blue, and Lolo
produced cones. At Spring, however, 24 percent
of the trees produced cones. The 1982 and 1983
larch cone crops had about half as many trees
produce cones as in 1980. As more trees produced
cones, the average number of cones per tree also
increased .

The percentage of trees that had cones increased
in the larger diameter classes. An average of
only 13 percent of the larch in 1981 produced
cones; in the small, mid, and large diameter
classes, 5, 18, and 32 percent of the trees
yielded cones (table 5). Cones were on 49 per-
cent of the Douglas-fir and 48 percent of the
larch in 1980. In Douglas-fir, 45, 60, and 86
percent of the small, mid, and large diameter
classes produced cones (table 5). In western
larch 38, 50, and 84 percent of the small, mid,
and large diameter classes had cones. The

Diameter

Dougl

as-

fir

Western

larch

Year

class

0 1

-99

>99

0

1-99

>99

1980

4- 6

55

33

12

62

35

3

8-10

40

29

31

50

32

18

12-14

14

22

64

16

14

70

1981

4- 6

100

0

0

95

5

0

8-10

>99

<1

0

81

18

1

12-14

100

0

0

68

32

0

1982

4- 6

78

16

6

84

15

1

8-10

40

29

31

78

14

8

12-14

25

35

40

48

29

23

1983

4- 6

98

2

0

91

9

0

8-10

98

2

0

68

30

2

12-14

95

2

3

36

42

22

The frequency of cone production varies by
species (table 6). During these 4 years, larch
had cones more frequently than Douglas-fir.
Although 34 percent of the trees of both species
failed to have any cones, 26 percent of the
larch and only 1 percent of the Douglas-fir had

66

cones in 3 or in all 4 years. The remaining 40
percent of the larch and 65 percent of the
Douglas-fir produced cones 1 or 2 of the 4 years.

Table 6. — Percent of Douglas-fir and western
larch trees in four western Montana
study areas that had one or more cones
during the 4 years of the study by
species and diameter class

Number of years
Diameter cones produced

Species

class

0

1

2

3

4

Douglas-fir

4-

• 6

51

38

11

0

0

8-

•10

24

35

40

1

0

12-

•14

7

25

65

3

0

Western larch

4-

• 6

54

27

11

7

1

8-

■10

22

17

24

27

10

12-

•14

6

4

36

33

21

As the diameter increased, the frequency of cone
production usually increased (table 6). For
example, 11, 41, and 68 percent of the Douglas-
fir and 19, 61, and 90 percent of the larch of
the small, mid, and large diameter classes had
cones 2, 3, or 4 years of the study. More than
half of the Douglas-fir and larch trees in the
small diameter class failed to produce cones any
of the 4 years; less than one-fourth and one-
tenth of the mid and large diameter classes
failed to produce cones in any of the 4 years.

MANAGEMENT IMPLICATIONS

Although Douglas-fir produced more cones per tree
than did western larch during the 4 years of this
study, larch produced some cones every year;
Douglas-fir nearly failed to produce cones in 2
of the years. Where seed production is an impor-
tant consideration, larger diameter full-crown
dominant and codominant Douglas-fir and western
larch should be reserved because larger trees
produce more cones more frequently than smaller
diameter trees. The number of mature Douglas-fir
cones tripled from the small to the mid diameter
class and tripled again from the mid to the large
diameter class. Western larch cones increased
fivefold through each of the diameter class
comparisons .

ACKNOWLEDGMENT

REFERENCES

Alexander, R. R. ; Noble, D. L. Production of
Engelmann spruce seed, Fraser Experimental
Forest, Colorado: a 5-year progress report.
Res. Note RM-324. Fort Collins, CO: U.S.
Department of Agriculture, Forest Service,
Rocky Mountain Forest and Range Experiment
Station; 1976. 4 p.

Boe, Kenneth N. Periodicity of cone crops for
five Montana conifers. Mont. Acad. Sci. Proc.
14: 5-9; 1954.

Fowells, H. A., comp. Silvics of forest trees of
the United States. Agric. Handb. 271.
Washington, DC: U.S. Department of
Agriculture; 1965. 762 p.

Fowells, H. A.; Schubert, G. H. Seed crops of
forest trees in the pine region of
California. Tech. Bull. 1150. Washington, DC:
U.S. Department of Agriculture, Forest
Service; 1956. 48 p.

Franklin, Jerry F. ; Carkin, Richard; Booth,
Jack. Seeding habits of upper-slope tree
species. I. A 12-year record of cone produc-
tion. Res. Note PNW-213. Portland, OR: U.S.
Department of Agriculture, Forest Service,
Pacific Northwest Forest and Range Experiment
Station; 1974. 12 p.

Hedlin, A. F.; Yates, H. 0., Ill; Tovar, D. C.j
Ebel, B. H.; Koerber, T. W. ; Merkel, E. P.
Cone and seed insects of North American coni-
fers. [Ottawa, ON] : Can. Forest Service, USDA
Forest Service, Secretaria de Agricultura y
Recursos Hidraulicos, Mexico; 1980. 122 p.

Pfister, Robert D.; Kovalchik, Bernard L. ; Arno,
Stephen F. ; Presby, Richard C. Forest habitat
types of Montana. Gen. Tech. Rep. INT-34.
Ogden, UT: U.S. Department of Agriculture,
Forest Service, Intermountain Forest and
Range Experiment Station; 1977. 174 p.

Shearer, Raymond C. Influence of insects on
Douglas-fir, Pseudotsuga menziesii (Mirb.)
Franco, and western larch, Larix occidentalis
Nutt., cone and seed production in western
Montana. In: Yates, Harry 0., Ill, comp. and
ed. Proceedings of the cone and seed insects
working party conference; 1983 July 31-
August 6; Athens, GA. Asheville, NC: [U.S.
Department of Agriculture, Forest Service],
Southeastern Forest Experiment Station; 1984:
112-121.

Research leading to this paper was funded in part
by a USDA Forest Service-sponsored program en-
titled Canada-United States Spruce Budworms
Program (CANUSA) .

67

'7) CONE PRODUCTION IN^PINUS ALB I CAUL IS FORESTS^
T. ^Weaver and F.jjorcella

ABSTRACT: Whitebark pine cone production was
estimated for a 6 to 8 year period in each of 29
stands widespread in the northern Rocky Mountains.
1) One-time sampling was possible since the
estimate was m|de by multiplying the number of
branches per m by an estimate of annual cone
production made from counts of conelets, mature
cones, or cone scars on successively older annual
increments of those branches. 2) Average c,one^
production ranged from 0.3 to 3.6 cones 'm *yr
and from 22-270 seeds "m-^ ' year--'- . 3) Regression
analysis was used to relate the variance observed
to time and place, a) Year-to-year variation in
the cone yield of branches, trees, and stands in a
region appears to be both internally and
externally controlled. Internal control is
suggested by the fact that good cone years were
usually preceded by poor cone years. While
external control is indicated by significant
correlations between growth and weather
conditions, control is not dominated by the effect
of any one factor or any particular developmental
stage, b) Although cone production of the average
branch varied significantly within 30 percent of
the trees and within 48 percent of the stands
observed, it did not vary significantly among
stands. c) Regressions relating stand cone
production to easily measured stand
characteristics such as canopy cover, fallen
cones, and/or stand size explain no more than
50 percent of the variance among stands.

INTRODUCTION

Pinus albicaulis (whitebark pine) is a dominant or
codominant tree in many high-altitude forests of
western North America (Weaver and Dale 1974; Arno
1986). Its large, well-provisioned seeds are
important not only for their reproductive
function, but also as food for man, bears,
squirrels, and nutcrackers (Blankenship 1905;
Forcella and Weaver 1980; Kendall 1980b; Hutchins
and Lanner 1982; Tombach 1983). Managers
considering either of these functions might ask:
What is average production? How does it vary
between stands? How does it vary between years?
And how might I estimate these variables in an
unstudied stand?

Paper presented at the Conifer Tree Seed in the
Inland Mountain West Symposium, Missoula, MT ,
August 5-6, 1985.

In 1976 T. Weaver and F. Forcella were Professor
of Botany and Graduate Research Associate, Biology
Department, Montana State University, Bozeman.
F. Forcella is now Research Agronomist, USDA
Agricultural Research Service, Morris, MN.

This paper (1) demonstrates a method for
estimating a stand's cone production over a 6 to
8 year period after a single sampling,

(2) reports estimates made in 29 stands
widespread in the northern Rocky Mountains, and

(3) uses that data set both to determine the
relationship of yield to easily measured stand
characteristics and to determine the relationship
of yield to regional weather conditions.

METHODS

Cone production. — Cone production in the 1969
through 1976 period was estimated for 29 P.
albicaulis stands from the Montana-Idaho border
[Bitterroot (5 stands) and Salmon River (2)
Mountains], from southwest Montana [Little Belt
(1), Big Belt (3), Castle (3), Elkhorn (2),
Pioneer (1), Tobacco Root (2), Madison (3), and
Bridger Mountains (1)] and from western Wyoming
[Absaroka (3) , and Wind River Mountains (3) ] .
More exact locations are given by Forcella
(1977) .

The measurements necessary for these estimates
were made as follows:

1. Current, future, and past production were
determined for 25 representative shoots, that is
five shoots each from the tops of five
representative trees.

2

2. The number of shoots per m in the canopy was
determined by counting che potential cone bearing
leaders on each of the five trees and dividing by
a canopy area calculated from the tree's greatest
and least canopy diameters.

3. Tree cover in the study area was determined
as a percent of 90 overhead points covered by the
canopy; the points were examined through a
vertical periscope held at meter intervals along
three 30 meter lines placed parallel to the slope
at three representative sites in each stand
studied .

Though our method has the advantage of allowing
one to sample production over a series of years
simultaneously, the following problems should be
recognized :

1. The trees were not sampled randomly because
some trees could not be climbed.

2. Shoots could not be sampled randomly because
some were in inaccessible parts of the crown.

68

3. Heavy bearing shoots may have been absent or
undersatnpled due to animal predation. Trees
raided by black bears (Ursus amer icanus) ,
recognizable by claw marks and broken branches,
were therefore not sampled. Less easily avoided
were the effects of red squirrels (Tamiasciurus
hudsonicus) who commonly cut leader shoots from
the trees (Schmidt and Shearer 1971).

4. On the other hand, shoots bearing cones at the
time of sampling are also apparent to man and are
likely to be over sampled; this may compensate for
underestimation due to predation.

5. Estimates of the coming cone crop will be high
if significant numbers of immature cones abort
(compare with Allen 1941, Finnis 1953, and Sarvas
1962); cone numbers predicted for 1977 were not
significantly higher than the actual crop — as they
would have been if abortion had been significant.

A second measure of cone production was made by
counting cones which fell to the ground in three
6.67 x 30 M plots in each stand. Estimates of the
1973 cone crop were made in 14 stands in 1974,
estimates of the 1974 cone crop were made in 28
stands in 1975, and numbers of old weathered cones
were recorded in both years.

Relating production to environment . — A nested
analysis of variance was used to evaluate the
effects of place and time on cone production. The
analysis detailed below determined the
significance (p<0.05) of variation between 5
branches in a tree, 5 trees in a stand, 6 to 16
stands in a f loristically defined subassociation
and 29 stands in the entire Pinus albicaulis-
Vacc inium scoparium association. Annual variation
in production was simultaneously tested as an
interaction in each nest of the analysis. The
three floristic regions included: t. the Wind
River and Absaroka Mountains; 2. central and
southwestern Montana; 3. the Bitterroot and Salmon
River Mountains (compare with Forcella 1977).

The significant effects of time detected in the
preceding analysis were tentatively attributed to
weather. To clarify the relationship between cone
yield and weather we regressed deviations of
normalized cone yields against deviations of
weather variables (monthly mean temperature and
monthly total precipitation) from their means in
each of the 46 months preceding abscission
(compare with Eis 1976). Cone yields were
normalized to eliminate the effect of large
differences in average yield between stands.
Regional average weather data were used (USDC
1966-1976, compare with Lowry 1966) because no
data are available from Pinus albicaulis stands,
because deviations in precipitation or temperature
from normal at low altitude usually parallel those
found at high altitude, and because regional
averages should eliminate weather station
peculiarities like frost pockets. Each regional
weather datum summarizes the data from dozens of
official weather stations.

Relating production to simple stand parameters. —
Simple or multiple regressions of cone production
(leader number x average leader yield, table 1)
against canopy cover, tree basal area, fallen cone
number, and stand size explained significant
amounts of the observed variation. On the other
hand, regressions of cone yield against stand age,
stand area, stand elevation, stand slope-aspect,
total productivity, total standing crop, and cover
indicator plants either alone or in combination
were never significant.

RESULTS AND DISCUSSION

Cone development. — Cone history is reviewed here
as a basis for understanding our method, for
relating seed production to climatic factors, and
as partial explanation for within-stand tree
distribution. The development of Pinus albicaul is
cones occupies parts of three summers and, due to
apical branch growth, evidence of progressively
older cones is displayed at nodes increasingly
removed from the branch tip.

1. Late in the first summer a bud develops at the
branch apex.

2. Pistillate conelets emerge from these buds in
the second summer about the time of snowmelt
(early July); they are found in the first lateral
branch whorl, are purple, are about 5 mm long, and
are easily recognized. July is therefore the best
month to analyze shoots to predict the two
forthcoming cone crops (compare with Allen 1941).
Staminate strobili mature in early July and
pollination occurs at this time. By the end of
the growing season (mid-September) the conelets
are approximately 1.5 cm long by 0.9 cm in
diameter.

3. Rapid growth begins again in July of the third
summer and full size (6.3 x 4.7 cm) cones appear
in the second branch whorl by mid-August. Cone
maturation continues until mid-September or early
October when abscission normally occurs.
Undisturbed cones abscise and fall to the ground
intact, but if Clark's nutcracker (N'uc i f raga
columbiana) removes the cone scales before
abscission, the cone axes may remain in lower
whorls of the branch for up to 10 years (compare
with Smechkin 1963) . Perhaps abscission depends
on a hormone produced in the cone scales shortly
before abscission so that if the cones are
destroyed prior to its synthesis abscission does
not occur.

The cones exude and are coated with a viscous
aromatic resin. The resin may reduce seed
predation by squirrels (Smith 1970) and
nutcrackers. It also bonds the cone scales
together (Clements 1910) until the resin
crystallizes or is consumed by fungi. When the
cone falls from the tree, cone scales and the
large wingless seeds fall away from the axis;
small clusters of seedlings found on the forest
floor may result either from in situ
disarticulation or from animal caches (compare
with Hutchins and Lanner 1982) .

69

Table 1. — Cone crops, locations, and stand characteristics for 28 Pinus albicaulis stands

STD1 MTN2 STATE1 CLIM4 COVER5 SIZE6 AGE7 1976 1975 1974 1973 1972 1971 1970 1969 MEAN+SE

16

WR

WY

WR

63

2

198

2. 7

2 .

5

0.3

2.5

1 . 0

2.5

1 . 9+0 . 4

17

WR

WY

WR

63

1

161

2.0

3.

8

1 . 6

1 . 6

1 . 4

1 . 1

1 . 0+0.4

18

WR

WY

WR

71

1

214

1 . 3

1 .

3

0. 7

0.8

0. 6

1 . 5

1 .0+0.2

19

AB

WY

WR

63

1

131

0.7

1 .

7

0. 2

0.9

0.9

0.7

0.9+0.2

20

AB

MT

SC

49

1

210

1 . 2

1 .

9

1 . 7

0. 7

0. 6

1.3

1 . 2+0 . 2

09

AB

MT

SC

54

2

307

0.9

1.

4

1.6

1.2

1.5

0.5

0.9

1 . 1+0. 2

07

BB

MT

CN

61

2

120

0. 0

0.

1

9 . 0

3 . 1

3. 7

3. 9

1 . 7

3.5

3. 1+1. 0

06

BB

MT

CN

51

3

364

0.5

0.

4

2 . 7

1 . 6

1 . 8

2.5

0.8

1 . 7

1 . 5+0 . 3

f\ o

Oo

BB

MT

CN

r r
00

3

55

1 . /

0.

4

2. 6

0. 2

0. 2

1 . 5

0.2

1 . 9

1 . 1+0 . 3

f\ o

UJ

LB

MT

CN

00

5

1 Jz

U. 2.

0.

0

3.8

1 o

1 . 8

0. 5

1 . 3

1 . 3+0. 6

05

CA

MT

CN

"7 r\
/0

1

330

0. 9

0.

0

1 . 8

0. 6

0. 9

0. 8

0 . 8+0 . 2

05

CA

MT

CN

10

1

330

0.9

0.

0

1 . 8

0.6

0.9

0.8

0.8+0 . 2

04

CA

MT

CN

1 "7

37

2

55

1 . 5

0.

3

1 . 2

0.7

0. 2

0.6

0 . 8+0 . 2

10

MD

MT

SW

52

1

205

0.6

0.

8

0.7

0. 2

0.6

0.9

0. 2

0 . 6+0 . 1

01

MD

MT

SW

73

2

150

0. 6

3 .

8

1 . 2

2. 2

0. 9

1.8-

0. 9

1 . 6+0 . 5

02

MD

MT

SW

97

2

240

1 . 1

4 .

9

5.4

4.6

3.1

4.6

1 . 2

3.6+0. 7

12

EH

MT

SW

73

2

135

1 . 5

6.

1

1 . 9

2.2

2.0

3.9

1 . 4

4.2

2 . 9+0 . 6

11

EH

MT

SW

40

3

53

0.3

2 .

1

0.6

0.5

0. 5

1 . 3

0.3

1 . 3

0.9+0. 2

14

TR

MT

SW

56

2

160

1 . 5

3 .

6

3.3

0. 5

2 . 7

3 . 8

0. 5

2 . 3+0 . 5

13

TR

MT

SW

47

2

643

0.3

0.

2

0.4

0. 3

1 . 3

0. 6

0.2

0. 5+0. 1

15

BR

MT

SW

53

2

320

1.0

0.

0

1.2

0.0

0.1

0.1

0.4+0.2

29

PI

MT

SW

69

3

182

1.1

2.

8

8.4

2.8

6.2

3.3

0.9

3.6

3.6+0.9

22

BT

MT

WE

50

2

423

2.5

1.

1

44.9

2.3

2.4

2.7

1.2

1.0

2.3+0.4

23

BT

MT

WE

18

2

29

0.5

0.

8

0.5

0.8

0.8

0.3

0.1

0.1

0.5+0.1

24

BT

MT

WE

13

1

32

0.7

0.

2

0.3

0.3

0.3

0.3

0.0

0.1

0.3+0.1

25

BT

MT

WE

22

2

115

0.7

3.

7

0.8

3.8

1.7

1.1

0.7

0.4

1.6+0.5

26

BT

ID

NV

32

2

59

0.1

2.

3

0.3

1.0

1.2

0.6

0.2

0.4

0.8+0.3

27

SR

ID

NV

30

2

94

1.1

2.

6

0.8

1.9

1.1

0.7

0.3

0.5

1.1+0.3

28

SR

ID

NV

31

2

188

0.1

0.

9

0.4

0.6

0.4

0.2

0.1

0.2

0.4+0.1

Stand number.

Mountain ranges are: Absaroka(AB), Big Belts(BB), Bitterroot(BT), Bridger(BR),
Castle(CA), Elkhorn(EH), Little Belts(LB), Madison (MD), Pioneer(PI), Salmon River(SR),
Tabacco River(TR), and Wind River(WR).
States are: Idaho(ID), Montana(MT), and Wyoming(WY) .

Climate regions are: Central(CN), Northeastern Valley(NV) of Idaho, South Central(SC),
Southwest(SW), Western(WE) Montana, and Wind River(WR) Wyoming.
Tree canopy cover (percent).

Stand size where 1=0 to 0.75 ha, 2=0.75 to 1.25 ha, and 3=1.25 ha +.
Age of dominant trees.
Cones per square meter.

Cone-and-seed-production. — Average stand co,ne _^
production ranged from 0.3 to 3.6 coijies'm 'yr
and averaged 1.4 ± 0.2 cones'm 'yr (table 1).
Among the 6 to 8 years studied, maximum
production in different stands ranged from 2.1 to
over 100 times minimum production; in the median
stand maximum production was 6.5 times minimum
production (table 1). The average coefficient of
variation (standard deviation/mean) between years
in the average stand was 0.7.

The average mature pistillate cone weighed 23 ± 7
grams (range 10 to 50 gm) with cones produced in
good years tending to be larger than those
produced in poor years. Average cone production
ranged, then, from a minimum of 7_£o an^average
of 32 and to a maximum of 83 gm'm 'yr on a
ground area basis. The range observed across all

years and stands was 0 to 193 gm'm 'yr , Cone
production was approximately 9 percent of total
above- and belowground production in 14 stands in
which total production was measured (Forcella and
Weaver 1977) .

The numbers of cones found on the ground were
usually fewer than scars counted in tree tops.
This relationship is demonstrated by the
regression equation:

C = 0.305 + 1.314 F

where C = the number of 1974 cone scars counted in
the treetop, F = the number cjf fresh cones counted
on the ground in 19 75, and r =0.51. Smechkin
(1963) compared similar methods in a Pinus

2

sibirica forest with the same result (r =0.62).

70

Table 2. — Comparison of cone and seed crops of several Plnus forest types

Forest Type Average Cones. m~z"yr Average Seeds'm- "yr~l Reference

t

grams

//

grams

y.4

r

Pinus contorta

4.

7

28

268

0.6

2

Smith 1968, 1970

1 -

contorta—

Purshla trldentata

4.8

24

120

0.4

2

Lotan 1967

D
1 .

con t or t a—

Geranium fremontli

Q

o •

n
u

40

13

-

Molr 1972

P .

contorta-

Vacclnlum royrtlllus

Q

0 .

U

40

5

-

Molr 1972

D

r •

cortorta bolanderl

52

20

Westman and

Whlttaker 1975

r •

s l Di r ica

—

100

30

■

Forraosof 1933

r •

s 1 b 1 r i c a

Vacclnlum myrtlllus

0.

1-

■2

4

—

—

—

™

Bolchenko 1970

p
1 .

sy 1 ve s t r 1 s—

Calluna vulgaris

1.

5

9

—

30

0.2

2

Sarvas 1962

p.

sylvestrls-

Vacclnlum myrtlllus

3.

0

18

—

60

0.4

2

Sarvas 1962

p.

sylvestris-

Oxalls acetosella

4.

5

27

—

90

0.6

2

Sarvas 1962

p.

monophylla-

Junlperus osteosperma

0.

1-

■1

8

2-34

1-26

1-9

28

Forcella unpubl.

edul is-

Junlperus osteosperma

0.

8

8

Forcella unpubl .

cerobroides-

Junlperus deppeanna

7.

3

17

35

6

35

Forcella unpubl.

p.

alblcaulls-

Vacclnlum scoparlum

0.

3-

3

6

6-84

10

20-250

2-25

30

Forcella and

Weaver 1977

Masses of seeds and cones not provided by the author were taken from Schopmeyer (1974).
Ranges in figures represent smallest and largest data provided.

Percent total productivity except third and fourth forest types which are percent aboveground
productivity only.

Percent total cone mass.

We attribute the deficiency of cones found on the
ground (about 25 percent) to Clark's nutcracker,
squirrel, and bear activitv.

Typical cones contain about 75 * 28 seeds, each
weighing 0.1 ± 0.02 gm dry. Seed live weights
are about 0.17 gm (Schopmeyer 1974). Seed mass
usually comprises 30 percent of cone mass and may
comprise 50 percent in an especially good cone
year; the proportion of cone mass devoted to
seeds in other conifers is usually lower (Smith
1970 and table 2). Average seed production
ranged, then, from a minimum of £.3 to an average
of 10.5 to a maximum of 27 gm'tj "vr , i.e.,
from 23 to 105 to 270 seeds'm "yr . The range
observed across all vears and s^tand^ was 0-63
gm'm 'yr and 0-630 seeds'm *yr . Seed
production was about 3 percent of total arboreal
production in 14 stands studied intensively
(Forcella and Weaver 1977a). Of the
approximately 75 scales on a typical cone, one
third, mostly apical and basal scales, were
inf ert ile .

Comparison of Pinus albicaulis forests with other
pine forests (table 2), leads to the following
conclusions: P inus albicaul is forests produce

T2 -i

normal cone crops (37 gm'm *yr average) .
Because the cones are heavy, ccjne numbers are
relatively small (1.4 cones'ra "yr average).
Since a large proportion of the cone is devoted to
seed , 2seed ^production is relatively high (10
gm'm 'yr average), yet because the ^eeds^are
large, seed numbers are normal (105 m 'yr
average). The net effect is to deposit normal
numbers of abnormally well provisioned seeds.

Variation in cone production potential. — Total
cone production is the product of mean shoot
production (=cone production potential) multiplied
by the number of fertile shoots per hectare. Yet,
since the density of fertile shoots in a stand may
be less than optimal, the yield of the average
fertile shoot may be a better index of site
production potential than is the total number of

71

cones actually produced in the stand. In the
following discussion we therefore compare
production potentials across vegetational units of
increasing size (trees, stands, sub-association,
and the entire association) by considering a
sample of branches of fixed size and ignoring the
actual number of shoots producing cones.

Cone crops varied significantly between branches
in 29 percent of the trees studied, and between
trees in 48 percent of the stands studied, but
they did not vary significantly between stands in
any region or between stands in all the different
regions. One might conclude:

1. that cone production varies little between
branches of a tree due to identical genetics and
similar mesoenvironmental conditions,

2. that it varies more between trees in a stand
due to greater dissimilarity in genetics and
mesoenvironmental conditions, and

3. that it differs relatively little between
stands in a region due to averaging of
between-tree heterogeneity in forests with
relatively constant genetic and mesoenvironmental
conditions .

The fact that fertile branches have similar cone
production potentials throughout a region or even
throughout an association supports our
expectation that the average cone production of a
stand is determined primarily by the number of
fertile shoots per hectare and their interaction
with weather conditions.

Variability in time. — In the analysis of cone
production potential just discussed, the
between-year effect was tested as an interaction
at each level. Branch production, tree
production, and stand production varied
significantly between years (p=0.05) in 60
percent, 78 percent, and 100 percent of the
cases, respectively. When all stands in the
three regions were considered simultaneously
there were no significant differences between
years .

Variability of production in time is apparently
due, in part, to internal factors. Excellent
cone years (with yields one standard deviation or
more above mean yield) were preceded in 20 of 29
cases by poor cone years (with yields equal to or
less than the mean) . Since the probability of a
poor seed year so defined is 50 percent, this is
a significant deviation (<0.05) from our
expectation. In other species poor fruit years
are also followed by good fruit years, apparently
because initiating fruits cannot compete with
maturing fruit for available carbohydrates during
initiation and development (Kozlowski 1971) . The
amplitude of natural cycles in fruiting might be
increased by natural selection if poor seed years
preceding excellent seed years resulted in better
establishment of the tree through temporarily
overprovisioning previously starved-out predator
populations (compare with Janzen 1971, Forcella

1980) ; such selection could occur only if the seed
predator used the subject as its principle source
of food. Other variations may not have internal
causes. For example, we see no physiological or
evolutionary reason for the fact that 16 of 21
excellent cone crops were preceded 4 years earlier
by a poor cone year (p<0.05).

Especially poor cone years (with yields more than
one standard deviation below the mean) were not
significantly correlated with yields in any
previous year; we therefore think a poor cone year
is more likely determined by weather rather than
by internal factors.

We hypothesized that the demonstrated synchrony of
variation in cone production within a stand and
within a region is due to weather; and similarly,
that the lack of synchrony between regions is due
to differences in weather between regions. To
clarify the relationship between cone yield and
weather we regressed deviations of normalized cone
yields against deviations of weather parameters
(mean temperature and total precipitation) from
their means in each of the 46 months preceding
cone abscission as explained under methods
(compare with Eis, 1976).

Cone production was correlated with preceding
weather conditions, but in no simple way (table
3); six observations follow:

1. In well-sampled regions (represented by six to
eight stands) correlations with temperature or
rainfall significant at the 0.05 level may occur
in half and correlations significant at the 0.001
level may occur in 20 percent of the 46 months
preceding cone maturation. While significant
correlations become progressively harder to detect
as sample sizes decrease to two stands per
climatic region, it might be possible to detect
significant correlations in every month if sample
sizes were increased sufficiently.

2. Though many correlations are highly
significant, few explain much of the observed
variation in cone yield. The average r is 0.21
for precipitation and 0.20 for temperature and the
r of correlations significant at the 0.001 level
are only 0.30 for precipitation and 0.35 for
temperature. Assuming that variation in regional
data parallels variation in higher altitude
conditions, this suggests that each of the

46 months preceding cone abscission plays a small
but important part in determining yield. If final
yields of P. albicaulis are, in fact, determined
by a summation of everyday conditions, unique
events are notably less important to yields than
they are for pinyon pine (Forcella 1981) .

3. Numbers of significant correlations for both
temperature and precipitation are similar in the
prebud, bud, pollination, and cone maturation
years .

4. Forty-five percent of the significant
temperature correlations and 48 percent of the
significant precipitation correlations occurred in
winter months (November-April) when 'inactive'

72

Table 3. — Statistically significant relationships of cone yield to weather of the 46 months
preceding cone drop In southwest Montana . Correlation coefficients are presented
with their signs; regressions significant at the 0.1 percent, 1 percent, and 5
percent level are indicated by a, b, and no letter respectively

Factor Temperature Precipitation

Year2

preb

bud

ju V

ma t

preb

bud

ju V

ma t

JAN

-29

+40b

-51a

+34b

FEB

-32

+43a

+42b

-40b

-35b

+4 8a

MAR

+39b

-45a

+54a

APR

+29

-31

MAY

+45a

-30

-37b

-36b

JUN

+50a

-26

+32

-49a

JUL

-40b

—J J

AUG

+33

+46a

-40b

SEP

-26

-56a

OCT

+41b

-32

+31

-31

+46a

NOV

+41b

-32

+31

-31

+46a

DEC

+32

-48a

+2 7

Normalized deviation of yield was regressed against deviation from average climatic data as
explained in the text. Correlations for other regions were generally of similar size, sign
and significance and generally showed similar seasonal distribution. They will be provided
upon request.

Years are those before bud formation (preb), of cone bud formation (bud), of juvenile
pollinated cones (juv), and of cone maturation (mat).

trees might be assumed to be little affected.
Correlations with winter precipitation could be
due to its effect on summer soil water supply.
The remaining significant temperature correlations
were 20 percent in spring (May-June) , 16 percent
in summer (July-August) and 19 percent in fall
(September-October). The remaining significant
precipitation correlations were 15 percent in
spring, 18 percent in summer, and 19 percent in
fall.

5. Negative correlations are slightly more common
than positive correlations between cone yields and
summer (80 percent negative in July-August) and
winter (57 percent negative in November-April)
temperatures, as they are with spring (70 percent
negative in May-June), fall (65 percent negative
in September-October) , and winter (62 percent
negative in November-April) precipitation. We are
not ready to conclude that high temperatures and
heavy precipitation lower yields.

6. The fact that highly significant correlations
between yield and a given developmental period may

differ in sign between different areas is
consistent with the observation that year-to-year
variability in cone production disappears when one
averages across regions.

Our results will frustrate anyone wishing to
predict future cone crops from weather data. The
highly significant, but weak, correlations
observed suggest that a complex physiological
model would be needed to make such predictions and
that its final prediction wouldn't be available
until shortly before the crop matured. Similar
results could be had by sampling branches in a
specific stand for numbers of mature or juvenile
cones and, since stands in a region behave
similarly, results from a representative stand
might predict regional crops reasonably.

Variability associated with easily measured stand
charac ter ist ics . — The considerable variability of
average cone production among stands must be
caused directly by site characteristics such as
climate and soils and stand characteristics such

~3

Table 4. — Regressions relating cone production to easily measured stand characteristics

Regression

equations .

r2

Ca =

0.

470

+

0.

2

,00032 canopy cover

0. 42

Ca =

0.

288

+

0.

023 basal area

0.28

C =
a

0.

812

+

0.

439 fallen cones

0. 32

Ca =

0.

450

+

0.

00023 canopy cover

+

0.

206

fallen cones

0.46

Ca =

0.

430

+

0.

2

00032 canopy cover

+

0.

470

stand size

0.52

C8 =

0.

515

+

0.

o

00039 canopy cover

+

0.

734

stand size

0.49

0.

269

+

0.

o

00011 canopy cover

+

0.

.001

stand size

0.32

Cone production in the average year (Cg) is expressed as cones'm yr . C and C
represent cone production in good and poor years respectively.

Units were percent of ground area for canopy cover, m /ha for basal area, and new plus
old cones/m on the ground for fallen cones. Stand size was scaled with 1=0= 0.75
ha, 2 = 0.75 to 1.25 ha, and 3 = more than 1.25 ha.

as age, density, and the genetic allocation of
photosynthate to reproduction. Few of these
variables were studied because their measurement
was too costly, time consuming, or difficult. On
the other hand, correlations of average cone
production with easily measured stand
characteristics were tested because any strong
correlations discovered would be useful for
managers who wish to predict production in
particular stands.

Measures of stand cover and fallen cones should
be correlated with cone production since they
index numbers of potentially fruiting branches
and their previous fruitfulness (table 4) .
Canopy cover alone explained 42 percent of
observed variation in cone production. Basal
area only explained 23 percent of the variance,
probably because branch numbers were more closely
related to tree cover or circumference than
cross-sectional areas. Perhaps due to uneven
consumption by animals, numbers of fallen cones
explained only 32 percent of the observed
variation. A regression combining canopy cover
and numbers of fallen cones was our second best
(r =0.46) .

Our best simple predictor of cone production
(table 4, r =0.52) involved stand size with
canopy cover. Besides explaining observed
variation best, this estimator has the advantage
of being least expensive to apply — both canopy
cover and stand size can be estimated from aerial
photos in the non-field season. The 'small stand
effect' is likely due to poorer fertilization in
small stands than in large ones; Sarvas (1962)
observed poor pollination in Pinus sylvestris
when stands were smaller than 2 ha. The fact
that yields of cones in good cone years
(C =yields greater than median yields) seem to be

more affected by stand size than yields in poor
cone years (C =yields less than in median years)
suggests that^other factors (probably weather) ,
override the pollination effect in poor cone
years .

Other potential estimators of stand production
studied were less useful. Regression of yield
against stand age, stand area, stand elevation,
stand slope-aspect, productivity, biomass, and the
cover of indicator plants alone showed no
significant correlations. Complex combinations of
these factors in multiple^regressions increased
attributal variability (r ) to 60 percent but
since these regressions are biologically
uninterpretable they are not reported.

SUMMARY AND CONCLUSIONS

Evidence of cone production — juvenile cones,
mature cones, or cone scars — located at
progressively older nodes of terminal branches can
be used to estimate cone yields of a branch over a
6 to 8 year period from a single observation. The
product of mean branch yields, so determined, and
fruiting branch density (branches per m ) provides
a measure of stand cone yield through the same
period .

Average reproductive production in the 29 stands
observed ranged from 0.3 to 3.6 cones'm 'yr and
23 to 270 seeds •m_2-yr~1. The coefficient of
variation (SD/mean) between years for cone
production in^the average stand was 0.7
cones 'm *yr . While seed numbers are comparable
to those observed in other pine forests their
weights were greater. Cone production comprised
about 9 percent of total arboreal production.

74

Cone yield potentials varied significantly
between branches in 30 percent of the trees
studied and between trees in 48 percent of the
stands studied, but never between stands.

Cone yields varied significantly between years in
branches, trees, and stands in a region. Some
internal control is suggested by the fact that
high-yield years usually followed low-yield
years. External control is indicated by
significant correlations between regional weather
and cone production. These correlations suggest
that temperature and precipitation conditions
have significant, but not dominant, effects in
prebud, bud, juvenile cone, and mature cone years
and all months of the year. Unique events seem
less significant, therefore, for Pinus alblcaulis
than for Pinus edulis (Forcella 1981).

Easily measured stand characteristics are never
more than mediocre predictors of average cone
production. Our best regressions were based on
canopy cover, canopy cover with fallen cones, and
canopy cover with stand size. They explained 42,
46, and 52 percent, respectively, of the observed
variation. Fortunately, the best predictor is
also least expensive to apply.

ACKNOWLEDGMENTS

The idea of estimating past cone production from
cone scars was suggested by D. Perry's
observation that past P. contorta cone production
could be so analyzed. L. Munn helped with
production-weather regressions. USDA Forest
Service Contract Number 12-11-204-12, Supplement
#33, for the study of grizzly bear food
production in whitebark pine forests, partially
supported the work.

REFERENCES

Allen, G. A basis for forecasting seed crops of
some coniferous trees. Journal of Forestry.
39: 1014-1016; 1941.

Arno, S. Pinus alblcaulis (Engelm.): whitebark

pine. In: Burns, R. Sllvics of trees of the
United States. USDA Agriculture Handbook.
Washington, DC: U.S. Department of
Agriculture, Forest Service; (in press).

Boichenko, A. Pine growth on the northern

boundary of its area in the Transurals.
Soviet Journal of Ecology. 6: 37-45; 1970.

Blankenship, J. Native economic plants of

Montana. Montana Agricultural Experiment
Station; Bulletin. Bozeman, MT; 56: 1-388;
1905.

Clements, F. The life history of lodgepole burn
forests. Bulletin 79. Washington, DC: U.S.
Department of Agriculture, Forest Service.
1910. 56 p.

Eis, S. Association of western white pine cone

crops with weather variables. Canadian Journal
of Forest Research. 6: 6-12; 1976.

Forcella, F. Flora, chorology, biomass, and
productivity of the Pinus alblcaulis -
Vaccinium scopar ium association. Bozeman, MT:
Montana State University; 1977. 99 p. M. S.
Thesis.

Forcella, F. Cone predation by pinyon cone beetle
(Conophthorus edul is ; Scoly t idae) : dependence
on frequency and magnitude of cone production.
American Naturalist. 116: 594-598; 1980.

Forcella, F. Ovulate cone production in pinyon:
negative exponential relationship with late
summer temperature. Ecology. 62: 488-491;
1981.

Forcella, F. ; Weaver, T. The biomass and

productivity of the Pinus albicaul is -
Vacc inium scopar ium association. Vegetatio.
33: 95-105; 1977.

Formosof, A. The crop of cedar nuts, invasions
into Europe of the Siberian nutcraker
(Nuci f raga caryocatoctes macrorynchus Brehm.)
and the fluctuations of the squirrel (Sciurus
vulgaris L.). Journal of Animal Ecology. 2:
70-81; 1933.

Hutchins, H; Lanner, R. The central role of
Clark's nutcracker in the dispersal and
establishment of whitebark pine. Oecologia.
55: 192-201; 1982.

Janzen, D. Seed predation by animals. Annual
Review of Ecology and Systematics. 2:
465-492; 1971.

Kendall, K. Use of pinenuts by grizzly and

black bears in the Yellowstone area. Inter-
national Conference on Bear Research and
Management. 5: 166-173; 1980.

Kendall, K. Bear-squi r rel-pi nenut interaction.
In: Knight, R. Blanchard, B. Kendall, K.
Oldenburg, L. eds. Yellowstone
grizzly bear investigations. Unpublished
Report of the Interagency Study Team 1978-
1979. Bozeman, MT: U.S. Department of the
Interior, National Park Service. 1980. 51-60.

Koslowski, T. Growth and development of trees.
New York: Academic Press, 1971. 514 p.

Lotan, J. Cone serotiny of lodgepole pine near
West Yellowstone, Montana. Forest Science.
1 3: 55-599; 1967.

Lowry, W. Apparent meterological requirements
for abundant cone crop in Douglas-fir.
Forest Science. 12: 185-192; 1966.

Moir, W. Litter, foliage, branch and stem
production In contrasting lodgepole pine
habitats of the Colorado Front Range. In:
Franklin, J. Depster, L. Waring, R. eds.
Research on coniferous forest ecosystems —
proceedings of a symposium. Portland OR: U.S.
Department of Agriculture, Forest Service,
Pacific Northwest Forest and Range
Experiment Station; 1972: 189-198.

75

Sarvas, R. Investigations on the flowering and
seed crops of Pinus syl vestris. Communi-
cations Institute Forestalls Fenniae. 53:
1-198. 1962.

Shopmeyer, C. Seeds of woody plants in the
United States. Agricultural Handbook 450.
Washington, DC: U.S. Department of
Agriculture, Forest Service; 1974. 883 p.

Smechkin, I. The method of calculating the
productivity of stone pine forests by the
remnants of cones on the ground. In:
Shmanuk, A. ed. Fruiting of the Siberian
stone pine in eastern Siberia. Akademiya
Nauk SSSR, Trudy Instituta Lesa i Dresiny.
Vol. 62 (NYIS translation TT6550123); 1963:
161-167.

Smith, C. The adaptive nature of social

organization in the genus of tree squirrels
(Tamiasciurus). Ecological Monographs.
38: 31-63; 1968.

Smith, C. The co-evolution of pine squirrel
(Tamiasciurus) and conifers. Ecological
Monographs. 40: 349-371; 1970.

Tombach, D. Nutcrackers and pines: coevolution
of coadaptation. In: Nitecki, M. Coevolution
Chicago: University of Chicago Press; 1983.
392 p.

U.S. Department of Commerce. Climatological
data. Washington, DC: U.S. Department of
Commerce, National Oceanic and Atmospheric
Administration, Environmental Data Service;
1968-1976.

Weaver, T.; Dale, D. Pinus albicaulis in

central Montana: environment, vegetation,
and production. American Midland
Naturalist. 92: 222-230; 1974.

Westman, W; Whittaker, R. The pygmy forest
region of northern California: studies on
biomass and primary productivity. Journal
of Ecology. 63: 493-520; 1975.

Section 2. Cone Prediction, Collection, and Processing

77

h

CONE PREDICTION, COLLECTION, AND PROCESSING

D.G.W.^E

Edwards

ABSTRACT: Methods of predicting the size of a
cone crop and the seed yield are reviewed. The
planning and methods of harvesting the crop, and
the procedures for seed extraction, seed cleaning
and sorting are discussed.

INTRODUCTION

The expansion of reforestation programs has
created a greater focus on the problems of seed
supply, particularly for specific provenances and
for genetically improved seeds. Rising costs for
all types of seedling production have emphasized
the need for high-quality seeds.

Seed production in most conifers is periodic and
intervals between good crops vary. In the years
between heavy crops, few or no cones may be
produced. So that seeds will always be available
for forest regeneration, the forester must be able
to predict a heavy crop, and must know how and
when to harvest the seeds and how they must be
processed. Where natural regeneration is planned,
advance knowledge of a heavy crop allows for
modification and timing of the logging methods, or
site preparation, so that the approaching seed
fall is used to the best advantage.

This paper provides a broad review of cone crop
prediction, cone collection and seed processing.
Crop prediction, or forecasting, is the means
whereby the forest manager looks ahead, sometimes
as much as one and a half years, for early signs
that collectable quantities of cones may develop.
Should the signals be positive, resources such as
manpower, equipment and funding can be organized
well ahead of the collection date. Since most
conifer seeds ripen and begin to disperse in a
relatively short period of time, cone collections
must be carefully timed. Optimally, seeds should
be mature, or nearly so, and free from insect or
disease damage. Several methods of cone
collection, some of them mechanized, have been
developed and the most appropriate one must be
chosen to suit the species, and stand and crop
conditions. Cone and seed processing involves
numerous steps that begin with seed extraction
from dried, opened cones. In many species the
seed wing must be removed before the seeds are
cleaned of non-seed debris and impurities and

Paper presented at the Conifer Tree Seed in
the Inland Mountain West Symposium, Missoula,
MT, August 5-6, 1985.

D.G.W. Edwards is Research Scientist, Pacific
Forestry Centre, Canadian Forestry Service,
Victoria, British Columbia, Canada.

sorted to remove empty or damaged seeds. Seedlot
quality has to be checked in germination tests so
that efficient use can be made of the seeds in
the nursery, and the seeds have to be prepared
and packaged for cold storage. While the
concepts and methodology discussed relate
principally to natural stands since these satisfy
the bulk of reforestation needs both at present
and for the immediate future, they can also be
applied in seed orchards.

CONE CROP PREDICTION

Accurate crop predicting (or forecasting) is
difficult since many factors affect the crop from
its initiation to seed maturation, and these are
incompletely understood. Successful predictions
are based on knowledge of the reproductive cycles
of the various species which have been described
in detail for PseudotsuRa menziesii (Mirb.) Franco
TsuRa heterophylla (Raf . ) Sarg . , T . mertensiana
(Bong.) Carr. , Pinus contorta (Dougl.), Thu.i a
plicata Donn. , Chamaecyparis nootkatensis (D. Don.
Spach, Picea enRelmannii (Parry) and P. Rlauca
(Moench) Voss (Allen and Owens 1972; Owens 1973;
Owens and Molder 1984a, 1984b, 1984c, 1984d) .
Predictions can be made at three main stages in
the reproductive cycle: (i) the crop year before
flowering, (ii) the early spring of the crop year
and (iii) after flowering, when conelets are
visible .

1. Predictions In The Crop Year Before Flowering

Early predictions are the most complex, since
they are based on factors influencing the
initiation of reproductive structures and their
development. They are also the least reliable
because many factors can subsequently damage the
crop, so such forecasts need periodic revision
and adjustment. Such predictions are based on
observations on the periodicity (frequency) of
cone crops over many years, the relationships to
weather conditions preceding the crops, and the
formation of reproductive stuctures in winter
buds .

Periodicity . --The period between good seed crops
varies within and among species (table 1) . The
phenomenon of alternate bearing is due to the
developing crop having a negative effect on the
subsequent year's crop. The reasons can be
morphological as, for example, in Picea glauca,
TsuRa heterophylla, Thu.j a plicata and other trees
that bear reproductive structures in terminal
positions on the shoots, since these species do

78

not have as many available locations for flower
production the year following a good crop. Thus
heavy crops are followed by light ones. The
effects can also be physiological, the presence of
a crop in one year influencing the initiation or
development of buds the following year. In species
that produce flowers on the previous year's shoots,
such as Abies balsamea (L.) Mill. (Morris 1951;
Powell 1977) or PseudotsuRa menziesii (Owens 1969;
Allen and Owens 1972), new shoots tend to be short
and few flower bud primordia develop, possibly
because of a lower level of available carbo-
hydrates during the year of a heavy crop. Lee and
others (1979) concluded that a high carbohydrate:
nitrogen ratio favored female strobilus initiation
in Pinus elliottii Engelm. The minimum period
between good crops in these and several other
species is therefore two years (Morris 1951; Baron
1969; Dobbs and others 1976; Powell 1977). There
is also some evidence that a maturing cone crop
has an effect on future crops in species with a
three-year reproductive cycle, such as some pines
(Wenger 1957; Lester 1967; Baron 1969; Eis 1976)
either by influencing the initiation and develop-
ment of reproductive primordia or by affecting the
developing conelets.

A heavy current crop, therefore, can be used in
many species as an indicator that the next crop
will be poor. However, instances of consecutive

Table 1. — Periodicity in some western conifer
species. (Source: Schopmeyer 1974)

Species Interval between good crops

(years)

Abies

amabilis

2-3

Abies

Rrandis

2-3

Abies

lasiocarpa

2-4

Juniperus communis

irregular

Juniperus occidentalis

Juniperus scopulorum

2-5

Larix

laric ina

3-6

Larix

lyallii

1-10

Larix

occidental is

1-10

Picea

enRelmanni i

2-3

Picea

Rlauca

2-6 +

Picea

mariana

4

Picea

punRens

1-3

Pinus

banks i ana

3-4

Pinus

contorta

1

Pinus

f lexilis

2-4

Pinus

monticola

3-7

Pinus

ponderosa

2-5

PseudotsuRa menziesii

2-11

Thu.ia plicata

3-4

TsuRa

heterophylla

2-8

TsuRa

mertensiana

1-5

good cone crops have been reported in a number of
species (Haig and others 1941; Fowells and
Schubert 1956; Maguire 1956; Eis and others 1965;
Lowry 1966; Franklin 1968; Rehfeldt and others
1971), although they are rare. In any year, wide
differences in crop size between stands of the
same species within a region may occur. Such
events demonstrate that environmental factors may
at times override the negative effects of previous
seed production.

In any given year the level of seed production
may vary from one species to ^the next, and may
vary from one region to another as well as among
stands within a region (Haig and others 1941;
McWilliams 1950; Rowe 1955; Waldron 1965b;
Franklin 1968; Bingham and Rehfeldt 1970). Within
a species, Franklin (1968) observed that stand to
stand variation was least in heavy or very light
crop years, and greatest when cone crops were
medium.

Among pines, most species flower every year, yet
some species are strongly cyclic in seed
production (Wright 1953; Fowells and Schubert
1956; Maguire 1956; Lester 1967; Franklin 1968;
Bramlett 1972). The main reason appears to be
conelet abortion which can account for between
40% and 70% of the loss of a potential crop
(Wright 1953; Sarvas 1962; Snyder and Squillace
1966; Lester 196 7; Wang 1970; Shearer and Schmidt
1971; Bramlett 1972). High conelet loss has been
related to low temperature at the time of
pollination failure (Wright 1953; Sarvas 1962;
Hard 1963; Boyer 1974; Schoenike 1955),
physiological causes (Wright 1953; Sarvas 1962;
Wang 1970; White and others 1977) and insect
damage by feeding (DeBarr and Ebel 1974; Kormanik
1974; DeBarr and Kormanik 1975; Neel and others
1979) .

Weather conditions .- -In different species and
climatic regions initiation and development of
reproductive primordia occur at different times
of the growing season (Gifford and Mirov 1960;
Allen and Owens 1972). This is when weather
conditions have been shown to be most critical.
However, in many studies correlations have been
found only when the weather has had a profound
negative influence on the reproductive cycle so
the use of climatic conditions in forecasting may
only be of use when records of previous cone
crops are considered (Calvert 1979). During
primordia initiation and development (i.e.,
during the year preceding pollination) positive
responses to warm, sunny weather have been noted
(Lester 19C7; Rehfeldt and others 1971; Eis 1973a,
1976) in many species with a two-year reproductive
cycle (table 2). Moisture deficit during this
period was linked by Ebell (1967) to increased
strobilus production in PseudotsuRa menzies i i .

In pines, interpretation of climatic effects on
seed production is complicated by the three-year
reproductive cycle and interactions with
physiological factors causing primordia or conelet
abortion (Calvert 1979). Positive correlations
with weather patterns occur in several species
(table 3) but only in Pinus monticola Dougl .

79

Table 2. — Species in which seed production

increases following a warm, sunny summer

have correlations been found during the critical
year of pritnordia initiation and the conelet and
crop years.

Winter buds. — Interaction between climate and
physiological processes becomes evident in the
number of reproductive buds in species with a
two-year cycle, or in the number of conelets in
pines in the fall and winter preceding cone
maturity. Numbers of male buds are closely
correlated with the abundance of female buds in
PseudotsuRa menziesii (Silen 1967) and Pinus
ponderosa Laws. (Roeser 1941) and a crop prediction
method based on male buds was proposed by Silen
(1967). This relationship does not hold in other
species (Wright 1953; Eis and Inkster 1972).
Predictions are most frequently based solely on
female bud abundance, using regression techniques
(Medvedev and Pal'gov 1971), sequential sampling
schemes (Roe 1966; Eis and Inkster 1972; Eis
1973b), or ratios of female to vegetative buds
(Allen 1941a) . The accuracy of these prediction
methods has varied from species to species (Eis
and Inkster 1972). Poor or nil crops can be
predicted with 1007,. accuracy while forecasting of
heavy crops may only be 70%-90% accurate. In pines,
surviving conelets at the beginning of the second
year may be used for crop predictions (Snyder and
Squillace 1966; Shearer and Schmidt 1971) provided
the size of the current cone crop is considered.

The use of reproductive buds in seed crop fore-
casting is a more direct and reliable approach

Table 3. — Positive correlations between weather and cone production in pines

Weather variable Species Source (Period of observation-years)

Year of primordia initiation:

Crop

warm spring

Pinus
Pinus
Pinus

banksiana
ponderosa
resinosa

Larson 1961
Maguire 1956 (23)
Lester 1967 (15)

warm, possibly
droughty summer

Pinus

Pinus
Pinus
Pinus

monticola

ponderosa

resinosa

sylvestris

Eis 1976 (20)

Rehfeldt and others 1971 (18)
Daubenmire 1960
Lester 1967 (15)
LaBastide and

van Vredenburch 1970 (38)

warm, possibly
wet fall

Pinus

monticola

Eis 1976 (20)
Rehfeldt and others

1971

(18)

of conelets:

rain following
pollination

Pinus

monticola

Eis 1976 (20)

warm fall

Pinus

monticola

Eis 1976 (20)

year:

warm dry spring

Pinus

monticola

Eis 1976 (20)

80

Species

Source (Period of
Observation - years)

Abies Rrandis
Abies sibirica

Larix leptolepis
1979)

Picea abies

Picea Rlauca
Picea mariana
Picea omorika

PseudotsuRa menziesii

Eis 1973a (37)
Miscenko 1963 (cited by
Calvert 1979)

Yanagihara and others 1960
(49) (cited by Calvert

LaBastide and van

Vredenburch 1970 (38)
Fraser 1958 (3)
Fraser 1958 (3)
Maurin and others 1970

Eis 1973a (37)
LaBastide and van

Vredenburch 1970 (38)
Lowry 1966 (48)

Tsufta canadensis

Maurin and others 1970

when bud types can be positively identified (Ei6
1967; Eis and Inkster 1972). Cone crop prediction
based on winter bud counts is the method now used
in British Columbia since detailed, illustrated
descriptions of reproductive bud and cone develop-
ment have been published (Canadian Forestry
Service 1983) for 18 western conifers. However,
counting of buds still presents practical problems
since representative twigs from the upper part of
the crown must be sampled (Finnis 1953). These
may be obtained by climbing, by shooting off the
ends of branches using a rifle, or removal from
recently felled trees. In some circumstances,
sampling by helicopter may be justified. Samples
are more readily obtained from seed orchards. Bud
examinations can be carried out in the field or in
the laboratory. In addition to these detailed
examinations, shoots placed in containers of water
can be "forced" (i.e., the buds will continue to
develop and burst to reveal whether they are male,
female or vegetative) under warm, humid well-
lighted conditions. This process may require two
to three weeks but it will confirm any diagnosis
based on bud morphology.

Reproductive bud identifications, or conelet
counts in pines, are carried out far enough in
advance of a developing crop to provide the
forester or seed orchard manager ample time to
plan for a large collection operation if a heavy
crop is indicated. But there is still time for
the crop to fail, so its development during the
spring and summer of the crop year should be
monitored. However, if winter bud appraisals
indicate a poor or nil crop, further preparations
are unnecessary.

2. Predictions in Early Spring of the Crop Year

During the spring of the year in which the seed
crop will mature, pollination and fertilization
occur in species with a two-year reproductive
cycle, while fertilization only takes place in
pines. After bud burst, potential seed crops can
be estimated from the abundance of developing
strobili (Allen 1941a; Silen 1967; Eis and Inkster
1972). Usually, abundant megastrobili indicate a
large seed crop, even in some pines (Lester 1967;
Shearer and Schmidt 1971) but weather conditions
before, during and after bud break may cause major
losses (Wright 1953; Matthews 1963).

combined with moisture stress, has been known to
arrest pollen development completely in
Pseudotsuga menziesii (Chira 1967). Since
vegetative buds and shoots are less sensitive than
reproductive structures to low temperatures (Hard
1963: Zasada 1971) strobilus damage may be over-
looked unless the reproductive structures
themselves are examined. Rain has damaged the
quantity and viability of pollen in Picea glauca
(Nienstaedt 1958) and Pinus ponderosa (Turner
1956). In contrast, wet weather has been reported
to have no effect on the pollen of Pinus
sylvestris L. or Pseudotsuga menziesii in other
studies (Sarvas 1962; Silen 1962).

Methods of prediction based on strobilus abundance
include regression techniques or sequential
sampling, similar to those based on reproductive
buds .

3. Predictions After Flowering

Predictions of the seed crop when the developing
cones are visible on the trees are the easiest to
use and the most accurate. Such surveys are the
preferred method of crop forecasting in Ontario
(Ontario Ministry of Natural Resources 1984).
These forecasts not only consider the relative
abundance of maturing cones, but they also take
into account the quantity and quality of seeds.
They allow minimum time for organization, however.

Cone crop rating . --Most rating methods are
quantitative and are intended to indicate where
crops are heaviest and if they are worth
collecting. One method, devised for California
conifers by Baron (1963) used five ratings against
which crops were compared (table 4). This was
based on a system described by Fowells and Schubert
(1956) whose rating classes used actual numbers of
cones, the numbers varying with the species.
Unfortunately, Baron introduced the terms "light,"
"medium" and "heavy" making the method entirely

Table 4. — Cone crop rating system for California
conifers (Source: Baron 1963)

Rating Category1' Definition

Date of flowering varies from species to species
(Wright 1953; Ahlgren 1957; Boyer 1978) but since
the relative order of flowering is usually the same
from year to year (Wright 1953), poor weather will
not necessarily affect all species. Strobilus
development can be seriously disrupted by cold, or
unusually dry weather at the time of bud burst,
and megastrobili are particularly susceptible to
frost damage during the receptive stage (Roeser
1942; Wright 1953; Barras and Norris 1964;
Bramlett 1972; Eis and Inkster 1972; Timmis 1977).
Environmental stresses during the pollination and
fertilization sequences are major contributors to
seed crop periodicity in many trees (Haig and
others 1941). Cold weather may cause pollen cone
drop in Picea abies (L.) Karst (Sarvas 1968) and
P. glauca (Zasada 1971) and, in addition to or

1 -- None

- no cones on any trees

2 -- Very light - few cones on less than 25%

of the trees

3 -- Light - few cones on more than 25%

of the trees

4 -- Medium - many cones on 25% to 50% of

the trees

5 — Heavy - many cones on more than 50%

of the trees

TV Cones on full-crowned trees over 30 cm d.b.h.

81

subjective. For example, a given number of cones,
say 80-100, on Pinus monticola would probably be
categorized as "many" and might indicate a heavy
crop, while the same number of cones on a mature
Picea Rlauca cr PseudotsuRa menziesii tree might
go unnoticed. By the same token, a thousand cones
on Picea enRelmanii (Parry) Engelm. might
constitute "many" but would be classified as "few"
on a mature Thu.j a plicata (Dobbs and others 1976).
"Medium" and "heavy" crops (table 4, categories 4
and 5) are generally considered collectable.
Despite the subjectivity, similar systems have
been applied elsewhere. A method using seven
rating categories is used in British Columbia,
while ten rating categories are used in Alaska
(Zasada and Viereck 1970) as well as in Arizona
and New Mexico (Schubert and Pitcher 1973). When
rating a crop, attention has to be confined to
that portion of the crown expected to bear cones.
In Pinus monticola and Abies spp. this is limited
to the top four or five whorls of branches, whereas
in Picea Rlauc a /enRelmanii and PseudotsuRa
menziesii the upper two-thirds of the crown is
potentially cone bearing; in Thu j a plicata and
Pinus contorta cones may be found over the entire
crown .

Other methods based on total cone counts, as well
as that of Fowells and Schubert (1956), have been
developed (Haig and others 1941; Wright 1953;
Franklin 1968). In most species, especially those
that bear crops over a large portion of the crown,
such indices are limited in value, because cone
production per tree increases with age, tree
diameter and crown size (Haig and others 1941;
Garman 1951; Crossley 1956; Roe 1963; Waldron
1965; Lotan and Jensen 1970; Stiell 1971). They
may be used, however, to estimate cone yield to
determine if a collection quota can be met. One
exception is Pinus strobus L. which, past a certain
size, does not increase its cone bearing crown with
tree size so relatively small trees produce the
same crop as larger ones (Wright 1953) except at
high stand densities (Garber 1970) .

Waldron (1965) used a subjective cone abundance
rating which was then multiplied by an estimate of
the surface area of the cone bearing portion of
the crown to give a cone production index in Picea
glauca. In Great Britain, Seal and others (1962)
recommended that for Abies spp., Psejjdotsuga
menziesii , Picea sitchensis (Bong.) Carr. and
Pinus sylvestris . the cones visible through 6x or
8x binoculars from a distance roughly equal to the
tree's height can be counted and multiplied by
four to estimate the total cones on the tree.
Winjum and Johnson (1962) preferred to view a
PseudotsuRa menziesii tree from its south side and
to count the number of cones on one branch in each
whorl. The total number of cones on the tree was
estimated from a mathematical equation. A more
complex method was proposed by Lotan and Jensen
(1970) who used regression equations involving
diameter at breast height, live crown ratio, stump
height, age and partial cone counts in Pinus
contorta . Diameter and age were found to influence
cone production in Pinus ponderosa and PseudotsuRa
menziesii by Linhart and others (1979) .
Relationships between seed production and basal
area have been found in Picea enRelroannii (Roe

1967), Pinus resinosa Ait. (Roe 1964; Stiell
1971), P. palustris Mill. (Crocker 1973) and P.
strobus (Garber 1970).

The most useful quantitative rating methods are
based on percentages of full crop values. These
methods were first described for use with broad-
leaved species (Sharp 1958; Grisez 1975). They
compare current fruit counts with the maximum
counts ever made, and offer promise in making
estimates from different sources comparable.
Although they have application in research
(Calvert 1979), they require more effort to apply
than qualitative systems and may not be much more
advantageous for large-scale predictions.

Two common errors made in rating cone crops are
counting old cones that have shed their seeds and
evaluating roadside trees which, because of their
increased exposure to sunlight, often bear a
heavier crop than trees further in the stand
(Dobbs and others 1976).

Filled seed counts . --Final decisions on whether to
collect a crop should be based on the amount of
sound seeds forming in the cones. In many
species, cones will develop without pollination
but no seeds will be produced (Allen 1941b;
Orr-Ewing 1954; Meagher 1974) so an inspection of
the developing seeds is essential. Abundant
pollen is required for a good seed set regardless
of female strobilus production (Sarvas 1957, 1968;
Boyer 1974; Fechner 1979). Heavy male and female
flowering usually occur in the same year and this
is reflected in a higher yield of sound seeds
(Cayford 1964; Roe 1964; Sarvas 1968; Zasada and
Viereck 1970; Shearer and Schmidt 1971). Good
seeds may be found in relatively small areas even
in poor years, but these crops should be carefully
inspected before collections are undertaken.

The most common inspection method is to slice the
cones longitudinally and count the number of
filled seeds so exposed. Minimum filled seed
counts to set collection standards (table 5) have
been established (Buszewicz and Holmes 1956; Seal
and others 1965; Meagher 1974; Dobbs and others
1976) and regression equations based on such data
have been developed to predict the amount of good
seeds in a cone crop (Buszewicz and Holmes 1956;
McLemore 1962; Calvert 1978).

Insect damage to cones and seeds, as well as the
presence of disease such as cone rust, should also
be evaluated since they can seriously affect seed
yields. In light crop years, insects and diseases
can destroy the entire crop. The presence of
insects is often, but not always, signalled by
premature browning of the cones, small holes in the
cone scales, accumulations of frass, pitch-like
exudations, and a general disfigurement of the
cone. If the cone-cutting test reveals that more
than half of the visible seeds have been affected,
collection is probably not worthwhile.

For Pinus contorta and other species whose cones
are serotinous and difficult to slice, it is
easier to extract the seeds by dipping the cones

82

Table 5. — Average number (and range) of filled seeds exposed per half cone, by crop year
(Source: British Columbia Ministry of Forests, Silviculture Branch 1985)

Species

Crop year

Minimum count for
collectable cropi'

Abies amabilis
Abies Rrandis
T.arix

Pinus contorta

Pinus monticola

Pseudotsuga

menziesii
Thu.j a plicata

Tsufta heterophylla

Tsuga mertensiana

1980
9

(5-12)

1981

16
(11-24)

3

(1-5)
7

occ idental \ s
Picea r1 auca/

Picea enRel^'annii (4-10)
Picea sitchensis

(see text)

11

(7-15)
5

(1-7)
7

(2-12)
3

11

(7-13)
1

2

(1-4)
1

(1-2)

1982
10

14

(12-18)

6

(3-8)

6

(4-8)
4

(1-7)
9

(6-11)
5

(3-7)
7

(4-10)

1983

12
(9-12)

4

(2-7)
5

(3-6)

1984

(11-24)
1

1980-1984

11
(5-16)
15

3

1-5
6

(2-10)
5

(3-6)

10
(4-15)
5

(1-7)
8

(1-12)
4

(1-7)
7

(4-10)

8-12

12-14

20 per^/
whole cone

90 per
whole cone
5

4-6 per
whole cone
3

h? Figures applicable just prior to collection because insects or disease may decrease counts Tf~
there is a significant time lag between examination and collection.

if Dobbs and others 1976.

briefly in boiling water then oven-drying them at
65*C for three to four hours. Extracted seeds may
then be cut or crushed to reveal a firm white
megagametophytic tissue (endosperm) if they are
filled. For Pinus contorta a minimum of 20 filled
seeds per entire cone indicates a collectable crop
in British Columbia (Dobbs and others 1976) but
experience in Alberta suggests collectability is
indicated by a count of only six to eight filled
seeds (Helium and Wang 1985).

CONE COLLECTION

The basic objective of any collection method is to
get the cones from the tree tops and into sacks in
the most efficient and safe manner without damaging
seed quality. All cone collections require advance
planning and organization. The larger the collec-
tion operation, the more its success depends upon
staff training, especially that of the supervisors
(Dobbs and others 1976; Calvert 1985). Besides
considerations of which species and provenances
are to be targeted, cone quotas and the amount of
manpower, equipment and storage facilities must be
calculated. These have all been reviewed by Dobbs
and others (1976) .

Since most reforestation seeds will continue to be
derived from natural forests, stands of locally

important species selected specifically for seed
production purposes should be reserved. These
reserved stands should be at appropriate elevation
intervals in seed zones where substantial
reforestation is expected to occur (Pitcher 1966;
Dyer 1968; Holmes 1972; Rudolf and others 1974).
All collection sites must have good access and an
adequate number of well- formed trees of the
required species bearing a collectable crop.

The choice of collection method is influenced by
several species characteristics, such as cone
shape and size, tree crown shape, and the position
of the cones on the tree. The collection method
must also consider whether the cones occur at the
branch ends or along its length, whether they are
erect or pendant, and whether old cones are
persistent. Cone persistency can be either a
problem, as in Larix species, or an advantage, as
in Pinus contorta . since their serotinous nature
allows collection of more than one year's crop at
a time. Other more minor characteristics, such
as the extreme pitchiness in Pinus mont icola or
Abies cones, and the tendency for true fir cones
to disintegrate at maturity, must also be weighed.
The method of collection also depends on the number
of crop-bearing trees and their accessibility;
scattered crop trees require a different approach
from many crop trees along a road or around a

8 3

clearing. Age and height of the trees, whether
they are in a sub-dominant or lower crown position,
and the evenness of the canopy level, if aerial
collections are contemplated, must also be
considered.

1. Collection Methods

There are four general methods of cone collection:
climbing, felling or topping, aerial raking and
clipping, and collections from squirrel caches.
The advantages and disadvantages of different, cone
collection methods have been compared by Cornell
(1984) (table 6).

Climbing. — Climbing is the oldest traditional
method. It uses few items of accessory equipment
compared to other methods, and causes minimum
damage to the trees, but requires trained personnel
skilled in safety precautions and comfortable
working at heights. A climber begins picking at
the top of a tree and works his way down and
around the crown, using a cone hook to pull up
branches bearing cones beyond reach. Cones are
usually placed in a bag hanging from the worker's
safely belt. Cones should not be thrown to the
ground, even in sacks. Even mature cones,
especially those of Abies species, should be
handled gently to avoid bruising or damage to
fragile seed coats (Edwards 1982). Collection
costs are minimized by climbing only those trees
bearing a heavy cone crop (Goddard 1958) .
Individual trees should be selected on the basis

of phenotype and safety criteria. Climbing for
cones has been found to be productive when the
crop trees are immature and tree heights are less
than 15 m, the cones are medium to large, the
stand is fairly open, and the crowns full, with
well-spaced, sturdy branches. The species
suitable for climbing include Pseudotsuga
menziesii . Larix occidentalis Nutt. , Pinus
ponderosa and P. monticola.

Felling and topping. — Tree felling to collect
cones has become common in British Columbia.
Collections should be coordinated with logging or
clearing operations, but trees may be felled
specifically for cone harvest if plans are made to
recover the merchantable wood later. Felling
requires experienced fallers capable of placing
the trees so that the tops are readily available
for picking. Cones should be removed from only
the phenotypically best trees. For safety reasons,
all felling must be completed before pickers are
allowed to begin work, but picking is then faster
than by climbing, and no special tools are
required. Collecting from felled trees is best
when the crop trees are mature, where there is
good access to an adequate number of trees that
are not required for future collections, and when
capable fallers are available. It can be used for
all species except Abies, the cones of which
shatter on impact with the ground, and the method
is particularly useful for winter collection of
Pinus contorta cones, which are more easily
detached in cold weather, provided snow does not
cover the downed tops .

Table 6. — Some comparisons of cone collection methods (Source: Cornell 1985)

Method

Advantages

Disadvantages

CLIMBING

Best phenotypes selected.
Cones picked when ripe.
Minimum damage to trees.

High climber hazard.
Picking limited to nearby roads.
Staff limitation-may not reach
all areas at peak ripeness.

FELLING

Faster, less expensive.
Less hazard than climbing.
Best phenotypes selected.

Some felling hazard.
Felled trees need harvesting.
Meed coordination with logging.
Crop trees lost to further

collection.
Cones need to be picked

promptly once trees felled.

AERIAL

SQUIRREL CACHES

Rapid .

Limited surface access

required.
Best phenotypes selected.
Cones picked when ripe.
Access to all areas.
Best method for some

species .

Low hazard.

Requires limited staff
training and little
special equipment.

Expensive .

Requires large operation and

extensive planning for efficiency.
High pilot hazard.

Limited control over seed ripening.

No phenotype selection.
Quotas may not be met.

84

If trees would not otherwise be felled, or where
subsequent utilization of the timber is not
feasible, tree topping may be a more suitable
method, especially in species where cones are
confined to the uppermost crown as in Pinus
monticola. Tops may be cut out with a saw by
climbers, or brought down with a high-powered rifle
using soft-nosed bullets for maximum effect (Dobbs
and others 1976), or removed by aerial clipping.
Slayton (1969) found climbing and topping Picea
glauca trees, followed by stripping the cones by
hand, was cheaper than climbing alone. This
conclusion was also reached by Calvert (1985) for
Picea glauca and P. mariana Mill. (Britt.).
Whether trees are felled or tops removed, prompt
picking (within 2 or 3 days) may be necessary to
forestall cone opening caused by high soil tempera-
tures, or to prevent losses to birds and animals
(Stein and others 1974) .

Interest in machinery for stripping cones from tops
and slash of Picea mariana and Pinus banks i ana
began in Ontario in the mid-1960s (Haig 1969) and
equipment has been developed which can economically
strip cones, particularly in Picea mariana , even
in light crops (Horton 1984). The increasing
demand for tree seeds has spurred other attempts to
mechanize cone harvesting. The use of tree shakers
has been successful with some species, especially
in the southern United States, but this '"ethod
cannot be employed except where the terrain will
allow easy access and operation of the equipment,
such as in seed orchards or some seed stands.
Tree shakers arc most effective on species with
cones that are easily detached such as Pi cea
engelmannii , Abies grandis . Pseudotsuga menziesii
or Pinus ponderosa (United States Dept.
Agriculture 1972), and can remove cones very
rapidly; 75% of the total cones on a Pseudotsuga
menziesii tree 48 cm d.b.h. and 30 metres tall
were removed in 21 seconds (United Nations Food
and Agriculture Organization 1968) . Shaking may
be effectively used on smaller cone crops that
would be uneconomic to collect by climbing
(Richardson 196 7 ) .

Aerial clipping and raking.- Technology is also
being developed for aerial collections of cones
(Dobbs and others 1977; Apt and others 1979; Hedin
1983) and two systems, clipping and raking, have
been approved for use in Canada. Clipping employs
a two-man team of a helicopter pilot and a clipper
operator who is secured by harness to the air-craft
and operates a hydraulic anvil-type pruner or a
special electric chain saw. Depending upon the
species and crown shape, either cone laden
branches or tops are removed and stored inside the
aircraft as it moves from tree to tree. Aerial
raking uses a device, suspended below the heli-
copter, that is lowered over the target tree until
the cone bearing top protrudes through the center
of the cutter head. When the rake is lifted, the
head severs branches which fall into a collection
basket. As with clipping, the aircraft moves from
tree to tree until a load has been gathered, and
flown to a landing for off-loading. For safety
reasons, pickers are permitted on the landing only
when aerial delivery is complete.

Raking and clipping are suitable for use on mature
trees in mixed stands when the cones are in the
upper third of narrow, tapered crowns. Clipping
is best used when the target trees are dominant or
codominant , while raking may also be applied to
lower canopy trees when they bear cones. All
aerial systems require favorable weather
conditions: steady winds of less that 15 kmh, and
little or no rain. Both systems can be used for
collecting Pseudotsuga menziesii or Picea cones.
If used on other species, clipping can become
prohibitively costly, but raking is ideal for true
firs, and some rakes have been devised for use on
Thu.j a plicata and Thu.j a heterophyl la .

Aerial methods are excellent for harvesting cones
from inaccessible stands, from trees in stands
protecting streams, or from areas which migratory
animals traverse. Because of tariff rates, equip-
ment rentals, fuel costs, time spent training
people and organizing and coordinating the collec-
tion program, aerial operations are expensive.
For cost-effectiveness, both productivity and
resulting seed quality must be high, which can
only be achieved when there is a heavy cone crop
with good seed counts. Nevertheless, the overall
costs of helicopter collections have been found to
be competitive (Hedin 1983; Cornell 1984),
especially for species such as true firs
(Wallinger 1985). Since larger quantities of
cones can be harvested per day, helicopter collec-
tions can be completed in a shorter time frame,
which allows them to be started closer to full
ripeness of the seeds.

Squirrel caches.- Provided the source stand is of
good phenotypic quality, the reforestation value
of the seeds obtained from squirrel caches may not
suffer unduly since caches are located in the
general vicinity of the trees from which the cones
are removed. Should the stand contain poor
phenotypes, the method should not be used. In
light crop years squirrels can seriously deplete a
developing seed crop (Larson and Schubert 1970) ,
in some instances leaving very little for the
forester to collect (Shearer and Schmidt 1971).
Fears that Pseudotsuga menziesii seeds from
squirrel-cached cones may be of poor quality
because they were cut from the trees before
becoming fully mature have been dispelled by
Lavender and Engstrom (1956) who observed that
some cones were cut prior to full ripeness, but
that the squirrels cut cones in quantity, and
began caching them, only when the seeds were ripe,
and that no significant increases in seed quality
occurred with later cutting. Wagg (1964) observed
that seed quality in cached cones of Picea glauca
was higher than in cones collected from the tree
because some of the mature seeds had fallen from
partially opened cones on the tree, resulting in
an increase in the proportion of under developed
seeds in later cone collections. However, since
caches are typically found in damp areas in decayed
wood or duff, or around old stumps or logs, cones
should be checked for the presence of pathogens.
Commercial seed merchants frequently collect from
caches, yet their product can be of exceptionally
high qua! ity .

B5

2. Seed Maturation

One of the crucial considerations in any
collection is that the seeds have achieved either
full ripeness or a threshold of maturation from
which they will continue to develop in the cones
before seed extraction begins. A detailed review
of seed maturation and its effects on seed quality
can be found in Edwards (1980a) .

While viable seeds can be obtained long before the
cones are ready to open, seed maturity in conifers
is usually associated with seed dispersal. Unfor-
tunately not all the cones mature simultaneously,
and calendar dates vary with locality and weather
patterns. There may be variations in ripening
among the cones on any one tree (Ching 1960;
Fowells 1949; Maki 19A0) , among trees in the same
stand (Allen 1958b), between different stands in
the same year (Fowells 1949) and considerable
variation from one collection year to another
(Allen 1958a; Fowells 1949). Most conifers
release their seeds quickly once maturity has been
attained so the period for collecting mature seeds
is no more than two weeks in most species. The
general conclusion drawn from a multitude of
investigations is that the more mature the seeds
are when collected, the greater their vigor and
potential for establishment of seedlings (Pollock
and Roos 1972) . Therefore the timing of seed
collection has to take into account numerous
considerations and if collection is delayed, even
by a day or so, the bulk of the crop may be lost.

The cones of some species such as Pinus contorta,
P. banks i ana Lamb, and Picea mariana usually are
serotinous, that is the edges of the cone scales
are bonded together by resin. Cones of these
species may remain intact on the tree for several
years after maturation, providing a longer
collection "window." In Pinus albicaulis Fngelm. ,
intact cones fall from the tree and the seeds are
released only after the cones have lain on the
ground and have disintegrated over several months
(Krugman and Jenkinson 1974).

Among the consequences of collecting cones too
early is a high moisture level that favors mold
growth. Unless the cones are well ventilated
during storage before seed extraction, they may
suffer from heating that can cause direct damage
to the seeds as well as exacerbate mold activity.
The higher moisture content of immature cones
requires longer kilning, thereby increasing
extraction costs (Roe 1960), and normal kilning
temperatures may be lethal to the immature seeds
(Matyas 1973). Immature cones may fail to expand
fully when kilned and remain closed or semi <• 1 osed ,
thus reducing seed yields (Maki 1940) . Immature
seeds generally do not retain their viability in
storage as well as mature ones (Holmes and
Buszewicz 1958), are usually lower in dry weight,
may germinate slowly, are more susceptible to
disease and produce a higher proportion of
abnormal germinants (Edwards 1980a). Olson and
Silen (1975) concluded that immature Pseudotsusa
menziesii seeds collected around mid-August from a
seed orchard required considerably more work to
extract them from the cones and produced very
light seeds. Nearly all the seeds that were

collected early germinated below 10% and germinant
mortality was high. They estimated that immature
seeds cost 10 times more to produce seedlings than
seeds collected just prior to seedfall.

Maturation indices . --Numerous indices of seed
maturity have been developed (Edwards 1980a) .
Physical indices such as cone color, seed color,
cone moisture content or specific gravity, seed
brittleness, and tissue shrinkage when a cut seed
is exposed to the air for several hours, as well
as chemical indices including the level of fat,
sugar, starch, protein and other constituents in
the seeds have been used as indicators of the
progress of ripening in many tree species,
including broadleaves . Dobbs and others (1976)
suggested that embryos should have extended to
fill at least 75% of the cavity within the endo-
sperm before cone collections are started. Embryo
extension can be easily determined in the field or
it can be recorded on x-ray film (Wang 1973).
Recent experience has shown that early collections
may cause germination problems in the nursery
unless the cones are properly stored to allow seed
ripening to be completed before the seeds are
extracted. For British Columbia species it is now
recommended that cone collections be delayed until
embryo extension is at least 90% complete, and
until the endosperm has changed from a viscous,
milky consistency to a white, firm state
resembling the edible portion of a coconut (Rooke
1985). Embryo maturity is related to summer heat
and in severe climates the best seeds may be found
on southern aspects or on the south-facing side of
the crowns. The utility of heat sums to predict
the Lime of cone collection in Pinus ponderosa has
been discussed by Tanaka and Cameron (1979).
Degree day summations for judging seed maturity
have been used in Scandinavia but are not widely
used in North America (Edwards 1980a) .

Other characteristics that should also be checked
when judging cone ripeness have been compiled by
Wallinger (1983) for west coast conifers. An
illustrated guideline for estimating when to
collect seeds of some eastern species including
Picea glauca, P. mariana, P. punsens and Pinus
banks i ana has been recently produced also (Ontario
Ministry of Natural Resources 1984).

Artificial ripeninR.- Prematurely collected cones,
containing immature seeds, require special care
prior to seed extraction. The seeds of several
conifers, including PseudotsuRa menziesii (Silen
1958), Picea Rlauca (Zasada 1973; Edwards 1980a;
Winston and Haddon 1981,) Larix occidental is
(Shearer 1977), as well as several Abies species
(Rediske and Nicholson 1965; Pfister 1966; Oliver
1974) and the major southern pines (Wakeley 1954)
will continue to ripen in the cones after harvest
if they are properly stored. The conditions for
successful artificial ripening remain ill-defined,
but where success has been obtained in the Pacific
Northwest, air temperatures between 5° and 10°C,
relative humidities of 65%- 75%, and good air
circulation around the cones have all been
implicated (Edwards 1980a). In other words the
cones should be kept cool and well ventilated but

86

should not dry out too quickly. Cones collected 4
to 6 weeks prior to natural seedfall have produced
high quality seeds when artificially ripened for 1
to 2 months. The earlier cones are collected,
however, the more sensitive they are to storage
conditions or, conversely, the more easily seed
quality can be damaged (Zasada 1973). The
provision of adequate artificial ripening facili-
ties in the field raises additional problems, and
expenses, but when large collections are contem-
plated, an early start might make the difference
between sufficient cones being collected and quota
shortfall .

3. Cone Storage in the Field

Even when cones with fully mature seeds have been
harvested, seed quality can be impaired at almost
every subsequent step. Cones may be transported
to the processing plant immediately after harvest,
but more likely they will be assembled at a
shipping depot in the field. This provides an
opportunity to clean them of excessive debris
which, if it is not removed, may complicate later
seed cleaning.

If the cones were wet when sacked, they can be
spread out at the shipping depot and air-dried,
then resacked. Proper interim storage at the
shipping depot allows the cones to "cure," that
is, lose some moisture as they continue to ripen.
For this, individual cone sacks should be exposed
to free movement of air. Additional heating, by
direct sunlight for example, and rewetting must be
avoided. Protection from rodent depredations may
also be required. Portable racks (Stein and
others 1974) or trestles (Dobbs and others 1976)
set up in well- shaded locations, or open sheds,
some of which are portable (Wallinger 1982), can
readily provide the right conditions. Cones of
Abies spp. are particularly susceptible to heat
buildup and molding, especially if moist when
sacked. They should be placed in screen-bottomed
trays at the interim storage shed, or as soon as
they reach the processing plant. Fans may be
required to provide adequate ventilation. Since
the cones of most conifers expand as they lose
moisture, sacks should not be overfilled in the
field, otherwise the scales may acquire a set that
severely impairs seed extraction (Stein and others
1974; Dobbs and others 1976).

4. Cone Transportation

Seeds can be injured during transportation if the
cones are freshly picked, and more so if they were
picked prematurely. Allowing the cones to air dry
at the interim storage depot, for 4 to 6 weeks for
some species, makes transportation conditions less
critical. Even so, cones remain perishable. Even
after sufficient drying, travel times to the
processing plant must be kept to a minimum, and
the cones kept cool and well ventilated. The use
of refrigerated trucks for cone shipment is now
advocated in British Columbia (Johnson 1984) and
for moving tops of trees in Ontario (Horton 1984).
Truck drivers should be informed of the perishable
nature of their loads, the need for proper care,

and prompt delivery of their cargo. Field
collection supervisors should advise the
processing plant in advance of shipments so that
the necessary staff will be available for prompt
unloading of the cones (Dobbs and others 1976).

5. Cone Storage at the Processing Plant

Some of the cones cannot be processed immediately
because of equipment limitations, but in many
cases cones are stored for a further period to
ensure additional drying. Specially designed
sheds are normally used for this purpose (Stein
and others 1974). For most species there is a
period of safe storage (table 7) and seed extrac-
tion schedules need to be prepared accordingly.
Cones of several British Columbia conifers have
been stored for up to 6 months without loss of
seed quality (Leadem 1982), but shorter periods
are preferred since there is a danger of seed
mortality or germination in the cones. For
example, seeds in cones of Tsuga heterophylla and
Thu.j a plicata may germinate before they can be
extracted, so these species should be scheduled
for early processing. Cones of Abies spp., on the
other hand, are often stored for 2 months or
longer, until they have completely disintegrated
and do not have to be kilned. By then the seeds
will be fully ripened.

Table 7. — Safe storage periods in sacks for
cones of western conifers
(Source: British Columbia Ministry of
Forests 1985)

Species Storage period (months)

Abies
Abies
Abies

amabi 1 is

grandis

lasiocarpa

2- 4

Larix

occidentalis

3-5

Picea
Picea
Picea

engelmannii

glauca

sitchensis

3-5

Pinus

contorta

4 +

Pinus
Pinus
Pinus

f lexilis

monticola

ponderosa

3-5

Pseudotsuga menziesii

3-5

Thu.ia plicata

1

Tsuga heterophylla
Tsuga mertensiana

1

b7

CONE AND SEED PROCESSING

Technically, processing includes all treatments
applied from the time the cones arrive at the
processing plant until the seeds are prepared
for sowing in the nursery. For the purposes of
this review, however, only those steps leading
up to cold storage will be discussed.

Upon delivery to the plant, cone sacks are iden-
tified by seedlot and stored until scheduled for
processing which begins with cone drying or
kilning, continues with tumbling and dewinging,
and ends with cleaning and sorting. Depending
upon the species and the methods employed, up to
five stages may be involved in processing
extracted seeds (fig. 1).

1. Kiln Drying

The objective of kilning is to open the cones
quickly without damaging seed viability, so the
heat must be carefully controlled (Rietz 1941;
Carmichael 1958). Usually, the scales of most
cones have already begun to open, but additional
drying is needed to fully flex the cones into an
open position to enable maximum seed recovery.
Cones should be exposed to an air flow of
gradually decreasing moisture content and rising
temperature up to a maximum between 40° and 50°C
for most species near the end of the drying
period. Rapid kilning of cones still high in
moisture content removes moisture from the outer

layers of the scales, which partially flex and
then set in a semi-open position preventing seed
release. This condition is known as "case-
hardening" (Edwards 1981). The risk of case-
hardening is reduced if the cones have been well
cured, i.e., air-dried during storage. In
addition, wet seeds exposed to high kiln tempera-
tures at the beginning of the process would
likely be scalded. If the cones were too tightly
packed in the sacks, full opening of the scales
may not occur even if drying is slow and pro-
longed. In some processing plants cones may be
moved from the storage sheds to a ventilated loft
above the kiln where escaping heat aids the
drying process. This reduces kilning time and
healing costs, and minimizes the chance of seed
damage (Allen 1957). Once again the importance
of proper cone curing, or conditioning, between
the time of harvest and the start of processing
is emphasized; properly handled cones require
less heat so less damage to the seeds is likely,
resulting in a high quality product at lower cost.

Various kiln schedules (Wang 1973; Schopmeyer
1974) have been designed to dry and open cones in
the shortest possible time without damaging seed
viability or causing case-hardening. Most cones
in British Columbia are dried in less than 16
hours (table 8) , but immature cones and those
attacked by insects usually do not completely open
even after extended kilning. When cone scales
have fully flexed the seeds fall out easily during
tumbling and should be removed from the heat as
soon as possible thereafter.

EXTRACTED SEEDS

EZfl

a

SCALPING DEWINGING

DUST SEEDS

Figure 1. — The five main stages of seed processing: scalping, dewinging, drying, sizing and gravity
separation. Some stages may be omitted depending upon the species and methodology. (Diagram based on
the Hilleshog, Sweden, processing system which moves seeds by manual means between the early stages and
pneumatic means between later stages.)

88

The outer scale edges of serotinous cones of Pinus
contorta var. latifolia are bonded together by a
resin-like material, the exact nature of which is
not known (Helium and Wang 1985), although it
melts at temperatures above 45°C (Cameron 1953).
A widely used method for breaking the bond has
been to dip the cones for about one minute in
water at 80°-90°C, after which they are kilned for
several hours at 60°C (table 8). A newer method
of breaking the resin bond uses a scorching device
(MacAuley 1975) that briefly exposes cones to high
temperatures. Using this process, Helium and Wang
(1985) have recommended that Pinus contorta cones
be "cracked open" by using a flash of heat
(210°- 230°C for 1.5 minutes), after which they are
lightly misted with water, then kilned. A cone
moisture content of 15-20% at the start of kilning
is optimal for good seed yields regardless of the
condition of the cones up to that time (Helium
1981; Helium and Barker 1980; Helium and Wang
1985). Most seeds are released within about 6
hours of kilning at 60°C; longer schedules
increase the amount of empty seeds (Helium 1981).

Kilns are essentially over-sized ovens. A basic
type consists of a large chamber equipped with
healers, humidifiers and fans. Cones are spread
evenly on shallow trays that are stacked on
movable carts or dollies. The kiln is loaded with
a batch of cones which are processed for the
prescribed time, removed when cool and a fresh
batch loaded. Such kilns can process large
volumes of cones at one time, but all cones in a
batch are subject to the same kilning conditions.
AnoLher disadvantage is that seeds that fall out
of the cones must remain in the kiln exposed to
the high temperature until the end of the batch
schedule .

In some plants, kilning and tumbling are performed
simultaneously in a rotating drum kiln that
contains, within the heating chamber, a tumbler
that shakes the seeds loose from the cones (T.owman
1975). This type permits loose seeds to be
removed from the heat at the earliest possible
time, but like the chamber kiln, this is also a
batch-type unit. Other kilns provide continuous,
on-line drying by employing a tunnel along which
the temperature progressively rises. Carts laden
with cones on trays move slowly from the cooler
entrance of the tunnel to its hotter exit. A
variation of this type consists of a conveyor belt
on which thinly spread cones move slowly through a
long heater box (Gradi 1973). Various types of
kilns were reviewed by Sziklai (1981). McConnell
(1973) described a small portable kiln designed to
dry small lots of pine cones featuring
economy, safety, portability and versatility.

Any artificial heating involves a fire hazard and
the dust, resin and dry cone scales are particu-
larly inflammable (Morandini 1962; Stein and others
1974). Stringent fire precautions, including a
ban on smoking, should be enforced, and fireproof
construction materials should be used throughout.
Arrangements for removal, by vacuum or other means,
of inflammable dust and debris are essential, and
dust masks should be worn by operator to reduce
the health hazard.

2. Cone Tumbling

Immediately after kilning, cones are tumbled in
horizontally rotating, screened drums or cylinders
to shake the seeds free. In rotating drum kilns
the tumbler is programmed to revolve at intervals
during the drying process so that the seeds can be
removed from the heating chamber at the earliest
possible time . If tumbling is delayed too long
after kilning the open cones may reclose if they
are exposed to moist air (Morandini 1962). During
tumbling, seeds and small-sized debris fall
through the drum screens into a collector or onto
a conveyor belt. While some tumblers are enclosed
at both ends and must be stopped for loading and
unloading (Morandini 1962) , more modern designs
employ an open-ended, inclined cylinder. Cones
are fed in at the higher end and the inclination
of the drum can be altered to control the time the
cones are tumbled (Turnbull 1975) so the operation
can be continuous. Laboratory- sized tumblers for
use in research have been described by Winjum and
Ellis (1960) and Harris (1970).

Tumbling should be as brief and as gentle as
possible. Prolonging the action tends to shake a
higher proportion of poorly developed seeds loose
and increases the amount of debris caused by cone
breakage (Morandini 1962). Speed of rotation and
time of tumbling must be adjusted to the cone and
seed characteristics of the species. In Larix
dec idua Mill, and Picea abies (L.) Karst, for
example, long periods of tumbling may be necessary
since the seeds are frequently tightly held
(Aldhous 1972).

With some species, remoistening the cones after
one tumbling then redrying and additional tumbling
has improved seed yields (Eliason and Heit 1940).
After a first tumbling, Pinus sylvestris cones
were soaked in water at 30°C for some 30 minutes,
until they softened and began to reclose, and were
then air-dried so that the scales opened again.
The yield of seeds from the second tumbling
averaged 36% of that of the first, thereby justi-
fying the extra cost of the process (Van Haverbeke
1976). A large proportion of the seeds removed in
the second tumbling are empty or poorly developed
and difficult to remove in seed sorting. Seeds
processed this way should be kept separate and
used quickly since they may not store well
(Baldwin 1942) .

Some processors use screen-bottomed, tray shakers
about 15-20 cm deep, the seeds falling through
into a catchment bin beneath. This method is
ideal for collections from seed orchards or other
small lots.

3. Scalping

Following their removal from the cones, seeds are
processed to remove unwanted debris, to remove the
membranous wings which greatly increase their bulk
and make nursery handling very difficult, and to
separate empty or otherwise non- viable seeds.

B9

Table 8. — Cone-processing schedules used in British Columbia (Source: British
Columbia Ministry of Forests 1985)

Species

Time in

80°-90°C

water

Cone dryins schedule

Air-
drying
period

Kiln-drying period
Time Temperature

seconds

Abies amabilis
Abies grand is
Abies lasiocarpa

days

60-180
60-180
60-180

hours

6-14
6-14
6-14

29-30
29-30
29-30

Larix laricina
Larix occidentalis

8

7-9

49
43

Picea enselmannii
Picea Rlauca
Picea mar i ana
Picea sitchensis

Pinus albicaulis
Pinus contorta
var . contorta
var . latif olia
Pinus f lexilis
Pinus monticola
Pinus ponderosa

30-60

20-50
20-50

20-50

15-30

2-20
2-30
15-30

6-24
6-24

5- 11

6- 24

96
6-8

0
14

3

38-49
38-49

54
38-49

49
60

43
49

PseudotsuRa menziesii

Coast (var. menziesii)
Interior (var. glauca)

Thu j a plicata

8-21
14-60

2-10
16-48

24-36

32-43
38-43

33

TsuRa heterophylla
Tsuga mertensiana

16-48
16-48

30-43
30-43

Immediately after tumbling the seeds still have
their wings attached, and to separate them from
the debris they are passed over a scalper, which
consists of two or more vibrating, inclined
screens, one above the other, of progressively
finer mesh from the upper to the lower screen
(Lowman 1975) (fig. 2). Seeds are usually
retained on the intermediate screen. "Tappers"
are often used to keep the particles in motion,
while moving brushes under the screens reduce seed
lodging in the perforations. Dust and chaff are
removed by an exhaust hood.

4 . Dewinging

Seedwing removal serves primarily to facilitate
subsequent cleaning and to reduce the volume of
material placed in cold storage and to improve the
field sowing characteristics of the seedlot. In
older processing plants, seed dewinging was the
operational step during which seed damage through
crushing, cracking or abrasion was most likely to
occur (Eliason and Heit, 1940; Allen 1957; Kamra

1967; Wang 1973), but newer equipment has minimized
or eliminated the danger. Excessive dewinging must
be avoided since germination can be seriously
damaged, the seeds seem to be more easily contamin-
ated by mold and the resulting seedlings are weak
(Baldwin 1942) .

Wings are attached to the seeds by different means
in different species; for example, in Psjeudotsuga
menziesii seeds the wing is an integral part of
the seedcoat and must be mechanically broken off,
while in pines and spruces the wings weakly encase
the seeds and can be removed with much less
abrasive action. Seeds of Abies spp. are particu-
larly susceptible to damage to their fragile
seedcoats (Edwards 1982), the reduction in
viability being related to rupturing of the several
resin vesicles (Gunia and Simak 1970; Kitzmiller
and others 1973, 1975). Allen (1957) observed
that seeds of Abies lasiocarpa that were passed
three times through a brush dewinger lost 50% of
their original viability. For Abies concolor
(Gord. and Gledl.) Lindl. and A. masnif ica A. Murr.

90

exTMCTto secos

I MANUAL rCEOIMCI

Figure 2. --Seed scalping to remove particles
larger and smaller than the seeds. The inter-
changeable screens are vibrated electromagneti-
cally. (Diagram based on the the Hilleshbg Co.,
Sweden, equipment.)

seeds, Kitzmiller and others (1975) recommended
the use of a scalper treatment followed by
pneumatic separation as the least damaging method.

Wings are small or impractical to remove from the
seeds of several genera including Thu.j a . Cupressus ,
Chamaecyparis and Libocedrus . Seeds of Juniperus
are wingless, and since they are produced in fleshy
indehiscent strobili, commonly called "berries,"
quite different processing procedures must be
adopted. These methods have been reviewed by
Stein and others (1974) and Johnsen and Alexander
(1974) .

Dewinging methods for other conifers are generally
categorized as either dry or wet.

Dry dewinging.- Dry dewinging involves rubbing the
wings off mechanically, and the simplest and safest
method is to gently hand rub the seeds in a sack,
but this is practical for small quantities only.
One of the simplest mechanical devices is a wire
screen or perforated plate, the holes of which are
large enough to allow the seeds to pass through but
not the wings. A soft brush works the seeds
against the screen while a draft of air draws off
the wing fragments. Many types of mechanized
dewingers, mostly rotating devices, have been
developed (Lowman 1975). These may have brushes,
rotating knobs or paddles which force the seeds
through narrow outlets, breaking off the wings
(Morandini 1962). Distances between the knobs,
paddles, brushes and cylinder wall must be adjusted
so that there is neither too much pressure that
could crack or break the seedcoats nor too much
friction that could cause heat damage. Lovnan and
Casavan (1978) designed a small-lot dewinger

consisting of a rubber- 1 ined , inclined cylinder
with a rotating central shaft to which are attached
pure gum flaps. Another type of dewinger employs
a vertical cylinder inside of which is an auger
that lifts and rubs the seeds against one another.
Some operators add a small amount of coarse debris
to enhance the rubbing action that breaks off the
wings. In British Columbia this type of dewinging
has proven very effective on seeds of Pseudotsuga
menziesii and Larix occ idental is .

Wet dewinging . --The wings of Pinus spp . and Picea
spp. are attached by means of a two-pronged
depression that grips the seed (pine) or by means
of a spoon-shaped hollow partially enclosing the
seed (spruce). Since the wings are more hygro-
scopic than seeds, they expand when wet and loosen
their grip on the seeds from which they can be
separated cleanly with minimal agitation in a
short time (Wang 1973). This is the basis of wet
dewinging, a method formerly believed to impair
seed quality and storability. But only a small
amount of moisture is required, and since the
separation process is quite rapid, the moisture
content of the seeds does not increase markedly.
A jet of air can be employed to blow the wings out
of the mixer and simultaneously begin to redry the
seeds (fig. 3) .

Several types of dewingers are in use. Many are
cement mixers with modified paddles, the slow
rotation of which rubs the wings loose while they
are lightly sprayed with water. Prolonged wetting
should be avoided and it is important that the
seeds be redried promptly to avoid any diminution
of viability. Seeds are usually dried in a thin
layer spread on fine-screened trays through which
warm (<30°C) air is blown (fig. 4). It can also
be carried out in a cabinet-type dryer (Lowman
1975), or a chamber-type cone kiln if it is not in
use. New forced-air seed drying equipment has
been developed recently in Great Britain (Waddell
1984), while a compact, multiple compartment
tumbler drier was described by Leadem and Edwards
(1984) .

5 . Seed Cleaning and Sorting

In some processing plants, seed cleaning is
inseparable from sorting the better grades of seeds
from incompletely developed or empty ones. After
dewinging there may still be some detached wings
among the seeds and further separation is required.
This increases the precision with which the seeds
can be mechanically sown in the nursery.

Three cleaning and sorting systems are commonly
used; screens, air columns and gravity (vibrating)
tables .

Screens . - One of the simplest cleaning systems is
based on particle size, which is the principle of
the scalper described earlier. When this is
combined with an air current so that dust, chaff
and light, empty seeds are blown away from the
oscillating screens, the machine becomes a fanning
mill. Air flow, mesh size, screen inclination,

91

DEWINGING

REMOVING
WINGS

MANUAL
REMOVAL

EMPTYING
SEEDS

Figure 3. — Wet dewinging using a rotating barrel
similar to a cement mixer. Water is added to
swell the hygroscopic wings which release the
seeds. An air jet blows out the loose wings and
begins to redry the seeds. (Diagram based on the
Hilleshog Co., Sweden, equipment.)

PNEUMATIC
CONVEYOR TO
SEED SI2ER

SEED DRYER
TRAYS

DRYING

HEATER
AND FAN

Figure 4. --Seed drying using a tray system
through which warm air is forced. The drying
bench is designed for easy handling of the
trays. (Diagram based on the Hilleshog Co.,
Sweden, equipment.)

distance travelled and rate of oscillation are
all usually adjustable and in some instances the
fanning mill may produce the final product. For
some species, or seedlots, the fanning mill is
only an intermediate step and more precise
cleaning has to be obtained by other means.

Air columns . -These employ a vertical air current.
Separation depends on the relative rate of fall

of seeds with the same surface area but different
weight, or those of uniform weight but different
surface area. Filled seeds sink while empty
seeds are carried away by the rising air current,
the velocity of which can be adjusted to suit the
species. For this method to work, it is important
that the seeds have been completely dewinged,
since, any wing remnants will increase the
proportion of seeds blown away. The South Dakota
blower (Erickson 1944) is one of several devices
(United States Department of Agriculture 1952;
Silen 1964; Hergert and others 1966; Woolard and
Silen 1973; Lowman 1975; Edwards 1979) based on
this principle. The efficiency of air separation
is improved if the seedlot has been previously
sorted into uniform size classes (fig. 5) . After
separation, the filled seeds from all size classes
are recombined (fig. 6).

Gravity tables. --An apparatus originally developed
by the mineral industry to separate ore from clay
and Lo grade ore has been adapted by the seed
industry (Lowman 1975). The specific gravity table
comprises an oscillating, inclined, perforated
deck, the adjustable slant and vibratory motion of
which causes the seeds to move while air is forced
up through the perforations, separating the seeds
into layers, or strata, of different densities.
Heavier particles, such as stones or dried pitch
fragments "track" uphill, while the airstream
"floats" lighter materials down slope. Three
basic rules govern this type of sorting: a) seeds
of the same size but different densities can be
separated, b) seeds of different sizes but the
same densities can be separated, but c) seeds of
different sizes and different densities cannot be
readily separated (Vaughan and others 1968; Thomas
1978) . Movable dividers on the discharge edge of
the table allow the seeds to be divided into a
number of different density fractions. The process
is rapid and efficient (Switzer 1959) and is
continuous as long as the feed hopper contains
seods .

There is little evidence that gravity sorting or
air separation causes seed damage, although either
process might exacerbate injuries caused by other
treatments. Pneumatic separation inflicted some
resin vesicle damage in Abies concolor and A.
maRnif ica seeds but it eliminated impurities and
empty seeds, and was less damaging than other
cleaning methods (Kitzmiller and others 1975).

Other cleaning and sorting methods for tree seeds
have been described, including an inclined belt
(Hergert 1971; Lowman 1975), magnetic, electro-
static, and conductivity devices (Bonner 1978;
Karrfalt and Helmuth 1984), and flotation in
various liquids (Baldwin 1932; McLemore 1965;
Lebrun 1967; Barnett and McLemore 1970; Simak
1973; Barnett 1976). Barnett (1971) and Edwards
(1980b) reported reductions in viability when
organic solvents were used as seed separating
agents. Attempts to use water as the separating
medium were only partially successful (Edwards
1980b) until Simak (1981; 1983; 1984) described
the "IDS" method. In the IDS process, seeds are
soaked in water, then incubated for several days,
then dried. Under the appropriate drying regime,

92

GRAVITY SEPARATION

SPECIFIC GRAVITY SEPARATOR

Figure 5 , --Seed sizing, shown diagrammat ical ly
using a single, vibrating screen with
progressively larger perforations. Several
screens stacked vertically are typical. Kach
seed-size class is then sorted separately .
(Diagram based on the Hilleshbg Co., Sweden,
equipment . )

SEED SIZING

\i/v,6RATORy \V \*/ ♦
crpnpo "V — - . , — . . — .

heavy seeds- all size classes light seeds dust, debris

all'size empty seeds

CLASSES

Figure 6. --Seed sorting using aspirator
separators. Seeds from each size class are fed
into devices from which air is drawn by vacuum.
Air speed through each separator is controllable
so that heavy (filled) seeds fall against the air
stream while lighter seeds are drawn into a
separate chamber. Dust and debris are drawn into
a third chamber. Seeds from all size and weight
classes are recombined. (Diagram based on the
Hilleshog Co., Sweden, equipment.)

viable seeds retain all or most of their moisture,
while dead seeds dry out. This difference in
moisture content causes live seeds to sink, and
dead seeds to float, when the mixture is again
placed in water. Using this method, Pinus contorta
seeds were improved from 67% to 96% germination
capacity, retaining 72% of the original seed bulk
(Simak 1984), and from 85.0% to 92.5%, retaining
91.5% of the original bulk, and from 77% to 96.5%,
retaining 78% of the original bulk (Edwards and
others 1985). IDS processed seeds can be stored
for at least two years (Edwards, unpublished). The
method is being tested on other British Columbia
conifers with a view to developing a procedure for
use on a large scale (Edwards, unpublished).

The efficiency of all seed cleaning and sorting
methods can be checked by periodically cracking or
cutting samples of the processed seeds. X-ray
techniques (Belcher 1973; Eden 1965; Edwards 1982)
are faster and less destructive. Partially filled
seeds should be removed since these are often a
cause of fungal contamination of the seedlot,
especially in pine seeds (Rowan and DeBarr 1974).
After cleaning, seeds must be thoroughly r> i xed to
ensure a homogeneous seedlot, and moisture levels
must be adjusted to ensure that they meet required
standards before they are placed in low tempera-
ture storage.

As noted earlier, processing normally continues
with seed preservation in cold storage until needed
for sowing in the nursery, with germination testing
and, in some seed plants, with preparation of the

seeds for sowing. However, these topics are beyond
the scope of this review. Interested readers will
find that tree seed storage has been well covered
by Holmes and Buszewicz (1958), Stein and others
(1974) and Wang (1974), and it was the subject of
a recent international symposium (Wang and Pitel
1982). Tree seed testing was thoroughly described
by Bonner (1974) and the special needs for the true
firs were detailed by Edwards (1982). Presowing
treatments have been reviewed by Bonner and others
(1974) .

RBFRRRNCBS

Ahlgren, C. E. Phenological observation of

nineteen native tree species in northeastern
Minnesota. Ecol. 38: 622-628; 1957.

Aldhous, J. R. Nursery practice. Bull. no. 43.
London, U.K.: Forestry Commission; 1972:

47 18 .

Allen, G. S. A basis of forecasting seed crops
of some coniferous trees. J. Forest. 39:
1014 1016; 1941a.

Allen, C. S. Douglas-fir seed from young trees.
J. Forest. 40: 722-723; 1941b.

Allen, C. S. Better handling of a scarce

commodity. British Columbia Lumberman 41(7):
32 -36; 1957.

93

Allen, C. S. Factors affecting the viability and
germination behaviour of coniferous seed.
Part I. Cone and seed maturity, Tsuga
heterophylla (Rafn.) Sarg. Forest. Chron.
34: 266-274; 1958a.

Allen, G. S. Factors affecting the viability and
germination behaviour of coniferous seed.
Part II. Cone and seed maturity, Pseudotsufta
menziesii (Mirb.) Franco. Forest. Chron. 34:
275-282; 1958b.

Allen, G. S.; Owens, J. N. The life history of
Douglas-fir. Ottawa, ON; Environment Canada,
Forestry Service; 1972. 139 p.

Bonner, F. T. New developments in seed

processing. In: Belcher, E. W. , compiler and
ed. Small lot forest seed processing
workshop: proceedings; 1977 October 18-20;
Macon, GA. Macon, GA: U.S. Department of
Agriculture, Forest Service; Southeastern
Area, State and Private Forestry; 1978: 19-23.

Bonner, F. T. ; McLemore, B. F. ; Barnett, J. P.
Presowing treatments of seed to speed
germination. In: Schopmeyer, C. S., technical
coordinator. Seeds of woody plants in the
United States. Washington, DC: U.S.
Department of Agriculture, Forest Service,
Agric. Handbook 450: 126-135; 1974.

Baldwin, H. I. Alcohol separation of empty seed
and its effects on the germination of red
spruce. Amer. J. Bot. 19: 1-11; 1932.

Baldwin, H. I. Forest tree seed of the north
temperate region with special reference to
North America. Waltham, MA: Chronica Botanica
Co. ; 1942. 240 p.

Barnett, J. P. Flotation in ethanol reduces

storability of southern pine seed. For. Sci.
17: 50-51; 1971.

Barnett, J. P. Collection and processing methods
affect pine seed germinability . In: Lantz, C.
W. , ed. Southeastern Area [Forest]
Nurserymen's Conferences: proceedings; 1976
August 3-5; Charleston, SC; August 17-19;
Mobile, AL. Washington, DC: U.S. Department
of Agriculture, Forest Service; 1976: 70-72.

Boyer, W. D. Longleaf pine cone production

related to pollen density. In: Kraus, J. , ed.
Seed yield from southern pine seed orchards
colloquium: proceedings; 1973; Macon, GA.
Macon, GA: Georgia Forest Research Council;
1974: 8-14.

Boyer, W. D. Heat accumulation: an easy way to
anticipate the flowering of southern pines.
J. Forest 76: 20-23; 1978.

Bramlelt, D. L. Cone crop development for six
years in shortleaf-pine. For. Sci. 18: 31-33;
1972.

Buszewicz, G; Holmes, G. D. Preliminary trials
of a quick method for estimating the seed
content of cones. Rep. For. Res. for year
ending Mar. 1956. London, U.K.: Forestry
Commission: 98-106; 1956.

Barnett, J. P.; McLemore, B. F. Storing southern
pine seeds. J. Forest. 68: 24-27; 1970.

Calvert, R. F. A method to evaluate white spruce
cone crops. Forest. Chron. 54: 88-91; 1978.

Baron, F. J. California cone crop-1963. Res.

Note PSW-25. Berkeley, CA: U.S. Department of
Agriculture, Forest Service, Pacific
Southwest Forest and Range Experiment
Station; 1963. 11 p.

Baron, F. J. Ten years of forest seed crops in
California. J. Forest 67: 490-492; 1969.

Barras, S. J.; Norris, D. M. Effect of a late
May freeze on red pine ovulate cone
development. Forest. Res. Note No. 115.
Madison, WI: College of Agricultural and Life
Sciences; 1964. 3 p.

Belcher, E. W. Radiography in tree seed analysis
has new twist. Washington, DC: U.S.
Department of Agriculture, Forest Service,
Tree Planters' Notes 24(3): 1-5; 1973.

Bingham, R. T.; Rehfeldt, G. E. Cone and seed
yields in young western white pines. Res.
Pap. INT-79. Ogden, UT: U.S. Department of
Agriculture, Forest Service, Intermountain
Forest and Range Experiment Station; 1970. 12 p.

Bonner, F. T. Seed testing. In: Schopmeyer, C. S.,
technical coordinator. Seeds of woody plants
in the United States. Washington, DC: U.S.
Department of Agriculture, Forest Service,
Agric. Handbook 450: 136-152; 1974.

Calvert, R. F. Forecasting seed crops of forest
trees. In: Morgenstern, E.K., Carlson, L. W. ,
eds. National workshop on tree seed
production and tree improvement, in
Canada- Research and Development needs
1977-1987: proceedings; 1978 April 17-20;
Chalk River, ON. Chalk River, ON: Environment
Canada, Forestry Service, Petawawa Forest
Experiment Station, Info. Rep. PS-X-74:
11-30; 1979.

Calvert, R. F. Operational collection of spruce
seed in Ontario. In: Zsuffa, L. ; Rauter, R.
M. ; Yeatman, C. W. , eds. Clonal forestry: its
impact on tree improvement and our future
forests symposium: proceedings; 1983 August
22-26; Toronto, ON. 19th Meeting Canadian
Tree Improvement Association, Part TI.
Ottawa, ON: Environment Canada, Forestry
Service; 1985: 188-192.

Cameron, H. Melting point of the bonding
material in lodgepole pine and jack pine
cones. Silvic. Leaflet No. 86. Ottawa, ON:
Canada Department Resources and Development,
Forestry Branch, Division Forest Research:
1953. 2 p.

Canadian Forestry Service. Reproduction of
conifers. A handbook for cone crop

94

assessment. Forest. Tech. Rep. 31. Ottawa,
ON: Environment Canada, Forestry Service;
1983. 46 p.

Carmichael, A. J. Determination of the maximum
air temperature tolerated by red pine, jack
pine, white spruce and black spruce seeds at
low relative humidities. Forest Chron. 34:
387-392; 1958.

Cayford, J. H. Redpine seedfall in southern
Manitoba. Forest. Chron. 40: 78-85; 1964.

Ching, T. M. Seed production from individual

cones of grand fir (Abies grandis Lindl . ) . J.
Forest. 58: 959-961; 1960.

Chira, E. One of the reasons for poor seed yield
from cones of larch and Douglas-fir. Les
(Bratislava) 23: 211-212; 1967. (Forestry
Abstracts 28, 5767; 1967.)

Cornell, B. L. Appraisal Fandrich cone rake
Stanislaus N.F. Memorandum to Regional
Forester, R-5 . Sonora, CA: U.S. Department of
Agriculture, Forest Service; 1984. 12 p.
Unpubl ished .

Crocker, T. C. Longleaf pine cone production in
relation to site index, stand age and stand
density. Res. Note SO-156. New Orleans, LA:
U.S. Department of Agriculture, Forest
Service, Southern Forest Experiment Station;
1973. 3 p.

Crossley, D. I. Fruiting habits of lodgepole
pine. Tech. Note 35. Ottawa, ON: Canada
Department Northern Affairs and Natural
Resources, Forestry Branch, Forest Research
Division; 1956. 32 p.

Daubenmire, R. A seven-year study of cone
production as related to xylem layers and
temperature in Pinus ponderosa . Amer. Midland
Nat. 64: 187-193; 1960.

DeBarr, G. L . ; Ebel , B. H. Conelet abortion and
seed damage of shortleaf and loblolly pines
by a seedbug, Leptoglossus corculus . For.
Sci. 20: 165-170; 1974.

DeBarr, G. L.; Kormanik, P. P. Anatomical basis
for conelet abortion on Pinus echinata
following feeding by Leptoglossus corculus
(Say). Can. Entomol. 107: 81-86; 1975.

Dobbs, R. C; Edwards, D. G. W.; Konishi, J.;
Wallinger, D. Guideline to collecting cones
of B.C. conifers. Joint Rep. No. 3. Victoria,
BC: British Columbia Forest Service/Canadian
Forestry Service; 1976. 98 p.

Dobbs, R. C; Silversides, C. E . ; Walters, J.
Development and evaluation of an aerial cone
rake. Info. Rep. FMR-X-100. Ottawa, ON:
Environment Canada, Forest Management
Institute; 1977. 17 p.

Dyer, W. Economics and mechanics of seed supply.

Pulp and Paper Mag. of Canada 69(3): 92-93; 1968.

Ebell, L. F. Cone production induced by drought
in potted Douglas fir. Ottawa, ON: Canada
Department of Forestry and Rural Development,
Bi-monthly Res. Notes 23: 26-27; 1967.

Eden, C. J. Use of x-ray technique for

determining sound seed. Washington, DC: U.S.
Department of Agriculture, Forest Service,
Tree Planters' Notes 72: 25-28; 1965.

Edwards, D. G. W. An improved air seed- sorter
for laboratory use. Info. Rep. BC-X-188.
Victoria, BC: Environment Canada, Forest
Service, Pacific Forest Research Centre;
1979a. 11 p.

Edwards, D. G. W. Maturity and quality of tree
seeds - a state-of-the-art review. Seed Sci.
and Technol.8: 625-657; 1980a.

Edwards, D. G. W. Tree seed research, Pacific
Forest Research Centre, B.C., 1977-1979. In:
Yeatman, C. W., ed . 17th Meeting Canadian
Tree Improvement Association, Part I:
proceedings; 1979 August 27-30; Gander, NF .
Ottawa, ON: Environment Canada, Forestry
Service; 1980b: 231-236.

Edwards, D. G. W. Cone collection and processing
effects on seed quality and yield. In: Huher,
R. F., compiler. High-quality collection and
production of conifer seed workshop:
proceedings; 1979 November 14; Edmonton, AB.
Edmonton, AB: Environment Canada, Forestry
Service, Northern Forest Research Centre,
Info. Rep. NOR-X-235; 1981: 12-37.

Edwards, D. G. W. Collection, processing,
testing and storage of true fir seeds - a
review. In: Oliver, C. D.; Kenady, R. M. ,
eds . Biology and management of true fir in
the Pacific Northwest symposium: proceedings;
1981 February 24-26; Seattle, WA. Seattle,
WA; University of Washington, College of
Forest Resources, Institute of Forest
Resources, Contribution No. 45; 1982: 113-137.

Edwards, D. G. W.; Taylor, D. W.J McKendry, E.

C. Enhancing seedlot quality in lodgepole
pine by flotation sorting. In: Baumgartner,

D. M.; Krebill, R. G.; Arnott, J. T.;
Weetman, G. F., compilers and eds. Lodgepole
pine, the species and its management
symposium: proceedings; 1984 May 8- 10;
Spokane, WA ; 1984 May 14 16 ; Vancouver, BC .
Pullman, WA: Washington State University,
Cooperative Extension; 1985: 375-376.

Eis, S. Cone crops of white and black spruce are
predictable. Forest Chron. 43: 247-252; 1967.

Eis, S. Cone production of Douglas-fir and grand
fir and its climatic requirements. Can. J.
For. Res. 3: 61-79; 1973a.

Eis, S. Predicting white spruce cone crops. Rep.
BC-P 7. Victoria, BC: Environment Canada,
Forestry Service, Pacific Forest Research
Centre: 1973b. 4p .

^5

Eis, S. Association of western white pine cone
crops with weather variables. Can. J. For.
Res. 6: 6-12; 1976.

Eis, S.; Gat-man, E. H. ; Ebell, L. F. Relation
between cone production and diameter
increment of Douglas-fir (Pseudotsuga
menziesii (Mirb.) Franco), grand fir (Abies
grandis (Dougl.) Lindl . ) and western white
pine (Pinus monticola DourI . ) . Can. J. Bot.
43: 153-159; 1965.

Eis, S; Inkster, J. White spruce cone production
and prediction of cone crops. Can. J. For.
Res. 2: 460-466; 1972.

Eliason, E. J.; Heit, C. E. The results of
laboratory tests as applied to large scale
extraction of red pine seed. J. Forest. 8:
426-429; 1940.

Erickson, E. L. The South Dakota seed blower.
In: Munn, M. P., ed. 35th Annual Meeting
Association Official Seed Analysts 1943/44:
proceedings; 1944 July 17-21; Columbus, OH.
Geneva, MY: Association Official Seed
Analysts; 1944: 91-95.

Fechner, G. H. The biology of flowering and
fertilization. In: Bonner, F. T. ; Krugman,
S.; Yates, H.O. Ill; Turner, B. J., eds.
Flowering and seed development in trees
symposium: proceedings; 1978 May 15-18;
Starkville, MS. Starkville, MS: U.S.
Department of Agriculture, Forest Service;
1979: 1-24.

Garman, E. H. Seed production by conifers in the
coastal region of British Columbia related to
dissemination and regeneration. Tech. Pub. T.
35. Victoria, BC: Department of Lands and
Forests, British Columbia Forest Service;
1951. 47 p.

Gifford, E. M.; Mirov, N . T. Initiation and
ontogeny of the ovulate strobilus in
ponderosa pine. For. Sci. 6: 19-25; 1960.

Goddard, R. E. Additional data on cost of

collecting cones from a seed production area.
J. Forest. 56: 846-848; 1958.

Gradi , A. New techniques and progress in

processes of extraction of forest seeds. In:
Seed problems, international symposium on
seed processing: proceedings; 1973; Bergen,
Norway. International Union Forest Research
Organizations, Working Party S2.01.06.
Stockholm, Sweden: Royal College of Forestry;
Volume I, Paper No. 7; 1973. 14 p.

Grisez, T.J. Flowering and seed production in
seven hardwood species. Res. Note NE-315.
Upper Darby, PA: U.S. Department of
Agriculture, Forest Service, Northeastern
Forest Experiment Station; 1975. 8 p.

Gunia, S.; Simak, M. Effect of damaging resin
vesicles in the seedcoat on the germination
of silver fir (Abies alba Mill.) seeds. In:
Seed physiology of woody plants symposium:
proceedings; 1970; Kornik, Poland; 1970:
79-83.

Finnis, J. M. A note on the bud count method of
forecasting cone crops of Douglas-fir.
Forest. Chron. 29: 122-127; 1953.

Fowells, H. A. An index of ripeness for sugar
pine seed. Res. Note No. 64. Berkeley, CA:
U.S. Department of Agriculture, Forest
Service, California Forest and Range
Experimental Station; 1949. 5 p.

Fowells, H. A.; Schubert, G. H. Seed crops of
forest trees in the pine region of
California. Tech. Bull. 1150. Washington, DC:
U.S. Department of Agriculture; 1956. 48 p.

Franklin, J. F. Cone production by upper slope
conifers. Res. Pap. PNW-60. Portland, OR:
U.S. Department of Agriculture, Forest
Service, Pacific Northwest Forest and Range
Experiment Station; 1968. 21 p.

Fraser, D. A. The relation of environmental

factors to flowering in spruce. In: Thimann,
K. V., ed. The physiology of forest trees.
New York, NY: Ronald Press Co.; 1958: 629-641.

Garber, R. E. Natural seedfall in white pine

(Pinus strobus L.) stands of varying density.
Res. Note NE-119. Upper Darby, PA: U.S.
Department of Agriculture, Forest Service,
Northeastern Forest Experiment Station; 1970.
6 p.

Haig, I. T.; Davis, K. P.; Weidman, R. H.
Natural regeneration in the western white
pine type. Tech. Bull. No. 767. Washington,
DC: U.S. Department of Agriculture; 1941. 99 p

Haig, R. A. Trials of mechanical collection of
black spruce and jack pine cones. Info. Rep.
0-X-112. Sault Ste. Marie, ON: Department of
Fisheries and Forestry, Canadian Forestry
Service, Forest Research Laboratory Ontario
Region; 1969. 15 p.

Hard, J. S. Frost damage to red pine conelets.
Res. Note LS-5. St. Paul, MN. U.S. Department
of Agriculture, Forest Service, Lake States
Forest Experiment Station; 1963. 2 p.

Harris, S. F. A compact laboratory seed

extractor. Washington, DC: U.S. Department of
Agriculture, Forest Service, Tree Planters'
Notes 21(3): 8-9; 1970.

Hedin, I. B. Aerial cone collection techniques
in British Columbia. Tech. Note TN-69.
Vancouver, BC: Forest Engineering Institute of
Canada; 1983. 55 p.

Hergert, G.: Zillinsky, F. J.; Kemp, J. K. An

aspirator for cleaning small seed samples. Can
J. Plant Sci. 46: 570-572; 1966.

Hergert, G. B. ; Wang, B. S. P.; Yeatman, C. W. A
belt cleaner for small samples of tree seed.
Forest. Chron. 47: 40-41: 1971.

96

Helium, A. K . Lodgepole pine seed extraction.
In, Huber, R. F. , compiler. High-quality
collection and production of conifer seed
workshop: proceedings; 1979 November 14;
Edmonton, AB. Edmonton, AB: Environment Canada,
Forestry Service, Northern Forest Research
Centre. Info. Rep. NOR-X-235; 1981: 38-53.

Helium, A. K . ; Barker, N. A. Cone moisture
content influences seed release in
lodgepole pine. Can. J. For. Res. 10: 239-244;
1980.

Helium, A. K . ; Wang, B. S. P. Lodgepole pine

seed: seed characteristics, handling and use.
In: Baumgartner, D. M.: Krebill, R. G.; Arnott,
J. T.; Weetman, G. F., compilers and eds .
Lodgepole pine, the species and its management
symposium: proceedings; 1984 May 8-10; Spokane,
WA; 1984 May 14-16; Vancouver, BC . Pullman, WA:
Washington State University, Cooperative
Extension; 1985: 187-197.

Holmes, G. D.; Buszewicz, G. The storage of seed
of temperate forest tree species. Forestry
Abstracts 19: 313-322, 455 476; 1958.

Holmes, T. R. Management possibilities and some
results for coastal seed production areas.
Vancouver, BC: University of British Columbia;
1972. 35 p. BSF Thesis.

Horton, K. W. Mechanizing cone collection of
black spruce in Ontario. Joint Rep. No. 2.
Sault Ste. Marie, ON: Gt . Lakes Forest Research
Centre/Ontario Ministry of Natural Resources;
1984 . 16 p.

Johnsen, T. N. ; Alexander, R. A. Juniperus L.
Juniper. In: Schopmeyer, C. S., technical
coordinator. Seeds of woody plants in the
United States. Washington, DC: U.S. Department
of Agriculture, Forest Service, Agric . Handbook
450: 460 469; 1974.

Johnson, C. M. Personal communication. Victoria,
BC: British Columbia Ministry of Forests,
Silviculture Branch; 1984 October 26.

Kamra, S. K. Studies on storage of mechanically
damaged seed of Scots pine (Pinus sy lvestris
L.). Stud. For. Suecica 42: 1 18; 1967.

Karrfalt, R. P.; Helmuth, R. E. Preliminary

trials on upgrading Platinus occ idental is with
the Helmuth electrostatic seed separator. Tn:
Murphy, P. M., compiler. The challenge of
producing native plants for the Intermountai n
area conference: proceedings: Intermountain
Nurseryman's Association; 1983 August 8-11; Las
Vegas, NV. General Technical Report THT-168.
Ogden , UT: U.S. Department of Agriculture,
Forest Service, Intermountain Forest and Range
Experiment Station; 1984: 79-81.

Kitzmiller, J. M.; Battigin, J. M.; Helms, J. A.
Effect of resin vesicle damage on gerrni nat i on
of stored Abies concolor seed. Internal Rep.
No. 1. Berkeley, CA: True Fir Management

Cooperative, School of Forestry and
Conservation, University of California; 1973.
16 p.

Kitzmiller, J. M.; Helms, J. A.; Stover, P.

Influence of seed treatment on germination of
Abies concolor and Abies magnif ica. Internal
Rep. No. 3. Berkeley, CA: Industry/University
True Fir Management Cooperative, School of
Forestry and Conservation, University of
California; 1975. 39 p.

Kormanik, P. P. Conelet abortion in pines. In:
Kraus, J., ed. Seed yield from southern pine
seed orchards colloquium: proceedings; 1973;
Macon, GA. Macon, GA: Georgia Forest Research
Council; 1974: 42-48.

Krugman , S.L.; Jenkinson, J.S. Pinus L. In:

Schopmeyer, C.S., technical coordinator. Seeds
of woody plants in the United States.
Washington, DC: U.S. Department of Agriculture,
Forest Service, Agric. Handbook 450: 598-638;
1974.

LaBastide, J. G. A.; van Vredenburch, C. L. H.

The influence of weather conditions on the seed
production of some forest trees in the
Netherlands. Med. St. Bosbouwproef st .
Dorschkamp Wageningen Nr. 120: 1-12; 1970.

Larson, M. M.; Schubert, G. H. Cone crops of
ponderosa pine in central Arizona, including
the influence of Abert squirrels. Res. Pap. No.
RM-58. Fort Collins, Co: U.S. Department of
Agriculture, Forest Service, Rocky Mountain
Forest and Range Experiment Station; 1970. 15 p.

Larson, P. R. Influences of dates of flushing on
flowering in Pinus banks i ana . Nature 192:
82 83; 1961.

Lavender, D. R. ; Engstrom, W. H. Viability of

seeds from squirrel- cut Douglas- fir cones. Res.
Note No. 27. Corvallis, OR: Oregon State Board
of Forestry; 1956. 19 p.

Leadem, C. L. Seed viability of Abies , Picea and
Tsuga after storage in the cones. In: Wang, B.
S. P.; Pitel, J. A., compilers and eds. Forest
tree seed storage international symposium:
proceedings; 1980 September 23-27; Chalk River,
ON; International Union Forest Research
Organizations, Working Party S2.01.06. Ottawa,
ON: Environment Canada, Forestry Service; 1982:
57-67.

Leadem, C. L.; Edwards, D. G. W. A

multiple compartment tumbler- drier for tree
seeds. Washington, DC: U.S. Department of
Agriculture, Forest Service, Tree Planters'
Notes 35(3): 23-25; 1984.

Lebrun, C. (Separation of full and empty seeds
by spec if ic- gravity measurement through
immersion in liquids.) Rev. For. Franc. 19:
786-789; 1967.

Lee, K . J.; Hollis, C. A.; Goddard, R. E.

Strobilus initiation in slash pine in relation

97

to shoot morphology and physiology of the
terminal bud. In: Bonner, F. T. ; Krugman, S . ;
Yates, H. 0. Ill; Turner, B. J., eds . Flowering
and seed development in trees symposium:
proceedings; 1978 May 15-18; Starkville, MS.
Starkville, MS: U.S. Department of Agriculture,
Forest Service; 1979: 359.

Lester, D. T. Variation in cone production of
red pine in relation to weather. Can. J. Bot .
45: 1583-1691; 1967.

Linhart, Y. B. ; Mitton, J. B.; Bowman, D. M. :
Sturgeon, K. B. ; Hamrick, J. L. Some genetic
consequences of differential reproduction in
populations of forest trees. In: Bonner, F. T. ;
Krugman, S.; Yates, H. 0. Ill; Turner, B. J.,
eds. Flowering and seed development in trees
symposium: proceedings; 1978 May 15-18;
Starkville, MS. Starkville, MS: U.S. Department
of Agriculture, Forest Service; 1979: 41-50.

Lotan, J. E. ; Jensen, C. E. Estimating seed

stored in serotinous cones of lodgepole pine.
Res. Pap. INT-83 . Ogden , UT: U.S. Department of
Agriculture, Forest Service, Intermountain
Forest and Range Experiment Station; 1970. 10 p.

Lowman, B. J. Catalog-equipment for processing
small seed lots. Missoula, MT: U.S. Department
of Agriculture, Forest Service, Equipment
Development Center; 1975. 57 p.

Lowman, B. J.; Casavan, K. Dewinger for small
seed lots. Washington, DC: U.S. Depart "en t of
Agriculture, Forest Service, Tree Planters'
Notes 29(4): 21-23; 1978.

Lowry, W. P. Apparent meteorological

requirements for abundant cone crop in
Douglas-fir. For. Sci. 12: 185-192; 1966.

MacAuley, J. D. Mechanical seed extraction of
lodgepole pine. Vancouver, BC: University of
British Columbia: 1975. 172 p. Ph.D.
Dissertation, unpublished.

Maki , T. E. Significance and applicability of
seed maturity indices for ponderosa pine. J.
Forest. 38: 55-60; 1940.

Maguire, w. P. Are ponderosa pine crops

predictable? J. Forest 54: 778-779; 1956.

Matthews, J. D. Factors affecting the production
of seed by forest trees. Forestry Abstracts 24:
i-xiii; 1963.

Matyas, C. Handling of autumn harvested cones in
Scotch pine seed orchards. In: Seed problems,
international symposium on seed processing:
proceedings; 1973; Bergen, Norway. Interna-
tional Union Forest Research Organizations,
Working Party S2.01.06. Stockholm, Sweden:
Royal College of Forestry; Volume I, Paper No.
13; 1973. 11 p.

Maurin, A; Liepa, I; Drike, A. (Mathematical

model for predicting seed production by Tsuga
canadensis and Picea omorika. ) Uc . Zap. Latv.

Univ. 122: 21-36; 1970. (Forestry Abstracts 32,
2392; 1971.)

McConnell, J. L. A portable kiln for drying

small lots of pine cones. In: Seed problems,
international symposium on seed processing;
proceedings; 1973; Bergen, Norway. Interna-
tional Union Forest Research Organizations,
Working Party S2.01.06. Stockholm, Sweden:
Royal College of Forestry; Volume I, Paper No.
11; 1973. 8 p.

McLemore, B. F. Predicting seed yields of

southern pine cones. J. Forest. 60: 639-641;

1962.

McLemore, B. F. Pentane flotation for separating
full and empty longleaf pine seed. For. Sci.
11: 242-243; 1965.

McWilliams, H. G. Collecting seed from forest
trees. Victoria, BC: British Columbia Forest
Service, Publication B.41; 1950. 17 p.

Meagher, M. Surveys for seed crops. Victoria, BC:
British Columbia Forest Service, Reforestation
Notes 1(1): 16-19; 1974.

Medvedev, A. N. ; Pal'gov, I. A. (Predicting seed
production in Picea schrenkiana by reproductive
bud counts). Auyl. Sharuash. Gylym. Khabar 1:
85-89; 1971. (Forestry Abstracts 33, 544; 1972).

Miscenko, B. P. (Prediction of seed yields in
Abies sibirica.) Lesn. Hoz. 16: 28-29; 1963.
(Forestry Abstracts 25, 3435; 1964).

Morandini, R. Forest seed handling, equipment

and procedures: I. Seed production, collection
and extraction. Unasylva 15(4): 185-199; 1962.

Morris, R. F. The effects of flowering on the
foliage production and growth of balsam fir.
Forest Chron. 27: 40-5 7; 1951.

Neel, W. W. ; DeBarr, G. L.; Lambert, W. E.

Variability of seedling damage to pine seed
following different methods of applying
carbofuran granules to a slash pine seed
orchard. In: Bonner, F. T. ; Krugman, S; Yates,
H. 0. Ill; Turner, B. J., eds. Flowering and
seed development in trees symposium:
proceedings; 1978 May 15-18; Starkville, MS.
Starkville, MS: U.S. Department of Agriculture,
Forest Service; 1979: 314-322.

Nienstaedt, H. Receptivity of female strobili of
white spruce. For. Sci. 4: 110-115; 1958.

Obydennikov, A. I. (Trial in predicting good
seed years of Amur larch.) Rast. Resursy
(Leningrad) 5: 77-82; 1969. (Forestry Abstracts
30, 5553; 1969.)

Oliver, W. W. Seed maturity in white fir and red
fir. Res. Pap. PSW-99. Berkeley, CA: U.S.
Department of Agriculture, Forest Service,
Pacific Southwest Forest Experiment Station;
1974. 12 p.

98

Olson, D. L. ; Silen, R. R. Influence of date of
cone collection on Douglas fir seed processing
and germination: a case history. Res. Pap.
PNW 190. Portland, OR: U.S. Department of
Agriculture, Forest Service, Pacific Northwest
Forest and Range Experiment Station; 1975. 10 p.

Ontario Ministry of Natural Resources.

Guidelines for tree seed crop forecasting.
Toronto, ON: Ontario Ministry of Natural
Resources; 1984. 141 p.

Orr-Ewing, A. L. Inbreeding experiments with the
Douglas fir. Forest. Chron . 30: 7-20; 1954.

Owens, J. N. The relative importance of

iniliation and early development on cone
production in Douglas-fir. Can. J. Bot. 47:
1039-1049; 1969.

Owens, J. N. The reproductive cycle of Douglas-fir.
Publication BC-P-8. Victoria, BC: Environment
Canada, Forestry Service, Pacific Forest
Research Centre; 1973. 23 p.

Owens, J. N. ; Holder, M. The reproductive cycles
of western and mountain hemlock. Victoria, BC:
British Columbia Ministry of Forests, Research
Branch; 1984a. 34 p.

Owens, J. N. ; Holder, M. The reproductive cycle
of lodgepole pine. Victoria, BC: British
Columbia Ministry of Forests, Research Branch;
1984b. 29 p.

Owens, J. N.; Holder, H. The reproductive cycles
of western red cedar and yellow-cedar.
Victoria, BC: British Columbia Kini.Ht.ry of
Forests, Research Branch; 1984c. 28 p.

Owens, J. N. ; Holder, H. The reproductive cycle
of interior spruce. Victoria, BC: British
Columbia Hinistry of Forests, Research Branch;
1984d. 30 p.

Pfister, R. D. Artificial ripening of grand fir
cones. Northwest Science 40: 103-112; 1966.

Pitcher, J. A. Cone and seed yields in white
spruce seed production areas. In: Second
genetics workshop of Society of American
Foresters and seventh Lake States forest tree
improvement conference: proceedings; 1965
October 21-23. Res. Pap. NC-6 . St. Paul, MN:
U.S. Department of Agriculture, Forest Service,
North Central Forest Experiment Station; 1966:
76- 77 .

Pollock, B. M. ; Roos , E. E. Seed and seedling

vigor. In: Kozlowski, T. T., ed . Seed biology.
Vol. I. Importance, development, germination.
New York, NY: Academic Press Inc.; 1972:
313- 387 .

Powell, G. R. Patterns of development in Abies
balsamea crowns and effects of megastrobilus
production on shoots and buds. Can. J. For.
Res. 7: 498-509; 1977.

Rediske, J. H . ; Nicholson, D. C. Haturation of
noble fir seed-a biochemical study. Forest.
Pap. 2. Centralia, WN: Weyerhaeuser Co.; 1965.
15 p.

Rehfeldt, G. E . ; Stage, A. R.; Bingham, R. T.
Strobili development in western white pine:
periodicity, prediction and association with
weather. For. Sci. 17: 454-461; 1971.

Richardson, B. Y. Shaking trees to collect pine
cones. Southern Lumberman 215: 140-141: 1967.

Rietz, R. C. Kiln design and development of
schedules for extracting seed from cones.
Tech. Bull. No. 773. Washington, DC: U.S.
Department of Agriculture; 1941. 70 p.

Roe, A. L. A procedure for forecasting western
larch seed crops. Res. Note INT-49. Ogden, UT:
U.S. Department of Agriculture, Forest Service,
Intermountain Forest and Range Experiment
Station; 1966. 7 p.

Roe, A. L. Seed dispersal in a bumper spruce
seed year. Res. Pap. INT- 39. Ogden, UT: U.S.
Department of Agriculture, Forest Service,
Intermountain Forest and Range Experiment
Station; 1967. 10 p.

Roe, E. I. August- collected cones yield poor red
pine seed. Tech. Note 575. St. Paul, HN: U.S.
Department of Agriculture, Forest Service, Lake
States Forest Experiment Station; 1960. 2 p.

Roe, E. I. Seed stored in cones of some jack
pine stands, northern Hinnesota. Res. Pap.
LSI. St. Paul, HN: U.S. Department of
Agriculture, Forest Service, Lake States Forest
Experiment Station; 1963. 14 p.

Roe, E. I. Heavy crop of red pine cones yields
many thousands of good seeds. Res. Note LS-36.
St. Paul, HN: U.S. Department of Agriculture,
Forest Service, Lake States Forest Experiment
Station; 1964. 4 p.

Roeser, J. Some aspects of flower and cone

production in ponderosa pine. J. Forest. 39:
534-536; 1941.

Roeser, J. The influence of climate on seed
production in Douglas- fir. J. Forest. 40:
304-307; 1942.

Rooke, H. Personal communication. Duncan, BC:
British Columbia Hinistry of Forests,
Silviculture Branch; 1985 June 20.

Rowan, S. J.; DeBarr, G. L. Holdy seed and poor
germination linked to seedbug damage in slash
pine. Washington, DC: U.S. Department of
Agriculture, Forest Service, Tree Planters'
Notes 25(1): 25-27; 1974.

Rowe, J.S. Factors affecting white spruce

reproduction in Hanitoba and Saskatchewan.
Tech. Note No. 3. Ottawa, ON: Canada Department
of Norlhern Affairs and National Resources,
Forestry Branch, Forest Research Division;
1955. 27 p.

99

Rudolph, P. 0.; Dorman, K. W. ; Hitt, R. G.;
Plummer, A. P. Production of genetically
improved seed. In: Schopmeyer, C. S., technical
coordinator. Seeds of woody plants in the
United States. Washington, DC: U.S. Department
of Agriculture, Forest Service, Agric. Handbook
450: 53-74; 1974.

Sarvas, R. Studies on the seed setting of Norway
spruce. Medd. Norske Skogf orsdksv . 14:
529-536; 1957. (Forestry Abstracts 19, 1628;
1958. )

Sarvas, R. Investigations on the flowering and
seed crop of Pinus sylvestris . Commun . Tnst.
Forest Fenn. 53.4; 1962. 198 p.

Sarvas, R. Investigations on the flowering and seed
crop of Picea abies . Commun. Inst. Forest.
Fenn. 67.5; 1968. 84 p.

Schopmeyer, C. S., technical coordinator. Seeds
of woody plants in the United States. Agric.
Handbook 450. Washington, DC: U.S. Department
of Agriculture, Forest Service; 1974. 883 p.

Schoenike, R. E. 1955. Why pine seed crops fail.
Forest Farmer 14: 10; 1955.

Schubert, G. M. ; Pitcher, J. A. A provisional
tree seed zone and cone crop

rating system for Arizona and Mew Mexico. Res.
Pap. RM-105. Fort Collins, CO: U.S. Department
of Agriculture, Forest Service, Rocky Mountain
Forest and Range Experiment Station: 1973. 8 p.

Seal, D. T.; Matthews, J. D. ; Wheeler, R. T.
Collecting cones from standing trees. Forest
Record No. 39. London, UK: Forestry Comrni ssion;
196S. 45 p.

Sharp, W.M. Evaluating mast yields in the oaks.
Bull. 635. University Park, PA: Pennsylvania
Stale University Agricultural Experiment
Station; 1958. 22 p.

Shearer, R. C. Maturation of western larch cones
and seeds. Res. Pap. INT-189. Ogden, UT: U.S.
Department of Agriculture, Forest Service,
Intermountain Forest and Range Experiment
Station; 19/7. 15 p.

Shearer, R. C; Schmidt, W. C. Ponderosa pine
cone and seed losses. J. Forest. 69: 3/0-372;
1971.

Silen, R. R. Artificial ripening of Douglas-fir
cones. J. Forest. 56: 410-413; 1958.

Silen, R. R. Pollen dispersal considerations for
Douglas fir. J. Forest. 60: 790-795; 1962.

Silen, R. R. A laboratory seed separator. For.
Sci. 10: 222-223; 1964.

Silen, R. R. Earlier forecasting of Douglas-fir
cone, crops using male buds. J. Forest. 65:
888 892; 1967.

Simak, M. Separation of forest seed through
flotation. In: Seed problems, international
symposium on seed processing: proceedings;
1973; Bergen, Norway. International Union
Forest Research Organizations, Working Party
S2.01.06. Stockholm, Sweden: Royal College of
Forestry; Volume I, Paper No. 16; 1973. 21 p.

Simak, M. (Removal of filled-dead seeds from a

seed bulk.) Sveriges Skogsvardsf orbunds Tidskr.
5: 31-36; 1981.

Simak, M. A new method for the improvement of
the quality of Pinus contorta seeds. In:
Murray, M. , ed. Lodgepole pine: regenerat i on
and management workshop: proceedings; 1982
August 17-19; Hinton, AB. Portland, OR: U.S.
Department of Agriculture, Forest Service,
Pacific Northwest Forest and Range Experiment
Station; 1983: 39-41.

Simak, M. A method for removal of filled-dead
seeds from a sample of Pinus contorta . Seed
Sci. and Technol . 12: 767-775; 1984.

Slay ton, S. H. A new technique for cone

collection. Washington, DC: U.S. Department of
Agriculture, Forest Service, Tree Planters'
Notes 20(3): 13; 1969.

Snyder, E. B.; Squillace, A. E. Cone and seed
yields from controlled breeding of southern
pines. Res. Pap. S0-22. New Orleans, LA: U.S.
Department of Agriculture, Forest Service,
Southern Forest Experiment Station; 1966. / p.

Stein, W. I.; Slabaugh, P. E. ; Plummer, A. P.
Harvesting, processing and storage of fruits
and seeds. In: Schopmeyer, C. S., technical
coordinator. Seeds of woody plants in the
United States. Washington, DC: U.S. Department
of Agriculture, Forest Service, Agric. Handbook
450: 98-125; 1974.

Stiell, W. M. Comparative cone production in

young red pine planted at different spacings.
Public No. 1306. Ottawa, ON: Department of the
Environment, Canadian Forestry Service; 1971.
8 p.

Switzer, G. L. Elimination of empty or blind
coniferous seeds by specific gravity
separation. Washington, DC: U.S. Department of
Agriculture, Forest Service, Tree Planters'
Notes 35: 15-16; 1959.

Sziklai, O. Seed processing. In: Pollard, D. F.
W.; Edwards, D. G.; Yeatman, C. W. , eds . Seed
orchards and strategies for tree improvement
symposium: proceedings; 1981 August 17-20;
Duncan, BC. 18th Meeting of Canadian Tree
Improvement Association, Part II. Ottawa, ON:
Environment Canada, Forestry Service; 1982:
148-158.

Tanaka, Y. ; Cameron, P. C. Maturation of

ponderosa pine seeds in southern Oregon. In:
Bonner, F. T.; Krugman , S.; Yates, H. O. Ill;
Turner, B. J., eds. Flowering and seed
development in trees symposium: proceedings;

100

1978 May 15-18; Starkville, MS. Starkville, MS:
U.S. Department of Agriculture, Forest Service;
1979: 21R-225.

Thomas, J. Principles of gravity separation. In:
Belcher, E. W. , ed . Small lot forest seed
processing workshop: proceedings; 1977 October

18- 20; Macon, GA. Macon GA: U.S. Department of
Agriculture, Forest Service; Southeastern Area,
State and Private Forestry; 1978: 44-56.

Timmis, R. Critical frost temperatures for

Douglas-fir cone buds. Can. J. For. Res. 7:

19- 22; 1977.

Turnbull, J. W. Seed extraction and cleaning.
In: Report on FAO/DANIDA Training Course on
Forest Seed Collection and Handling. Rome,
Italy: United Kations, Food and Agriculture
Organization. Vol. 2; 1975.

Turner, R. M. A study of some features of growth
and reproduction of Pinus ponderosa in northern
Idaho. Kcol. 37: 742-753; 1956.

United Nations Food and Agriculture

Organization. Vehicle-mounted trunk shaker for
tree seed harvesting. For Equip. Note, FAO NO.
A. 55. 68. Rome, Italy: United Nations Food and
Agriculture Organization; 1968. 1 p.

United States Department of Agriculture. Manual
for testing agricultural and vegetable seeds.
AgrLc. Handbook 30. Washington, DC: U.S.
Department of Agriculture; 1952: 16-19.

United States Department of Agriculture, Forest
Service. An evaluation of two tree shakers for
harvesting cones. Equip. Develop, and Test Rep.
No. 2400-4. Missoula, MT: U.S. Department of
Agriculture, Forest Service; 1972. 27 p.

Van Haverbeke, D. F. Soaking and retumbling
controlled-pollinated Scots pine cones to
increase seed yields. Washington, DC: U.S.
Department of Agriculture, Forest Service, Tree
Planters' Notes 27(4): 8-9, 33. 1976.

Vaughan, C. E. ; Gregg, W. R.; Delouche, J. C,
edo. Seed processing and handling. Seed
Technol. Lab. Handbook No. I. Starkville, MS:
Mississippi State University; 1967. 295 p.

Waddell, T. Gathering in the cones harvest.
Slasher, Issue No. 318. London, UK: United
Kingdom Forestry Commission; 1984: 6.

Wagg, J. W. B. Viability of white spruce seed
from squirrel-cut cones. Forest. Chron. 40:
98-110; 1964.

Wakcley, P. C. Planting the southern pines.
Agric. Monog. No. 18. Washington, DC: U.S.
Department of Agriculture, Forest Service;
1954. 233p.

Waldron, R. M. Cone production and seedfall in a
mature white spruce stand. Forest Chron. 41:
316-329; 1965.

Wallinger, D. Personal communication. Victoria,
BC: British Columbia Ministry of Forests,
Silviculture Branch; 1981 January 5.

Wallinger, D. Characteristics of mature cones
and seeds. Victoria, BC: British Columbia
Ministry of Forests, Silviculture Branch; 1983.
2 p. Unpublished brochure.

Wallinger, D. Personal communication. Victoria,
BC: British Columbia Ministry of Forests,
Silviculture Branch; 1985 July 10.

Wang, B. S. P. Collection, processing and

storing tree seed for research use. In: Seed
problems, international symposium on seed
processing: proceedings; 1973; Bergen, Norway.
International Union Forest Research
Organizations, Working Party S2.01.06.
Stockholm, Sweden: Royal College of Forestry;
Volume I, Paper No. 17; 1973. 12 p.

Wang, B. S. P. Tree seed storage. Publ . No.

1335. Ottawa, ON: Department of Environment,
Canadian Forestry Service; 1974. 32 p.

Wang, B. S. P.; Pitel, J. A. , compilers and eds .
Forest tree seed storage international
symposium: proceedings; 1980 September 23-27;
Chalk River, ON; International Union Forest
Research Organizations, Working Party S2.01.06.
Ottawa, ON: Environment Canada, Forestry
Service; 1982. 243 p.

Wang, C. W. Cone and seed production in

controlled pollination of ponderosa pine.
Station Pap. No. 7. Moscow, ID: University of
Idaho, Forest, Range and Wildlife Experiment
Station; 1970. 15 p.

Wengcr, K. F. Annual variation in the seed crops
of loblolly pine. J. Forest. 55: 567-569; 1957.

White, T. L.; Harris, H. G. ; Kellison, R. C.

Conelet abortion in longleaf pine. Can. J. For.
Res. 7: 378-382; 1977.

Winjum, J. K. ; Ellis, I. D. Miniature cone

tumbler for seed research. Forest Res. Note No.
26. Centralia, WN: Weyerhaeuser Co.; 1960. 2 p.

Winjum, J. K. ; Johnson, N. E. Estimating cone
crops on young Douglas-fir. Forest Res. Note
No. 46. Centralia, WN: Weyerhaeuser Co.; 1962.
12 p.

Winston, D. A.; Haddon, B. D. Effects of early
cone collection and artificial ripening on
white spruce and red pine germination. Can. J.
For. Res. 11: 817-826; 1981.

Woolard, R. F . ; Silen, R. R. All-pneumatic

laboratory seed cleaner successful. Washington,
DC: U.S. Department of Agriculture, Forest
Service, Tree Planters' Notes 24(4): 15-17;
1973.

Wright, J. W. Flowering and fruiting of

northeastern trees. Station Paper 60. Upper
Darby, PA: United States Department of

101

Agriculture, Forest Service, Northeastern
Forest Experiment Station; 1953. 38 p.

Yanagihara, T; Tochiaki, K. ; Arai , K. (On the

relation between the harvest of Japanese larch
seed and meteorological conditions.) J. Japan.
Forest. Soc. 42: 347-351; 1960. (Forestry
Abstracts 23, 1866; 1962.)

Zasada, J. C. Frost damage to white spruce cones
in interior Alaska. Res. Mote PNW-149.
Portland, OR: U.S. Department of Agriculture,
Forest Service, Pacific Northwest Forest and
Range Experiment Station; 1971. 7 p.

Zasada, J. C. Effect of cone storage method and
collection date on Alaskan white spruce (Picea
glauca) seed quality. In: Seed problems,
international symposium on seed processing:
proceedings; 1973; Bergen, Norway.
International Union Forest Research
Organizations, Working Party S2.01.06.
Stockholm, Sweden: Royal College of Forestry;
Volume I, Paper No. 19; 1973. 10 p.

Zasada, J. C; Viereck, L. A. White spruce cone
and seed production in interior Alaska. Res.
Note PNW-129. Portland, OR: U.S. Department of
Agriculture, Forest Service, Pacific Northwest
Forest and Range Experiment Station; 1970. 11 p

5j

CAN UE ATTAIN BRITISH COLUMBIA SEED PRODUCTION GOALS FOR INTERIOR SPRUCE?

Paul J.lBirzins

ABSTRACT: The British Columbia seed production
target is A ,000 viable seeds per interior spruce
ramet per year by age 15 (15 years from grafting).
In 1983, 10-year-old ramets produced an average
of 65. A cones and 27.6 filled seeds per cone,
resulting in a mean filled seed production per
ramet of 1,805. One year later production was
substantially lower at 46.46 filled seeds per ramet
when the average number of cones per ramet was
10.1 and filled seeds per cone was 4.6. Based on
this information, orchard location, and associated
cone induction trials, the chances of attaining
the seed production target are discussed.

INTRODUCTION

Estimated cone and seed yields from grafted ramets
and future seedling requirements are used to
determine the size of seed orchards in the coopera-
tive British Columbia Tree Improvement Program.
Failure to meet seed production targets will limit
the positive impact of our reforestation and tree
improvement programs on the forest economy. There-
fore, the accuracy of seed production estimates
must be continually updated to determine the level
of seed production from orchards in relationship
to seedling demand. If these figures are not
updated, seed collections from natural stands and
from orchards may become out of balance with
seedling demand, resulting in considerable financial
loss. For example, if seed production from the
orchards falls below target levels, collections
from natural stands will be required to reach the
seed production goal. If orchard production is
not determined in time to collect seed from the
natural stands or there is not a good inventory
of seed from the designated seed zone, the seed
shortfall could result in a slowdown in refores-
tation. This would result in a loss of desirable
species on productive growing sites, which would
eventually lead to a decrease in the annual
allowable cut for the area.

Recent data for the BC interior show that antici-
pated annual demand by the year 2000 for interior
spruce will be 86.1 million seedlings (Albricht
1985) . This high interior spruce seedling demand
has resulted in the allocation of a major part of
tree improvement funding into grafted clonal
seed orchards and associated breeding activities
for the species. To date about 30 ha (96 acres)
of spruce seed orchards have been established.

Our spruce seed production estimates are based on
extrapolation from extremely limited data.
Expectations are that by age 15 (15 years from
grafting) each orchard ramet should be producing
an average of 100 cones with 40 filled seeds per
cone for a total of 4,000 viable seeds per ramet.
Relatively recent literature indicates that this
estimate may be high. Ten- to 19-year-old white
spruce ramets had considerably lower production
of viable seeds (2,262 per ramet) but higher cone
yields of 174 cones per ramet (McPherson and
others 1982) . The relatively small number of
filled seeds per cone (13) has also been reported
for black spruce (P. mariana) (McPherson and
others 1982; Verheggen and Farmer 1983).

Our first clonal orchard was established at
Skimikin, BC (lat. 50°47', long. 119°14') in 1979
and therefore is too young to provide meaningful
seed production information. Fortunately, in
1976 G. Kiss, BC Ministry of Forests spruce breeder,
established a breeding arboretum at Vernon (lat.
50°15\ long. 119°14') which is located in the
hot, dry Okanagan Valley. The spruce program
was originally located considerably farther north
(3 of latitude); however, greater strobili
production was anticipated in the Okanagan Valley,
and this has been confirmed (Kiss, 1978). Based
on this fact, the majority of spruce seed orchards
are located in the Okanagan Valley.

WHAT ARE CURRENT SEED PRODUCTION YIELDS?

Projected annual seedling requirements for British
Columbia (BC) indicate that interior spruces
(Picea glauca, P. engelmannii , and their hybrids)
will continue to be the major reforestation species.

Paper presented at the Conifer Tree Seed in the
Inland Mountain West Symposium, Missoula, MT,
August 5-6, 1985.

Paul J. Birzins is Seed Orchard Projects Coordinator
British Columbia Ministry of Forests, Kalamalka
Research Station and Seed Orchard, Vernon, British
Columbia, Canada.

To update spruce seed yield estimates, cone and
seed data were collected from clones in the oldest
section of the East Kootenay breeding arboretum
in 1983 and 1984.

Materials and Methods

The ramets in the East Kootenay breeding arboretum
located at Vernon, BC were planted at a 5.5- by
5.5-m (18- by 18-ft) spacing on an orthic black
chernozem soil (Grandview Series) from 1976 through
1979. The 1.5 ha (3.7 acre) arboretum consists
of 127 clones planted in clonal row plots (4 ramets
per clone). Scions were collected from ortets

103

located between latitudes 49°01' and 50°52',
longitudes 114 18' and 116 36', elevations
between 854 and 2 012 m (2,800 - 6,597 ft) and
grafted onto 2-year-old rootstock using the side
veneer technique. The 235 ramets selected for
the study were grafted from 1972 through 1974,
planted in 1976, and were about 2m (6.6 ft)
tall in 1983.

Bulk collections of 10 cones per ramet were made

in the fall of 1983 from those ramets that had

produced a sufficient number of cones. Cones

o o

were oven dried for 12 hours at 50 to 70 C
(122° to 158°F) . Seed was extracted by hand,
dewinged, and the filled seed fraction was
determined using a single-tube South Dakota
blower. A sample of filled seeds in this fraction
was counted and weighed. This weight was then
divided into the total seed weight to determine
the total number of filled seeds. Since this
figure represented 10 cones, the total weight
was divided by 10 to determine the number of
filled seeds per cone. The study was repeated
in 1984.

Results and Discussion

In 1983 the average number of cones per ramet was
65.4 (range 0-500) and the average number of
filled seeds per cone was 27.6 (range 0-89).
This resulted in a mean filled seed production
per ramet of 1,805, which contrasted with the
significantly lower production figure of 46.46
filled seeds per ramet that occurred in 1984
when the average number of cones per ramet was
10.1 (range 0-250) and the average number of
filled seeds per cone was 4.6 (range 0-20). This
cone and seed production information is summarized
in table 1. According to local observations,
in 1983 there was good male strobili production
and the cone crop was the largest since the
establishment of the arboretum. The 1984 cone
and pollen crop was classified as very light,
using the classification criteria outlined by
Dobbs and others (1976) . About 97 percent of
the total seed production during these 2 years
occurred in 1983. Cone crops in spruce stands
were of similar proportion. In 1984 at a
15-year-old East Kootenay spruce clone bank near
Prince George (about 3 of latitude north of
Vernon) there were absolutely no cones and in

Table 1. — Summary of cone and seed production from 10-
located at Vernon, BC

1983 there was a medium crop. This clone bank
consists of ramets from the same clones as the
ramets used in this study. This supports the
inverse relationship between cone production and
latitude .

As illustrated in figures 1 and 2 there was an
extremely unequal clonal contribution to seed

6000 9000
SEEOS/RAHrr

Figure 1. — Mean seed contribution per ramet by
clone in 1983.

year-old and 11-year-old interior spruce ramets

Variable

10-year-old ramets
(1983 data)

11-year-old ramets
(1984 data)

No. of clones

79

85

No. of ramets

235

235

Average no. of cones

65

4

(range

0-500)

10

1 (range

0-250)

Average no. of filled

seed per cone

27

6

(range

0-89)

4

6 (range

0-20)

Mean filled seed

production per ramet

1805

46

46

104

CONE PRODUCTION FROM TRANSPLANTED RAMETS

n

I0«

JO W to M

ioo iio no ito in loo j jo j»o ito in x» no mo ho >n

Figure 2. — Mean seed contribution per ramet by
clone in 1984.

production in both the marginal and bumper crop
years at the arboretum. The discrepancy between
clonal number (79 clones in 1983 compared to 85
clones in 1984) is attributed to seed extraction
difficulties during 1983. The ANOVA for both
years indicates significant differences in cone
production per clone (0.01 level). In an un-
managed seed orchard the pollen mix and cone
production can be very poor, as shown in the
1984 data.

Ideally, we want equal amounts of pollen and cone
production per clone to maximize the genetic
quality of the seed. An unequal strobili distribu-
tion per clone implies increased selfing levels
and therefore decreased genetic quality. This
problem could also be aggravated by nonsynchronous
flowering in the orchard or lack of a pollen crop
of significant size. Utilization of supplemental
mass pollination techniques rapidly becomes an
essential orchard management tool for maximizing
the quantity and quality of seed.

Our seed orchard staff has consi
substantial increases in spruce
in the year following seed orcha
For instance, no cones were obse
3-year-old ramets when planted a
One year after transplanting the
seed orchard, each ramet had an
cones. A similar number of rame
clones that had been established
2 years earlier had an average o
per ramet .

stently observed
cone production
rd establishment,
rved on 111
t a nursery .

ramets into a
average of 88
ts from the same

in the orchard
f only two cones

Average heights of the two groups of ramets were
quite similar (transplanted trees 134 cm [4.4 ft],
orchard trees 140 cm [4.6 ft]). Obviously,
transplanting accounted for the additional cone
production. Eighty-one of the transplanted trees
were over 100 cm (39.4 ft) tall and had an average
of 14.9 cones per tree. The data from this study
are summarized in table 2. With increasing size
of transplanted ramets there was a pronounced
increase in cone production in the year following
transplanting .

DISCUSSION AND CONCLUSIONS

If the goal of 4,000 filled seeds per ramet by age
15 is to be reached, a 67 percent increase in
seed production will be required in 4.5 years.
The breeding arboretum is not managed as a
production seed orchard; however, the cone
production data suggest that unmanaged seed
orchards could yield well below their biological
potential. Further decreases in seed production
could occur if cone and seed insects "discover"
our seed orchard sites. Presently the seed orchards
are isolated from natural spruce stands and are
relatively free of insects and disease.

Each interior spruce cone could potentially produce
about 200 seeds (Owens and Molder 1984) . Up to
100 filled seeds per cone have been counted in
local natural spruce stands. A conservative
assumption as to the potential for filled seed
production in a managed spruce seed orchard might
be 90 seeds per cone. Using methods discussed
by Bramlett and Godbee (1982), "seed efficiency"
averaged over both the study years was only 18
percent ([16.1 seeds per cone realized - 90 seeds
per cone potential] x 100). Supplemental mass
pollination would be a useful tool to increase
seed set.

Once we have the cones we must be able to harvest
them. Flower and cone abortion losses of up to
75 percent of the crop have been observed for
other species such as coast Douglas-fir (Pseudotsuga
menziesii) (Bartram 1982). We don't know the
magnitude of the losses for interior spruce.

In spite of the potential problems, our seed orchard
managers should be able to meet our seed production
goals. We know that transplanting shock substan-
tially increases cone production. Our physiolo-
gists are attempting to duplicate these results
in operational seed orchards using various
combinations of root pruning, drought stressing
(heat and water), GA 4/7 treatments, and fertilizing.

105

Table 2. — Cone production from 5-year-old transplanted interior spruce ramets

Variable

Ramets planted in 1981,
data collected in 1984

Ramets planted in 1983,
data collected in 1984

Seed orchard planning zone

No. of ramets

Avg. Ht . of ramets

Avg. no. of cones per ramet

No. of ramets > 100 cm in ht .

Avg. cone production per
ramet < 100 cm

No. of ramets <_ 100 cm in ht .

Avg. cone production per
ramet < 100 cm

Shuswap Adams
111

140 cm
2

101

2.198
10

Shuswap Adams
111

134 cm

88
81

115.07

30

14.9

In combination with booster pollination and insect
and disease control we have the potential to
exceed our seed production target. With a good
quantity of genetically superior seed we are off
to a good silvicultural beginning.

ACKNOWLEDGMENTS

The author is indebted to C. Bartram for data
assistance, J. Konishi and M. Albricht for tech-
nical advice, and G. Kiss for providing the
ramets .

REFERENCES

Albricht, M. A. Personal communication. Vernon,
BC: Ministry of Forests, Silviculture Branch,
Kalamalka Research Station and Seed Orchard;
1985. December 20.

Bartram, C. Quantification of orchard seed produc-
tion efficiencies and absolute seed losses
from the time of flowering through to seed
germination. Internal report SX82609-Q. Victoria,
BC: Ministry of Forests, Silviculture Branch;
1982. 24 p.

Bramlett, D. L.; Godbee, J. F. Inventory-monitoring
system for southern pine seed orchards. Res.
Pap. 28. Georgia Forestry Commission; 1982. 19 p.

Dobbs, R. C; Edwards, D. G. W.; Konishi, J.;
Wallinger, D. Guideline to collecting cones
of BC conifers. Joint report no. 3. British
Columbia Forest Service/Canadian Forestry
Service; 1976. 98 p.

McPherson, J. A.; Morgenstern, E. K. ; Wang, B. S.
Seed production in grafted clonal orchards
at Longlac, Ontario. For. Chron. 58(1): 31-34;
1982.

Kiss, G. Interior spruces (E.P. 670). In: Forest
Research Review. Victoria, BC: Ministry of
Forests; 1978: 62 p.

Owens, J. N.; Molder, M. The reproductive cycle
of interior spruce. Victoria, BC: Ministry
of Forests; 1984. 30 p.

Verheggen, F. J.; Farmer, R. E., Jr. Genetic and
environmental variance in seed and cone charac-
teristics of black spruce in a northwestern
Ontario seed orchard. For. Chron. 59(4): 191-19
1983.

106

f

COLD STRATIFICATION FOR LODGEPOLE PINE SEED.

Ik' 1 1 urn and I

ABSTRACT: In this study, cold stratification did
not increase total germination of immature seeds
(August) of lodgepole pine, had little effect on
seeds at their peak of germination (September) and
elicited no or negative responses from seeds frozen
in fall before fully ripe (October) . The germina-
tion rate was hastened consistently by cold strati-
fication and the 42-day treatment lead to more
rapid germination than did 21 days. The need for
cold stratification apparently increases again as
cones and seeds remain on the tree over winter.

INTRODUCTION

Cold stratification of tree seed implies temporary
storage in a moist medium at temperatures just
above freezing. This allows the breaking of dor-
mancy (Copeland 1976) so that germination can
proceed. Cold stratification is also done to
allow after-ripening to take place (Edwards 1980)
and it is accepted that seeds absorb moisture
during this time or the treatment would serve no
useful purpose. Seeds that cannot take up water
do not respond to moist storage or cold stratifi-
cation. The stratification is usually carried
out for a minimum of 15 days for north temperate
pines (Krugman and Jenkinson 1974) and must, in
some species, last for up to 270 days.

Some authors claim that cold stratification is not
needed for the full germination of lodgepole pine
(Pinus contorta Dougl . var. latifolia Engeltn.)
seeds from southern sources (Cri tchf ield 1976)
but it is considered as a basic need for more
northern sources (Thompson 1984). It is carried
out routinely on all lodgepole pine seeds used in
nursery practice in Alberta. (See also Wheeler
and Critchf ield 1985) .

Paper presented at the Conifer Tree Seed in the
Inland Mountain West Symposium, Missoula, MT,
August 5-6, 1985.

A. K. Helium is Professor, Dept. of Forest
Science, The University of Alberta, Edmonton,
AB, Canada; I. Dymock is Research Scientist,
Canadian Forestry Service, Northern Forest
Research Centre, Agriculture Canada, Edmonton,
AB.

. J^Dymock

Our knowledge about what happens to seeds during
cold stratification is incomplete (Tanaka 1984) ,
but there is an understanding that if they are in
need of cold stratification to germinate fully
then they are not fully ripe (Edwards 1980) . It
is generally held that cold stratification is not
harmful to germination of lodgepole pine seeds
(Wang 1978).

In contrast, cold stratification normally hastens
germination. The time to reach 50 percent of
total germination may be shortened by days. Rapid
germination in nurseries under controlled moisture
conditions is a distinct advantage because it leads
to the production of homogeneous crops of seedlings
(Tinus and McDonald 1979). Spontaneous germination
is an advantage and a requirement here, but it is
certainly not desirable under variable field condi-
tions. The need for cold scratif ication is there-
fore a built-in safety mechanism in seeds, which
helps ensure that germination occurs at the right
time in the field.

Cones of both spruce and pine can be collected soon
after the embryo is fully developed, judging by
x-ray photography, and provided the seeds are left
in the cones for a month at about 10° C (50° F)
and at about 65 percent relative humidity (Edwards
1978, 1980). The ripening which is taking place
in such storage is therefore similar to that which
takes place on the tree given that fall weather
will allow this process to occur. Simak (1966),
Kardell (1973), and Helium and others (1983) have
demonstrated that about 4 weeks are needed, after
complete embryo development, before the seeds of
Scots pine (P. sylvestris L.) and lodgepole pine
reach their full germination potential.

The main purpose of this study was to test the
relationship between seed ripening and the need
for cold stratification. The effect of temporary
cone storage before extraction was also investi-
gated to evaluate the potential effects of after-
ripening on need for cold stratification.

METHODS

The cones for this study were gathered from 20
felled trees at each of four collection times at
three different altitudes (table 1) on the Procter
& Gamble Cellulose (Canada) Ltd. lease area in the
Grande Prairie forest of Alberta.

107

Table 1. — Lodgepole pine cone collections made
south of Grande Prairie, Alberta, at
four different times and three
different altitudes between July and
October 1984

Sample location Altitude Collection Total

times trees

(53

54°55'N. ,118°50'W.

670

July

17-20

80

(Lower location)

Aug

20-23

Sept

17-20

Oct

29-31

54°38'N. ,119°05'W.

970

July

17-20

80

(Middle location)

Aug

20-23

Sept

17-20

Oct

29-31

54°25°N. ,119°39'W.

1515

July

17-20

80

(Upper location)

Aug

20-23

Sept

17-20

Oct

29-31

Cones collected in July 1984 represent the 1983
cone crop while cones collected in August to
October represent the 1984 cone crop. Only trees
with an estimated 30 cones or more of the desirable
kind were felled. There were more trees without
than with the required cones on these sites, so
binoculars were used to assess each tree before
felling. The 1984 cone crop was particularly
poor at all three altitudes compared to 1983
and earlier crops.

Cones were kept separate by tree, site, and time
of collection and seeds were extracted from 17 of
the cones per tree, cleaned, and counted. The
remaining cones were needed for other tests. The
seeds were then pooled to make certain that the
same number of filled seeds was pooled for each
tree and stand per collection time. All germin-
ation tests were therefore run on fully balanced
samples .

The cones were cracked open at 180° C (356° F)
for about 3 minutes and then left overnight at
40° C (104° F) in a gravity vented kiln. Cones
were then tumbled and seeds extracted, wet dewinged
by hand, and cleaned in a North Dakota blower.
The seeds were then left to air-dry for 24 hours
and subsequently stored at about 3° C (37° F)
until testing. Seed moisture contents were not
measured after each collection trip and extrac-
tion and cleaning run because experience has
shown seed moisture contents in general do not
exceed 8 percent under similar conditions.

The cone moisture contents were determined within
48 hours of collection and five cones per tree,
location, and collection time were dried at 105° C
(221° F) for 24 hours. Moisture conditions were
calculated on a dry-weight basis. Cones were
dried intact without seeds being extracted first.

Eight cones were stored in paper bags in a cold
room kept at 5° C to 10° F (41° F to 50° F) for
7 months from time of collection to test possible
effects of after-ripening on seed dormancy. Four
weeks should be long enough to satisfy after-
ripening needs after which seeds should be able to
withstand drying to 5 to 7 percent moisture.
Seven months were used in this study due to work
pressures .

The extraction, cleaning, and counting of seed
took 4 weeks, and a maximum of 42 days of cold
stratification was used. Thus, 70 days lapsed
between collection and the onset of germination
tests .

Cold stratification was done between layers of
moist peat at 2° C for 21 and 42 days. Four
replicates of 100 seeds were first x-rayed to
determine empty and damaged seed counts for all
tests. They were then placed on plastic frames
22 by 22 cm (8.8 by 8.8 inches) (approximately)
which were covered with Kimpak and 1 cm of moist
peat moss. The seeds were spread on top of the
peat in four distinct replicates. These frames
were stacked one on top of the other and all were
covered with 2-mil plastic. These stacks, of
about 20 frames, were opened only to remove frames
to start the germination tests or to remoisten
peat every 14 days .

Dry seeds (controls) were set to germinate at the
same time as the 21-day and 42-day stratified
seeds. The frames, Kimpak, peat, and seeds were
placed in clear plastic boxes with lids and put
in a Conviron germinator run at 30° C (86° F)
during 8-hour days followed by 20° C(68° F) and
16-hour nights. Tests were terminated after the
21st day. Tap water was added as needed in mist
form. The seeds were kept remarkably free of
fungus or other pest problems.

Seeds were classified as germinated when the pro-
truding radicle/hypocotyl was four times the
length of the seed coat. The seeds were then
removed. Cutting tests were not performed after-
ward because all replicates were x-rayed before
tests started. Percent germination was therefore
always calculated based on full seed only. Empty
and damaged seed in these tests rarely exceeded
5 percent. Germination was counted daily at the
same time.

The germination rate (R^q) is defined here as the
time needed to reach 50 percent of the total ger-
mination of a sample. It was calculated in days
for each replicate, source, and collection time
by lineal interpolation on germination curves.

The radiographs were taken using 20 s exposures,
15 kV, 5mA (Milliampere) and a distance of 47 cm
(18.8 inch). Seeds were put on top of the
packaged film for exposing and films were then
developed right away. This made it possible to
make sure that no replicate used had more than
about 10 percent empty or damaged seed.

108

RESULTS AND DISCUSSION
Cone Moisture Content

Average cone moisture contents dropped from about
60 percent of ovendry weight in mid-August to just
over 20 percent in late October 1984 in all three
sample stands regardless of altitude. This drying
did not follow the expected concave asymptotic
curve (Helium and others 1983) observed earlier.
It resembled more the right-hand part of a convex
parabola showing very rapid moisture loss after
rather than before the September collection
(table 2). No doubt, the -20° C (-4° F) weather
experienced just before the October collection
affected cone ripening, but freezing tests on
collected cones failed to yield useful data
because the tetrazolium test was used to evaluate
damages and this proved too unreliable. October
was reportedly the coldest month in 30 years in
Alberta .

Table 2. — Cone moisture contents for lodgepole
pine at time of collection south of
Grande Prairie, Alberta in 1984

Altitude Collection times (X ± 1SD)

of

sample July August September October

1983 cones

1984

cones

670

12.711.22 62

0±4

.40 47

0±13.

70

22

2±3

12

970

12.4±1.43 59

1±5

.26 54

1+ 7.

18

21

8±2

56

1,515

12.8±1.44 62

5+6

.09 59

3± 5.

21

20

7±2

76

Natural cone drying proceeded most rapidly at the
low site between August and September (15.0 percent)
and slowest (3.2 percent) at the highest site.
This trend was reversed between September and
October. Now the cones at the highest site lost
most moisture (38.6 percent) and the low site
cones the least (24.8 percent). Judging by a
study of 1980 cones from the same stands (Helium
and others 1983) the variability in cone moisture
content should have been expected to be greatest
during July and late September and least in August
and in the following spring. The very rapid rates
of cone drying from September to October 1984 are
interpreted here as having been hastened by the
-20° C (-4° F) temperatures and heavy snowfall
experienced about 2 weeks before cone collection
could proceed.

Coefficients of variation (table 3) suggest that
tree-to-tree variability in cone moisture contents
was greatest in the September collection at the
low site and that August cones were quite homo-
geneous among trees at all altitudes. The October
cones were only marginally more variable than
the July cones of 1983.

The moisture contents for cones collected in July
1984, that is, cones ripened in 1983, were low
(12.4 to 12.8 percent). They were between 7
and 9 percent lower than expected judging by a
month-to-month survey done of cone moisture
contents on the campus of the University of
Alberta in 1982-1983 (Helium unpublished). The
cone moisture on campus dropped to about 15

Table 3. — Coefficients of variation for data in

table 2 on lodgepole pine cone moisture

Altitude Collection times (1984)

of

sample July August September October

(m)

1983

cones

1984 cones

670

9

5

7.1

20.1

14

0

970

11

5

8.9

13.3

12

1

1,515

11

.2

9.7

8.8

13

3

percent in April and May in this period while it
remained between 20 and 22 percent for the rest of
the year. The low moisture values for July 1984
cones are interpreted as being indicative of very
warm and dry forest conditions in the summer of 1984.

Total Germination

Germination reached 92 percent and above for all
seeds collected in September whether cold-stratified
or not and regardless of location or altitude (fig. 1)
The cold-stratified seeds did not reach higher levels
of germination than the unstratified seeds. Seeds
stratified for 42 days at 2° C (36° F) showed an
average 3 percent less germination than those stra-
tified for 21 days or not stratified at all (95
percent level of confidence).

Cold stratification at 2° C clearly damaged the
high-altitude seeds collected in August (an 11.5
percent loss in total germination) . The longer
the cold stratification period the less germination
obtained (99.9 percent level of confidence). No
damages were observed in seeds collected at the
mid and lower altitudes where seeds could reach
95-96 percent germination at this time.

Seeds collected in October, however, were signifi-
cantly (95 percent level) less germinable (85.0
percent ±5.5 percent) than those collected in
September (95.2 percent ±4.0 percent). Even though
the seeds collected in August and September could
germinate to above 90 percent the seeds collected
in October never exceeded 85 percent and fell, on
average, about 10 percent short of the September
total. The -20° C (-4° F) temperatures in mid-
October are assumed to have caused this 10 percent
loss in total germination at all altitudes. It is
surmised that the seeds from some trees, where cones
dried more slowly, were more severely damaged than
those from trees where cones had dried more rapidly
(were more mature) .

The cones from 1983, collected in July of 1984,
showed clearly that cold stratification aided total
germination in 1984 significantly (by 5 to 6 percent)
at both the middle and higher altitudes (fig. 2).
The trend was significant at the 99.4 percent level.
The pretreatment did not influence total germination
in seeds from the lower collection site. This
response pattern agrees well with the 1984 cone
data and the data from 1980 (Helium and others
1983) . It appears that an altitude of 670 m
(2,211 ft) above sea level is low enough to allow
relatively complete seed and cone ripening to
occur before fall and winter frosts arrive.

109

Upper Location

Cone Collection 1984

August

C = Control

21 D = 21 -day cold stratification
42D = 42-day cold stratification

September

100 r

Days

Middle Location

O 20

Figure 1. — Germination curves for seeds of lodge-
pole pine collected at three times and at three
different altitudes in the Grande Prairie forest
of Alberta. Seeds were cold-stratified for 21
and 42 days.

1 983 Seed
Collection: July 1984

C = Control

21 D = 21 day cold stratification
42D = 42-day cold stratification

Days

Figure 2. — Germination curves for 1983 lodgepole
pine seed collected in July of 1984 from three
different altitudes in the Grande Prairie forest
of Alberta.

Germination Rate

Germination rates were improved significantly
(99.9 percent level) with cold stratification
compared to controls. The 42-day treatment gave
significantly better (99.9 percent level),
faster, germination (5.04 days ± 0.45) than the
21-day treatment (5.90 days ± 0.53) regardless
of altitude of sample (fig. 1). The control took
7.49 days ± 0.53.

Seeds collected in September had the most rapid
rate of germination (5.84 days ± 0.91), while
the August seeds used 6.31 days ± 1.32 to reach
the same point. The October seed was interme-
diate at 6.03 days ± 1.20. There was only a
weak difference (significant at 90 percent level
only) among germination rates over the collection
period. The 1983 seeds, collected in July 1984,
germinated to 50 percent in 6.40 days ± 1.03 and
proved to be slower than the seeds collected in
August and September, but the September seeds
were similar to the October seeds (90 percent
level of confidence) .

The germination rate was therefore only slightly
more rapid at the time when the 1984 seeds were
ripest than before or after this time.

After-Ripening

Without exception, seed germination dropped sig-
nificantly as a result of cone storage after
August collection even though the relative humidity
was maintained around 65 percent and temperatures
kept at between 5 and 10° C (41 and 50° F) . The
attempted after-ripening led to a loss of over
31 percent germination at the lower altitude,
26.5 percent loss at the middle and 14.3 percent
loss at the highest altitude site in August 1984.
These losses were undoubtedly due to drying
damages. A considerable proportion of the after-
ripened seeds had cracked endosperms. Because
after-ripening implies additional maturation such
seeds should have been able to withstand long-
term storage ( 7 months) without showing drying
damages after the first 4 weeks.

The after-ripened seeds did not benefit consis-
tently from cold stratification except in rates
of germination. Total germination was sometimes

1984 Seed

Collection: August 20-23, 1984

C

21D
42D

Control

21-day cold stratification
42-day cold stratification

Days

Figure 3. — Germination curves for after- ripened
lodgepole pine seed collected in the Grande
Prairie forest of Alberta in August of 1984.

110

slightly better in controls than in stratified
seed and sometimes slightly worse (fig. 3). See
also Winston and Haddon (1981).

CONCLUSIONS

In this 1-year study cold stratification hastened
germination in lodgepole pine seeds. Cold stra-
tification also elicited three kinds of responses
in the same seed regarding total germination:

1. Immature (August) seeds were harmed by
cold stratification at 2° C (36° F) and the
longer the period of cold stratification the
lower the total germination. The treatment effect
was significant at the 95 percent level of confi-
dence .

2. Once the seeds had reached their peak ger-
mination, which happened in late September in 1984,
little or no additional germination could be
obtained by cold stratification. In fact, the
42-day stratification had a slight negative
effect on germination (99.3 percent level).

3. Seeds from 1983 (collected in July of
1984) benefitted by cold stratification and seeds
collected in October 1984 could tolerate a 21-day
period of stratification but not a 42-day period.
It appears, therefore, as if seeds enter dormancy
again gradually once winter sets it. There was a
clear indication in these samples that altitude
played a role.

A relative humidity of about 65 percent was not
adequate for maintaining necessary cone moisture
contents in order to allow after-ripening to
proceed in lodgepole pine seed in this study.

ACKNOWLEDGMENT

We wish to thank Lisa Hackett and Lori Siemens
for their invaluable help and good work on this
project. Without their careful lab work the
results would surely not have been so clear and
concise. Frank Pendwick, Peter Nosco, and the
temporary staff at the Northern Forest Research
Centre of the Canadian Forestry Service helped
with collection of cones each month. We want to
thank Procter and Gamble Cellulose Co. for per-
mission to collect cones on their lease area
and for their help in cone collection as well.

REFERENCES

Copeland, L.O. Principles of seed science and
technology. Burgess Publishing; Minneapolis,

MN: 1976. 369 p.

Critchfield, W. B. The distribution, genetics
and silvical characteristics of lodgepole pine.
In: Proceedings of the IUFRO joint meeting of
working parties, (no dates) Vancouver, Canada,
Ministry of Forests, Prov. of British Columbia,
1978: 1: 65-94.

Edwards, D. G. W. Maturity and seed quality; a
state of the art review. In: Bonner, F., ed .
Flowering and Seed Development in Trees Symposium:
IUFRO proceedings; 1978: May 15-18; Starkville,
MS.; 1979: 233-263.

Edwards, D. G. W. Maturity and quality of tree
seeds - a state-of- the-are review. Proc . Int.
Seed Testing Assn. 8(4): 625-657; 1980.

Helium, A. K . , Loken, A., and Gjelsvik, S. Seed
and cone maturity in Pinus contorta from
Alberta, Canada, 1980. Agric. and For. Bull.
6(3): 24-30; 1983.

Kardell, L. Studies of pine seeds from northern
Sweden. Investigation on storage of pine cones
and pine seeds (Pinus silvestris L.) in northern
Sweden, Allmanna Forlaget , Lund, 2: 1-70, 1973.

Krugman, S. L. and Jenkinson, J. L. Pinus

L. Pine. In: Schopmeyer, C. S. ed. Seeds
of woody plants in the United States. USDA
Forest Service Handbook 450, 883 pp. 1974.

Simak, M. The ripening process in pine seed
(Pinus silvestris L.) Res. Note Dept. of
For. Genet. Royal Coll. of For., Sweden 2:
411-426; 1966.

Tanaka, Y. A. Assuring seed quality for seedling
production: Cone collection and seed pro-
cessing, testing, storage and stratification.
In: Duryea, M. L., and Landis, T. D. ed.
Forest Nursery Manual. Production of bareroot
seedlings. Martinus Nijhoff/Dr. W. Junk Publ.
Holland, 1984; 27-40.

Thompson, B. E. Establishing a vigorous nursery
crop: Bed preparation. Seed sowing and early
seedlings growth. In: Dureya, M. L., and
Landis, T. D. ed . Forest Nursery Manual.
Prod, of bareroot seedlings. Martinus Mijhoff/
Dr. W. Junk Publ. Holland, 1984; 41-49.

Tinus, R. W. and McDonald, S. E. How to grow tree
seedlings in containers in greenhouses. Gen.
Tech. Rep. RM-60. Fort Collins, CO: U.S. Depart-
ment of Agriculture, Forest Service, Rocky
Mountain Forest and Range Experiment Station;
1979, 256 p.

Wang, B. S. P. Seed yield and germination

requirements of Alberta white spruce and lodge-
pole pine. Alberta For. Dev. Res. Trust Rept.
No. 92; 1978: 26 p.

Wheeler, N. C. and Critchfield, W. B. The

distribution and botanical characteristics of
lodgepole pine: Biogeograph ical and Manage-
ment Implications. In: Baumgartner, D. M.,
Krebill, R. G., Arnott, J. T. and Weetman,
G. F. ed . Lodgepole pine, the species and
its management. Symposium proceedings;
1984 May 8-10 Spokane and May 14-16 Vancouver,
B.C.; Cooperative Extension Service. Wash.
State Univ.; 1985: 1-14.

Winston, D. A. and Haddon, B. D. Effects of

early cone collection and artificial ripening
on white spruce and red pine germination.
Can. J. For. Res. 11(4): 817-826; 1981.

Ill

EFFECT OF STRATIFICATION TIME AND SEED TREATMENT ON

GERMINATION OF WESTERN WHITE PINE SEED

R. J. \Hoff

/r

ABSTRACT: Germination of intact western white
pine seeds increased from 7 percent for unstratified
seeds to 83 percent for seeds that were stratified
under cold-wet conditions for 15 weeks (105 days) .
Fifty percent of the seeds for which the seed coat
and papery membrane were removed or pieces were
cut away germinated with no stratification and
98 percent germinated after 15 weeks of stratifica-
tion. Dormancy is determined mainly by the papery
membrane and physiological elements of the embryo
or gametophyte . The relative importance of these
determinants changes with stratification time up
to 105 days, after which only the papery membrane
remains a barrier to the germination of about
15 percent of the seed. Families also had an
impact on germination; family heritability of
79 percent was calculated.

INTRODUCTION

Cold-wet stratification has been the principal
method for breaking dormancy in western white pine
(Pinus monticola) seed. And yet high germination
was seldom assured. Germination sometimes would
be high, other times low; worst of all, there was
no predictability even when the seed could be
shown to be viable, for example by using seed
cutting tests (Hoff and Steinhoff, in press).

Several kinds of treatments have been tried; for
instance, alternating cold and warm stratification,
acid soak, freezing, cutting, scarification,
sodium hypochlorite soak, long-term soaking in
water, and infrared light exposure (Larsen 1925;
Anderson and Wilson 1966; Partridge and others
1985; Works and Boyd 1972; Malone 1983) have been
used with variable success.

METHODS AND MATERIALS

The seeds came from 20 individual trees (families)
located in the Palouse River drainage, ID,
latitude 46°59'N, longitude 116°33'W., elevation
823 m (2,700 ft). Cones were harvested from
individual trees when they had become
flaccid — the cones were no longer hard and they
could be bent back and forth with little or no
crackling. In this state the cones had not dried
very much, but the scales were no longer stuck to
each other. All cones were picked within 7 days
in September 1984. Cones were dried in a
greenhouse, seeds were extracted and then stored
at 3 °C (37 °F) . Seed germination tests were
started November 1984.

Five hundred seeds of each family were placed in
a small plastic mesh bag; seed surfaces were
sterilized in a 0.25 percent sodium hypochlorite
solution for 10 minutes. Seeds were rinsed three
times with water and then given a 24-hour running
water rinse.

The seeds were then subjected to five seed
treatments and five stratification times: 0, 21,
42, 90 and 105 days.

To stratify, each of the 20 family seed lots was
rolled up in a wet paper towel and placed in an
unsealed small plastic bag. An insulated box
inside a refrigerator was used as a stratification
chamber and temperature was monitored remotely.
Stratification temperature was maintained at
3 °C ±1 (37 °F ±1) .

After each stratification time the seeds were
subjected to five treatments:

The objectives of this research were to 1) determine
the effect of stratification time on germination
when physiological ripeness of cones and conditions
of stratification were closely controlled; 2) to
assess the interactions with length of stratifica-
tion of various seed treatments involving the
removal of or cutting of the seed coat and papery
membrane; and 3) to assess the impact of families
on germination.

Paper presented at the Conifer Tree Seed in the
Inland Mountain West Symposium, Missoula, MT,
August 5-6, 1985.

R. J. Hoff is a Research Plant Geneticist,
Intermountain Research Station, Forest Service,
U.S. Department of Agriculture, Moscow, ID.

1. Seeds intact (fig. 1A) .

2. A part of the seed coat was carefully
removed so as not to damage the membrane that
lies between the seed coat and gametophyte
(fig. IB).

3. The entire seed coat was removed. Because
the membrane is attached to the seed coat at both
ends of the seed, part of the membrane was also
removed (fig. 1C) .

4. A slice was cut out of the seed. This cut
included the seed coat, membrane, and a small
part of the gametophyte (fig. ID) .

5. The tip (radicle end) of the seed was cut
off. This could be only 1-2 mm (1/16-inch) from
the tip so as not to cut the radicle or at least
to minimize cutting it (fig. IE).

112

B

Figure 1. — Western white pine seed treatments: A, intact; B, part of the seed coat was removed leaving the
papery membrane intact; C, all of the seed coat was removed which tore part of the membrane away with it;
D, seed was cut on the side, cutting through the seed coat, membrane, and into the gametophyte ; E, the tip
(radicle end) was cut, cutting seed coat, membrane, and a very small piece of the gametophyte.
(magnification = 7x)

The seeds were incubated in 6-cm (2.4-inch)
plastic petri dishes containing two filter paper
pads that were kept moist. Each test was laid out
on a laboratory bench maintained at 20 °C (68 °F) .
Germination data were taken 21 days after incubation
began. A seed was counted as a germinate when the
radicle had grown at least one-half the length of
the seed.

The analysis of variance and expected mean
squares are shown in table 1. The model assumes
that treatments are fixed and stratification time
and families are random. Before analysis, the
data were transformed to arcsin and differences
among treatment and stratification means were
determined with Duncan's New Multiple Range Test
(Steel and Torrie 1960) .

Table 1. — Model for analysis of variance and expected mean squares

Source of variation

Degrees

°f ]
freedom Expected mean squares

Block

1

o2
e

+ tsf a2
b

Treatment (trt)

4

O2
e

+

bo2 ,
tsf

+

bso2 +
ts

bfo2 + bsfo2
ts t

Stratification time (str)

4

o2
e

+

bta2
sf

+ btfo2

Families (fam)

19

a2
e

+

bta2
sf

+ btso2
ts

Trt x str

16

o2
e

+

tsf

+

bfo2
ts

Trt x fam

76

a2
e

+

bo2 ,
tsf

+

bso2f

Str x fam

76

a2
e

+

bta2
sf

Trt x str x fam

304

o2
e

+

tsf

2

Error

499

o2
e

Where: b=2, t=5, s=5, f

20.

Contains all sources of variance involving interactions of blocks.

113

Table 2. — Average percentage germination of seed from 20 families of western white pine after various
treatments and stratification times

Days of cold-wet stratification

Seed treatment

21

42

90

105

Mean

Intact

Seed coat cut - membrane intact
Seed coat removed - membrane broken
Side of seed cut - membrane cut
Tip of seed cut - membrane cut

Mean

7
16
49
50
51

35

36
33
61
62
62

51

44
47
69

77
71

62

81
81
98
94
94

89

83
84
98
99
99

93

50
50
75
76
75

66

Family heritability was calculated using the
formula :

f sf

e
bts

Stratification time

0 days 21 days 42 days 90 days 105 days
x germination 35 51 62 89 93

RESULTS

An increase in cold-wet stratification time
resulted in an increase in germination for
all seed treatments (table 2) . This varied
from 35 percent for no stratification over
all treatments to 93 percent after 105 days'
stratification. These means were significantly
different at the 1 percent level of probability
(table 3) . Stratification time accounted for
54 percent of the variation.

Results of Duncan's New Multiple Range Test of
the germination means for stratification time
were :

Means underscored by the same line are not
significantly different at the 5 percent level of
probability.

The effect of seed treatment is also shown in
table 2. These means were significantly
different at the 1 percent level of probability
(table 3) . Treatments accounted for 14 percent
of the variation. Duncan's test at the 5 percent
level of probability indicated that treatments 1
and 2 (seed coat intact and membrane intact) were
not significantly different and that treatments 3,
4, and 5 (those that disrupted the seed coat and
membrane either by removal or cutting) also did
not differ, but that these two groups differed.

Table 3. — Analysis of variance

and variance components

for percent germination

of western white

pine seed

Degrees

of

Mean

Variance

Percent of

Source of variance

freedom

square

component

variance

Block

1

0

008

-.000

Treatment (trt)

4

6

881**

.033

14

Stratification time (str)

4

25

827**

.129

54

Families (fam)

19

0

675**

.012

5

Trt x str

16

0

202**

.004

2

Trt x fam

76

0

118**

.006

3

Str x fam

76

0

082**

.005

2

Trt x str x fam

304

0

056**

.010

4

Error

499

0

036

.036

15

**Statistically significant at the 1 percent level of probability.

114

Treatments 3, 4, and 5 averaged 50 percent germina-
tion with no stratification compared to 12 percent
for treatments 1 and 2. The differences in seed
treatment decreased with stratification time. But
Still, at 105 days' stratification, germination of
the seeds with intact coats and membranes had not
equaled that of seeds whose coats and membranes
were disrupted.

Seed lot also had a significant effect on germina-
tion (table 3) . Table 4 combines treatments 1
and 2 and shows the frequencies of family
germination means for each stratification time.
Intact seeds from two trees had more than
30 percent germination with no stratification, and
there were two other trees with very low seed
germination, even after 105 days' stratification.
Nearly 100 percent of the seeds of all four of
these trees germinated after 105 days' stratifica-
tion when the seed coats and membranes were
disrupted.

Table 5 combined treatments 3 , 4 , and 5 to show
frequencies of family germination means for each
stratification time. The lowest average germination
was 22 percent for no stratification and the
highest was 75 percent. After 105 days, the
lowest germination was 92 percent.

Family heritability was:

.012 (.90)

hf

.012 + .005 + .036

= .79

50

Additive genetic variation was decreased by

10 percent to account for average inbreeding in

natural stands .

Table 4. — Frequency of average germination by

family for each stratification time for
treatments 1 and 21

Days of

stratification

Percent

class

0 21

42

90

105

1-10

11 1

1

11-20

7 3

2

21-30

6

2

1

1

31-40

2 3

4

41-50

4

3

1

51-60

3

4

1

61-70

1

3

2

71-80

1

2

2

81-90

2

5

5

91-99

5

6

100

3

3

^"Treatment 1 =

= seed coat intact

r

treatment 2 =

membrane intact.

Table 5. — Frequency of average germination by

family for each stratification time for

treatments 3, 4, and 5

Days of stratification

Percent
class

21

42

90

105

1-10
11-20
21-30
31-40
41-50
51-60
61-70
71-80
81-90
91-99
100

2
15
3

7
13

Treatment 3 = seed coat and part of membrane
removed, treatment 4 = seed coat and membrane cut,
treatment 5 = seed tip cut.

DISCUSSION

The effect of stratification time was obvious and
expected. But still 105 days were not enough to
overcome all the components of dormancy for the
intact seed. Seeds from two trees had rather low
germination (30 and 42 percent) after 105 days'
stratification, and nearly every seed lot had some
seed that did not germinate after 21 days' incuba-
tion. Cutting these ungerminated seeds after 21
days' incubation resulted in rapid germination.
Thus, these seeds were physiologically viable.

Considering the various seed treatments together
with stratification time reveals some of the
"sites of dormancy." With no stratification,
7 percent of the intact seed germinated; 16 percent
germinated with an intact membrane; and 50 percent
of the seed with disrupted seed coat and membrane
germinated. Interpretation in terms of dormancy
indicates that 7 percent had no dormancy, 9 percent
(16-7) had seed coat dormancy, 34 percent had
dormancy due to the papery membrane (50-16) , and
50 percent (100-50) had physiological elements in
the embryo, gametophyte, or both, that prevented
germination. Even though Duncan's test indicated
that the intact seed and intact membrane treatments
did not differ at all stratification times, it is
possible that they do with no stratification, and
thus they were kept separate. The treatments
that disrupted the seed coat and membrane were
averaged for each of the stratification times.
Table 6 shows the proportional change of these
sites of dormancy over stratification time.

115

Table 6. — Change in percentage of western white
pine seed with no dormancy, with seed
coat, membrane, and physiological
dormancy for five levels of cold-wet
stratification

Days of cold-wet stratification

Site of

dormancy

0

21

42

90

105

Percent -

None

7

35

44

81

84

Seed coat

9

0

0

0

0

Membrane

34

22

6

14

15

Physiological

50

38

28

5

1

Physiological dormancy has probably been overcome
by 90 days' stratification and possibly much
earlier — sometime between 42 and 90 days. Seed
coat dormancy appears to be eliminated by about
21 days of stratification. Thus, the main
element of dormancy that is present after 90 and
105 days involves the papery membrane.

The membrane or seed coat do not appear to be
physical barriers, because seeds that are physiolog-
ically ready to germinate and that are inhibited
by the seed coat/membrane will readily germinate
through the seed tip (normal area for emergence of
the radicle) when the seed is cut on the side.
These structures are most likely to be barriers to
water or gas exchange. For some trees, the seed
coat/membrane is thought to be a barrier to oxygen
transport (Stone 1957; Kozlowski and Gentile 1959).

That seed germination is strongly influenced by
family is indicated by the high heritability . To
assure equal representation of all families in a
particular seed lot, seeds of each family should
be kept separate, given the specific treatment
needed to break dormancy, sown separately, and
then combined after the seeds have germinated.

REFERENCES

Anderson, H. W.; Wilson B. C. Improved stratifica-
tion procedures for western white pine seed.
Report 8. Olympia, WA: State of Washington,
Department of Natural Resources; 1966. 11 p.

Hoff, R. J.; Steinhoff, R. J. Cutting stratified
seed of western white pine to determine
viability or increase germination. Tree
Planters Notes. [In press] .

Kozlowski, T. T.; Gentile, H. C. Influence of the
seed coat on germination, water absorption,
and oxygen uptake of eastern white pine seed.
For. Sci. 5:389-395; 1959.

Larsen, J. A. Methods of stimulating germination
of western white pine seed. J. Agric. Res. 31:
889-899; 1925.

Malone, P. L. Effects of seed coat, short stratifi-
cation, and hydrogen peroxide on germination
of western white pine seed. In: Proceedings;
western white pine management symposium; 1983
March 8-10; Coeur d'Alene, ID. Missoula, MT:
U.S. Department of Agriculture, Forest Service,
Northern Region; 1983: 106-120.

Partridge, A. D.; Woo, J. Y . ; Advincula, B. A.
Improved germination of western white pine
seeds. Tree Planters Notes. 36:14; 1985.

Steel, R. G. D.; Torrie, J. H. Principles and

procedures of statistics. New York: McGraw-Hill;
1960. 481 p.

Stone, E. C. Embryo dormancy of Pinus jef f ereyi
Murr. seed as affected by temperature, water
uptake, stratification, and seed coat. Plant
Physiol. 32:93-99; 1957.

Works, D. W.; Boyd, R. J., Jr. Using infrared

irradiation to decrease germination in various
species of western conifer trees. Trans. ASAE
15: 760-762; 1972.

CONCLUSIONS

1. Cold-wet stratification resulted in high
germination for most seed lots, but 105 days were
not enough for some seed lots.

2. Disrupting the seed coat and membrane will
result in moderate germination even without
stratification .

3. The sites of dormancy observed were: seed
coat, papery membrane, and physiological elements
of the embryo , gametophy te , or both .

4. Family heritability was high (0.79) indicat-
ing a substantial genetic component.

116

SEED DORMANCY IN THREE PINUS SPECIES OF THE INLAND MOUNTAIN WEST,

<5r

Carole L. ILeadem

!_

ABSTRACT: Several seed sources of Pinus
a 1 bi caul i s , Pinus contorta , and Pinus
monti col a were given various treatments to
determine the degree and type of dormancy
restricting germination in each species. The
occurrence of physiological and mechanical
dormancy was investigated using stratification
(moist chilling at 2*C) and partial removal of
the seedcoat. Development of embryo and
gametophyte tissue was assessed using X-rays.
Where incomplete development was suspected, a
combination of warm (20°C) and cold (2°C)
stratification was also employed.

Physiological dormancy, measured as the degree
of response to various stratification periods,
was greatest in P_. al bicaul i s and P.
monti col a , and least in P. contorta . Seed
coats imposed significant restraints to
radicle protrusion in P. al bicaul i s and P.
monticola, but only slightly affected
germination of P. contorta . Germination of P.
al bi caul i s is also commonly restricted by seed
immaturity due to short growing seasons found
in subalpine regions.

The results are discussed with regard to the
dormancy strategies and the habitats in which
each of the species is generally found.

INTRODUCTION

Seeds are said to be dormant when they are
placed under conditions favourable for growth,
yet fail to germinate. Dormancy, found in
most tree seeds, is an important adaptive
mechanism because it ensures the survival of
the species by delaying germination until
conditions in the external environment are
conducive to active growth (Osborne 1981).
The expression of dormancy is under genetic
control (Naylor 1983), but it is also strongly
influenced by environmental factors (Steinhoff
and others 1983; Rehfeldt 1983, 1985).
Through variations in environment, plant
populations become physiologically specialized
to segments along the environmental gradient.
Dormancy is part of the life strategy by which
species adapt to the local environment. Since
the most successful species are those which

Paper presented at the Conifer Tree Seed in
the Inland Mountain West Symposium, Missoula,
MT, August 5-6, 1985.

Carole L. Leadem is Tree Physiologist,
Research Branch, British Columbia Ministry of
Forests, Victoria, British Columbia, Canada.

are most suited to a particular habitat,
specializations in the manner by which
dormancy is controlled result in a more
successful life strategy. Control of dormancy
can be exerted exogenously by physical,
chemical, or mechanical means, but can also be
achieved endogenously via morphological or
physiological traits (Nikolaeva 1977). The
type of control varies by species, even within
members of the same genus.

Pinus al bicaul i s Engelm. , Pinus contorta var.
latifol ia Engelm., and Pinus monticola Dougl .
are three pines found primarily in mountainous
areas throughout the western United States and
Canada (fig. 1). P. a 1 bicaul is (whitebark
pine) is a subalpine species common at high
elevations growing in shallow soils on exposed
slopes and rocky ridges. Climatic conditions
are characterized by cool summers and cold
winters with deep winter snowpack. Trees have
high frost resistance but low shade tolerance
(Krajina 1969; Franklin and Dyrness 1973). P.
contorta (lodgepole pine) is an extremely
adaptable tree which occurs in habitats from
dry forest to swamp, and from lowlands to
subalpine. Climatic requirements are also
wide, but shade tolerance is almost nil, even
in the driest habitats (Krajina 1969). It is
usually considered a pioneer species, although
it can be a persistent serai or climax species
under certain conditions (Volland 1985). P.
monticola (western white pine) is a relatively
diverse species growing anywhere from mesic
forests to swamps and bogs. It is found in
lowlands, but is also well adapted to interior
montane areas. Trees are moderately shade
tolerant, but are not very frost resistant.
P. monticola requires at least 50 cm
precipitation in the interior and 100 cm along
the coast (Krajina 1969).

Figure 1. — a; cone lengths, clockwise from left,
P. monticola (13-20 cm), P. contorta (2-6 cm),
and P. al bicaul i s (4-8 cmTT b; seed lenqths,
left to right, P. albicaul is (8-13 mm) P.
monticola (5-8 mm), and P. contorta (3-4 mm).

117

The three species can occur sympatrical ly
(Sowell and others 1982), but in zones where
they coexist, such as the Abies lasiocarpa
zone of Oregon and Washington, P. monticola
tends to be distributed lower in the zone,
whereas P. albicaulis is conspicuous higher in
the zone (Franklin and Dyrness 1973). In
addition to their distribution along
environmental clines, there are indications
that these three species have variable
requirements for breaking seed dormancy
(Anderson and Wilson 1966; Pitel and Wang
1980; Helium and Wang 1985). The questions to
be addressed in this study are how dormancy
mechanisms vary among closely related members
of the same genus, and if control of dormancy
can be related to their habitat of origin.

MATERIALS AND METHODS
Pinus al bicaul is Engelm.

In 1981 and 1982 P. albicaulis seeds (lots P81
and P82) were collected from a small stand of
trees on Baker Mountain (latitude 49°28',
longitude 115°38', elevation 2200 m) near
Cranbrook, British Columbia. Seeds were
X-rayed prior to treatment to remove empty
seeds and to assess gametophyte and embryo
development. For experiments in which they
were sterilized, seeds were soaked for 5
minutes in a 4% solution of sodium
hypochlorite (NaOCl), rinsed three times with
deionized water, then soaked for 48 h in fresh
deionized water. Cold stratification to
release seeds from physiological dormancy was
conducted at 2°C for either 30 or 60 days, as
indicated in tables 1 and 2. Compound
stratification was used to promote development
and break dormancy of immature seeds. In
compound stratification seeds receive 30 or 60
days warm stratification at 20°C followed by
30 or 60 days stratification at 2°C.

For germination tests seeds were incubated for
30 days at 20°C with continuous light (Pitel
1982). Four replications of 25 seeds each
were used for each treatment. In instances
where seeds were clipped, 1 mm of the seedcoat
was cut from the radicle end just prior to
incubation. Data were analyzed by Analysis of
Variance and differences between treatment
means were determined using Duncan's Multiple
Range Test.

Imbibition studies were performed on 20-seed
samples soaked in deionized water at 20° C.
Water uptake was determined from the means of
four samples which were weighed at 0, 1, 2, 4,
8, 24, 48, 72, and 96 h after the start of
soaking. Moisture content (m.c.) on a fresh
weight basis was calculated after seeds had
been dried for 24 h at 105°C. Determinations
were made on a total of three southern B.C.
interior seed sources (including P82).

Pinus contorta Var. latifol ia Engelm.

Seeds were obtained from four seed sources
collected in the interior of B.C. between 1967
and 1973.

Prior to germination testing, seeds were
soaked for 24 h and stratified for 3 wk at
2°C. Seeds were incubated at 30°C/20°C with
8 h light for 3 weeks. Four replications of
100 seeds were used for each treatment.
Germination data were analyzed and water
uptake was measured as described for £.
albicaul is.

Pinus monticol a Dougl .

Three southern B.C. interior seed sources
collected in 1964 and 1977 were used for
experiments. To determine the most effective
means of breaking dormancy, seeds were soaked
for 24 h, then given five treatments: (1)
stratification at 2°C for 60 days, (2)
stratification at 2°C for 90 days, (3)
stratif ication-redry, i.e., stratification for
30 days at 40% m.c, followed by 90 days at
30% m.c, (4) stratification for 0 days, (5)
stratification for 30 days at 20°C, followed
by 60 days at 2°C. To ascertain the degree of
mechanical restraint imposed by the seedcoat,
all stratification treatments were combined
with a seedcoat treatment. Either 1 mm of the
radicle end of the seed coat was cut off, or
the coat was left intact. After
stratification and seedcoat treatments, seeds
were incubated at 30°C/20°C for 3 weeks with
8 hours light during the high temperature
period. Four replications of 100 seeds were
used for each treatment, and data were
analyzed as for P. albicaulis. Water uptake
was measured as described for P. albicaulis.

RESULTS

Pinus albicaulis

Morphological dormancy, defined by Nikolaeva
(1977) as under-development of the embryo, was
clearly evident in P_. albicaul is. Embryo and
gametophyte tissue incompletely filled the
seeds prior to treatment (fig. 2a), but
maintaining imbibed seeds under warm
conditions for 30 to 60 days effectively
promoted development of the immature tissue.
This was visually demonstrated in X-rays taken
following stratification (fig. 2b) and by
results of germination tests (table 1).

The effects of cold temperatures on
germination also indicated the presence of
physiological dormancy. Warm temperatures
enhanced seed development, but seeds still
required an additional 60 days at 2°C to break
dormancy (table 1). However, it was first
necessary to remove mechanical restraint by
clipping the coats, for even with the longest
compound stratification, only 4% of the seeds

118

Figure 2. --Seed development of Pinus albicaul is
as determined by X-ray. (a) prior to treatment
(b) after 30 days at 20°C plus 60 days at 2°C.

with intact coats were able to germinate. A
high incidence of microbial growth was noted
in all treatments during the first test, so a
second experiment was devised to examine if
surface sterilization would reduce mold and
thus improve germination. Seeds were
sterilized, rinsed, and imbibed for 48 h prior
to receiving either 60 days cold, or 60 days
warm plus 60 days cold stratification. All
coats were clipped prior to incubation to
remove mechanical restraint. Sterilization
only slightly increased germination of seeds
which received cold stratification, but
significantly improved germination of seeds
which received the combined warm and cold
treatment (table 2). Mold was still apparent
during testing, but in view of the gains made
in germination, sterilization was continued as
the standard procedure.

A third experiment was performed to see if
there was any benefit to modifying the
procedure by completely removing the seed
coat. Seeds were sterilized, then stratified
using the combined warm and cold treatment.
After stratification, coats were either left
intact, clipped, or removed. In intact seeds,
germination was 8%, but both seedcoat
treatments resulted in higher emergence (table
3). There was apparently no advantage to
removing the coat, since germination was about
30% regardless of whether the coat was clipped
or entirely removed.

P. al bicaul i s is a hard-coated pine seed, but
coats did not appear to be impermeable to
water. Early water uptake was rapid, reaching
30% m.c. within 24 h, but seeds were not fully
imbibed even after 96 h (fig. 3).

Pinus contorta

Most British Columbia sources of P. contorta
will germinate moderately well without
stratification. However, stratification
usually increases both germination rate and

total germination, so chilling for three weeks
at 2°C is commonly recommended (International
Seed Testing Association 1976).

Table 1. --Effects of clipping and

stratification on germination of
Pinus albicaulis (Lot P81)

Treatment

Germination

percent

Days Days
(a 20°C @ 2°C Clipping

0

60

0.0

0

60

+

0.0

30

30

0.0

30

30

+

0.0

30

60

0.0

30

60

+

8.0b

60

60

4.2°

60

60

+

30. 0a

Germination after 30 days incubation at 20°C
with continuous light. None of the treatments
was sterilized prior to treatment. Means with
the same letter are not significantly different
at p=0.05.

Table 2. --Effects of sterilization and strati-
fication on germination of Pinus
albicaulis (Lot P82)

Treatment

Germination
percent

Days Days
p 20°C @ 2°C Ster. Clip

0 60 - + 6.7C

0 60 + + 9.9C

60 60 - + 20.2°

60 60 + + 31. 4a

All seeds were clipped at the radicle end
just prior to incubation. Seeds were
incubated for 30 days at 20°C with continuous
light. Percentages followed by the same
letter are not significantly different at
p=0.05.

119

Table 3. --Effects of the seedcoat on germin-
ation of stratified P. albicaul is
seeds (Lot P82)

Germination

Treatment

percent

Intact coat

8.0b

Clipped coat

32. 2a

No coat

30. 0a

stratified for 60 days at 20°C plus 60 days at
2°C. Incubation was for 30 days at 20°C with
continuous light. Means with the same letter
are not significantly different at p=0.05.

SOAK TIME (H)

Figure 3. --Moisture content of P. albicaulis,
P. contorta, and P. monticola seeds as
affected by length of imbibition: □ £.
albicaulis, A £. contorta , O £. monticola.
Data points are based on the means of 12
separate moisture determinations. Note that
only half of the standard error bar is shown
for each point.

To assess the degree of dormancy in £.
contorta, stratified seeds were compared to
unstratified controls. Mean germination of
the stratified and unstratified seeds was not
significantly different, indicating that
physiological dormancy was not present in the
lots used in this study.

Generally, it would be expected that
germination of clipped seeds by removal of
mechanical restraint would be the same or
slightly higher than -that of intact seeds, but
cutting the seeds decreased average
germination from 60% to 42% (table 4).
Cutting the seeds not only destroyed the
integrity of the coats but probably damaged
the seeds in other ways, although mold was not
a problem in intact seeds, substantial fungal
growth was present on seeds which had been
cl ipped.

Seedcoats of £. contorta do not appear to
mechanically restrict germination, nor do they
act as a barrier to water. A study of the
water uptake performed on three seed sources
showed that seeds were almost completely
imbibed within one day (fig. 3). Seeds
reached 28% m.c. during the first 24 h and
gained only an additional 3% m.c. during the
remainder of the 96 h period.

£. monticola

Relative to undipped controls, clipping the
seedcoats increased germination from 10% to
40% (fig. 4). On average, unstratified seeds
germinated only 8% without clipping, but about
38% with clipping.

Table 4.— Effects of clipping and

stratification on germination of
P. contorta

Germination percent

Seedlot + Strat - Strat - Strat
- Clip - Clip + Clip

1427
1806
2112
2238

MEAN

54. 3a
62. 0a
57.8ab
59. 8a

54. 5a
60. 3a
68. 5a
57. 5a

45. 5a
36. 3b
42. 5b
41. 5b

58.5

60.2

41.5

Seeds were soaked for 24 h, stratified for
3 wk at 2°C, and incubated at 30°C/20°C with
8 h light for 3 wk. Means with the same
letter are not significantly different at
p=0.05.

120

Stratification consistently increased
germination capacity but the degree of
response varied with the treatment and seed
source. For lot 633, stratification for 60
days at 2°C was most effective, but for lots
849 and 3249, the best treatment was 90 days
at 2°C. For stratified and clipped seeds,
germination was 73%, 82%, and 92% for lots
633, 849, and 3249, respectively.

The two alternate stratification techniques
were less effective than the standard method.
In the stratif ication-redry technique
(treatment 3), moisture is controlled during
stratification by chilling seeds for 30 days
at 40% m.c, then for 90 days at 30% m.c.
(Danielson and Tanaka 1978). This treatment
has proven effective in releasing Abies
species from dormancy (Edwards 1982; Leadem
1985). However, performance of P. monti col a
was very poor using this treatment, averaging
only about 10% germination (fig. 4). It
appears that the redry method either requires
some modification before being applied to P.
monticola , or alternately, is entirely
unsuitable for breaking dormancy of these
seeds.

Compound stratification was included because
it is frequently used for immature seeds to
enhance embryo growth, however, when tested on
P. monticola, germination was usually not
greater than that of unstratified seeds (fig.
4). However, response to a stratification
method promoting embryo growth was not
expected, because embryos appeared to be fully
elongated in X-rays taken prior to treatment.

Although seed coats imposed a significant
obstacle to radicle protrusion (fig. 4), they
did not constitute an appreciable barrier to
water uptake (fig. 3). Water uptake by P.
monticola seeds, although less than the other
two pines, was virtually complete within 48
h. Moisture contents of intact seeds which
averaged 26% at 48 h had only increased to
28% by 96 h.

DISCUSSION

Seed immaturity, as well as physiological
dormancy, was responsible for the poor
germination of P. albicaulis. Because the
species is found at high elevations, the

□ 60 da.

■ 90 da.

[£] 30 da. + 90 da.

j\v] o da.

M 30 da. + 60 da.

O
<

ui

o

1 00r

80

60

40

20

b c

ab

cd

de

1 §

d «■:■:•: d «-v

1

be

de

e

d

:

J.

633

8 4 9
SEEDLOT

3 2 4 9

Figure 4. --Effects of clipping and stratification on germination of P_. monticol a .
Stratification treatments are in order as follows: 60 days at 2°C; 90 days at
2°C; 30 days at 40% m.c, followed by 90 days at 30% m.c; 0 days at 2°C; 30 days
at 20°C, followed by 60 days at 2°C. Black and hatched or dotted treatment
bars represent germination of clipped seeds, while blank bars are corresponding
percentages for intact seeds. Data points are means of six 100-seed replicates.
Means with the same letter are not significantly different at p=0.05.

121

growing season may be limited to a few months,
leaving very little time for seeds to mature
in the cones. Also contributing to poor
development is premature harvesting by the
Clark's nutcracker and small rodents such as
chipmunks and squirrels. P_. albicaul is cones
are indehiscent, lacking the tracheid cells
■which in other pines cause scales to reflex
and release the seeds (Lanner 1982). Thus,
seed dispersal is dependent upon the
dissection of cones by animals foraging for
food (Tomback 1981). Unfortunately, foraging
animals are not concerned with embryo
development and will harvest seeds long before
they are mature. As a result, it is often
necessary to harvest cones while they are
immature to ensure that they are still intact.

The lack of development in P. albicaul is was
partially overcome by exposing imbibed seeds
to 20°C to enhance embryo elongation, but warm
temperatures alone were inadequate to effect
dormancy release. Seeds required at least 60
days of cold temperatures (2°C) to overcome
physiological barriers to growth. Clipping
the seedcoat was also a necessary prerequisite
for germination. However, the fact that
clipping was necessary indicates that
stratification treatments did not fully
satisfy the germination requirements of P.
albicaul is. Although clipping is considered a
measure of mechanical restraint, I feel that
mechanical and physiological dormancy are not
separate entities. Rather, dormancy should be
considered a single phenomenon representing a
balance between the mechanical constraint of
the coat and the expansive force of the
embryo (Chen and Thimann 1966; Barnett
1972). Stratification is an effective
dormancy-breaking technique which initiates
metabolic processes that may alter the outer
seed layers and increase the growth potential
of the embryo. It is this increased growth
capacity, and to a lesser degree a weakening
of outer tissues, which enables the embryo to
rupture the seedcoat and germinate. It should
be noted that increased growth capacity was
not verified in this study because embryo
growth tests were not performed. How
stratification affects growth potential may be
explored in future studies.

The £. contorta lots employed in this study
did not exhibit dormancy since unstratified
seeds germinated as well, although not as
quickly, as stratified seeds. The seedcoat
did not prevent radicle emergence, nor did it
restrict the entry of water.

For monticola, the failure to germinate
appeared to be primarily due to physiological
dormancy because all three lots used in this
study responded best to stratification
conducted at 2°C for 60 to 90 days. Two
modified stratification techniques were tested
in addition to the above methods: combined
warm/cold stratification, which is often
beneficial for enhancing germination of
immature seeds, and the stratif ication-redry
method, which has been shown to be successful

when extended stratification periods must be
used to overcome dormancy. However, neither
of these alternate methods were as effective
as the standard stratification treatment at
2°C. A strong component of dormancy seemed to
be the restraint imposed by the coat, since
clipping the seedcoats increased germination
regardless of the type of stratification
used. As with the other two pines, seedcoats
only affected protrusion of the radicle and
did not constitute barriers to water. Thus,
as noted earlier, it is not the inability of
the seeds to imbibe water which is responsible
for lack of germination, but other physiological
factors which prevent rupture of the coat.

How can the preceding observations be related
to what is known about the habitats and life
strategies of the three pines? P. albicaul is
is a species which is able to survive in
marginal subalpine habitats with short growing
seasons. The lack of seed wings and
indehiscence of the cones has resulted in
dependence on animals for dispersal of the
seeds. Seeds frequently do not mature on the
tree during the short growing season, but the
caching of the seeds in the soil by animals
allows the seeds to continue ripening during
long storage under the snow. X-ray evidence
and results from germination tests reveal
large variability in seed development and
dormancy. These traits would suit a life
strategy whereby only a few individuals
germinate at a time so that all regeneration
efforts do not occur at once, in the event
that conditions are unfavourable. This
strategy may be conservative, but is more
likely to ensure continuation of the species
in a harsh environment.

P_. contorta is a pioneer species which is
opportunistic in nature. Although it can be
found anywhere from dry to wet sites, it is
generally a shade intolerant species which
starts to decline in importance once the
canopy begins to close. As a prominent member
of the initial serai stage of ecosystems in
which it is found, it must germinate quickly
to fill the pioneer niche. Therefore, most
seed populations of P_. contorta tend not to be
dormant. For this species, there would be
little advantage to delaying germination since
habitats would tend to become less favourable
with time.

£. monticola is a widespread species found in
mesic environments of the western United
States and Canada. Relative to P. albicaul is
and £. contorta , P. monticola exists in a
stable environment where water, temperature,
and shade conditions are generally
favourable. There is no necessity for rapid
germination, nor must germination be extended
over a long period, thus dormancy is probably
governed by other factors.

The results of this study indicate that the
type of dormancy exhibited by seeds of
P. albicaul is, £. contorta and P. monticola
can be related to their habitat of origin.

122

Each species occupies a different ecological
niche, and as life strategies vary according
to changes in local environments, so would
dormancy control mechanisms also be expected
to differ. The data show that dormancy does
vary between species, although the release
mechanisms are presently unknown. The nature
of these mechanisms may be difficult to
determine because of the variable responses to
stratification as shown, for example, by
different seed sources of P. contorta (Kamra
1982; Helium and Wang 19857- Notwithstanding
these difficulties, understanding the control
of dormancy in coniferous seeds would have
significant impacts in both theoretical and
applied forestry. Further research in this
field is therefore strongly encouraged.

ACKNOWLEDGMENTS

Sincere thanks are given to Sue Baker and
Carl White for their expert technical
assistance in performing all the experi-
ments reported in this paper.

Thanks are also given to Chris Thompson, B.C.
Ministry of Forests, Nelson, B.C. for col-
lecting the Pi nus albicaul is seeds from Baker
Mountai n .

REFERENCES

Anderson, H.W.; Wilson, B.C. Improved
stratification procedures for western
white pine seed. Res. Rep. No. 8 (For.
Land Man. Res. Rep. No. 2). State of
Washington, Department of Natural
Resources; 1966. 11 p.

Barnett, J. P. Seedcoat influences
dormancy of loblolly pine seeds.
Can. J. For. Res. 2:7-10; 1972.

Chen, S.S.C.; Thimann, K.B. Nature of seed
dormancy in Phanel ia tanaceti fol ia .
Science. 153:1537-1539; 1966.

Danielson, H.R.; Tanaka, Y. Drying and
storing stratified ponderosa pine and
Douglas-fir seeds. For.Sci. 24(1): 11-16;
1978.

Edwards, D.G.W. Storage of prechilled Abies
seeds. In: Wang, B.S.P.; Pitel, J. A.,
ed. IUFR0 international symposium of
forest tree seed storge: proceedings; 1980
Sept. 23-27; Chalk River, Ont.: Petawawa
Natl. For. Institue, Chalk River, Ont.;
1982: 195-203.

Franklin, J.F.; C.T. Dyrness. Natural

vegetation of Oregon and Washington. Gen.
Tech. Report PNW-8. Portland, OR: U.S.
Department of Agriculture, Forest Service,
Pacific Northwest Forest and Range
Experiment Station; 1973. 417 p.

Helium, A.K.; Wang, B.S.P. Lodgepole pine
seed: seed characteristics, handling and
use. In: Baumgartner, D.M.; Krebill,
R.G.; Arnott, J.T.; Weetman, G.F., ed.
Lodgepole pine, the species and its
management symposium: proceedings; 1984 May
8-10; May 14-16; Spokane, WA. Vancouver,
B.C.; 1985:187-197.

International Seed Testing Association.

International rules for seed testing. Seed
Sci. Technol. 4: 3-177; 1976.

Kamra, S.K. Seed biology of lodgepole pine
(Pinus contorta Dougl.). Report 3. Umea,
Sweden: Swedish University of Agricultural
Sciences, Department of Forest Genetics and
Plant Physiology; 1982. 51 p.

Krajina, V.J. Ecology of forest trees in
British Columbia. Ecol . North America.
2(1): 1-147; 1969.

Lanner, R.M. Adaptations of whitebark pine
for seed dispersal by Clark's Nutcracker.
Can. J. For. Res. 12: 391-402; 1982.

Leadem, C.L. Stratification of Abies amabi 1 is
seeds. Can. J. For. Res. In press; [1985].

Naylor, J.M. Studies on the genetic control
of some physiological processes in seeds.
Can. J. Bot. 61:3561-3567; 1983.

Nikolaeva, M.G. Factors controlling the seed
dormancy pattern. In: A. A. Khan, ed.
The physiology and biochemistry of seed
dormancy and germination. Amsterdam,
Neth.: El sevier/North-Hol 1 and Biomedical
Press; 1977. 51-74.

Osborne, D.J. Dormancy as a survival

strategem. Ann. Appl . Biol .98: 525-531 ;
1981.

Pitel, J. A. Improved germination of whitebark
pine (Pinus albicaul is Engelm. )seeds.
Chalk River, Ont: Petawawa National
Forestry Institute, Can. For. Serv.;
1982. 5 p. Unpublished report.

Pitel, J. A.; B.S.P. Wang. A preliminary
study of dormancy in Pinus albicaul is
seeds. Can. For. Serv. Bi-Monthly Res.
36 (1): 4-5; 1980.

Rehfeldt, G.E. Adaptation of Pinus contorta
populations to heterogeneous environments
in north Idaho. Can. J. For. Res.
13:405-411; 1983.

Rehfeldt, G.E. Ecological genetics of Pinus
contorta in the Wasatch and Uinta
Mountains of Utah. Can. J. For. Res.
15:524-530; 1985.

123

Sowell, J.B.; Koutnik, D.L.; Lansing, A.J.
Cuticular transpiration of whitebark pine
(Pinus albicaulis) within a Sierra Nevadan
timber! ine ecotone, U.S.A. Arctic and
Alpine Res. 14(2): 97-103; 1982.

Steinhoff, R.J.; Joyce, D.G.; Fins, L.
Isozyme variation in Pinus monticola.
Can. J. For. Res. 13:1122-1132; 1983.

Tomback, D.F. Notes on cone and vertebrate
mediated seed dispersal of Pinus abicaulis
(Pinaceae). Madrono. 23(2): 91-94; 1981.

Volland, L.A. Ecological classification of
lodgepole pine in United States. In:
Baumgartner, D.M.; Krebill, R.G.; Arnott,
J.T.; Weetman, G.F., Ed. Lodgepole pine
the species and it's management symposium:
proceedings; 1984 May 8-10 and May 14-16;
Spokane, WA and Vancouver B.C.:
Washington State University, Pullman, WA;
1985: 63-75.

f

9 NET RETRIEVAL
James L.J McConnell

ABSTRACT: The Net Retrieval Seed Collection
System is a mechanical seed harvesting system.
It has been used successfully in southeastern
seed orchards for the past several years.
Techniques, machinery, and costs for the system
are described.

INTRODUCTION

There are well over 6,000 acres (2 428 ha) of
loblolly (Pinus taeda), 600 acres (243 ha) of
short leaf (P. echinata) and 300 acres (121 ha) of
Virginia pine in the Southeastern United States.
These species are all commercially important,
with the loblolly pine far and away the most
important. Also the seed is considered to be
hard to collect because the cone is not easily
freed from the limb. Traditionally, collecting
cones has been the primary means of collecting
seed .

With such large areas of seed orchards to
collect during the short period between cone
ripening and seed fall, seed orchard managers
have sought a better method of collecting the
high-value pine seed needed to reforest the
southern pine wood basket.

The netting-seed collection concept, which
originated with the Georgia Forestry Commission,
shows promise as a method to collect seed. The
Commission spread net material over the seed
orchard floor and waited for the cones to ripen
and the seed to fall. A hand labor force was
used to windrow the material that had accumulated
on the net. The windrowed material was hand-fed
through a peanut harvester. The peanut harvester
did a good job of separating the seed from the
pine straw and other trash. Six years ago the
USDA Forest Service's Southern Region and
Missoula Equipment Development Center began
development of a system that would mechanize the
concept developed by the Georgia Forestry
Commission. The initial goals were to provide
a system that could be operated by five to seven
people, harvest the seed in a relatively short
period of time, and do all this in a safe manner.

Paper presented at the Conifer Tree Seed In The
Inland Mountain West Symposium, Missoula, MT,
August 5-6, 1985.

James L. McConnell, Regional Geneticist, Jerry
L. Edwards, Mechanical Engineer, USDA Forest
Service, Southern Region.

SEED COLLECTION

4

and Jerry L. ^Edwards

The initial goals have been met. One net
retrieval machine can safely harvest seed from
40-60 acres (16-24 ha), using only five to six
persons and do the entire job. Elimination of
the need for tree climbing creates a safer, more
productive, and less labor-intensive method of
seed collection (fig. 1). Except for poor seed
years, the system has proven to be cost-effective.
The seed orchard managers are extremely pleased.
The system is presently on-line at three Southern
Region orchards with the fourth scheduled this
year. Several State and industrial orchard have
begun the use of parts of the system.

Figure 1. — Netting retrieval and seed separation
equipment .

125

the net used in the system is a polypropylene
plastic fabric manufactured initially as a
backing for carpets. Knitted monofilament
polyethylene fabric designed as shadecloth has
also been used successfully (fig. 2).

dumped onto the retriever conveyor as the net is
recovered and fed into the seed separator. The
seed separator separates the seed from unwanted
material (pinestraw, cones, sticks, etc.). The
large amount of relatively clean pine straw
processed can be bailed and used as mulch in
nurseries or sold for landscaping.

The seed, as it comes out of the seed separator,
is only roughly cleaned and requires additional
cleaning and processing before being placed in
cold storage. The seed receives additional
processing at a conventional seed extraction
facility. The seed requires fine cleaning and i;
introduced into the conventional processing
equipment just after the cone tumbler and goes
through the remainder of the seed
cleaning/processing equipment. There is a
substantial savings in cost since the expensive
cone drying-opening process is eliminated.

Figure 2. — Net (polypropylene plastic).

Optimum seed fall in the South usually occurs in
November, but as with most outdoor activities,
it depends on the weather. As weather fronts
move through, rain occurs; after the front
passes, several cool, dry, sunny days occur
bringing good collecting conditions. The
netting is spread over the orchard floor several
'7eeks before cone opening (fig. 3).

Figure 3. — Orchard floor covered with netting.

As the cones open, orchard trees are shaken with
mechanical tree shakers similar to those used to
shake pecan trees. The specially designed
cushioned clamping pads, when attached to the
tree, tend to protect the tree and, at the same
time, transmit the shaking energy to the tree.
The omni-direction shaking characteristics
developed by the shaker head removed nearly all
the seed from the open cones. Experience has
taught us that minimizing tree damage depends on
a competent, well-trained tree shaker operator.
A second shaking prior to retrieval yields
additional seed from the cones. When the seeds
have been dislodged from the cones and have
fallen into the net, one end of the net is
attached to aluminum pipe core which has been
secured to the net retrieval machine. The
retrieval machine operates by turning the core
and rolling the net onto the core. Seed, pine
straw, and other debris on the netting is

EQUIPMENT

Power requirements to operate the netting
retrieval and seed separation equipment are less
than 30 PTO horsepower (34.4 kilowatts). The
PTO power is used to drive a hydraulic pump that
supplies the power to operate the machinery.

An operator's station is located at the rear of
the retrieval equipment. From this location, all
machine functions can be observed and controlled.

COLLECTION COSTS

In 1983, the net collection process produced
1,649 lb (748 kg) of loblolly pine seeds
from 141 acres (57 ha) at three Southern Region
seed orchards. Overall, 1983 was not a good seed
collection year; total production and yield were
down. The seeds were collected at the following
orchards: Francis Marion, South Carolina;
Erambert, Mississippi; and Stuart, Louisiana,

Variable Costs. — Labor and general-type equipment
(vehicles, wheel tractors, etc.) costs totaled
$29,920.

Fixed Annual Costs. — Fixed costs for 1983 were:

Item Total cost

Netting $316,214
Cores 14,040
Retrieval 144,480
Equipment

Expected

life
10 years
20 years
20 years
Total

Annual
amortization
$31,621
702
7,224
$39,547

Cost of net collection, 1983. — Total costs of net
collection for the year were:

Category
Variable cost
Fixed cost

Total
Cost

$/acre
$212!

$29,920

39,547 280

Total $69,467 $492

Number of acres = 141 (57 ha)

$/ha
524
692
$1216

126

1983 COLLECTION

Cones collected in 1983 from the same orchards
yielded 1.1 A pounds of seeds per bushel (0.12
kg/dl). The seeds collected from the netting
system weighed 1,649 lb (748 kg). The equivalent
number of bushels of cones required to yield the
seeds obtained from the netting is 1,446 bushels
(5 096 dl). The collection of cones obtained
by contract or force account (using Forest
Service workers) would have been $33 per bushel
($9.35 per dl):

Collection $30
Drying and extraction 2
Transportation to extractory 1

$33/bushel

($9.35/dl)

The cost of cone collection FY 1983 (hypothetical)
was $47,718 (volume of seed X cost/unit).

The comparison of costs (cone collection versus
netting system) was:

$69,467 Cost of net collection
-47,718 Cost of cone collection
$21,749

This proves that cone collection would have been
more economical.

net, the lower the cost per pound (kg) of the seed.
Retrieval and quality of separation of the seed is
virtually unaffected by the volume of seed. On
the other hand, the volume of debris (pine straw,
twigs, etc.) on the net has a measurable effect on
the rate of separation.

In general, the smaller the seed and cone crop,
the more the advantages are in favor of cone
collections. When the crop is small, the cones
can be selectively collected; however, it is nearly
impossible to selectively deploy netting and still
catch the seed fall.

The equipment now used for the net retrieval system
is considered a first-generation production
prototype. Many improvements will be made to
produce a more efficient and compact system. The
cost of equipment may continue to rise, but
probably not as fast as the cost of labor,
especially the trained labor force that is required
in cone collection.

The initial costs of the net retrieval system are
high. Therefore this system would not be
economical in young orchards or an orchard with a
low production capacity. The following factors
can be used to decide whether to use the net
system or harvest cones in a particular year:

Net Seed Collection System

1984 COLLECTION

In Fiscal Year 1984, 4,529 lb (2 054 kg) of
loblolly pine seeds were collected from 216
acres (87 ha) within the netting systems from
the three Forest Service seed orchards mentioned
earlier. Seed production was spotty; east coast
collections were light, but Culf coast collections
were good to heavy.

Cost of net collection in 1984 was:

Category
Variable cost
Fixed costs

Total

Total cost

$48,575
39,547
$88,122
^No. of acres = 216

$/acre
$225'
183
$408

($/ha)
$ 558
454
$1012

The cost of cone collection in 1984 (hypothetical)
was $104,511 (volume of seed X cost/unit).

La Number of acres (ha) of orchard under
considerat ion .

2. Calculated production capacity of the
orchard (number of bushels (dl) of cones or pounds
(kg) of seed in the orchard under consideration);
end product will be pounds (kg) of collectible
seeds.

3. Cost of deploying and recovering the net
seed collection system on the following basis:
a five-person crew operates at the rate of 0.25
acres (0.11 ha) per crew per day. This work
includes the total job of deploying the net,
shaking the trees (twice), retrieving the net,
processing the seed, and returning the rolls of net
to storage. The time sequence becomes relatively
unimportant, because much of the job takes place
before and after the cone ripening period. With
this information, the orchard manager can calculate
the price of seeds per pound.

Cone Collection System

The comparison of cost (cone collection versus
netting system) was:

$104,511 cost of cone collection
- 88,122 cost of net collection
$ 16,389

This proves that net collection would have been
more economical.

RESULTS AND DISCUSSION

The cost of the net seed collection system is
greatly affected by the volume of seed
available. The larger the volume of seed on the

La Price per bushel (dl) to collect cones,
transport them and extract the processed seed.

2. Can the cone crop be picked before cones
mature to the point of opening?

3. Is an adequate supply of collection
equipment available?

4. Is there an adequate pool of people to do
the work safely?

If the price per bushel (dl) for the netting
system is higher than that for harvesting cones,
the orchard manager would then decide that, in all
probability, cone picking will be the most
economical method.

127

Every orchard and organization will generate a
different number, but we feel the net retrieval
seed collection system is a viable alternative to
a difficult job. Netting material and tubing
specifications are shown in the appendix.

APPENDIX

Netting material specifications

Material: Polypropylene/polyethylene plastic
Width: 16 ft, 5 inches (4.75 m)

Length: 350 ft (106.16 m)1

Salient characteristics:

Color - Black

Weave count - 6 X 8 per in^

Weight - Minimum 2.1 oz per yd^

(50 g per m2)
Maximum 3.0 oz per yd^
(17.4 g per m2)

Core (tubing)

Material: Aluminum Alloy 6063 T6

Diameter: 4.0 inches (10,6 cm)

outside diameter

Length:
Weight :

17 feet, 3 inches (5.23

0.73 lb per lineal foot
(56.5 g per lineal m)

Specification: Federal Specification
QQ-A-200/9c

Cost !

$20/piece (approximate)
— 1982 contract price

Source of supply:

Reynolds Metals
6601 Broad Street
Richmond, VA 23261
USA (804) 281-2655

Tensile strength - Minimum 60 lb warp (27 kg)
warp (length)
Minimum 70 lb (31.5 kg)
fill (width)

Burst strength - Minimum 175 lb per in^
(12.2 kg per cm^)

Yarn stability - Minimum 1 oz (250 g)

Cores - All cores to be continuous

in lengths 17.25 ft
(5.23 m) overall

Outdoors wearing - Minimum of 70 percent

retention after 400 hr in
weather-o-met er .

Selvedge edge - Minimum of 0.25 inch (0.64
cm) selvedge area for each
edge.

Cost - $1.62/lineal yd

($1.47/lineal m)
@16'5" (4.75 m) width
$1418/acre ($3545/ha)
— 1982 contract price

Source of supply:

Amoco Fabrics Company
Patchogue Plymouth Division
550 Interstate North Parkway
Atlanta, GA 30339
(404) 995-0935

Weathashade

568 West Orange Blossom Tr.

Apopka, FL 32703

(305) 889-3692
-^Length can be varied to meet needs of the
individual orchard.

128

DOUGLAS-FIR SEED COLLECTION AND HANDLING ^
Richard M.j^Schaefer III and Daniel L. jMiller

ABSTRACT: High-quality conifer seed is a must for
successful and economical seedling production,
especially in containerized operations. High-
quality seed is characterized by high percent
germination (90+), uniform germination and insig-
nificant mold and disease problems. High-quality
seed can be assured by attention to details during
collection, handling, and processing. Collections
should be made only after the seed has fully
matured. Immature collections result in reduced
and erratic seed germination. Proper cone handling
and storage methods will prevent problems associa-
ted with cone heating and molding. Skillful
processing will produce clean, dry seed and will
preserve the germination potential obtained
through proper collection and handling.

INTRODUCTION

Seed collections must stress quality as well as
quantity. Seed quality is especially important in
greenhouse nurseries. As containerized seedling
production increases so does the need for high-
quality seed. For instance, using a seed cost of
S5.00/M seedlings, increasing germination from 50 to
95 percent increases seed value by five times
(table 1). Actual growing costs associated with the
lower germination potential seed are considerably
higher than the single-sown 95+ percent germination
seed because the increased seed volume means higher
stratification, sowing, and crop thinning costs.
Additional savings for 95+ percent seed could be as
much as $5-$10/M in container nurseries. The 95+
percent germination seed also reduces grower risk
from additional failures caused by fungus and low-
vigor seedlings resulting from poor quality seed.

Table 1. — Douglas-fir seed values per pound for a
containerized nursery assuming a $5.00/M
seedling seed cost

Germination
Percent

Seeds per
Cavity1

Seeds Per Pound

25M 35M ASM

High-quality seed also reduces seed inventory and
storage costs. Prior to implementing the procedures
described here, our Douglas-fir seed germination
rates ranged from 12 to 39 percent. Assuming this
seed is usable in our nursery, we would need an
additional $675,000 in seed inventory to operate
at current capacity. High-quality seed is also
important in bare root nurseries, especially as
precision seeders come into use.

These dollar values become more important when you
consider that obtaining high-germination seed costs
no more than obtaining low-germination seed. The
secret to high-quality seed is attention to detail.
Cones must be harvested when seed is ripe and han-
dled and stored so as to preserve seed quality. It
costs just as much to harvest immature seed with 30
percent germination as it does to collect mature
seed with 90 to 100 percent germination. In fact,
immature or insect damaged seed is often more diffi-
cult to extract. This increases processing costs
and reduces seed yields.

The methods described here were developed from
communications with the late Mr. Charles Brown of
Brown Seed Company, Vancouver, Washington. Charlie's
methods were developed during more than 30 years of
working with conifer seed. His goal was to produce
the best possible conifer seed. To do this, he
found that the cones must be fully ripe when collec-
ted and stored under proper conditions. If not,
seed viability was reduced. His cone maturity indi-
cators are the result of years spent observing cone
ripening. These cone handling procedures were
developed to prevent molding which reduces seed
germination and avoid cone heating which causes
varying dormancy levels. Under Charlie Brown's
direction, Brown seed earned the worldwide reputa-
tion of being the best available. This is ample
testimony that his procedures work.

The following are Brown's methods that Potlatch
foresters are using. These methods have resulted
in the collection of consistently high-quality
Douglas-fir seed with germination rates of 85 to 99
percent .

95+ 1 $125 175 225

75 3 42 58 75

50 5 25 35 45

Assumes a sowing rate that will produce at least
95 percent stocked containers.

Paper presented at the Conifer Tree Seed in the
Inland Mountain West Symposium, Missoula, MT,
August 5-6, 1985.

Richard M. Schaefer III is Seedling Production
Supervisor, and Daniel L. Miller is District
Forester, Potlatch Corporation, Lewiston, ID.

SEED INVENTORY PLANNING

Douglas-fir usually produces a medium to heavy cone
crop once every five to seven years in northern
Idaho. At Potlatch, we strive to maintain a five-
year seed supply on hand for each seed zone. We
have delineated seed transfer zones by ± 500-foot
elevation classes and by 15-mile geographic distance
within the same habitat type. Five-year harvest
plans are used to estimate the number of seedlings
and amount of seed needed for each zone. Collections
are scheduled only during heavy cone crop years

129

unless seed inventory is low for specific zones.
Estimated seedling needs are converted to pounds of
seed and bushels of cones for planning and budget-
ing purposes. These estimates are developed during
the winter so that selected areas can be checked
for potential cone crops in early spring. This
planning also allows scheduling logging to prevent
harvest in selected collection areas prior to cone
ripening. This is especially necessary for fall-
and-pick cone collections, our preferred method.
Also, other stands may be reserved as permanent
cone collection areas. This is especially impor-
tant if there are few trees of the desired species
of cone bearing age within the seed zone.

EARLY CROP ANALYSIS

Potential cone crops can be evaluated as early as
mid-June. By then, the female cones have been
pollinated and are large enough to be easily seen.
Also, cones damaged by spring frosts can be easily
identified. These turn red-brown and shrivel.
Cone crops can be rated as follows:

Light crop - few scattered cones, usually
near tree top and limb ends.

Medium crop - several cones per limb, majority
in the upper 1/3 of crown.

Heavy crop - 10-20 cones per limb and extend-
ing down most of the crown.

For a crop to be rated heavy, most of the trees in
the stand should have heavy crops. Scattered trees
with heavy cone crops don't necessarily indicate a
good collection year. Unless seed is badly needed,
collections should not be scheduled in light to
medium crop years. Seed yield is generally poor in
light to medium crop years because of poor pollina-
tion and increased insect damage. Seed and cone
insects may destroy nearly all seed during light
crop years. Since it costs as much to collect
wormy cones as good ones, collections should be
scheduled when you can get the most seed for your
efforts. An elementary but important fact to
remember is that collection costs and extraction
costs are rather fixed; the amount and quality of
seed recovered is the major variable. So unless
seed inventories are critically low, only moderate
to heavy crops should be collected to avoid higher
costs and lower percent germination lots.

Seed count monitoring should begin in mid- July. By
then, seeds have developed far enough to produce
accurate counts. Cones may be picked or shot down
for inspection. Check cones for insect activity.
Mid (July) season insect attack shows as curled,
brown tipped or dead, dry cones. Insect bore holes
and frass may also be evident. Insect damage may
not be easily noticed, however, so cut tests should
be a standard part of crop evaluation procedures.
The cut test is made by slicing cones longitudinally
down the center with a sharp knife, hatchet or cone
cutter .

Tests are conducted as follows:

• Sample at least 10 cones from each of 4 to 5
trees selected at random in the stand.

• Count all sound filled seed on one face of
the cut cone. Filled seed have white centers
(endosperms) . Aborted seed are darkened or
shriveled. Look for insect activity inside
the seed .

• If more than half of the cones sampled have
insect damage, subtract one sound cut seed
from the count on each damaged cone. Insect-
damaged cones often do not fully open during
processing, preventing extraction of all
sound seed. Adjusting the seed count compen-
sates for this.

• The number of filled seeds should average at
least 5 for economic harvest. Lower counts
can be accepted depending on the need for
the particular source.

• A 6 or better count is considered adequate
for a large collection effort.

Samples should be collected on at least two-week
intervals to monitor maturity and insect damage.
Insect damage will increase as the summer progresses.
Some crops that appear good in July may be completely
destroyed by harvest time.

COLLECTION PLANNING AND PREPARATION

Planning cone collection activities involves deter-
mining the type of collection. Preparation involves
obtaining needed equipment and preparing storage
facilities. The type of cone collection — climb
and pick, fall trees and pick, or squirrel cache —
will have an effect on seed quality. Unless abso-
lutely necessary, squirrel cache collections are
avoided because it is impossible to consistently
get 90+ percent germination from squirrel-cut cones.
This is due to two factors:

1. Squirrel cache collections are characterized
by inconsistent maturity levels and large
variations in dormancy, resulting in lower
and extended germination. The squirrels do
not always wait until seeds are fully mature
before cutting cones.

2. Squirrel cache collections have the high
probability that cones and resulting seed
will be highly contaminated with various
disease-causing fungi. This is due to the
cool, moist environment in the cache. Tests
run at the Potlatch greenhouse facility have
shown up to 14 times more germinate mortality
in Douglas-fir squirrel cache seed lots than
in hand-picked seed lots.

In many of the squirrel cache lots, correlation
between germination tests and greenhouse performance
is very poor. Hand-picked cones, on the other hand,
usually have operational performance similar to that
of current germination tests.

130

Preparation is one of the most important steps
following confirmation of the collectible cone crop.
This involves:

1. Having adequate field personnel to handle cone
maturity checks and clean, bushel, and tag
cones .

2. Providing a secure area adequate for daily
cone storage.

3. Having enough clean 1.5-2 bushel burlap
collection bags. Dirty or previously used
bags may reduce quality enough to lower the
percentage of the germination testing.

4. Determining numbers, volumes, and species to
be collected in each area.

5. Giving proper notification to the required
number of cone pickers to handle the harvest.
This would include the type of harvest, prices
paid to collect, cut test specifications,
acceptable maturity, etc.

6. Obtaining adequate field storage racks and
developing cone transportation plans. Long-
term storage requires a covered area to pro-
tect cones from rain and rodents that allows
adequate air flow for drying.

CONE MATURITY MONITORING

Cone maturity monitoring is a prerequisite for
scheduling collections. Maximum seed germination
is possible only when cones are harvested when the
seed is fully ripe. Early harvests reduce germin-
ation. Immature seed are more difficult to extract,
have lower germination and highly variable dormancy
and may lose viability in storage sooner than mature
seed. If in doubt , wait . It is wiser to lose some
seed due to cone opening than to waste time and
dollars processing immature seed. Cone and seed
characteristics provide our most reliable maturity
indicators. Year-to-year fluctuations in weather
patterns affect cone ripening and make it impossible
to set calendar dates for harvest. Elevation also
affects maturity dates. Therefore, on-site cone
inspections at each collection point provide the
most reliable information for scheduling cone har-
vest. Remember, a sound regeneration system is
based on mature, high-quality seed.

Cones should be checked at least every two weeks
during July and early August. These checks will
identify insect losses. Weekly or twice weekly
checks should be made as cones mature. Cone color
is not the best indicator of seed maturity. Examine
the seed. Often it will mature before cone color
changes. Mature seed endosperm is quite firm, nut-
like and not milky or runny. Squirrel activity
should not be used as a basis for starting collection
activities. Squirrels are interested in food, not
90+ percent germination seed. A good example of the
squirrel's inability to evaluate seed maturity
occurred in 1980. Several seed companies collected
Abies grandis cones in north Idaho from squirrel
cache sources. The seed extracted from several
hundred bushels had such low germination percentages

that the seed was not marketable. This could have
easily been avoided by people trained to personally
evaluate cone maturity.

The following Douglas-fir seed and cone maturity
descriptions were developed by Charlie Brown and
have been used successfully by Potlatch foresters
since 1977.

Seed

Immature seed are white to cream-colored with clear
to white wings. As the seed matures, the seed coat
turns tan then dark tan as the wing turns brown.
Mature seed has a golden brown seed coat with com-
pletely brown wing.

Cones

Immature cones are green. As they ripen, they
acquire a yellow tinge like a ripening banana — not
bright yellow but slightly yellowish. As they lose
their yellow tinge and begin turning brown, they
puff up. Bracts turn brown first and are the first
indicator of approaching maturity.

Seed is ripe when cones puff. This is weather
related. High humidity will keep ripe cones tight.
Cones won't drop seed until they turn brown and open.

Cutting mature cones logitudinally through the center
will reveal brown lines (seed wings) running from the
seed to the scale tip. If the brown wing isn't
obvious, the cone isn't mature. The cut surface of
immature cones will turn brown (like a cut green
apple) within five minutes. Near ripe cones will too,
but not as rapidly.

CONE HANDLING AND STORAGE

The following guidelines will ensure quality seed from
the cone receiving station to the processor:

1 . Cone pickers are required to turn in harvested
cones daily. Filled sacks should be kept in the
shade during the day and not stacked together,
especially not in car trunks. Cone heating
must be prevented .

2. All cones are run over a cleaning table to
remove debris, check maturity, perform cut test
counts, measure for payment, and label. Bags
are double tagged — one inside the bag and one
tied on the closure (figure 1).

3. A maximum of one bushel of cleaned cones is
placed into 1.5-2 bushel sized loose-knit bur-
lap bags. This allows room for air circulation
as cones expand and open. The bag is placed as
quickly as possible on a portable field drying
rack. Quite often, short distance transport of
cones to a more central area is required. Since
fresh-picked cones are quite susceptible to
heating, arrangements must be made to have the
green cones shipped immediately, unloaded, and
re-racked before heating damage occurs. Green
cones will heat, just like piles of wet hay or

131

Dennuon Eastman

CONE PICKERS CERTIFICATION

J certify and honestly state:
The cones within this container
were collected or picked by me,
or through my personal super-
vision from the following location:

LOCAL NAME OF COLLECTION AREA *

ELEVATION - ■ NEAREST 100 FEET ♦

LOT NO ♦

TREE|NO

Field Verified Report

BU.

Bushel Measure

STATION

SPECIES +

SEED ZONE ♦

ELEVATION +

DATE +

CERTIFICATION ♦ CLASS

CONE COLLECTOR'S
CERTIFICATION

I certify the cones within this con-
tainer were collected by me, or
thru my personal supervision,
from the following location. This
information must be filled in and
placed in sack before cones are
loaded into vehicle at point of
collection.

Local Name of Collection Area

DATE COLLECTED *

SIGUA TURE OF PICKER (OR WORKING SUPERVISOR! 4

Signature ot Buyer 4

Certifying Agency 4

Figure 1. — Cone bag tags used successfully in Potiatch operations.

Est. Elevation

Date Collected

Signature of Collector

POTLATCH CORP.
Lew is ton, ID

grass. Green cones can be stacked together
for 1-4 hours; you can test for heating by
putting your hands between the piled sacks.
Cone heating can reduce dormancy homogeneity
that affects stratification time and produces
mold problems.

4. To promote seed lot homogeneity, the fresh
field-racked cones should be allowed to

"af ter-ripen. " This simply means the cones
should be kept for approximately a two-week
period in a dry, shady, well-ventilated, cool
area. This allows the cones to slowly begin
drying. During this period, the seeds in
each cone have time to reach almost equal
dormancy levels; if done properly, this can
apply to the entire seed lot. After proper
stratification, the result will be quick,
uniform seed germination.

5. Following after-ripening, cones may be air-
dried before shipment to the processor. If
early shipment for long distances is required,
precautions are needed to avoid cone heating.
Potiatch Corporation has found, in our north-
ern Idaho climate, that two months of air
drying brings the seed moisture levels down

to approximately 15-20 percent. At this level,
there is no trouble with cone heating if the
bags are tightly stacked for 24 hours. Each
person responsible for collections will have
to determine how best to avoid cone heating.

The final link in the process of obtaining high-
quality seed is the extracting and processing
facility. Improper cone and seed processing can
negate all the care taken in the collection process.
Only processors with a good reputation for producing
clean, high-quality seed should be used. If
possible, visit the facility and observe their
procedures. High-quality seed is clean. It does
not contain pitch nodules or other inert debris.
It also does not contain damaged seed. Blank seed
should be removed during cleaning and proper
processing will not produce cracked or damaged
seed. Properly handled and processed seed should
smell good, not sour.

TEAMWORK NEEDED

To produce high-quality seed, all steps from
inventory planning through collection and cone
storage must be done correctly. Improper methods
applied anywhere in the process can substantially
reduce both seed yield and germination. Since
vigorous, high-quality seed is necessary to produce
seedlings that meet rigid specifications, both
field personnel responsible for seed procurement
and nursery managers must work together to assure
that proper seed collection procedures are adhered
to.

132

6

0 AERIAL CONE HARVESTING IN BRITISH COLUMBIA,

D.P.^W

Wal linger

ABSTRACT: Early attempts to devise an efficient
means of harvesting cones with the use of a
helicopter were not totally successful but the
results were promising enough to encourage
continuing development. At present, two
methods — raking and clipping — meet air
transport and safety guidelines, and are in
common use.

In raking, a basket-like device is lowered over
the upper tree crown and the cone-bearing
branches are stripped from the stem as the unit
is raised by the vertical lift of the
helicopter. In clipping, the helicopter hovers
at the upper tree crown while an operator,
using a hydraulic pruner or electric chainsaw,
cuts and recovers the cone-bearing top or
branches.

Each of the methods has specific applications
and limitations regarding stand type, species,
and weather. Operating procedures and training
requirements for clipping have been identified
and documented. Productivity of aerial cone
harvesting has been shown to be competitive
with traditional collection methods and a
significant volume of cones is now collected
aerially in British Columbia.

device for collecting scion material.
However, the "Cage" suffered from cold weather
stress and was scrapped.

In 1976, two forest companies auapted the
clipping system for operational collection of
Abies amabilis fir. In one case, the
cone-bearing branches were clipped and stowed
in the poly-lined passenger compartment. On
the other project, an operator hooked a small
choker on the stem and cut off the top with a
small chainsaw. The recovered top was
suspended from the cargo hook as other trees
were harvested.

Concurrent with these developments, Jack
Walters at the University of British Columbia
Research Forest had designed an aerial cone
rake. Jack's unit was developed under a
service contract with the Canadian Forestry
Service and was operationally tested in fall
1976. True to Jack's philosophy of "think
big," the rake was enormous. It stood 7 m
(23 ft) high, was 4.5 m (15 ft) in diameter at
the base, and weighed over 800 kg (1800 lb).
While the unit's performance was not favorable
on l. cost-benefit basis, Jack's creation
provided the spark that was needed to get
things "off the grounu."

BACKGROUND

The harvesting of cones by aerial means had
probably been on quite a few minds for quite a
few years, but it wasn't until the mid- 1970' s
that somebody did something about it.

Our first knowledge of an aerial system was the
scion collecting device conceived by Pete
Theissen of Pacific Northwest Region, U.S.
Forest Service, in 1974. This system has come
to be known as aerial clipping and will be
discussed in more detail later. It is also the
standard method of scion collecting in British
Columbia today.

In 1974, Okanagan Helicopters, with Forest
Service input, fabricated a suspended, unmanned

Paper presented at the Conifer Tree Seed in the
Inland Mountain West Symposium, Missoula, MT,
August 5-6, 1985.

D.P. Wallinger is a Registered Forester,
Silviculture Branch, Ministry of Forests,
Victoria, BC.

In the next year or two, the sky was filled
with all kinds of raking hardware — kind of like
a forestry space program. Most of this metal
joined the "Cage" on the scrap heap. Two
commercial developers did design successful
rakes, but only one lias survived.

In the meantime, aerial clipping on a
production scale was becoming reasonably
successful. Equipment was more sophisticated
and procedures refined. An electric chainsaw
was introduced as an option to the hydraulic
pruner.

The "flying gondola" was conceived afl r a
monsoon bucket had been used for samp ng
during the spruce budworm outbreak ol
1975-1976. First thought of as a ineana of
scion collecting, the gondola quickly became a
vehicle for cone harvesting. The bucket was
slung beneath a Long Ranger and maneuvered into
the treetop. The saw operator, riding in the
bucket, set a small choker around the stem and
sawed off the top. The severed top hung from
the bucket while subsequent trees were
harvested. This was a highly productive system

133

and two or three units were built and used in
1979. Unfortunately, because a single-engined
helicopter was used, the method was grounded by
the Federal Ministry of Transport. Its use
would likely be permitted if two-engined
aircraft were used but the economics of the
method would be unfavorable.

So that leaves us, at present, with two
approved aerial harvesting systems — raking and
clipping.

RAKING

The term "rake" loosely refers to branch
collectors, strippers, or cutters. The
configuration is generally a circular or
hexagonal fiberglass cone at the top of which
is a large opening with wide, radiating slots,
each of which is bordered by upward-oriented
cutting edges. Around the circumference of the
base is a vertical fence of expanded aluminum
welded to a frame. The diameter of the rake is
normally about 2 m (6.6 ft). From three points
around the circumference of the upper frame,
light cables rise to a common ring to which the
helicopter's cargo hook is attached. The unit
may weigh between 130 and 150 kg (285-330 lb).

In practice, the rake is centered over the
target tree and lowered until most of the
cone-bearing branches protrude through the
upper opening of the rake. The helicopter then
lifts the rake and the cutting edges sever the
limbs which are then retained by the
surrounding mesh. Two or more lowering and
lifting sequences may be required to harvest
the majority of available cones from a tree.
Harvesting takes about 30 seconds per tree.
The pilot then proceeds to other trees and
repeats the procedure until he has a safe
maximum pay load.

At the landing, while the helicopter hovers,
the dumping mechanism is activated by the
ground crew and the collected material falls
free of the mesh enclosure. Any pieces hung up
in the rake teeth are cleared by hand. The
dumping device is then secured and the machine
leaves for another cycle. A cycle is defined
by a series of harvesting events — lift-off,
fly-out, harvest, fly-in and unload. Cycles
are usually under 15 minutes during which time
eight to twelve trees may be harvested. Pilot
fatigue is a safety consideration and two
pilots usually alternate the flying task every
hour or so. A day's collecting lasts about
7 1/2 hours or 3 1/2 hours per pilot.

The Bell 206L-1 (Long Ranger) or Hughes 500 are
adequate machines for raking. In 1979, it was
determined that, to be successful, aerial rakes
must be:

a) simple in design and easily maintained;

b) light in weight but strong;

c) suitable for several species and
capable of harvesting several trees in
one flight;

d) stable in flight and easily maneuvered;
and

e) efficiently unloaded.

The rakes in use today meet these criteria and
they are very efficient for harvesting the
species for which we have a large demand,
namely true firs, spruces, and Douglas-fir.

Rakes can be used in all types of terrain.
They can be ferried short distances between
collection areas or can be transported over
longer distances by pickup truck. New
unloading mechanisms enable harvested material
to be dumped into the box of a large truck and
hauled to a central picking facility.

With improvements in rake design and
experience, productivity has become quite
favorable. In fact, collecting true firs by
conventional methods could never match the
productivity of the rake.

The only drawback to raking is that the
selection of phenotypes is left up to the
pilot. In our situation, where considerable
emphasis is placed on collecting only from
better phenotypes, we are dependent on
carrier-employed pilots. They are usually
anxious to maximize productivity, and we all
know that the most cones are not always on the
best trees.

Another concern with raking is cone recovery.
Our best estimate is that only about 50 percent
of the cones on a tree are recovered. As long
as there are lots of trees available, this is
not a problem, but when a quota must be taken
from a stand with a limited number of
cone-bearers, cone recovery is important.

CLIPPING

The development of clipping to its present
state has been a cooperative effort between the
Ministry of Forests and the carriers who have
done the flying for us.

The system involves some modifications to the
aircraft itself. It must have:

a) low skid gear - no bear paws;

b) front and rear doors on the right side
removed;

c) live-mike capability between pilot and
clipper operator;

d) a barrier between the rear and forward
cabin areas;

e) the rear cabin lined with poly;

f) an E.L.T. (Electronic Locator
Transmit ter) ;

g) a hanger for the clipper installed by
the right hand step; and

h) a counter-balancing weight secured in
the left side of the forward cabin to
maintain center of gravity.

Following several incidents in 1982, the
Ministry of Forests set up a committee to
review the aerial clipping system and to
establish operating procedures. In cooperation

134

with the air carriers, an operations manual and
a formal training program were produced by the
Ministry and are now in place. Because clipping
demands precise flying, pilot requirements and
qualifications are stringent. Clipper
operators must undergo an intensive "hands-on"
training course. The pilot and operator must
work as a team, so their personalities must be
compatible.

Briefly, the clipping unit has three
components. The auxiliary power unit consists
of a battery pack and is independent of the
helicopter's electrical system. However, the
battery pack can be replaced by connecting the
hydraulic pump unit directly into ship's power.

The hydraulic pump consists of an electrically
driven hydraulic unit, a hydraulic reservoir,
and switching valves. The electrical circuit
can be activated by a master switch.

The third component consists of the
hydraulically operated clipper head, hydraulic
hoses, and an electrical cable which leads to
an operating switch on the handle of the
clipper. The clipper must cut through a 6-cm
(2 1/2-inch) stem in three seconds.

The electric chainsaw option is made up of two
components. A generator is mounted to the left
front floor of the aircraft and is wired to the
aircraft's electrical system and produces
110 volts to power the chainsaw. The electric
chainsaw is equipped with a short bar and a
special guard which covers the entire chain
except for a 6-cm (2 1/2-inch) opening along
the underside of the guidebar.The operator
wears natural-fiber coveralls, appropriate
footwear, suitable gloves, and a secureu and
visored helmet fitted with voice-operated
headset. In the clipping position, the
operator is seated on the step of the right
rear door with his feet on the skid. He is
secured to the aircraft by a special body
harness attached to an interior hard point and
also by the regular seat belt extended by a
lanyard.

Prior to each day's operation, there is a
briefing session and inspection of equipment.
During the day, whenever the aircraft is
refuelled or shut down, there is a routine
14-point check of the helicopter, equipment,
and safety gear. The daily routine also
includes a full debriefing at which discussion
is recorded and is often very candid, to say
the least. These sessions are valuable for
pointing out potential hazards or improving
technique.

In flight, the clipper operator is responsible
for tree selection but the pilot has the final
say on the basis of safety. When agreement on
a tree is reached, the pilot checks local winds
and moves the aircraft into position while the
operator checks the area around the tree for
any hazards and advises on tail rotor
clearance. As the machine is brought up to
the tree, the operator checks his harness and

readies the clipper. When the skid touches the
tree stem, the cut is made and top recovered.
As the pilot pulls away, the top is stowed in
the rear cabin. This is all done in a series
of smooth actions with constant communication
between pilot and operator. Ten to twelve tops
may be clipped during an 8-10 minute cycle.
The "what ifs" associated with in-flight
clipper problems, loss of communication,
injury, or helicopter malfunction are a part of
the daily briefing. Anything which occurs that
disrupts the smoothness of the operation is
considered to be an incident and is discussed
at the debriefing.

Aerial clipping is both physically and mentally
demanding, so that rest periods for both pilots
and operators are rigidly enforced and daily
flying times, using two air crews, are the same
as for raking. Normally, the Bell 206B (Jet
Ranger) is used on the clipping operation.

Clipping is suited to harvesting where the
cones are concentrated in the top of the
crown. Spruce, especially, is conducive to
clipping because practically all the cones are
located in the top 3 m ( 10 ft) of the narrow
conical crown. The method is also used
effectively on immature Douglas-tir, but when
more than one approach to the same tree is
necessary, as for true firs or white pine,
efficiency declines.

Providing the clipper operator is an agency
employee, clipping affords the best opportunity
for tree selection on the basis of phenotype,
cone ripeness, and the incidence of
insect-infested cones. The quality of cones
harvested is, therefore, as good or better than
when trees are felled or climbed. However,
there are disadvantages. The method is not too
safe in steep terrain, especially if there is a
high frequency of snags. The size of tops
which can be harvested is limited to about
2. A m (8 ft) and to a stem diameter which the
clipper will accommodate. Because of rotor
wash, holding hover while at the tree is often
difficult. The cost of fabricating and
installing the clipper unit, preparing the
helicopter, and training and outfitting
operators is considerable. However, much of
this cost is a one-time investment.

AERIAL SYSTEMS IN GENERAL

The use of helicopters provides a little
excitement into an activity which, we all know,
is far from glamorous. Consequently, we are
finding that there is a tendency for the aerial
systems to be prescribed for situations where
they are not always warranted. People seem to
think that aerial systems are the panacea which
will end all their problems. This is not so;
the criteria for success are demanding.

A collection is measured by the amount of clean
seed it yields, so right off the top, high
filled-seed counts are a must.

135

The economic success of the collection
operation itself is predicated on maximizing
the volume of cones delivered per flying hour.
Therefore the crop must be heavy and the
species suitable. Present aerial systems are
marginal for the cedars and western hemlock, and
are definitely not suited to ponderosa and
lodgepole pine or to western larch. Harvesting
is from dominant and codominant trees which
have narrow, conical crowns.

Stands should be even-aged and at the peak of
their cone-producing years, and should be
within about 4 km (2 1/2 miles) of road access
or a log landing so that fly-in and fly-out
times are short.

Aerial collections require good planning to
acquire helicopters and equipment, schedule
areas, and to coordinate operations so as to
minimize ferry time. Pre-organization is based
on a good knowledge of the stand and the
surrounding area. This involves a
reconnaissance flight to:

a) establish geographical and elevational
boundaries;

b) confirm with the pilot the species and
types of trees to be harvested;

c) select and check approaches to dump
sites and fuelling areas;

d) note hazards to the flying operation
such as snags, large birds, powerlines,
and industrial or recreational
activities; and to

e) develop a pattern for harvesting the
stand.

The delivery of helicopter fuel is one item in
particular that has to be checked; if the truck
delivers to the wrong area, you have a costly
problem.

Dump and pick sites must be separate from the
fuelling area. They must be accessible, firm,
and free of dust or debris. Helicopter
approach and departure paths must be flat and
clear of obstructions. A few simple
wind-indicators (flagging tape) are desirable.

Pickers should not be allowed to work on a
landing that helicopters are using. Therefore,
several dump sites are advisable or the picking
can await the completion of the aerial phase.
The other option mentioned earlier is to haul
the harvested material to a central facility
where the pickers can enjoy shelter, and
freedom from noise, dust, and bugs. This
requires extra monitoring — one person checking
seed set, ripeness, and insect damage at the

Table 1. — Percentage of cones harvested by aerial methods in British Columbia, 1979-83 (registered seedlots
only)

Species

1979
Total %
1

1980
Total %

1981
Total %

1982
Total %

1983
Total %

(HI)

Aerial

(HI) Aerial (HI) Aerial

(HI)

Aerial

(HI)

Aerial

True firs
Alpine
Amabi lis
Gr and

22.0
1833.6

91.6

7.8
45.8
13.5

100.0
81.5

115.9

100.0

18.0
707.8
121.3

93.4
81.5

350.0
25.0

99. 1
100.0

Douglas fir

Coast

Interior

621.0
311.6

2.9

213.7
822.6

85.6

1431.3
1354.3

46.1
13.9

276.6
42.0

Spi

Sitka
Interior

1153.2
10074.4

2.7
8.7

1.3

302.4

13.6

0.4

2.3 100.0
2762.7 48.1

207.3
4308.4

40.4
89.2

Hemlocks

Western

Mountain

228.3
83.4

4.4
14.3

1.0

21.1

319.9 59.6
52.2 38.5

11.6
5.2

43.1
80.8

Cedars
W. red
Yellow

93.5

13.6
1.0

3.3
5. 1

73.7 19.0

52.6
1.8

83.3

Pines
Lodgepole
Ponderosa
W. white

899.3
4.0
17.5

1356.8
11.8
19.6

587.9
74.0

6.8

148.9
32.5 25.2

1020.8
50.6
4.8 100.0

Larch

195.7

32.5

20.8

Total/

Mean 15341.8 20.7 3006.6 3.2 893.3 14.0 7292.3 43.5 6377.5 67.7

ll HI (Hectolitre) = 2.75 (Imp.) = 2.83 (U.S.) Bushels, approx.

136

dump site, while another monitors cone
cleanliness at the picking site.

Aerial harvesting permits collections to be
made from areas which are presently
inaccessible but likely to be logged in the
near future. This means that cones can be
processed, seed cleaned, and the stock grown
for the area by the time it is logged and
prepared. Cone crops can also be harvested
from desirable stands which have been reserved
for streamside protection or game corridors.
Damage to tree crowns and whether or not the
mutilated tops become entry points for diseases
or insects is also a consideration. Our
philosophy is that in most cases, the trees
from which cones have been harvested will
themselves be harvested in the next few years.
In addition, many trees suffer similar damage
naturally from snow, Ice or wind breakage, and
insect populations. Where clipping is used,
this consideration is at least minimized by
having only one clean cut.

As with any air operation, good weather favors
safety and efficiency. Helicopters working in
a hover position are particularly sensitive to
wind, and to rain when it affects visibility.
On cone harvesting operations, winds must be
steady and less than 32 km (20 miles) per hour;
the ceiling must be at least 150 m (500 ft)
above ground; horizontal visibility must be at
least 0.8 km (one-half mile) and there must be
no impairment due to rain on the bubble.

As mentioned earlier, safety is the controlling
factor in any aerial method. We are well aware
that if a tragedy occurs on any aerial
harvesting operation, then all operations will
likely be shut down — at least temporarily — by
the federal Ministry of Transport or the
Workers' Compensation Board — or both. We
believe that this unfortunate situation can
best be prevented by:

a) employing certified, capable, and
well-trained practitioners;

b) maintaining established and documented
proceuures backed up by strict
discipline;

c) using safe, efficient, and
well-maintained equipment; ana by

d) continually monitoring the program with
debriefings and post-collection "beef"
sessions.

EXTENT OF PROGRAM IN BRITISH COLUMBIA

Table 1 outlines the percentages of various
species harvested aerially in B.C. during the
past five crop years.

Aerial systems have had a high profile
(over 90 percent) in the collection of true
firs, and are increasing at a steady pace in
the harvest of what we call interior spruce
which is white, Engelmann, and their natural
hybrids. In the last good crop year, over
3800 hectolitres (10,500 bushels) of interior
spruce were collected aerially. The
reforestation program in B.C. is expanding and
will soon reach 200 million trees annually, so
it looks like cone harvesting from natural
stands will be around for a few years yet.

PRODUCTIVITY

I do not want to discuss dollar costs because
they change every year and are meaningless
unless the input factors and collection
circumstances are similar. Besides, some of
our costs would likely make you people south of
the 49th parallel shake your heads. Remember,
however, that in B.C. we are talking about
hectolitres, which are about 2.8 times the U.S.
bushel. We are also operating in an economy
where we're paying the equivalent of $1.86 for
the same gallon of gas that you pay $1. 15 for.
What we do know is that with aerial
collections, there is considerable economy in
scale, that is, large collections.

Table 2. — Proportional costsl of stand establishment estimated for 1983-84 (108 million seedlings)

Activity

$/ 1000
seedlings

@ 1150 Seedlings
$/hec tare

per Hectare2
tl seed ling

% of

lolal cost

Cone collection

3.00

3.45

0.3

0.5

Seed processing

0.42

0.48

0.04

0.1

Nursery

17 5.00

201.25

17.5

29.7

Site preparation

166.90

191.93

16.7

28.4

Planting

242.60

279.00

24.3

41.3

Total

587.92

676. 11

58.8

100.0

lAll species, all stock types, all methods; and including production overhead

(excluding administration)
^Hectare = 2.47 acres (approx.)

137

Predictably, the costs of delivering the top
material to the ground and the shucking and
bagging of the cones account for 80-90 percent
of the total collection cost. The relative
proportion of these two major items appears to
depend on cone size and, to some extent, on the
size of the branch pieces delivered.

Shucking of true firs is least costly, with
Douglas fir, spruce, hemlock and cedar
following in that order. In general, clipped
tops are easier to pick from because pickers
handle only one piece and more cones are
intact. Raked material is more fragmented and
many smaller pieces must be sorted through.

Our philosophy concerning collection costs is
that they are a very minor proportion of the
total cost involved in stand establishment.
Table 2 indicates that (in 1983-1984) cone
collection costs represent about one-half of
one percent of the total cost of our restocked
hectare.

Actually, the proportion is a bit lower because
the seed used has usually been collected and
processed in an earlier year when costs would
have been less. So we feel that what may seem
to be a large increase in collection cost is
really only a small increase in total cost of
the planted tree. If aerial collection
improves seed quality or yield, then we are

actually money ahead. Cones from the topmost
portion of the crown are likely to have a high
seed-set, and the propensity for "selfing" is
minimized.

Productivity is best measured by the volume of
cones delivered per flying hour, and depends on
three factors.

The first, cone volume delivered per cycle, is
a function of the:

a) abundance of the crop, (the number of
cones per tree and the frequency of
cone-bearing trees) ;

b) the size of the cones;

c) the percentage of cones recovered;

d) the design and efficiency of the
equipment;

e) the ability of the air crew; and

f) the size and maneuverability of the
helicopter.

The second, elapsed time per cycle, depends
largely on the fly-in and fly-out distance.
Short cycle times reflect a sound
reconnaissance, good pre-organization, and an
efficient team operation.

The third factor, picker productivity, is
influenced by the species (size) of the cones
and the size of the material delivered. The
pickers' diligence and attitude is important

Table 3. — Productivity of aerial harvesting (based on data acquired to 1984)

Species

Cone
crop

Number
of

Total cones
collected

Productivity
Hi/harvesting hour

Leve 1

collections

(HI)

Lowest

Highest

Average

Aerial Raking

Coast D-Fir

Med-Hvy .

3

81.2

4.3

True firs

Hvy.

13

1663.0

5.7

20.2

14.2

W. hemlock

Med-Hvy.

7

106.2

2.2

M. hemlock

Med-Hvy.

2

7.8

3.0

W.R. cedar

Med.

3

13.7

2.7

Interior spr.

Med.

3

473.2

1.3

2.7

2.6

Hvy.

2

752.4

2.0

3.6

3.4

Aerial Clipping

Interior spr.

Med.

3

172.0

2.0

2.2

2.1

Hvy.

5

993.2

3.1

5.3

4.0

Int. D-fir

Med.

3

149. 1

1. 1

2.7

2.1

*1 hectolitre = 2.75 bushels (Imp.) = 2.83 bushels (U.S.) approx.
^Clean, sacked cones from material delivered.

138

and chey must be kept happy with a suitable
wage, lots of cones to pick, and plenty of bug
dope. Picker productivity may be enhanced
where material has been hauled to a covered and
cool central picking area.

The productivity data we have managed to
collect so far provides what we think are
fairly credible averages for at least the true
firs and interior spruce (table 3). For other
species, more information must yet be collected.

Good-sized collections of interior spruce have
been made from both medium and heavy crops, and
some evidence of the effect of crop abundance
is beginning to emerge. We also found that
productivity of clipping was higher, and costs
lower, after formal training had been given to
clipper operators.

At present, we are reasonably pleased with the
progress made with aerial harvesting and we are
confident that new developments will bring
their rewards. The helicopter people have
gained considerable experience and have learned
a lot about cone picking. We feel that several
of the carriers are capable and knowledgeable
enough to undertake projects on their own.
Accordingly, we are working with them to
develop a contract document which will not only
meet their needs but also serve the interests
of their clients. We are sure that it will all
work out just fine.

RECONDITIONING STRATIFIED SEED AT PINE RIDGE FOREST NURSERY

0

Katherine A.\Yakimchuk

ABSTRACT: Lodgepole pine (Pinus contorta) and
white spruce (Picea glauca) seed is stratified
(moist, prechilling treatment) before sowing at
Pine Ridge Forest Nursery, Alberta, Canada, to
break dormancy and produce uniform germination.
If any stratified seed remains after sowing it
can be saved. It should be dried gradually to
a storable moisture content (between 4 and 8
percent) , placed in a sealed container, and
stored at 0 F (-18 C) for later use. Stratified
seed that has been reconditioned and stored has
maintained its germinability and has produced
strong, healthy seedlings.

To stratify the seed it was first soaked in
distilled water for 24 hours. It was then
drained and each seedlot was placed in a plastic
screen bag. The seed was spread in a thin layer
to a depth of 0.25 inch (6 mm) inside each bag.
The bags of seed were placed between layers of
damp vermiculite in waxed cardboard boxes.
' Damp paper towelling was placed around the seed
bags so vermiculite particles would not get

The seed was stored at
After 21 days the
seed was removed from storage and a 0.5 ounce
(15 gram) sample was taken from each seedlot
for germination and moisture testing.

mixed in with the seed .
40°F (4.5°C) for 21 days,

INTRODUCTION

Lodgepole pine (Pinus contorta) and white spruce
(Picea glauca) seed used for sowing at Pine
Ridge Forest Nursery is stratified by a moist,
prechilling treatment 3 weeks before sowing.
Standard practice is to discard any seed
remaining after sowing. The assumption is that
this seed cannot be stored and maintain high
quality. Seed has great value when stocks
are low; therefore, it would be ideal to save
this seed for future use.

Initial trials on reconditioning stratified seed

at Pine Ridge Forest Nursery on five seedlots

indicated little or no change in germination.

The seedlots have retained their viability after
o o

being stored at 0 F (-18 C) for 4 years
following stratification and reconditioning.
The purpose of this paper is to describe tests
of a procedure used for reconditioning stratified
seed. This seed has maintained its germinability
and has produced strong, healthy seedlings.

METHODS AND MATERIALS

Seed used for testing came from the seed
inventory kept at Pine Ridge Forest Nursery. To
perform this test, 10.5 ounces (300 grams) of
seed were withdrawn from each seedlot. Ten white
spruce and one lodgepole pine seedlots
originating at various locations in Alberta
(table 1) were tested. Germination and moisture
tests were performed on the seed before
stratification as a control.

Paper presented at the Conifer Tree Seed in the
Inland Mountain West Symposium, Missoula, MT,
August 5-6, 1985.

Katherine A. Yakimchuk is Seed Program Supervisor
at Pine Ridge Forest Nursery, Alberta Forest
Service, at Smoky Lake, Alberta, Canada.

The remaining seed was air dried by pouring
each seedlot out of its mesh bag onto a tray,
and spreading it into a thin layer. An
oscillating fan was placed by the seed to assist
drying. The seed was dried for ,15 to 20
minutes, until no surface moisture was visible.
This is the condition of seed used in regular
nursery sowing operations. A 0.5 ounce (15 grams)
sample was taken from each seedlot for
germination and moisture testing.

The seed remaining on the trays was left to dry
for 3 days at room temperature, until it
reached a storable moisture content of 6 to 8
percent. After this drying process, a 0.5
ounce (15 gram) sample was taken from each
seedlot for germination and moisture testing.

Seed remaining was sealed in plastic bags and
placed inside waxed cardboard boxes Jor
storage in seed freezers at 0 F (-18 C) .
Following this the seed was sampled each month
for germination and moisture testing to assess
its quality.

Testing followed International Seed Testing
Association rules (International Seed Testing
Association 1976) . Moisture content was
analyzed using two 0.18 ounce (5 grams)
replicates of seed each. The seed was oven
dried at 220 F (105 C) for 16 hours. The
percent moisture content was calculated on a
wet-weight basis. Germination tests were made
by placing the seed on moist kimpac paper
towelling inside covered, clear, polycarbonate
trays. Four replicates of 100 seeds each were
used per test. Test seeds were germinated in
Conviron germination cabinets for 21 days at
77 F (25 C) with a 12-hour photoperiod.
Germination counts were made twice per week
using the vigor class system developed by Wang
(1973). During seedling evaluations, only the
vigor class 1 seedlings were removed from the
trays. The remaining seedlings were left to
develop further.

140

Table 1. — Origins and collection years of seed samples used from the Alberta seed inventory for
reconditioning tests

^ ______________ Location -----------

Seedlot Species Collection Township Range Meridian Longitude Latitude

year ( w) ( M)

1 Sw

79

110

16 5 116° 45' 58° 35'

2 Sw

75

94

11 4 111° 50' 57° 10'

3 Sw

82

80

25 4 114° 05' 55° 55'

4 Sw

83

80

25 4 114 05' 55 55'

5 Sw

79

77

11 6 119 55' 55 45'

6 PI

82

66

10 5 115 45' 54 45'

7 Sw

79

62

20 5 117 25' 54 25'

8 Sw

82

43

o o
10 5 115 45' 52 45'

9 Sw

79

41

15 5 116 35' 52 35'

10 Sw

79

35

8 5 119 25' 52 0'

11 Sw

71

8

4 5 114 45' 49 40'

^"Sw = white spruce, PI =

lodgepole pine

Final germination percentages were calculated to

RESULTS AND DISCUSSION

a vigor class 4. At Pine Ridge Forest Nursery it

is felt that a seedling

not having reached

this

Visual assessment showed that seedlings from

size in 21 days will not

survive in the field or

the control (no treatment) for all seedlots were

greenhouse. Vigor classes used are as follows:

strong and healthy but showed an unevenncss of

1. Normal, fully developed seedlings with

growth. Heights of seedlings within each

seedcoat completely shed

replicate varied. Seedlings from stratified

2. Normal, well-developed seedlings with

seed showed uniformity in height and were

seedcoat almost shed.

generally tall, strong, and healthy. This was

3. Normal, we 11 -developed seedlings with

seen at the onset of the trial as well as after

visible cotyledons and seedcoat partly shed.

the seed had been kept in cold storage.

4. Normal seedlings

with moderately developed

hypocotyl and cotyledons

barely visible.

The initial moisture content of all but one

seedlot was between 5 and 7 percent (table 2) .

Analysis of variance was

performed on the

After stratification, moisture contents

germination test data for each treatment.

increased to 31.4-34.5 percent. When the seed

Germination percentages

calculated to vigor class

was air dried for removal of surface moisture,

4 and vigor class 2 were

analysed. Duncan

' s

moisture contents were lowered to 28.6-32.2

multiple range test was

run to test the differences

percent. After air-drying the seed for 3 days,

in germination among treatments.

moisture contents were reduced to 5.9-7.7

percent. This is the recommended moisture

Visual assessments of seedling growth were

made

content for long-term storage of white spruce

during seedling counts.

These were based

on

and lodgepole pine seed (Wang 1974) . The seed

height, strength, color.

and uniformity of

retained this moisture content during the time

growth among all seedlings.

it was kept in storage.

Table 2. — Moisture content (percent) of 11

Alberta white

spruce and lodgepole pine seedlots after four

treatments

- - - Treatment ----------------

Seedlot Control

Wet

Air Dry

Dried for storage Stored for 1 year

1 6.0

31.4

29.7

7.7 7.0

2 3.8

33.1

31. 1

6.3 6.5

3 5.6

31.9

31.5

6.6 6.6

4 6.3

32.6

29.0

5.9 6.2

5 5.9

31.3

30.1

6.9 6.8

6 6.3

34.5

32.2

- 7.7 7.6

7 5.7

31.9

30.5

7.5 7.2

8 5.2

32.5

28.6

6.1 6.4

9 6.4

34. 1

31.5

7.7 7.3

10 6.1

32.9

30.9

7.6 6.6

11 5.7

32.9

31 .9

6.1 6.3

141

Analysis of variance at a 5 percent confidence
level, based on germination results to a vigor
class 4 and 2, indicated that eight of the 11
seedlots showed significant differences among
treatments. In most of these cases, the
control germination percentages are lower than
those of the treated seeds. This indicated
that stratification is necessary for maximum
germination. Germination percentages varied
very little among the stratified, air dried,
and dried- for-storage trials. Sampling and
testing the seed monthly while it was in
storage indicated that the germination
percentages for most of the seedlots remained
very close to the original stratified
percentages. For seven seedlots, the analysis of
variance indicated there was a significant
difference in germination in the last months
of storage. Viability of the seed may have

been deteriorating, but the seedlings produced
from these seedlots were still very strong
and healthy. Five of the seedlots that showed
a reduction in germination had a lower
germination percent originally. The other
seedlots appear to be just as vigorous after
being kept in storage as they were at the start
of the trials.

By comparing germination percentages to vigor
classes 2 and 4 (table 3) , it is evident that
the seed has retained its germination capacity,
as the seed germinated quickly and reached an
acceptable size in 21 days. The majority of
the seedlings reached the vigor class 2 size
within 21 days.

The germination data show that stratifying

o o

the seed, drying it, and storing it at 0 F (-18 C)

Table 3. — Comparison of germination (percent) of 11 Alberta white spruce and lodgepole pine seedlots
after four treatments and seven storage times3

k , __________ storage time (months) -----------

Seedlot Control Wet Air Dry Dry 2 3 4 5 6 7 8 9 10 11 12

GERMINATION

TO VIGOR CLASS 4 IN 21 DAYS

1

862

91

90

86

93

90

88

90

91

89 . 5

91

90

.5

7

86

871

856

2

685

82

82

83

80

5

76.5

75

76

78.

5

76

77.5

72

3

695

723

6

66.5

3

672

78.

5

77

72

73

74

70.5

73

74.

5

75

72

70

1

662

672

682

4

674

80

77

74

77

5

74

67.54

70.51

76

656

684

67

.54

66.

54

674

711

5

88. 511

96

94

94

96

95

93

94.5

96

92

913

94

92

93

93

6

93

94.

5

96

94

93

93

91

95

93

92

92

92

.5

95

95

91

7

48. 514

72

67

65

70

5

631

622

592

64.

5

592

602

62

.51

59.

52

622

66

8

91. 58

97

95

97

96

94

94

95

96

96.5

97

96

94.

5

932

933

9

79

83

81

80

5

80

725

741

76

79

751

81.5

77

.5

751

726

751

10

75

81

76.

5

74

5

76

76

78

74

73

72

70

70

73.

5

78

71''

11

81

84

83

82

86

84

79

82

81.

5

84

78

83

82.

5

85

79

GERMINATION

TO VIGOR CLASS 2 IN 21

DAYS

1

80

90

90

85

91

5

87.5

85

88

90.

5

86

86

88.

5

83.

5

85.

75

82

2

52. 57

64.

5

4

60

68

74

5

66.5

4

58

4

59

71.

5

72.5

72

4511

51.

7

5

61.

2

0

46.75

3

4

60.5

71.

5

74

69

5

67

71

68

68.5

71

68

67

65.

25

58.

757

61.

75

4

64.75

4

596

77

74.

5

72

73.

5

70

643

67. 51

72.

5

644

661

61.

255

61.

55

62.

25

5652

5

83"

94

93.

5

93

5

94

93

91

91.5

94

4

88.5

87.5

389.

5

89

93

90.25

6

92

93

5

95.

5

93

5

92

5

92

91

94

92.

5

89.5

91.5

92

94.

75

94.

5

89.75

7

4614

64

5

64

62

68

57.51

57.51

533

61.

5

553

56.5

^4.

253

54.

53

57.

25

"""eo^s

8

90.5

95

95

95

94

90

91.5

93.5

95

94.5

94.5

92.

25

92

92

93

9

80

5

80

78

78

71.52

72

702

77

72

78

72.

5

71.

75

67.

75

671.5

10

6

64.5

77

76

72

5

75

72.5

74.5

673

69.

5

64?

63?

68.

51

68.

751

74.

5

65.25

11

6911

80

81

80

83

81

73. 51

79

80

81.5

75

78.

25

78.

75

79

73. 52

8

5

Raised numeral in table indicates number of times treatment is significantly lower than others in the
trial at a 5 percent level determined by Duncan's multiple range test.
^No treatment.

cStratified seed before air drying.

^Seed air dried for removal of surface moisture.

eSeed dried to a moisture content of 6-8 percent.

142

does not destroy seed quality. All seedlots
reacted similarly, regardless of their origin
in Alberta.

CONCLUSIONS

1. White spruce and lodgepole pine seed must
be given a moist prechilling treatment before
sowing, to attain maximum germination and
uniform growth.

2. Stratified seed that has not been sown
should be dried down gradually to a moisture
content of about 6 percent^ sealed in a
container, and stored at 0 F (-18 ) .

3. Reconditioned seed should be used within 1
year to ensure high quality. Some seedlots
may have lower germination, as determined by
periodic testing, after being reconditioned
and stored for several months. Seed should

be used as soon as possible.

4. Seedlots that initially have lower germination
potential deteriorate sooner than seedlots
having high germination potential.

5. Saving unused stratified seed eliminates
waste of costly seed, especially in areas
where the seed inventory is low.

6. Testing will continue to determine whether
reconditioned seed can be stored longer than

1 year and still maintain its viability.

REFERENCES

International Seed Testing Association.

International rules for seed testing. Seed
Science and Technology. 4:3-177: 1976.

Wang, B. S. P. Laboratory germination criteria
for red pine (Pinus resinosa) seed.
Proceedings fo the Association of Official
Seed Analysis. 63:94-101: 1973.

Wang, B. S. P. Tree seed storage. Publication
no. 1335: Ottawa: Department of the
Environment, Canadian Forest Service: 1974.
32 p.

Section 3. Seed Orchard and Seed Production Area
Management

144

REVIEW OF SEED PRODUCTION AREA AND SEED ORCHARD

MANAGEMENT IN THE INLAND MOUNTAIN WEST

Jenji^Koni

ishi

ABSTRACT: Based on survey data from various
agencies in the Inland Mountain West, which
covers that region between the east slopes of the
Cascade Mountains, the Rocky Mountains, through to
the Plains, the current scale of the planting pro-
gram is identified. The proportion of cones
collected for various species from natural
stands, seed production areas, and seed orchards
is estimated.

Current practices in cone crop induction, cone
and seed pest control, and cone harvest and
processing methods employed to procure the
necessary seed supply are reviewed.

The future trend for certain portions of the
Inland Mountain West region is to establish and
intensively manage seed orchards to produce seed
of superior genetic quality for the major
reforestation species.

INTRODUCTION

The area surveyed to address my topic includes
the three (3) western Canadian provinces and
thirteen (13) of the western and mid-western
states of the United States (fig. l). The area
west of the Cascade Mountains has been excluded
for the purposes of this review.

Within the Inland Mountain West Region the
current annual planting program is estimated to
total 201,681,000 trees. The total planting in
the Canadian portion of the Region is
127,333,000 trees, and in the 13 states
74,348,000 trees (table 1).

This Region is characterized by diversity in
climate, terrain, and economically important
conifer tree species.

Paper presented at the Symposium on Conifer Tree
Seed in the Inland Mountain West, Missoula,
MT, August 5-6, 1985.

Jenji Konishi is a Registered Professional
Forester, Silviculture Branch, Ministry of
Forests, Victoria, B.C.

■ Coniferous Forest

Figure 1. — Inland Mountain West Region

(The dashed line marks the divide of the Cascade

Mountains) .

The major species planted in the provinces are
engelmann and white spruce, lodgepole pine,
Douglas-fir and jack pine (table 2). The major
species planted in the states are Engelmann
spruce, Douglas-fir, ponderosa, lodgepole,
white, and sugar pine, grand fir, and larch.

145

PROVENANCE

Table 1. — Production of tree planting stock for
the Inland Mountain West Region

Region Area

USA (1984)

Arizona

California

Colorado

Idaho

Montana

Nevada

New Mexico

North Dakota

South Dakota

Oregon

Washington

Utah

Wyoming

Total 13 states

Canada (1985)
British Columbia
Alberta
Saskatchewan

Total 3 provinces

Production

190,000
10,000,000*

2,716,000
25,349,000
2,643,000
237,000
7,641,000
3,558,000
1,734,000
10,000,000*
10,000,000*
280,000
No data

74,348,000

87,000,000*

30,333,000

10,000,000

Total estimated planting
- Inland Mountain West Reg.

127,333,000

201,681,000

*Estimates for east of Cascades
For source of data see references.

GENETICS

SEED STORAGE
AND PREPARATION

TIMING OF CONE
COLLECTION

HIGH
QUALITY
SEEDS

/

CONE/ SEED
PROCESSING

POST-COLLECTION
HANDLING

Figure 2. — Factors contributing to hi*
quality.

seed

For the provinces, based on the above species
ratio and planting rate, a total of approxi-
mately 3^05 hectolitres of cones (9,350
bushels ) are required to annually sustain a
program of this scale. These cone volumes do
not consider needs of direct seeding programs
in Alberta and Saskatchewan. Considering the
species mix for the mid-western states it is
estimated that an equivalent volume is needed to
sustain the United States' portion of the
program in this Region.

In order to meet the above demands a systematic
approach to seed procurement was undertaken. It
included the establishment and management of
seed production areas (SPA's) and seed
orchards. The objective of these concepts was
to secure and provide high-quality seeds
(fig. 2) needed for the expanding planting
programs. In some of the species, collectible

Table 2. — Major species planted in western Canadian Provinces (millions of trees sown for in nurseries
- 1985)

Spruce

Lodge-

(White &

pole

Douglas-

Jack

Other

Engelmann)

Pine

fir

Pine

Species-'-

Totals

B.C. 2

56.5

22.0

5-1

3-4

87.0

Alberta^

26.0

4-2

0.1

30.3

Sask.4

6.3

3-7

10.0

Totals

88.8

26.2

5-2

3-7

3-4

127-3

69-7$

20. 6$

4.1%

2.9%

2.1%

100. 0%

Other species includes Western Red Cedar, Western Larch, Western White Pine and Ponderosa Pine.
Silviculture Branch, Ministry of Forests, Province of B.C., Victoria. Excludes planting west of
Cascade Mountains.

Reforestation and Reclamation Branch, Energy, Mines and Natural Resources, Province of Alberta,
Edmonton.

Forestry Division, Parks and Renewable Resources, Province of Saskatchewan, Prince Albert.

146

cone crops occur at infrequent intervals in
natural stands. For example, in British
Columbia, seven collectible interior spruce cone
crops have occurred during the past 26 years
(fig. 3)« This equates to a collectible crop
every 3*7 years but experience indicates there
can be intervals for up to 10 years between
collectible crops. Cone crop periodicity in
some species makes procurement of a sufficient
inventory of seeds difficult.

SOURCES OF SEEDS

Over the past 20 years, most provinces, states
and larger private land agencies within the
region have established SPA's for the major
planting species (Appendix I). In general,
SPA's have been established to produce seed in
the interval required to establish seed orchards
and bring them into production.

Seed orchard programs have been implemented
recently (within the last 10 years) in the
interior region of B.C. and Alberta, and are in
the planning stage in Saskatchewan. In the
western states, orchards have been established
in some areas for those species with larger
planting demands (Appendix II). Over the next
10 - 20 years as established orchards mature and

COLLECTIBLE
NOT COLLECTIBLE

additional orchards are planted, this concept
will significantly increase as a source of
genetically improved seeds for future
reforestation programs.

Based on a Canada-wide tree seed survey (for
1980-81) by P. S. Janas and B. D. Haddon over
88% of seed used for major reforestation species
originated from unimproved natural stands,
slightly over 11% came from seed collection and
seed production areas and only 0.2$ was from
seed orchards. A similar pattern exists for the
provinces within the Inland Mountain West
Region. Data of this nature were not obtained
from the Western States, but again, a similar
distribution in terms of source of seeds is
suspected .

In summarizing, despite the establishment of
SPA's and orchards over the past 20 years, the
majority of cone collections are currently being
harvested from unmanaged natural stands.

In some provinces and states, provenance tests
have been established and reviewed (10 years+
after establishment). For example, in B.C. a
review of provenance tests for interior spruce
and lodgepole pine has resulted in the revision
of seed zonation and transfer rules as well as
the identification of superior provenances to
serve as priority areas for operational cone
collection (figs. 4 and 5). It is of interest
to note that in both spruce and lodgepole pine
some of the best-performing provenances
originated from low to middle elevation in the
moist transition zone between the dry and wet
interior forests west of the Rocky Mountains.

MANAGEMENT PRACTICES - SPA'S

Stand Selection

NONE

YEAR

Figure 3« — Spruce cone crop rating 1959-1984,
central interior region of British Columbia.

Stand selection has for the most part been done
without the benefit of advance provenance
testing work. Some of the criteria used for
selection can be listed:

(a) thrifty young stands (25 - 30 years+) with
good tree form,

(b) appropriate geographic and elevational
distribution within a seed planning zone,

(c) gently sloping terrain and south facing
aspects are preferred, and

(d) stand should be accessible by all weather
road .

Roguing, Thinning, and Brush Control

Selected stands are brushed out to reduce weed
competition and thinned and/or rogued of
overstocked stems. Inferior pheno types are
removed to enable genetic improvement both
within the SPA and within a perimeter buffer
strip (usually 5 times the average tree height
in width) to serve as a pollen isolation zone.

147

Figure h. — Interior spruce provenance and progeny test sites and location of
good seed sources.

Figure 5. — Lodgepole pine provenance test sites and location of good seed sources.

148

In thinning, care needs to be exercised to
prevent windthrow and sunscalding of residual
trees .

Thinning provides for free growing trees which
produce large crowns conducive for optimizing
cone production. It also provides for access by
equipment for cultural and protection operations.

Fertilizing

To improve the nutritional status of the soil
and the trees and to induce cone production,
fertilizer applications are made usually in the
form of ammonium or calcium nitrate.
Fertilizing is done manually, by tractor powered
equipment, and in some instances, by aircraft.
In British Columbia, despite fertilizing
treatments in interior Douglas-fir and interior
spruce SPA's the response in terms of cone
induction has been inconsistent.

Protection

During brushing and thinning operations the
resulting slash is disposed to reduce the fire
hazard. It is also important that forest land
management staff are advised of the location of
SPA's so that they can be protected where
possible from fires and/or land alienation for
rights-of-ways or for other purposes.

Experiments to control cone and seed pests on
SPA's have been limited and the results
inconsistent .

Cone Harvest

Cone harvesting is usually done by climbing. In
some SPA's where the trees have grown tall the
tops have been removed at time of cone harvest.

Summary of Current SPA Performance

In some species such as ponderosa, lodgepole,
and white pine, seed production from SPA's has
been consistent. In many of the other major
species such as engelmann and white spruce,
Douglas-fir, and larch, production has been
sporadic. In B.C. only one or two cone
collections have been made from the interior
spruce and Douglas-fir SPA's established 20
years ago. Where poor seed production has been
experienced, the concept needs to be evaluated
and perhaps phased out. In some instances it is
doubtful that the costs of SPA establishment and
management can be justified in light of the low
seed production and limited genetic improvement
resulting from this approach. In the survey, an
attempt was made to gather seed yield data from
SPA's. While many returns were received there
exists a scarcity of yield data. This recent
experience in data collection was also echoed
previously (1974) by Paul Rudolph, Keith Donran,
Robert Hitt and Perry Plunnner in Chapter III,
Production of Genetically Improved Seeds, U.S.
Woody Plant Seed Manual.

MANAGEMENT PRACTICES - SEED ORCHARDS
Planning

Where large scale planting programs are to be
sustained over a long term, seed orchard
programs have been implemented and/or are under
consideration to provide both abundant
quantities of seeds as well as superior genetic
quality.

Long-term forecasting of species planting
requirements (20 - 50+ years) is fundamental
to priorizing tree breeding and seed orchard
projects for a species. To optimize genetic
gains it is essential that orchard programs be
integrated with breeding programs.

Tree breeding and associated seed orchard
programs represent a substantial long-term
investment where the returns are not realized
until harvest of the next forest crop. In view
of this point, it is important when implementing
a tree improvement program for a species that
favourable cost-benefit studies be provided to
ensure long-term and sustained support from
financial authorities (i.e., Government or
forest company).

Once the orchard project is decided upon further
decisions must be made as to the type of orchard
to establish (i.e., seedling vs. clonal).
Experience in coastal Douglas-fir indicates that
seedling orchards outproduce clonal orchards
2:1. The generation of orchard needs to be
selected (1st vs. 1.5 vs. 2nd generation
orchard). An orchard design must be prepared to
define tree spacing and clonal or family
arrangement to enable optimal panmixia. An
orchard working plan document which covers the
above topics as well as those which follow can
aid orchard staff in management operations.

Within the region surveyed orchard management
varies from extensive to intensive. The points
which follow apply to intensive management.

Site Selection

Orchard site selection is of crucial importance
in that it can determine the success or failure
of the project. Some important factors to
consider are listed:

(a) climate - warmer and drier site than that
of the parent tree's site region. Avoid
sites which are subject to late spring or
early autumn frosts.

(b) soil and topography - well drained soils on
level to very moderate sloping ground which
can be easily worked by equipment should be
selected. Avoid sites characterized by
strong prevailing winds.

149

(c) services and economy of scale - to enable
efficient management of the orchard the
following points need to be considered:

(i) availability of an abundant suitable
water supply,

(ii) good access road and close proximity to
hydro, labour, equipment and supplies,

(iii) plan for enough orchards to be located
on a site to justify capital
improvements; orchards in combination
with nursery operations also assists in
meeting economy of scale.

(d) freedom from pests - areas free of known
pests and diseases should be selected. For
example, root rots and tree rusts can cause
problems .

(e) isolation - the site should be isolated
from foreign pollen and also from
residential areas and lakes or streams
which may restrict use of pesticides within
the orchard .

Irrigation

Most orchards are equipped with either drip or
solid-set irrigation systems.

A drip system is cheaper than the overhead
system and requires much less water. The
irrigation line can be either laid on the
surface or buried.

The solid set system can be used to cool the
orchard in the spring and thereby delay the
flowering of orchard stock. This delay in
flowering can reduce pollen contamination (where
the same species occurs adjacent to the orchard)
as well as risks from frost and cone and seed
insect damage.

Drainage, Site Preparation, and Cover Crop
Establi shment

The orchard site should include installation of
water drainage systems where required. The soil
should be well prepared prior to planting to
permit good drainage and allow optimal root
development. Establishment of a cover crop can
aid in controlling erosion and reduce soil
compaction and add organic matter.

Soil and Tissue Analysis and Fertilizer
Prescriptions

Soil analysis should be completed early in the
planning stage to determine nutrient status
levels and pH. Examination should also be made
for soilborne pests.

Soon after orchard stock is planted, twig and
needle samples should be taken at a specific
time each year (e.g. early October in B.C.
orchards) and the tissue material analyzed at a
laboratory. This data should then be
interpreted to develop appropriate fertilizer
prescriptions to keep orchard stock in a healthy
and vigorous condition at all times.

Pruning

Pruning may include removal of small inter-nodal
branches in some species to improve light
penetration and aeration. Top pruning may be
required to control height where orchard
irrigation systems are installed and to minimize
cone picking height.

Cone and Seed Pest Control

A sound knowledge of and the ability to identify
pests at an early stage as well as availability
of effective pesticides and control methods are
basic to controlling loss of seeds from pests.

As cone and seed insects can be one of the major
sources of seed loss, consideration should be
given to organizing conelet sampling and
analysis for each orchard to determine insect
infestation levels as conelets develop.
Accurate assays done at the right time can
assist the orchardist in deciding whether to
spray or not to spray for control.

Sanitation practices include picking all the
cones from the trees regardless of crop size.
Leaving cones on trees and allowing them to drop
to the ground only serves as a brooding base to
increase insect populations.

Pollen Handling

Techniques have been developed to harvest,
extract, store, test, and reapply pollen on
orchards. This enables improvement of the
genetic makeup as well as yield of seeds,
especially in the early phase of seed
production. Once sufficient within-orchard
pollen is produced wind pollination is
sufficient, however, booster pollination to the
very early and late flowering trees can optimize
seed set. To be of benefit it is most important
that pollen be applied when flowers are in the
optimal receptivity phase of conelet
development, (i.e., just after bud burst).

Cone Induction

While choosing the appropriate site for orchard
location can provide abundant and frequent cone
production, experience indicates that many trees
in an orchard do not produce cones at the
frequency and quantities desired. Consistent
annual or biennial seed production of all trees
is desirable.

150

Some treatments being used operationally and/or
experimentally to induce cone production include
root pruning, girdling, fertilizing, and
application of gibberlic acids.

Cone Harvesting

Most cone harvesting in orchards is done with
the aid of ladders. In some orchards,
hydraulically operated man-lift units used in
the fruit industry are also utilized in seed
orchards. These units have a slow ground speed
but are versatile in that they can be used for
conducting orchard surveys, enable pruning, pest
control and pollination as well as for cone
harvest.

Alternatives to Conventional Orchards

Field orchards require careful site selection,
investment in land area sufficient to enable
economy of scale of operations and 10 - 15 years
are required prior to production of significant
quantities of seeds. Container-based seed
orchards in greenhouses can accelerate abundant
and early seed production, reduce the land area
required and may be a cost effective alternative
to field-based orchards.

Another alternative is hedging orchards whereby
selected superior clones are established in
hedges to serve as cuttings for asexual
production of planting stock.

Summary of Current Orchard Programs and
Performance

In the provinces, orchards are in the very early
stage of production and many orchards are under
development or in the planning stage. Early
production experience in lodgepole pine
indicates that orchards can consistently produce
seeds germinating at 95/&+ and the seeds are
larger than those from natural stands. In the
North Idaho and Montana area most of the seed
production to date is rust resistant white
pine. Many other orchards are established or
under development for the other major
reforestation species. Based oh survey returns
orchards have not yet been established in
Colorado, South Dakota, Wyoming, Arizona,
New Mexico, Utah and Nevada. In these states
the scales of the planting programs are
considered too small to develop an economic tree
improvement and orchard program. Also, for this
area, much reliance is placed on natural
regeneration. In the Pacific-northwest area
(east of Cascade Mountains) in Washington and
Oregon the main production to date has been in
blister rust resistant white pine. Early seed
production is being experienced in grand fir,
Shasta and white fir, larch, ponderosa, sugar,
and lodgepole pine, and many orchards are in the
development stage. In this area it is intended

that as seed orchards produce sufficient seeds

to meet program needs the SPA's will be phased out.

It is most important that seed orchard programs
be integrated with tree breeding programs and/or
utilize the information provided by thorough
provenance studies for a species. In this way
maximum genetic improvement of the seed produced
is realized.

While many seed orchard management practices
have been developed during the past 20 years
there exist opportunities for improvement in
the following areas:

- genetic quality improvement through early
establishment of 1.5 or 2nd generation
orchards linked to long-term breeding
plans ,

- carefully evaluate the performance of 1st
generation orchards and from this
experience make improvements in orchard
site selection, and in design of 1.5 or
2nd generation orchards so that both
genetic gain and seed productivity is
optimized ,

- a significant increase in orchard
development and establishment is needed
forthwith if orchard seed is to be the
major source of seeds for future planting
programs,

- expediting earlier seed production
through development of new techniques
such as containerized orchards,

- establishment of hedging orchards which
serve to produce rooted cuttings,

- research is required on cone induction to
ensure that all trees produce abundant
and regular crops, and

- research is required on pests and their
control to minimize loss in orchards.

ACKNOWLEDGMENTS AND CONCLUDING REMARKS

In concluding I wish to thank the organizers of
this important symposium for the opportunity to
review SPA and seed orchard management in the
Inland Mountain West. I also wish to thank
those persons who responded to my questionnaire
on SPA's and seed orchards. I am sure this
symposium and the resulting proceedings will
contribute most significantly towards extending
knowledge of seed biology, of operational cone
collection and seed handling practices as well
as management of SPA's and seed orchards. I
cannot overemphasize the importance of
knowledgeable, well-trained, enthusiastic and
dedicated silviculture staff in meeting the
objective of supplying high-quality seed for
forest renewal programs.

151

REFERENCES

Text References:

Respondants to Survey Questionnaire, May, 1985:

(i) Dhir, Narinder, K. , Head, Genetics

Section, Reforestation and Reclamation,
Energy and Natural Resources, Edmonton,
Alberta.

(ii) Thompson, John D. , Head, Forest
Nurseries, Forestry Division,
Saskatchewan Parks and Renewable
Resources, Prince Albert, Saskatchewan.

(iii) Howe, George, Regional Geneticist,

Northern Region, Region 1, USDA Forest
Serv-'.ce, Missoula, Montana.

(iv) Thiesen, Peter, Regional Geneticist,
Region 6, USDA Forest Service,
Portland, Oregon.

(v) Fins, Lauren, Executive Director,

Inland Empire Tree Improvement Council,
University of Idaho, Moscow, Idaho.

(vi) Wilson, Boyd, Geneticist, State of
Washington, Department of Natural
Resources, Olympia, Washington.

(vii) Hamilton, Ron, Regional Geneticist,
Intermountain Region, Region 4, USDA
Forest Service, Ogden, Utah.

(viii) Jeffers, Richard M. , Regional

Geneticist, Rocky Mountain Region,
Region 2, USDA Forest Service,
Lakewood, Colorado.

(ix) Bickford, Munroe, Wenatchee National

Forest, USDA Forest Service, Wenatchee,
Washington.

(x) Hughes, M. Q., Director, Timber

Management, USDA Forest Service, Region
3, Albuquerque, New Mexico.

(xi) Bordelon, Mike, Industrial Forestry
Association, 135 Nisqually Cut-off
Road, South East, Olympia, Washington,
98503.

(xii) Nicholson, Lou, Reforestation Forester,
Timber Management, Region 6, USDA
Forest Service, Portland, Oregon.

Source of Data: Table 1

- 1984 US Forest Planting Report, USDA,
Forest Service, February 1985

- Silviculture Branch, Ministry of Forests,
Province of British Columbia, Victoria

- Reforestation and Reclamation Branch,
Energy and Natural Resources, Province of
Alberta, Edmonton

- Forestry Division, Parks and Renewable
Resources, Province of Saskatachewan,
Prince Albert

Bartram, C; M. Crown. Cone Production

Highlights from Coastal Douglas-Fir Seed
Orchards - Silviculture Branch, Ministry of
Forests, Province of B.C. A poster
presented at the 1984 Western Forest
Genetics Association Annual Meeting at
Victoria, B.C.

Crown, M.; I. Karlsson; M. Meagher. Stand

Management for Conifer Seed Production in
British Columbia. British Columbia Forest
Service and Canadian Forestry Service;
1978. Unpublished Report.

. Ebell, L. F. Cone Induction Response of Douglas-
Fir to Form of Nitrogen Fertilizer and Time
of Treatment. Pacific Forest Research
Centre, Canadian Forestry Service, Victoria,
B.C.; 1971 October 15.

Fashler, A. M. K.; W. J. B. Devitt. A practical
solution to Douglas-Fir Seed Orchard Pollen
Contamination. Forestry Chronicle.
56(5): 1980.

Hanson, Pauline. Seed Orchards of British
Columbia. British Columbia Ministry of
Forests; 1985-

Janas, P. S.; B. D. Hadden. Canadian Forest
Tree Seed Statistics: 1980-81 Survey
Results. Information Report PI-X-41,
Petewawa National Forestry Institute,
Canadian Forestry Service, Department of
Environment; 1984.

Jaquish, B. ; G. Kiss; C. Ying. An Evaluation of
Interior Spruce Seed Zones and Seed Transfer
Rules based on Genecological , Provenance,
and Progeny Test Results. British Columbia
Ministry of Forests, Research Branch; 1983
March. 21 p. Unpublished Report.

Miller, G. E. Pest Management in Douglas-Fir
Seed Orchards in British Columbia. In:
Yates, H. 0. III. ed. Proceedings of the
cone and seed insects working party
conference, Working Party S2. 07-01,
International Union of Forestry Research
Organizations; 1983 July 31 - August 6,
Athens, GA. Asheville, NC: U.S. Department
of Agriculture, Forest Service, Southeastern
Forest Experiment Station; 1984; 179-185.

Ministry of Forests, Province of British

Columbia, Forest Research Review 1983-84- A
report of the highlights of the Ministry of
Forests Research Program. Victoria, B.C.;
1985-

Niemann, Tom. Analysis of Douglas-Fir Tree

Improvement and Seed Orchard Options for the
Cariboo Transition Zone. British Columbia
Ministry of Forests, Strategic Studies
Branch, Victoria; 1983«

152

Owens, J. N. The Reproductive Cycle of

Douglas-Fir. BC-P-8, Pacific Forest Research
Centre, Canadian Forestry Service, Victoria,
B.C.; 1975- 23 p.

Owens, J. N.; M. Molder. The Reproductive Cycle
of Interior Spruce. British Columbia
Ministry of Forests, Victoria, B.C.; 1984.
29 P-

Owens, J. N.; M. Molder. The Reproductive Cycle
of Lodgepole Pine. British Columbia Ministry
of Forests, Victoria, B.C.; 1984. 30 p.

Ying, C; K. Illingworth; M. Carlson. Geographic
Variation in Lodgepole Pine and its

Implications for Tree Improvement in British
Columbia; British Columbia Ministry of
Forests, Research Branch. In: Lodgepole Pine
Symposium; 1984 May 14-16; Vancouver.
University of British Columbia.

United States Department of Agriculture, Forest
Service. Pollen Management Handbook,
Agriculture Handbook No. 587- Washington
D.C., Edited by E. Carlyle Franklin; 1981
December.

United States Department of Agriculture, Forest
Service. Seeds of Woody Plants in the United
States. Agriculture Handbook No. 450,
Washington D.C.; 1974.

153

APPENDIX I

Inland Mountain West Seed Production Areas (SPA's)

Ave

. Annual

Seed Produced

Amount Seed

4- f\ Tin -4-

to Date

Production

Species^ No. SPA's Area (ha/ac)

(kgs/lbs)

Agency

CANADA - (Metric)

British Columbia (East of Cascades)

Fdi 21

-

_

MOF

Lw 1

MOF

Py 2 3949-2

-

MAP

MUr

Pw 1

-

MOF

PI 2

MOF

Se 9

-

MOF

Sx 53 3478.3

MOF

Totals: 89 7427-5

b

Alberta

No SPA's considered to be active.

Government

Saskatchewan

Sw 5 52.0

15.6

4.0

Government

Grand Total

Western

Provinces 94 7479-5

_

USA - (Imperial)

Region 1 - Northern Region - USDA (N. Idaho, W. &.

Cen. Montana)

Fdi 1

57-4

14.4

Nat. Forest

Lw [

4.6

1.2

Nat. Forest

Pw 40° 842

85.1

21.3

Nat. Forest

PI 1

11.8

2.9

Nat. Forest

Pw 1

6

0.0

0.0

Montana State

Totals: 41 848

158.9

Region 2 - Rocky Mountain Region - USDA (Colorado,

South Dakota, Wyoming)

Py 5 95

535

134

Wyoming State

PI 14

0

0

Colorado

State

Py 1 142

Nat. Forest

Py 1 40

10.4

Nat. Forest

Totals: 8 281

a See Appendix III for species abbreviations.
" Minimal collections made each year.
c Fdi and Lw will predominate.

154

APPENDIX I (Cont'd)

Inland Mountain West Seed Production Areas (SPA '3)

Ave. Annual
Seed Produced Amount Seed
to Date Production

O »-\ jr\ AO

opeci es

No. SPA's

4 ron ( Vi a 1 a t* 1 flriTQ/TViai
n I era \ lla / J ^ivgo/XUo/

Agency

USA - (Impe

rial)

Region 3 —

Southwestern Region -

Tl^lnA i A t*i »7 r\ y\ a W au Moyt o ^ 1
UOUn \ rt I X Zi U I lei | llcW lie A 1 v U /

No SPA's

- candidate areas

are

anticipated to be submitted for

review and approval

in the next

few years.

Region 4 -

Intermountain Region

- USDA (S. Idaho, Utah, Nevada)

VA -i

ral

2

18

Nat .

Forest

Py

11

100

■

Nat.

Forest

pi

1

Nat.

Forest

Totals:

14

118

Region 6 —

Pacific Northwest

Region - USDA (Washington, Oregon -

Efl ^"t of* Cfl Qffl Hpq ]

Lw

5

22

Nat.

Forest

Py

2

5

Nat.

Forest

pi

1

22

Nat.

Forest

Se

1

15

Nat.

Forest

Totals:

9

64

Grand Total

USA

72

1311

Grand Total
Inland

Mountain West

(Can. <S USA) 166

155

APPENDIX IT

Inland Mountain West Seed Orchards

Total Seed Ave. Ann.

Produced Cur. Seed

Stage of Orchards5 (No. - ha/ac) to Date Production

Species Dev. Est. Prod. Totals (kgs/lbs) (kgs/lbs) Agency

CANADA - (Metric)

British Columbia (East of Cascades)

Pli

2- 8.6

4-14.7

6-23-3

5.62

1.87

MOF

1- 5-0

1- 5.0

Co-op. b

Sx

2- 7-7

8-23-6

10-31.3

2.32

MOF

1- 7-4

2- 5-6

3-13.0

Co-op.

Totals:

6-28.7

14-43-9

20-72.6

Alberta

PI

1-14.7

1-14.7

Co-op.

Sw

2- 3-5

1- 2.5

3- 6.0

Co-op/F.S.

Totals:

3-18.2

1- 2.5

4-20.7

Saskatchewan

No Seed Orchards established to date.

Grand Total
Western

Provinces 9-46.4 15-46.4 - 24-93-3

USA - (Imperial)

Region 1 - Northern Region - USDA (N. Idaho, W & Cen. Montana)

Fdi

5-37

5- 37

0

Nat. Forest

Bg

2-

30

2- 30

0

Nat. Forest

Lw

1-10

2-

24

3- 34

1.5

Nat. Forest

Py

4-

80

4- 80

0

Nat. Forest

Pw

1- 4

1-

16

5-80

7-100

275

40

Nat. Forest

pi

2-20

2- 20

0

Nat. Forest

Py

1-

12

1- 12

few cones

B.of L.M.c

Fdi

1-12

1- 12

0

State Forest

Lw

1- 8

1- 8

0

State Forest

Py

1-

12.5

1- 12.5

0

State Forest

Blue Spruce

1- 3

1- 3

0

State Forest

d

1- 3

1- 3

0

State Forest

Totals: 13-97 11-174-5 5-80 29-351-5 276.5

a Classification Stage

Developing site clearing or preparation and/or under propagation.

Established 80$+ planted.

Producing orchards that have produced greater than 40% of seed

target in any one year.

b Co-op = in British Columbia, cooperative seed orchards managed by forest companies under government

funding.
c Bureau of Land Management,
d Russian Olive, Siberian Elm, Green Ash.

156

APPENDIX II (Cont'd)

Inland Mountain West Seed Orchards

Species

Dev.

Stage of Orchards8 (No. - ha/ac)

Est.

Prod.

Totals

Total Seed
Produced
to Date
(kgs/lbs)

Ave. Ann.
Cur. Seed
Production
(kgs/lbs)

Agency

USA - (Imperial)

Region 2 - Rocky Mountain Region - USDA (Colorado, South Dakota, Wyoming)
No established Seed Orchards in these states.

Region 3 - Southwestern Region - USDA (Arizona, New Mexico)

No known Seed Orchards in ^he private sector, or municipalites of New Mexico or Arizona.

Region 4 - Intermountain Region - USDA (S. Idaho, Utah, Nevada)
No Seed Orchards established.

Region 6 - Pacific Northwest - USDA (Washington, Oregon - East of Cascades)

Fdc

48-

465

48- 465

2898

3-4

Nat.

Forest

Bg

3-

27

3- 27

8.9

Nat.

Forest

Bsr

1-

9

1- 12

2- 21

6.8

Nat.

Forest

Bw

9-

134

2- 14

11- 148

5.5

Nat.

Forest

Lw

16-

172

16- 172

1-3

Nat.

Forest

Py

51-

927

8-110

59-1037

8.9

Nat.

Forest

Pw

8-

91

1- 10

9- 101

426

13-1

Nat.

Forest

PI

25-

351

1- 11

26- 362

2.4

Nat.

Forest

Ps

2-

34

2- 34

21.6

Nat.

Forest

Se

1-

7

1- 7

3-7

Nat.

Forest

Ci

4-

38

4- 38

4-7

Nat.

Forest

Fdi

1- 45

1- 45

0

0

Private Land

py

1- 20

1- 20

0

0

Private Land

Fdi

1- 8

1-8

0

0

State Forest

Totals:

168-

2255

16-230

184-2485

Grand Total

USA

181-

2352

27-404.5

5-80 213-2836.5

Grand Total Inland Mountain West (Canada & USA)

190 42 5 237

Classi f ication
Developing
Established
Producing

Stage

site clearing or preparation and/or under propagation.
80%* planted.

orchards that have produced greater than 4C# of seed
target in any one year.

157

APPENDIX III

List of Species Abbreviations

Fdc

Douglas-fir (coast)

Pseudotsuga menziesii (Mirb.)

Fdi

Douglas-fir (interior)

Franco var. menziesii

Lw

Western larch

Larix occidentalis Nutt.

Py

Ponderosa pine

Pinus ponderosa Laws.

Pw

Western white pine

Pinus monticola Dougl.

PI

Lodgepole pine

Pinus contorta var. latifolia

Se

Engelmann spruce

Picea engelmannii (Parry) Engelm.

Sx

Interior spruce (commonly white

x engelmann hybrids)

Picea sp.

Sw

White spruce

Picea glauca (Moench.) Voss

Bg

Grand fir

Abies grandis (Dougl.) Lindl.

Bsr

Shasta red fir

Abies magnifica var. shastatensis Lemm.

Bw

White fir

Abies concolor (Gorde & Glend.) Lindl.

Ps

Sugar pine

Pinus labertiana Dougl.

Ci

Incense cedar

Libocedrus decurrens Torr.

158

In

<3 PRODUCTION OF IMPROVED WESTERN WHITE PINE AND DOUGLAS-FIR
AT POTLATCH CORPORATION'S CHERRYLANE SEED ORCHARD
Roger L.j^Blair

ABSTRACT: Potlatch Corporation's tree improvement
program includes grafted seed orchards for western
white pine and Douglas-fir. Production of improved
seed at these facilities is discussed in light of
the attributes of the orchard site, cultural
procedures designed to promote early and consistent
seed production, and supplemental mass pollination.
Implications of orchard planting design are dis-
cussed as they affect production of white pine seed
with varying degrees of rust resistance and Douglas-
fir seed adapted to a range of elevational zones.

INTRODUCTION

Potlatch Corporation's Western Division began its
artificial regeneration program in 1979. At that
time the first crop from the Lewiston seedling
production facility was outplanted. Since then the
regeneration program has grown to the production
of 1.3 million seedlings per year sufficient to
regenerate 4,000 acres with present planting
spacings .

Prior to beginning the artificial regeneration
program, an extensive planning process indicated
that sufficient acres of Douglas-fir and rust-
resistant white pine would be planted to justify
tree improvement programs in these species. Based
on an evaluation of programs elsewhere and existing
research data, we decided to develop the grafted
seed orchard of both species on a carefully chosen
seed orchard site.

LOCATION

The site on which the seed orchard is located is
critical to its long-term capabilities of high,
consistent cone production (Werner 1975). Ex-
perience in the Southeast and Pacific Coast as
well as in New Zealand has indicated that the
following factors contribute to a good seed
orchard location:

L. Warm, dry climate. Sweet (1975) summa-
rizes regional differences in flowering
by indicating that high annual flower
production and reduced periodicity appear

Paper presented at the Conifer Tree Seed in the
Inland Mountain West Symposium, Missoula, MT,
August 5-6, 1985.

Roger L. Blair is Forestry Research Director at
Potlatch Corporation in Lewiston, Idaho.

to be associated with hot, dry summers.
Searching for sites with these character-
istics has become a major factor in seed
orchard siting in the Pacific Northwest.
In addition to the promotion of abundant
flowering, the frequency of flower-
damaging late spring frosts is greatly
reduced or even eliminated. Potlatch
Corporation's Cherrylane site is located
in the Clearwater River valley east of
Lewiston, Idaho. Annual precipitation is
about 30 cm with typically hot, dry
summers. The growing season is substan-
tially longer than that which occurs at
the elevations where white pine is native.

2. Isolation from natural stands. Although
this site is well within the latitudinal
and longitudinal boundaries of the ranges
of interior Douglas-fir and western white
pine, the site is remote by several kilo-
meters from natural occurrence of either
species. This will greatly reduce the
potential for pollen contamination from
unimproved surrounding trees, a factor
particularly important for western white
pine where blister rust resistance is
critical. In addition, cone and seed
damaging insects should be much easier

to control without an uncontrolled popu-
lation in surrounding stands (see Haverty
and Shea, this publication).

3. Fertile, well-drained soils. While the
sites upon which cone production is best
do not always support maximum vegetative
growth, good fertility is required for
successful orchard establishment and good
early vegetative growth. The Cherrylane
site is a sandy loam which had been farm-
ed for many years prior to orchard estab-
lishment. Soil pH is about 5.5, satis-
factory for conifer growth. The site is
adjacent to the Clearwater River and is
hence alluvial in nature. At a depth of
one to two meters, the soil grades to
nearly pure river sand. These soil char-
acteristics allow the soil moisture regime
to be maintained at nearly any desired
level with the existing irrigation system.
Nutrients can be added or withheld as
needed.

LAYOUT

Since the orchard is located in a deep, narrow
valley, wind currents are greatly influenced by

159

topography. Although large weather systems can
cause unpredictable wind direction, the more usual
air drainage pattern is downriver, east to west.
The orchard has been laid out to utilize these
wind patterns (see figure 1). For example, in
establishing a small section of a Douglas-fir
seedling seed orchard, a gradient was utilized
from low elevation seed sources to high elevation
from the northern boundary of the orchard to the
southern boundary. Given that most pollination
occurs among neighboring trees, one can reforest
sites with differing elevations with seed collected
from the appropriate area of the orchard.

PREVAILING WIND

DOUGLAS-FIR WHITE PINE

Figure 1. — Use of prevailing wind flow patterns in
orchard design at Potlatch Corporation's Cherrylane
Seed Orchard.

This layout procedure can be applied to other
characteristics. For example, as data on rust
resistance become available for our western white
pine selections, a gradient from high resistance
(with less emphasis on growth and form) to lower
resistance (with more emphasis on growth and form)
will be established across the seed orchard. This
will be accomplished through the roguing process
in the existing orchard and in the establishment
of new orchard sections. Once again seed can be
collected from the appropriate area in the orchard
to match the requirement of the planting site.
Sites known to be particularly high in rust hazard
can be reforested with seed of maximum resistance.
This procedure maximizes the utilization of expensive
seed orchard sites (Blair 1983).

MANAGEMENT
Irrigation

Because of the orchard location and soil type, irri-
gation is essential for orchard establishment and
maintenance. An irrigation system, utilizing over-
head impact irrigation heads, was in place for the
farming operation. This system was utilized during
the first few years of orchard establishment and
growth. This system soon became inadequate because
of the labor-intensive nature of the movable pipe
system, the interception of irrigation water by the
developing crowns and the added cost of continued
vegetation control with the broadcast irrigation.

The system has been replaced with a drip system
designed to provide up to three gallons per hour
per tree. With the orchard in full production,
this system is sized to have the capacity to
provide 18 gallons per tree per day. As the
orchard is thinned and rogued, individual trees
can be provided more water if needed.

An unforeseen problem with the drip system was the
development of slime mold in the system and hence
plugging of the emitters. Because water moves
slowly through the drip irrigation pipes, incubation
conditions for the slime mold organism were right
for rapid proliferation. This problem has been
corrected by injection of chlorine in the final
stages of an irrigation set.

Fertilization

Both broadcast and individual tree fertilization
have been conducted during the early stages of
orchard establishment and growth. These methods
were utilized during the period when overhead
irrigation was in place. Injection of liquid
fertilizers is now being accomplished through the
drip irrigation system. The present fertilization
regime is designed to maximize vegetative growth
and uses a complete fertilizer (12-8-4, NPK) .

Fertilization has long been known to improve cone
production in western white pine (Barnes 1969) .
Timing and formulations of fertilizers and their
interaction with irrigation are not well understood,
particularly for site conditions similar to those
at the Cherrylane Orchard. Cooperative research
on this subject is planned for 1986.

Supplemental Mass Pollination

Seed cone production has been early and consistent
on many of the 60 clones in the white pine seed
orchard. Seed cone counts conducted in the spring
of 1985 showed that 44 of the 60 clones were
flowering by four years after grafting. An average
of nine seed cones per tree were produced.

A major, but not unexpected, problem has been
pollen production. Only a few of the 60 clones
have produced pollen in significant amounts. To
offset this imbalance, supplemental mass pollination
experimentation was begun this year. Pollen
collected at the U.S. Forest Service white pine
arboretum in Moscow, Idaho, in the spring of 1984,
was utilized. This pollen was collected from
individuals known to be resistant to blister rust.
An assembly used to inject a flux (powdered metal)
into an air stream (a system used in gas welding)
was modified to inject pollen. Commercially avail-
able high pressure nitrogen cylinders were utilized
as a compressed air source. Air flow was directed
at a seed cone cluster at 18 pounds per square inch.
Although hampered by cool, wet weather, this device
worked well mechanically and was exceptionally
efficient in the use of pollen. Over 1400 flowers
were pollinated, most two or three times, utilizing
less than a pint of pollen.

160

Supplemental mass pollination offers the opportunity
to control pollen source as well as to improve
flower set and seeds per cone. This technology
may be extremely important for western white pine
where maximum blister rust resistance is required
on many sites for successful plantation establish-
ment .

SUMMARY

The Cherrylane Seed Orchard is believed to be
ideally located for consistent high production of
improved white pine and Douglas-fir seed. Although
orchard establishment did not begin until 1979,
preliminary indications are that early and consis-
tent flower production will result. Orchard
management is simplified by the location and soil
conditions.

Orchard layout is designed to utilize prevailing
airflow patterns to produce seed tailored for the
site requirements. Supplemental mass pollination
should speed our ability to produce seed in higher
quantities and with desired characteristics.

REFERENCES

Barnes, Burton V. Effects of thinning and ferti-
lization on the production of western white
pine seed. Research Paper INT-58. Ogden, UT:
U.S. Department of Agriculture, Forest Service,
Intermountain Forest and Range Experiment
Station; 1969. 14 p.

Blair, R. L. Influencing gene flow patterns through
seed orchard design. In: Bordelon, Mike,
preparer. Northwest Seed Orchard Manager's
Association 1983 winter meeting report.
Olympia, WA: Industrial Forestry Association;
1983.

Haverty, Michael I.; Shea, Patrick J. Protection
of blister rust-resistant white pine cones
from insect damage with permethrin and
fenvalerate. 1986. [these proceedings].

Sweet, G. B. Flowering and seed production. In:

Faulkner, R. , ed. Seed orchards. U.K. Forestry
Commission Bulletin No. 54; 1975: 75-82.

Werner, M. Location, establishment and management
of seed orchards. In: Faulkner, R. , ed. Seed
orchards. U.K. Forestry Commission Bulletin
No. 54; 1975: 49-5 7.

161

<0

POTENTIAL AND ACTUAL SEED YIELDS FROM

A SOUTHERN PINE SEED ORCHARD

David L.|jBramlett

ABSTRACT. — Monitoring the seed production on
26 sample trees in a loblolly pine (Pinus taeda
L.) seed orchard in Georgia indicated that only
18.6 percent of the potential seed crop was
recovered. Cone survival of the initial flower
crop steadily decreased during the 18-month
observation period, with an October 1978 cone
efficiency of 0.585 of the original 1977 flower
crop. Seed yields per cone were lower than
expected for loblolly orchards and averaged only
50.4 filled seeds per cone and a seed efficiency
of 0.375. Extraction efficiency and germination
efficiency averaged 0.833 and 0.868, respectively.
Seed yields from this orchard could be increased
by increasing the size of the potential seed crop
(flowers) or by increasing the overall seed
orchard efficiency. Results indicate that a more
intensive pest management program would reduce
cone and seed losses and substantially increase
seed orchard yields.

INTRODUCTION

To effectively and efficiently produce high
levels of viable seed in southern pine orchards,
managers need to have an accurate inventory of
the annual seed production potential, and need to
monitor the actual percentage of the seed crop
that is harvested. An inventory-monitoring system
(IMS) is currently being used in Federal, State,
and industrial seed orchards (Bramlett and Godbee
1982). The system utilizes sample trees to
quantify total flower production in the orchard
and sample branches to periodically track the
developing cone crop.

The use of IMS allows seed orchard managers to
effectively allocate cone harvesting labor and
equipment and to efficiently manage a pest control
program on a cost-benefit basis. Effective and
efficient management of seed orchards means that
seed yields can be increased and that more
genetically improved southern pine seedlings will
be available for planting.

Paper presented at the Conifer Tree Seed in the
Inland Mountain West Symposium, Missoula, MT ,
August 5-6, 1985.

David L. Bramlett is Plant Physiologist,
Southeastern Forest Experiment Station, USDA
Forest Service, Macon, GA.

METHODS

Twenty-six loblolly pine (Pinus taeda L.)
sample trees were selected in the Georgia Kraft
Piedmont Seed Orchard, Greensboro, GA.
Sample trees were randomly selected by row and
column designations without prior visual
inspection. The sample trees averaged 18 cm
in diameter and 8.8 m tall. Tree ages ranged from
8 to 12 years old, with an average age of 10
years. In April 1977, approximately 2 weeks after
female flower receptivity, the total flower count,
including all healthy, damaged, or dead strobili,
was tallied for each sample tree. At the time of
the total flower count , eight sample branches were
tagged and their initial flower count recorded as
both healthy and dead flowers. The sample
branches were distributed throughout the tree
crown but were not randomly selected locations.

After the initial count, additional counts of
all live flowers, conelets, or cones on the sample
branches were completed in July and October of
1977 and in April, July, and October 1978. Mature
cones were harvested from the whole tree in
October of 1978 and classified as healthy,
damaged, or dead. A cone analysis (Bramlett and
others 1978) was completed on 10 random cones
from each ramet. In the cone analysis the
following properties of individual cones were
observed: fertile scales, aborted ovules, total
seed, filled seed, and germinated seed. Values
of seed potential (SP), seed efficiency (SE),
extraction efficiency (EE), and germination
efficiency (GE) , were calculated for each cone.

RESULTS

Flower and cone production data are presented
in table 1. For the 26 sample trees, the total
flower count ranged from 80 to 1,534, with a mean
value of 444 flowers per tree. On the sample
trees a total of 3,660 flowers were tagged for the
periodic sample branch observations. The
original flower count decreased with time from
April 1977 to October 1978. Individual trees had
varying survival rates ranging from 26.7 to 82.9
percent of the original flower count, with an
overall average of 58.5 percent.

A similar observation of cone survival could
be generated by comparing the total number of
healthy cones harvested to the total number of
flowers. For the 26 study trees an average of
63.4 percent of the flowers were harvested as
mature cones. A regression (R = 0.67) of the
observations demonstrated a highly significant
(P = .001) regression equation as follows:

162

Table 1. — Total flower count per sample tree and survival of flowers on
branches from April 1977 through October 1978

Sample
t ree

Total
f lowers

Sample
flowers

Survival year 1

Survival year

2

Apr

July

Oct

April

July

Oct

1

1 ,534

217

213

213

213

210

173

125

2

514

165

163

160

153

146

135

102

3

440

140

136

128

126

118

88

53

4

1 ,256

222

207

202

201

200

193

184

5

609

153

141

139

132

131

98

85

6

602

157

148

146

142

142

125

99

7

301

149

139

129

126

126

1 1 5

103

8

203

100

88

79

67

63

63

57

9

551

199

196

192

192

187

170

133

10

581

188

170

148

138

130

120

102

1 1

319

144

119

109

106

97

95

87

12

94

90

80

73

70

65

65

55

13

325

146

133

115

107

104

72

39

14

536

245

232

215

200

180

162

140

15

190

101

99

96

82

78

76

53

16

80

67

63

62

56

49

49

46

17

115

84

80

77

75

74

69

62

18

83

64

63

62

62

62

47

42

19

674

83

81

65

62

60

55

44

20

57 2

164

162

158

149

148

135

125

2 1

143

110

98

100

54

49

46

37

22

801

183

167

135

106

100

97

92

23

130

112

97

96

89

89

89

81

24

113

86

78

69

68

68

63

63

25

267

131

128

81

73

69

59

39

26

515

160

148

128

120

114

105

95

Sum

22,548

3,660

3,429

3,177

2,969

2,859

2,564

2,143

Mean

444

141

132

122

114

110

99

82

TREECE = 0.088 + 0.933 x 0CT2CE
Where :

TREECE = total healthy cones (whole tree)
divided by total flowers (whole
tree)

0CT2CE = Total healthy cones (sample
branches) divided by total
flowers (sample branches)

A more useful relationship for the orchard
manager would be to predict the number of
healthy cones ( PREDCONES) based on the
observed total tree flower count ( TOTALFLO)
and the sample branch cone survival for the
tree (0CT2CE).

Thus: PREDCONE = TOTALFLO x 0CT2CE

Actual observations from the 26 sample trees
indicated a very strong relationship between the
total cone production per tree (TOTCONE) and the
calculated PREDCONE value. No statistically
significant differences were present between
the mean values of PREDCONE and TOTCONE for
the 26 sample trees in this study. „ A very
highly significant relationship (R = 0.97)
can be described by the regression equation:

TOTCONE = 20.88 + 0.9612 x PREDCONE

Results of the cone analysis indicated that only
an average of 50.35 filled seeds were produced
per cone from the 260 sample cones (table 2).
The seed efficiency calculated by the filled
seeds per cone by the seed potential (Bramlett
and others 1978) averaged 0.376 for the sample
cones. Extraction efficiency calculated by
dividing the number of extracted seeds by the
total number of seeds per cone averaged 0.833.
Germination efficiency was calculated as the
number of filled seed per cone germinating in a
laboratory test divided by the total number of
filled seed per cone. The average observed value
was 0.868.

With the four major components of seed orchard
monitoring — CE , SE , EE, and GE — an overall seed
orchard-to-nursery efficiency value (SO-NE) can
be calculated as a product of the four
parameters:

SO-NE = CE x SE x EE x GE

The average efficiency values for each tree
are shown in table 3.

163

Table 2. — Cone analysis of 260 loblolly pine cones from Georgia Kraft
Piedmont Seed Orchard, Greensboro, GA

Standard

Variable

Mean

deviation

Cone

length (mm

)

82.13

13.08

Cone

width (mm)

38.22

3.93

Fertile scales

(no.

)

66.00

10.47

First

yr. aborted ovules (no.)

54.06

29.42

Second yr. aborted

ovules (no.)

4.95

8.73

Extracted seed

(no.

)

64.33

38.03

Total

seed (no.

)

73.00

37.28

Filled seed (no

• )

50.35

32.82

Empty

seed

22.65

76.73

Germinated seed

(no

.)

46.92

32.89

Seed

potential

(no.

)

132.01

20.95

Seed

efficiency

(SE)

0.376

0.23

Extraction efficiency

(EE)

0.833

0.21

Germination efficiency (GE)

0.868

0.25

Table

3. — Values for

cone efficiency

(CE),

seed efficiency

(SE),

extraction

efficiency

(EE), germination efficiency

(GE), and

seed

orchard-

to-nursery

efficiency (SO-NE) for

individual loblolly

pine sample

trees

in

the Georgia Kraft

Piedmont Seed

Orchard ,

Greensboro, GA

Sample

Efficiency value

tree

CE

SE

EE

GE

SO-NE

1

0.576

0.728

0.984

0.997

0.411

2

0.618

0.684

0.972

0.973

0.400

3

0.379

0.583

0.933

0.906

0.187

4

0.829

0.425

0.985

0.988

0.343

5

0.556

0.485

0.968

0.979

0.255

6

0.631

0.490

0.901

0.858

0.239

7

0.691

0.399

0.963

0.995

0.264

8

0.570

0.599

0.957

0.958

0.313

9

0.668

0.378

0.816

0.982

0.203

10

0.543

0.338

0.763

0.912

0.127

11

0.604

0.234

0.826

0.779

0.091

12

0.611

0.100

0.478

0.915

0.027

13

0.267

0.177

0.780

0.947

0.034

14

0.571

0.173

0.807

0.949

0.076

15

0.525

0.060

0.520

0.212

0.003

16

0.687

0.042

0.519

0.496

0.007

17

0.738

0.329

0.932

0.969

0.219

18

0.656

0.412

0.639

0.950

0.164

19

0.530

0.480

0.754

0.785

0.151

20

0.762

0.715

0.938

0.934

0.478

21

0.336

0.290

0.924

0.909

0.082

22

0.503

0.328

0.878

0.905

0.131

23

0.723

0.226

0.891

0.386

0.056

24

0.733

0.537

0.916

0.989

0.357

25

0.298

0.262

0.695

0.929

0.050

26

0.594

0.291

0.909

1.000

0.157

Mean

0.585

0.376

0.833

0.869

0.186

164

DISCUSSION

A mean CE of 0.585 indicates that only 58
percent of the potential cone crop produced
healthy cones. Although the monitoring system
does not quantify specific causes of cone loss,
observations during the periodic counts are used
to identify the types of cone loss. Insects are
generally the major cause of cone mortality, but
fungal, environmental, and physiological causes
are also known to cause mortality of loblolly pine
flowers, conelets, or cones. The observed CE
value for the 1978 cone crop appears to be about
average for moderately protected loblolly pine
seed orchards. Certainly the cone survival could
be Improved by increasing the degree of
protection in the orchard and a major gain in CE
would most likely occur from a more intense pest
management program.

The average seed efficiency value of 0.376 is
considerably below expected value for well-
protected seed orchards. In the 1978 cones,
72.5 percent of the potential ovules failed to
develop a filled seed. First-year aborted ovules
were the major type of seed losses, with an
average of 54.06 per cone. These losses could be
from insufficient pollen or from insect damage.
Of the two possible causes, insect damage by the
leaffooted pine seedbug ( Leptoglossus corculus
(Say) appears to be the most likely cause of ovule
mortality (DeBarr and Ebel 1973). Once the
ovules enlarged in the second year of development,
ovule abortion continued at a lower rate (average
4.95 per ovule). After seedcoat formation and
fertilization, embryo mortality produces an empty
seed. Causes of empty seed include insect damage,
fungi, embryonic lethal alleles, and perhaps
environmental stress. The total number of empty
seed per cone was 22.65 and account for 31.0
percent of the total developed seed per cone. To
increase seed efficiency, better insect control
would appear to be the highest priority. If
insects are adequately controlled, SE values
should approach 0.65 to 0.75.

The average extraction efficiency of 0.833 for
the sample cones is lower than the 0.90+
considered normal for the species. Unfortunately
laboratory extraction may not accurately simulate
operational orchard extraction. Thus the observed
EE value for this orchard should be verified by
extractions from the operational orchard
extractory. Low EE values could be caused by
early harvesting of cones and improper opening and
extraction. For southern pines, seed
extractories with recent innovations normally
recover a high percentage of seeds from harvested
cones .

Germination efficiency averaged 0.868 for the
sample cones. This value is somewhat lower than
normal for loblolly pine. No apparent cause of
the lower GE value was evident. Normal values are
0.90 to 0.95 for loblolly orchard seed.

The overall seed orchard-to-nursery efficiency
of 0.186 indicates that only 18.6 percent of the
potential seed crop was recovered as viable seed
for the nursery. Conversely, 81.4 percent of the
potential seed crop was lost to numerous
destructive causes. Based on the monitoring
values, seed production of the orchard could be
increased by reducing losses and subsequently
raising the SO-NE value. The most obvious change
in management strategy would be to increase the
intensity of insect protection. Several major
insect pests can be controlled with an integrated
pest management program (DeBarr 1981). Further
monitoring would then be required to compare the
effectiveness of the pest management program in
terms of a higher SO-NE value. The other
approach to increase the orchard seed yields
would be to increase the total female flower
production. Cultural practices including
fertilization, subsoiling, and thinning could be
used to stimulate increased flower production.
Even planting more acres of orchard is an
alternative. The most cost-effective procedure,
however, would be to evaluate the current
management program and to continue monitoring
annual seed and cone crops to measure the
effectiveness and efficiency of the seed orchard
management program.

ACKNOWLEDGMENTS

The author expresses appreciation to the Georgia
Kraft Company and to Bill Arnold and his staff at
the Briar Patch Seed Orchard, Greensboro, Georgia

REFERENCES

Bramlett, D.L.; Belcher, E.W. , Jr.; DeBarr, G.L.;
Hertel, G.D.; Karrfalt, R.P.; Lantz , C.W.;
Miller, T. ; Ware, K.D.; Yates, H.O. , III.
Cone analysis of southern pines: a guidebook.
Gen. Tech. Rep. SE-13. Asheville, NC: U.S.
Department of Agriculture, Forest Service,
Southeastern Forest Experiment Station; 1978.
28 pp.

Bramlett, D.L.; Godbee, J.F., Jr. Inventory-
monitoring system for southern pine seed
orchard. Ga. For. Res. Pap. 28. Macon, GA:
Georgia Forestry Commission; 1982. 17 pp.

DeBarr, G.L. Prospects for integrated pest
management in southern pine seed orchards.
In: Proceedings, 16th southern forest tree
improvement conference; 1981 May 27-28;
Blacksburg, VA; Southern Forest Tree Improve-
ment Committee Sponsored Publication 38;
1981: 343-354.

DeBarr, G.L.; Ebel, B. H. How seedbugs reduce
the quantity and quality of pine seed yields.
In: Proceedings, 12th southern forest tree
improvement conference; 197 3 June 12-13; Baton
Rouge, LA; Southern Forest Tree Improvement
Committee Sponsored Publication 34; 1973:97-103.

165

EFFECT OF NITROGEN FERTILIZER AND

GIRDLING ON CONE AND SEED PRODUCTION OF WESTERN LARCH

fr

Russell T. ^Graham

ABSTRACT: Western larch (Larix occidentalis
Nutt.) is a poor cone and seed producer
throughout northern Idaho. Because the species
is important commercially, techniques for
increasing seed production for both artificial
and natural regeneration are needed. A study
was conducted with western larch trees to
evaluate the effectiveness of fertilizing,
girdling, and a combination of fertilizing and
girdling, as a means of increasing seed
production. Girdling dominant and codominant
70-year-old western larch at the base of the
live crown was very effective in stimulating
cone production; the application of ammonium
nitrate fertilizer in combination with girdling
did not increase seed yield. Fertilizer alone
did not increase cone production; rather, it
appeared to decrease the number of seeds per
cone. Cones from trees fertilized, girdled, and
both fertilized and girdled had heavier seeds
than cones produced by untreated trees.

INTRODUCTION

Western larch (Larix occidentalis Nutt.) is an
infrequent and sporadic cone producer throughout
northern Idaho. Since 1970, inventories of
western larch seed for northern Idaho forests
have been inadequate to meet needs of artificial
regeneration programs. Because western larch is
an important commercial tree species and the
demand for seed exceeds the supply, a method of
increasing seed production is needed.

Thinning forest stands has been shown to
increase cone production in conifers. Wenger
(1954) and Bilan (1960) showed that thinning
increased cone production in loblolly pine
(Pinus taeda L.) , and Barnes (1969) found up to
a fourfold increase in female strobilus
production caused by thinning western white pine
(Pinus monticola Dougl.). Barnes (1969) also
found tree spacings of 9-m (29.5 ft) to be more
effective than 6-m (19.7-ft) spacings for
stimulating production of female strobili.

Paper presented at the Conifer Tree Seed in the
Inland Mountain West Symposium, Missoula, MT,
August 5-6, 1985.

Russell T. Graham is a Research Forester,
Intermountain Research Station, Forest Service,
U.S. Department of Agriculture, Moscow, ID.

Mineral nutrition may also directly influence
cone production of conifers. Nitrogen appears
to be the most important element in cone
production. Coast Douglas-fir (Pseudotsuga
menziesii [Mirb.] Franco.) (Ebell 1962, 1971),
western white pine (Barnes 1969) , and Rocky
Mountain ponderosa pine (Pinus ponderosa var.
scopulorum Engelm. ) (Heidmann 1984) increased
cone production in response to applications of
nitrogen fertilizer. In addition, the type and
timing of fertilizer application can be very
important for cone initiation. Ebell (1972)
found calcium nitrate fertilizer superior to
urea and ammonium sulfate fertilizers in
stimulating Douglas-fir cone production. He
also found that application of fertilizer after
5 percent of the vegetative buds had burst in
May stimulated cone production better than early
spring or late spring applications.
Fertilization rates of 225 kg per ha (201 lb per
acre) of nitrogen to 1 800 kg per ha (1,607 lb
per acre) of nitrogen have been found effective
in increasing cone production of conifers.

Girdling is an effective method of increasing
cone yields in some conifers. Bilan (1960)
showed an increase in cone production
attributable to girdling loblolly pine. Also,
Ebell (1971) found a 300 percent to 400 percent
increase in the number of cones on stems of
Douglas-fir girdled at breast height compared to
untreated stems. Hoist (1959) reported a
twentyfold increase in the number of cones
produced on red pine (Pinus resinosa Ait.)
because of girdling. He also found girdling
trees at the base of the live crown was four
times as effective as girdling at breast height
for stimulating cone initiation. Stephens
(1964) girdled individual branches of eastern
white pine (Pinus strobus L.) and found
300 percent more female cones on the girdled
branches than on the untreated branches.

Most girdles to stimulate cone production on
conifers have been overlapping cuts up to one
internode apart severing the phloem on two sides
of the trunk or branch. They range from single
saw cuts to complete removal of the bark.
Girdling at breast height by removing a 2.5-cm
(1-inch) strip of bark slightly more than half
the stem circumference with an overlapping strip
on the opposite side, one internode above the
first, was used successfully by Ebell (1971) to
increase the number of cones on Douglas-fir.

The purpose of this study was to examine the
effects of girdling and fertilization,

166

individually and combined, as a means of
increasing seed production of western larch.
Reported here are the results of such a
combination of treatments on cone and seed
production in a young western larch stand in
northern Idaho.

Table 1. — Analysis of variance table for a study
of effects of girdling and
fertilization on cone and seed
production of western larch in
northern Idaho

METHODS

In 1980, we selected for this study a
70-year-old stand of western larch located on
the Wallace Ranger District, Idaho Panhandle
National Forests, 8 km (5 mi) east of Mullan,
ID, near the Idaho-Montana border. The stand
was thinned to a residual basal area of 9.2 to
13.8 m2 per ha (40 to 60 ft2 per acre).
Logging slash was machine piled and burned
in the spring and early fall of 1981.

Because snow cover prevented applying fertilizer
in the spring before bud burst, the study was
established in the fall of 1981. The treatments
could then influence bud formation during the
1982 growing season. Seventy-two dominant and
codominant trees were chosen in the stand and
divided into six-tree groups. Only the best
formed and more vigorous trees were chosen for
inclusion in the study. Three treatments were
randomly applied to six-tree groups in three
replicates. The treatments included:

1. Fertilize: 336 kg per ha (300 lb per
acre) of N in the form of ammonium nitrate was
broadcast with a hand spreader under the crown
of each tree, all within the drip line. This
resulted in fertilizing a 4.5-m (14.8-ft) radius
around the base of each tree.

2. Girdle: The trees chosen for girdling
were climbed (using climbing spurs) to the base
of the live crown. Here, a hand pruning saw
was used to sever the phloem for one-half the
circumference. A similar cut was made on the
opposite side at least the length of the circum-
ference down the tree. This resulted in
overlapping cuts.

3. Girdle and fertilize: Trees were girdled
and fertilized using the same procedures as for
treatments 1 and 2 .

Eighteen trees were maintained as controls, with
no treatment. The treatment strategy resulted
in a randomized complete block experiment with
subsampling (table 1) .

In the fall of 1983 each tree producing cones
was climbed, and the cones were collected.
Seeds were extracted from the cones and cleaned
at the Coeur d'Alene Nursery at Coeur d'Alene,
ID. Cones per tree, sound seeds per cone, and
mean seed weight were recorded. In addition, a
germination trial was conducted on the seed with
0, 14, and 28 days of stratification. An
analysis of variance for a randomized complete
block design was used to analyze the data
(table 1). Duncan's multiple range test was
used to detect differences among the treatment
means .

Analysis of variance

Source

Degrees of
freedom

Expected
mean squares

3 Treatments
and control

3 Blocks

Experimental
error

Sampling
error

Total

-2 + 6o2 + 18Yt2 ./3
e L l

o2 + 6o2 + 24V82./2
e *■ i

o2 + 6o2

60
71

RESULTS

Girdling was highly successful in stimulating
cone production on western larch. Trees that
were only girdled had a mean of 832 cones per
tree, the untreated trees had a mean of 77 cones
per tree, the fertilized trees had a mean of 87
cones per tree, and the girdled and fertilized
trees had a mean of 16 cones per tree (fig. 1) .
There were no significant (P<0.05) differences
in mean cones per tree among the fertilized,
control, and the combination treatments, but the
mean for the girdled treatment was significantly
higher than the other means. Saw cuts on the
girdled trees healed over within 1 year.

The only treatment that influenced the number of
seeds per cone was fertilization. This
treatment appeared to decrease the number of
seeds per cone with a mean of only 0.63
(fig. 2); the control and other treatments all
produced from 7 to 8 seeds per cone. The mean
number of seeds per cone for the fertilized
treatment was significantly lower than the means
for the other treatments.

The treatments applied to stimulate cone
production also affected seed weight. Seeds
produced on the control trees were significantly
lighter (3.11 mg) than seeds produced on treated
trees (fig. 3) . The heaviest seeds were
produced on trees fertilized only, with a mean
seed weight of 4.59 mg.

Other tree and site variables were tested to see
if they had a significant relationship to cone
production. Tree height, diameter at breast
height (d.b.h.), crown ratio, crown density, and
surrounding basal area were all tested to see if
they were significantly related to cone
production in addition to the treatments. The
only variable close to being significant was
crown ratio at the 0.163 level (table 2).

167

Tree height, crown density, d.b.h., and residual
basal area were all nonsignificant in explaining
the variation of cone production in western
larch. In addition, there were few significant
differences found among the treatments for the
tree and stand characteristics (table 2) . There
were no differences in germination among the
treatments after stratification of 0, 14, or
28 days. Germination means were 12, 95, and 96
percent, respectively, for the different
stratification periods.

10r

tn
o

X

IU
UJ

as

O
o

832

87

Figure 1. — Mean cones per tree produced by
western larch after girdling, fertilization, and
both treatments and by controls. The mean for
cones after girdling was significantly higher
than control and other treatment means.
Differences among treatment means were not
significant (P<0.05) .

Figure 2. — Mean numbers of seeds per cone from
western larch after girdling, fertilization, and
both treatments and from controls.
Fertilization apparently decreased the seed mean
significantly. Other differences among means
were not significant (P<0.05).

6r

4.43

Figure 3. — Mean weights of seed produced
western larch after girdling, fertilizat
both treatments and by controls. The
difference between the control mean and other
means was significant (P<0.05) .

by

ion, and

Table 2. — Mean

tree and stand characteristics for western

larch in northern

Idaho by treatment

and their

significance

in explaining cone production

Crown

Residual

Crown

Treatment

d.b.h

Height

ratio

basal area

density

cm

m

percent

mz/ha

Control

134.0a

28.9ab

31.7ab

8.66a

Good

Fertilize

35.3a

29.6a

35.0a

8.43a

Good

Girdle

33.5a

27.5bc

31.7ab

'7.90a

Good

Fertilize

32.8a

27.0c

29.4b

6.89a

Good

and girdle

Significance

.752

.367

.163

.752

.583

^Dif f erent

letters

indicate

significant (P<).05) differences among treatments.

168

DISCUSSION

Girdling of western larch to increase cone
production appears to be an effective method for
use in seed production areas and for natural
regeneration. Trees that were girdled produced
more cones than trees that were fertilized only
or both fertilized and girdled. Number of seeds
per cone and mean seed weight appeared
unaffected by girdling, and girdle cuts were
healed within 1 year.

The application of 336 kg of N per ha (300 lb
per acre) in the form of ammonium nitrate seemed
to offset the stimulus to cone production caused
by tree girdling. The addition of nitrogen may
have increased tree vigor to the point that new
growth consumed the increased nutrients but
diluted any changes in hormones that may have
occurred. The addition of nitrogen appeared to
influence seed weight, but not to a greater
extent than girdling alone.

The sawcuts apparently disrupted the
translocation of organic solutes in the phloem
enough to stimulate cone production. When the
downward movement of organic solutes is blocked,
they tend to diffuse into the xylem and are
translocated back to the leaves and shoots of
the crown (Kramer and Kozlowski 1979) . These
solutes containing carbohydrates, auxins,
gibberlins, and other compounds can concentrate
in leaves for fruit and seed production.

The accumulation of carbohydrates in the crown
may not be directly associated with seed and
cone formation. Ebell (1971) did not find an
increase in the level of carbohydrates in the
crowns of Douglas-fir that had been successfully
girdled to increase cone production. He
concluded that there was a doubtful relationship
between carbohydrates and reproductive bud
survival .

Besides disrupting translocation in the phloem,
other physical conditions within a tree may be
altered by girdling. These include increased
moisture stress in tree crowns and changes in
other physiological and nutritional processes
that may increase cone and seed yields. In
addition, the wound caused by a girdle may
produce wound response compounds that could be
redirected toward the crown to influence cone
production .

Girdling trees at the base of the live crown
appears very effective in stimulating cone
production. The redirected movement of
carbohydrates and hormones caused by girdling
may be less diluted with girdles near the crown
than with girdles at or near breast height. If
there is a beneficial chemical response to the
girdling wounds, this would also be near the
cone-producing area of the tree.

Overlapping cuts through the bark at the base of
the live crown created by a handsaw increased
the number of cones produced on open grown
western larch trees. Sawcuts healed after
1 year, making it possible to re-treat the trees

relatively quickly with minimal permanent
damage. However, climbing trees is
time-consuming, and therefore expensive, and
spurs often cause extensive damage. Therefore,
with such good success by girdling at breast
height in other species and the excellent
results of this study, girdling at breast height
should also be considered. Girdling trees at
breast height may require more extensive bark
removal to sufficiently disrupt translocation in
the phloem. Heavier applications of fertilizer,
in the range of 1,000 to 2,500 kg per ha (893 to
2,232 lb per acre) of N, might also be effective
in stimulating cone production and should be
investigated.

ACKNOWLEDGMENTS

Several people have been instrumental in the
successful completion of this study including
the District Ranger of the Wallace Ranger
District and members of his staff, especially
Henry Nipp and Ron Gribble. Coeur d'Alene
Nursery personnel also deserve thanks for
processing the cones and seeds. Special thanks
to Jonalea Tonn of the Intermountain Research
Station for her invaluable assistance.

REFERENCES

Barnes, Burton V. Effects of thinning and

fertilizing on production of western white
pine seed. Res. Pap. INT-58. Ogden, UT: U.S.
Department of Agriculture, Forest Service,
Intermountain Forest and Range Experiment
Station; 1969. 14 p.

Bilan, M. V. Stimulation of cone and seed

production in pole-sized loblolly pine. For.
Sci. 6: 207-220; 1960.

Ebell, L. F. Growth and cone production
responses of Douglas-fir to chemical
fertilization. Branch Report B.C. 62-8.
Vancouver, B.C.: Canadian Department of
Forestry Research; 1962. 22 p.

Ebell, L. F. Girdling: its effect on

carbohydrate status and on reproductive bud
and cone development of Douglas-fir. Can. J.
Bot. 49: 453-466; 1971.

Ebell, L. F. Cone-induction response of

Douglas-fir to form of nitrogen fertilizer
and time of treatment. Can. J. For. Res.
2: 317-326; 1972.

Heidmann, L. J. Fertilization increases cone
production in a 55-year-old pine stand in
central Arizona. For. Sci. 30: 1079-1083;
1984.

Hoist, M. J. Experiments with flower promotion
in Picea glauca (Moench) Voss. and Pinus
resinosa Ait. In: Recent advances in botany,
vol. 2. Toronto, Canada: Univ. Toronto Press;
1959: 1654-1658

169

Kramer, P. J.; Kozlowski, T. T. Physiology of
woody plants. New York: Academic Press; 1979.
435 p.

Stephens, G. R. , Jr. Stimulation of flowering in
eastern white pine. For. Sci. 10: 28-34;
1964.

Wenger, K. F. The stimulation of loblolly pine
trees by pre-harvest release. J. For.
52: 115-118; 1954.

FLOWERING OF PINACEAE FAMILY CONIFERS WITH GIBBERELLIN A^/7 MIXTURE:

HOW TO ACCOMPLISH IT, MECHANISMS AND INTEGRATION WITH EARLY PROGENY TESTING

Richard P.lPharis and Stephen D.^Ross

//

ABSTRACT: Factors influencing the successful use
of Gh^j-j to promote early and enhanced flowering
in Pinaceae family conifers are reviewed, includ-
ing the mode and timing of application, together
with interactions with other growth regulators
and cultural practices. Prospects for evaluating
the inherent growth potential of progeny at a
very young age under phytotron conditions and
integration with indoor-potted breeding orchards
to accelerate the tree-breeding process are
discussed .

INTRODUCTION

It has been known since the late 1950s that
juvenility could be terminated (albeit tempor-
arily) and precocious flowering induced at will
in seedling members of many Cupressaceae and
Taxodiaceae family conifers through the exogenous
application of glbberellins (GAs) (see ref. in
Pharis and Kuo 1977). The potential usefulness
of this treatment for accelerating the excruci-
atingly slow processes of breeding and production
of genetically improved tree seeds was, of
course, immediately recognized. It appeared for
many years, however, that this effectiveness of
GAs for promoting earlier and enhanced flowering
in the Cupressaceae and Taxodiaceae did not
extend to Pinaceae family conifers, which
includes most of the commercially important
species for which tree improvement programs were
under way.

Then, in 1973, we discovered with Pseudotsuga
menziesii (Mirb.) Franco (Pharis 1976; Ross and
Pharis 1973) and later with other species
(references cited in: Pharis and King 1985;
Pharis and Kuo 1977; Pharis and Ross 1984; Ross
and others 1983; table 1); that the GA in
common use, GA3, was the wrong one for using with
the Pinaceae. Whereas GA3 was especially effec-
tive with conifers of the Cupressaceae and
Taxodiaceae, Pinaceae family conifers were found
to exhibit a specificity for certain GAs less
polar than GA3, most notably a mixture of GA4

Paper presented at the Conifer Tree Seed in the
Inland Mountain West Symposium, Missoula, MT,
August 5-6, 1985.

Richard P. Pharis is Professor of Botany,
Department of Biology, University of Calgary,
Calgary, AB; Stephen D. Ross is Tree Improvement
Physiologist, British Columbia Ministry of
Forests, Victoria, BC.

and GA7 (GA4/t) (Pharis and Ross 1984; Ross
and others 1983; table 1). In the past
decade over 80 research reports have noted a
positive flowering response to Gk^p for at least
18 Pinaceae species representing 5 of the 6
genera of this important family (table 1). The
genus Abies is not included in this list but it
has only been represented by a single study with
A. homolepsis which gave equivocal results
TKatsuta 1981).

The successful promotion of flowering is not,
however, just simply a matter of spraying trees
with GA4/7. For reasons not yet understood
(Pharis and Ross 1986) conifers of the Pinaceae
are not nearly as responsive to applied GAs
(GA4/7) as are those of the Cupressaceae and
Taxodiaceae (to GA3) . In this paper we will
consider those factors and conditions which if
met should ensure the successful promotion of
early and enhanced flowering in Pinaceae family
conifers using Gk^p.

FACTORS INFLUENCING Gk^p EFFICACY
Treatment Timing

As with other stimulation treatments, CA4/7 is
only effective if its application brackets the
period of cone-bud differentiation for the
species in question. Owens gives elsewhere in
this proceedings the times when seed- and
pollen-cone buds first become anatomically
distinct from vegetative buds for a number of
North American conifers. He notes, however, that
these times are the latest that treatments could
be expected to promote flowering. For maximum
effectiveness the treatment should be applied 2
to 3 weeks earlier to influence the biochemical
processes leading to anatomical differentiation
of bud types.

A few words of caution regarding treatment timing
are in order. Many workers make the mistake of
relating treatments to calendar date rather than
to stage of bud development. Depending on the
year, genotype, site and associated environmen-
tal/cultural conditions, etc., bud phenology may
vary by 2, A or more weeks in any given trial.
Consequently, where GA 4/7 applications are timed
to calendar date the treatment may be either too
early or too late to influence cone bud differen-
tiation (Ross 1985). Fortunately, it is not

171

Table 1. — A checklist for species of Pinaceae family conifers that will flower in response to application
of a mixture of the plant hormones, gibberellin Awy, with appropriate reference citations

Species

References

Larix leptolepis
L. decidua
Picea abies

P. engelmannii

P . glauca

P. mariana

P. sitchensis

Pinus banks iana

P. caribaea

P . contorta

P. densif lora

P. elliottii

P. palustris
P. radiata

P. sylvestris

P. taeda

P. thunbergii
Pseudotsuga menziesii

Tsuga heterophylla
Useful Reviews

(see ref. cited in Pharis and Ross 1984; Bonne t-Masimbert 1982)
(see ref. cited in Pharis and Ross 1984; Bonnet-Masimbert 1982)
(see ref. cited in Pharis and Kuo 1977; Pharis and Ross 1984; see also Chalupka
and Giertych 1977; Dunberg 1979; Dunberg and Oden 1983b)
(Ross 1985)

(Cecich 1985 ; Marquard and Hanover 1984a; Marquard and Hanover 1984b;
Pharis and Kuo 1977; Pharis and others 1985)
(Hall 1984)

(see ref. cited in Pharis and Ross 1984; see also Philipson 1981; Philipson
1983; Philipson 1985; Tompsett 1977)
(see ref. cited in Pharis and Ross 1984; see also Cecich 1982; Cecich 1983)
(Harrison 1985)

(see ref. cited in Pharis and Kuo 1977;

Pharis and Ross 1984)

(see ref. cited in Pharis and Kuo 1977; Pharis and Ross 1984; Katsuta 1981)
(see ref. cited in Pharis and Kuo 1977; Pharis and Ross 1984 and see also
Hare 1984)

(see ref. cited in Pharis and Ross 1984 and see also Hare 1984)
(see ref. cited in Pharis and Kuo 1977; Pharis and Ross 1984 and see also Ross
and others 1984)

(see ref. cited in Pharis and Ross 1984 and see also Chalupka 1978;
Chalupka 1984)

(see ref. cited in Pharis and Ross 1984 and also Hare 1984)

(see ref. cited in Pharis and Kuo 1977; Pharis and Ross 1984)

(see ref. cited in Pharis and Kuo 1977; Pharis and Ross 1984; Pharis and Ross

1976; Pharis and others 1980; see also Owens and others 1985; Pharis 1976;

Pharis 1978; Ross 1976; Ross 1977; Ross 1978; Ross 1979a; Ross 1979b; Ross 1979c;

Ross 1983; Ross and Pharis 1973; Ross and others 1983; Ross and others 1985;

Webber 1984; Webber and others 1985)

(see ref. cited in Pharis and Ross 1984; see also Pollard and Portlock 1984)
(Dunberg and Oden 1983; Pharis 1978; Pharis and King 1985; Pharis and Kuo 1977;
Pharis and others 1980; Pharis and Ross 1984; Pharis and Ross 1986; Ross and
Pharis 1982; Ross and Pharis 1986; Ross and others 1983; Zeevaart 1983;
Zimmerman and others 1985)

necessary to go through the laborious process of
dissecting axillary buds and potentially repro-
ductive terminal buds (in some species) to deter-
mine when they are at the proper developmental
stage for treatment. Reproductive bud develop-
ment is closely related to the seasonal pattern
of shoot elongation ( Owens, this proceedings).

For Pseudotsuga (Ross 1983) and Tsuga (Ross and
others 1980) GA4/7 treatment would begin in
spring at about the time of vegetative bud burst,
whereas for Picea (Marquard and Hanover 1984b)
it should be delayed until the new shoot is about
75-85 percent elongated. For these and most
other Pinaceae family conifers the difference in
timing of seed- and pollen-cone buds is generally
too small to be able to use treatment timing to
influence conebud sexuality (Owens, this
proceedings). However, this is not the case for
some species of Pinus, where it is possible to
preferentially promote either male or female
flowering through, respectively, early and late
applications of GA4/7 (Chalupka 1984). Another
point to consider is that all buds on a tree

do not develop at the same rate; in larger trees
there will be much greater variation in bud
development associated with crown position.
Thus, whereas 2 weeks or less of GA 4/7 applica-
tion may suffice to sexually differentiate a
given bud (Owens, this proceedings), a longer
treatment time is required to bracket the
differentiation for different buds throughout the
entire tree crown. The duration of treatment
will also depend on the population of trees being
treated and the magnitude of differences in their
bud phenology. For most seed orchard populations
it appears that 4 to 6 weeks of GA4/7 application
is about optimal for P. menziesii .

Mode of Application

Gibberellin A4/7 costs about $13 to $20 (Can.)
per gram, depending on source, so it is important
that the mode of application be both effective
and conserving of the hormone, as well as practi-
cal for operational use. The best mode of appli-
cation will depend on the species, the size of
the trees being treated and whether the objective

172

is to promote early and enhanced flowering for
breeding purposes or for volume seed production.
Thus, topical applications of GA^/yby micro
pipette may be highly cost effective for the
former but not the latter.

Foliar sprays are perhaps the most convenient
method for applying GA4/7 to large seed orchard
trees. And, aqueous sprays containing an approp-
riate cationic surfactant have proved to be
highly effective for certain conifers such as
Tsuga heterophylla (Raf .) Sarg. (Ross and others
1980). However, for other conifers, including
Pseudotsuga menziesii (Pharis and Ross 1976) and
Larix species (Bonnet-Masimbert 1982), such
foliar-applied GA4/7 seems to be only poorly
absorbed. Another problem with conventional
foliar sprays is that they are quite wasteful of
the GA4/7. Ultra low volume (ULV) sprayers
combined with anti-evaporant spray oils, which
also facilitate foliar absorption, may provide a
partial solution to both problems. Bower and
Ross (1986) provide preliminary results which
demonstrate its potential effectiveness for the
operational GA4/7 treatment of P. menziesii seed
orchards. In this particular study, ULV spray
applications in 2-percent spray oil elicited
nearly three times the female flowering response
at only 1/8 the dosage of GA4/7 as the
conventional aqueous surfactant formulation
applied by high-volume mist sprayer.

Still for many conifers stem injection remains
the most effective method for applying GAs
(Pharis and Ross 1976; Bonnet-Masimbert 1982).
As originally developed for Pseudotsuga
menziesii , this involved feeding an aqueous
solution of 25 to 100 mg L'^CA^^ from a modified
medical intravenous unit into a small (5/16-inch)
hole drilled into the stem at the base of the
live crown. Despite its effectiveness, the
method was laborious — injection holes plugged
up after about 2 weeks and had to be redrilled —
and really only practical for treating relatively
small trees. Philipson (1985) however, recently
described a simpler method of stem injection
which he found to be both quite convenient and
highly effective for promoting flowering in large
(6-m tall) Plcea sltchensls (Bong.) Carr. trees.
It involved drilling two shallow holes on
opposite sides of the stem and then injecting
with a syringe a concentrated ethanolic solution
of GA 4/7. New holes were drilled after 2 weeks
and the treatment repeated. Ross and Bower
(1985, unpublished) are presently evaluating this
quick method of stem injection (about 2 min/tree)
on 4- to 14-cm diameter P. menziesi i grafted
propagules .

Interactions with Other Regulators

The less polar GAs are the only growth regulators
that consistently promote flowering in Pinaceae
family conifers (Pharis and King 1985; Pharis
and Ross 1984; Pharis and Ross 1986; Ross and
Pharis 1986).

The auxin, napthaleneacetic acid (NAA), was found
by Hashizume (1967) to be an effective promoter
by itself of female flowering In girdled Larix
leptolepis saplings, but this was not the case
for P_. menziesii (Ross, unpublished results) or
two species of Picea (Tompsett 1977; Dunberg and
others, unpublished results). However, for these
latter species NAA and other synthetic auxins
were found to enhance the efficacy of applied
GAs. In the case of sexually mature grafts
of Picea sitchensis (Thompsett 1977) and In
immature P_. menziesii seedlings where the auxin
was used at a higher concentration (Pharis and
others 1980) the effect of applying NAA in
conjunction with GAs was to promote male flower-
ing at the expense of female flowering. However,
with younger seedlings (table 2) and grafts (Ross
1976) of P_. menziesii , synthetic auxins also
enhanced the female flowering response to applied
GA4/7 although not to the same extent as for male
flowering.

Table 2. — Female and male flowering responses by
potted Pseudotsuga menziesii trees to
stem injections of gibberellin A4/7
(GA4/7) alone and with the synthetic
auxins, napthaleneacetic acid (NAA) or
2 , 4 , 5-triphenoxypropionic acid (2,
4,5-TP) (Ross, unpublished results,
1979)1'2

Cone buds/tree (no.+ s.e.)
Treatment Female Male

Untreated

12

+

6

24

+

14

GA4/7 alone

67

+

15

50

+

17

GA4/7 + NAA

78

+

14

133

+

45

GAA/7 + 2,4,5-TP

100

+

18

138

+

38

^Plants were 5-year-old rooted cuttings of
physiological age 12-14 years at time of treat-
ment; 20 plants/treatment and grown outdoors in
17 L containers.

^Growth regulators were infused into the main
stem, GA4/7 at 100 mg L-' and each auxin at 5 mg
L-' in 0.02 percent ethanol :water , for 12 weeks
commencing approximately 2 weeks prior to
vegetative bud burst.

On the other hand, studies with Tsuga hetero-
phylla (Ross and others, unpublished results)
and Pinus radiata D. Don (G. B. Sweet, personal
communications) have found no response to NAA,
either by itself or with GA^,^ or GA^ . It seems
that further research is required to establish if
auxins and possible other growth regulators
interact with GAs to influence cone-bud sexuality
in conifers as is clearly the case for many
angiospermous plants (Pharis and King 1985).

173

Interactions with Cultural Treatments in the
Field

Promotion of flowering under field conditions is
a tricky business, especially where young trees
are involved. No single treatment, including
GA^y -j should be expected to work if site and
climatic factors are otherwise unfavorable for
flowering, as frequently appears to be the case
in many of our seed orchards. However, many
studies have shown that the probability of
success is greatest where GA4/7 is applied in
conjunction with a cultural treatment (for
example, nondestructive stem girdling, nitrate
fertilization, rootpruning, or drought) that by
itself is often ineffective in promoting
flowering .

Table 3. — Interaction between gibberellins A4/7+
and stem girdling on flowering in
7-year-old Pseudotsuga menziesii grafts
at the Weyerhaeuser Company, Rochester,
WA, seed orchard (Cade and others, un-
published results, 1975) 1

Clones Producing Cone buds/tree

Treatment

Females Males Seed Pollen

(percent) (percent) (no.) (no.)

Untreated 24 a

Girdled only 35a

GA^/7 + GA3 only 67b

GA4/7 + GA3 + gird. 97 c

26a 2.2a 62a

40b 1.7a 27a

30a 7.8b 20a

74b 50.7C 337b

^Values followed by the same letter do not dif-
fer significantly at P £ 0.05 based on Chi
square (percentages) and Duncan's multiple
range tests.

2irees received double overlapping, half-cir-
cumferential band girdles, 6 mm wide and 130 mm
apart, at onset of vegetative bud swelling.

3stem injections of a 1:1 (w/w) mixture of GA4/7
and GA3 in 0.02 percent ethanol : water also
commenced then and continued for 9 weeks, during
which time each tree received on average 1.13 g
of the GA mixture.

*Each treatment was tested on 1-7 (3.8 avg.)
ramets from each of 56 sexually mature parent
tree clones.

Table 3 provides a classic example of such syner-
gism, in this case between stem injections of
GA4/7 + GA3 and overlapping band girdles in a 7-
year-old grafted P. menziesii seed orchard in
western Washington. Wheeler and others (1985)

have shown that the same girdling treatment can
be highly effective in the promotion of flowering
for similar aged grafts in another orchard in
southern Oregon. However, the Washington orchard
was located on a relatively cool, moist site that
was generally not favorable for early flowering,
and here stem girdling by itself was totally
ineffective in increasing the production of seed-
or pollen-cone buds. The grafts responded with a
small (though significant) increase in female
flowering from stem injections of GA4/7 + GA3
alone. Yet, when combined, the hormone and gird-
ling treatments had a highly synergistic effect,
increasing the mean production of seed- and
pollen-cone cones relative to the otherwise best
treatment by factors of 5.6 and 44, respectively.
Ross and others (1980) report a similar
synergism between GA^/y and calcium nitrate
fertilization for field-grown rooted cuttings of
mature Tsuga heterophylla clones.

It should not be concluded from the above that
GA4/7 is always the most effective treatment.
There are examples for P_. menziesii where gird-
ling (Bower and Ross 1986) and root-pruning
(Ross and others 1985) were each more effective
in promoting flowering than GA4/7 alone. However,
the point we wish to make is that best results
have always been attained where the cultural
treatment was applied in conjunction with the
GA4/7. What is furthermore important is that
the magnitude of the synergism tends to be
proportionately stronger for those inherently
recalcitrant clones and families that without
treatment might not contribute to seed production
(Ross and others 1980; Ross and others 1985).

Benefits of Container Culture

The fact remains that, even with the best treat-
ments, flowering under field conditions will
still be subject to the vagaries of climate.
Ross and others discuss elsewhere in this
proceedings the practical advantages of managing
small trees indoors in pots for seed production,
central among which Is the ability to provide
optimal environmental conditions at the proper
time for influencing cone-bud differentiation.

With P. menziesii, simply restricting the root
volume of trees in small pots can by itself be
sufficient to cause precocious or enhanced
flowering. In one experiment grafts were either
outplanted into the seed orchard in the spring of
their second growing season or left outdoors in
2.1 L containers where they were kept well

watered, with and without stem injections of 25

1

mg L GA^/j for 6 weeks commencing at vegetative
bud burst. The following spring only 14 percent
of the outplanted grafts, but 80 percent of the
potted but otherwise untreated grafts, initiated
seed-cone buds, an average of 1.1 and 10.1 each,
respectively. Treatment with GA^/y nearly
trebled the production of seed-cone buds (to
28.1) by potted grafts and resulted in 94 percent
of the grafts flowering.

174

A second experiment with 2-year-old rooted
cuttings of P. menzlesli seedlings only 2 years
of age at time of propagation illustrates the
importance of container size on flowering. Trees
were repotted from 2.1 L into 6.3 , 19.7 or 76 L
containers in early spring and received stem
injections of GA4/7 as above plus a heavy dosage
of calcium nitrate. All trees were well watered
and the following spring those in the smallest
containers produced an average of 14.8 seed-cone
buds each, compared to only 3.0 and 0.8 for trees
in the 19.7 and 76 L containers, respectively.

Table 4. — Interaction between gibberellin A4/7
(GA4/7), water stress (WS) and stem
girdling on female flowering in young
potted Pseudotsuga menziesii trees
(Ross, unpublished results, 1976)^'^'

Treatment

Plants
flowering
(percent)

Seed-cone
buds/ tree
(no.)

Untreated

5a

0.1a

GA4/7 alone

50b

6.5b

GA^/7 + WS4

60b

19. ic

GA4 /7 + gird.5

55b

13.5c

GA4/7 + WS4+ gird

• 5 65b

7.0b

Iplants were 2-year-old rooted cuttings of
physiological age 4 years from seed at time of
treatment: 20 plants/treatment.

2values followed by the same letter do not
differ significantly at P<0.05 based on Chi-
square (percentages) and Duncan's multiple range
tests .

^Stem injections of GA^/7 at 25 mg L-1
in 0.05 percent ethanol : water began on 21 April
and lasted 12 weeks.

^Throughout the period of GA treatment water
stressed plants were allowed to attain an
average pre-sunrise shoot water potential of
-1.5 Mpa before watering to saturation.
Non-stressed plants were well watered.

^Girdled plants received double overlapping,
half-circumferential band girdles, 4 mm wide and
20 mm apart, beneath the lowermost live branch
on 20 April.

It is not clear to what extent this reflects a
more rapid build-up of favorable (to flowering)
internal water deficits in the small containers,
or reduction in root activity associated with pot
binding (Bonnet-Masimber t 1982; Philipson 1983).
Various studies on several conifers (table 4; Ross

1978; Brix and Portlock 1982; Bonnet-Masimbert
1982; Philipson 1983) have shown that drought pro-
duced by withholding irrigation frequently enhances
synergistically the flowering response to applied
GA4/7 in Potted trees. Note in table 4 that non-
destructive stem girdling was nearly as effective
in this regard as was drought, whereas the two
cultural treatments applied together were apparent-
ly overly stressful and antagonized the flowering
response to GA4/7.

CONCLUSIONS

With Regard to Flowering

It appears to us, based on the plethora of
successful reports noted in table 1 and in Pharis
and Ross (1984), and the examples given in tables
2-4, that virtually any Pinaceae family conifer
will be amenable to manipulation of early and/or
enhanced flowering by the use of GA4/7 + an
appropriate cultural treatment. Additional gains
in male flowering and increased female flowering
may also be made through the judicious use of the
auxin, NAA, given with the GA4/7 and cultural
treatment .

New species, and new or unusual climatic condi-
tions (for field seed orchards) may require some
additional empirical research with regard to opt-
imal timing of treatment, dosages of hormone(s),
and the most appropriate cultural treatment to
use. In some locations climatic conditions
(for example, wet cool, low solar insolation) may
preclude successful field cone induction most
years. In those cases, and perhaps in most
instances, we recommend the use of potted
propagules and heated plastic house environments
during treatment (see paper by Ross and others,
this proceedings).

In essence, there is now no excuse for a conifer
tree breeder to wait for nature to bring on the
'natural' flowering that results from termination
of the so-called 'juvenile phase'. All conifer
species should be manipulable, even at very early
ages, through the properly timed use of GA4/7
application combined with an appropriate cultural
treatment .

Additionally, enhanced production of seed from
genetically superior propagules/Fl seedlings is
now within our grasp (see Ross and others, this
proceedings), and we believe that such an
approach is, or can be, more cost effective than
field seed orchards.

With Regard to Early Progeny Testing

Finally, we now have before us the very real
possibility that inherently superior families, or
even genotypes, can be tested at an early age
(see table 5), and that young seedlings/germin-
ants from these superior families and/or geno-
types, can be multiplied clonally for out-
planting. It is not unrealistic to visualize
potted propagule seed orchards yielding, each

175

Table 5. — Effect of family and application of gibberellin ki /y mixture on the growth of
Pinus radiata seedlings at different ages in the phytotron compared to growth
rating of same families at age 9+ years in the field3

Stem volume (cm^) at ages^ from germination

Stem volume

138 dayse 175 days^ increase due

Growth rating to GA^y

Family at age 9+ Control Control Gk^p- treatment

code° years in field0 (no'GA^/y) (no Gk^p) treated^ (%)

R4 Fast 14.7(l)h

R3 Fast 13.1(2)

R5 Fast 11.2(3)

R6 Fast 10.9(4)

R9 Slow 9.0(5)

R8 Fast 8.8(6)

R7 Slow 8.7(7)

R2 Slow 8.7(7)

Rl Slow 8.3(8)

37.4(2)h

53.5(1)

43

36.2(3)

41.9(5)

16

39.4(1)

47.1(4)

20

29.5(7)

49.6(2)

67

29.8(6)

38.7(6)

30

27.8(8)

48.0(3)

73

34.4(4)

33.4(7)

-3

26.1(9)

32.4(8)

24

30.6(5)

27.1(9)

-11

Note: The effects of 'family' were evaluated statistically by ANOVA and the Duncan's
Multiple Range test. Solid vertical bars connect family means which are not signifi-
cantly different at P < 0.05. For GA^/y-treated families vertical bars are not
used, here R4 and R6 differed significantly from Rl at P < 0.05. Family R8 was
'bushy' (had a high proportion of d.w. in lateral branches) except for Gk^p-
treated plants, where volume (and also dry matter) was reallocated to the main stem.
See also table 2 in Ross and others (1983) for an example of a similar reallocation
in Pseudotsuga menziesii .

a Based on unpublished research results of R. Pharis, R. Griffin, K. Eldridge, M. Slee,
G. Nikles, P. Cotterill.

b Seed was obtained from controlled pollination crosses made at least 10 years earlier,
and was stored until germinated in November 1981. Parents are of New Zealand origin
and were all classed as superior phenotypes ("plus trees").

c Field growth ratings of "fast grower" or "slow grower" were given to each family on
the basis of volume and height growth at age 9+ over a wide range of progeny tests in
southeastern Australia.

" Plants were raised in a phytotron at 25°C day/20°C night under natural daylength

(November 1981 to May 1982); all plants received supplemental incandescent light for
16 hr/day in the phytotron.

e Stem volume was calculated from height and circumference (1 cm above cotyledons),
averages based on measurements of 10 plants for each family (age 138 days).

1 Stem volume was calculated from height and diameter (12 cm above cotyledons).

Averages based on measurements of about three plants for each of control and Gk^p
treatment groups, for each family.

8 Gibberellin k^p (about 55:45 GA^GAy), obtained gratis from Imperial

Chemical Industries, was applied as a root drench at 200 mg per L pH 8.0 (60 ml per
pot, every 6 days from age 132 days) to three seedlings of each family. As part of
the paired test, equivalent numbers of seedlings received as a "control"

treatment .

h Values in parentheses represent family ranking within each test or age class.

176

year, several hundred thousand to several million
'superior' seeds. Each of those seeds from
superior families could in turn yield 10 to
perhaps several hundred vegetative propagules for
out-planting. And, if the vegetative propagation
is accomplished during the first year after germi-
nation, then loss of juvenile growth potential due
to 'maturation effects' will be minimal or
negligible. The following discussion, and table
5, discuss briefly the possibility of early
progeny testing of families produced from Fl
controlled crosses. Although we do not discuss it
herein, a number of techniques exist to propagate
young seedlings, including traditional and tissue
culture techniques. Indeed, it now appears that
somatic embryos of Picea abies, and perhaps
Pseudotsuga menziesii, have been produced from
juvenile tissue, and if somaclonal variability is
not excessive, somatic embryogenesis may offer an
alternative and highly effective means of clonal
multiplication of superior families and/or geno-
types. (Durzan, D.J. personal communication).

The possiblity that inherent vigour in vegetative
growth may be 'testable' at a very early age is
shown by data presented in table 5 for Pinus
radiata . Early stem volume growth differed
significantly between families by day 138 from
germination (table 5), and if A stem volume
growth [Ross and others (1983, table 4)] or stem
d.w. is used, the significance of the differences
is even more pronounced (data not shown).
Similar tests on Pinus caribaea (Pharis and
others, unpublished) and Picea mariana (Williams
1985; Williams and others 1985, unpublished)
lead to the same conclusion — there are strong
and significant correlations between early growth
in a Phytotron or glasshouse environment and
growth in the field at about age 10 years.

Between age 138 and 175 days the P. radiata seed-
lings appeared to be getting 'pot-bound', and the
attractive correlation between rank order in the
phytotron and growth rating in the field is less
strong (for example, compare control values at age
175 days with values at age 138 days — Families
R6 and R8 have dropped to rank order 7 and 8
respectively, and Families R7 and Rl , both slow
growers in the field, have risen in rank order to
4 and 5). However, for those plants which were
treated with GA4/7, Phytotron rank order still
correlates very well with field growth rating (for
example, the five fastest growing families in the
phytotron are also fast growers in the field.

Hence, it appears that being 'pot-bound' may
reduce stem volume growth (and other parameters
also, data not shown) of some fast-growing
families to a greater degree than slow-growing
families. This lesion, however, appears to be
'cured' by application of GA^p (table 5).

The GA4/7 treatment causes an increased alloca-
tion of photo-assimilate to the main stem, often
at the expense of the lateral branches for P.
radiata (Pharis and others, unpublished) and
Pseudotsuga menziesii (see table 2 in Ross and
others 1983)"! Hence, families R6 and R8 may,

under pot-bound or even normal conditions, have
at least a modest deficit of endogenous GAs, or
other mobilizing/growth factors. The 'other
mobilizing/growth factors' may in turn be
produced by or from endogenous GAs/exogenous GA
application. The effect of GA^/^and/or high
endogenous GAs could be as straightforward as
providing more internodal volume in which to
store photosynthate . Or GAs may have other or
additional effects such as increasing ion uptake
or 'mobilizing' photosynthate.

Thus, GA4/7 application may be a useful addit-
ional tool to assist in discrimination between
fast- and slow-growing families of P. radiata,
and indeed GA4/7 was also shown to be useful for
this purpose with Picea mariana (Williams and
others [1985], unpublished).

REFERENCES

Bonnet-Masimbert , M. Effect of growth regulators,
girdling and mulch on flowering of young
European and Japanese larches under field
conditions. Can. J. For. Res. 12: 270-279;
1982.

Bower, R. C; Ross, S. D. Evaluation of ULV
sprays for seed orchard application of
GA4/7. In: 20th Mtg. Can. Tree
Improvement Assoc: proceedings Part 2; 1985
19-22 Aug. Laval, Quebec. Ottawa, Ont.: Can.
For. Service, (in press).

Brlx, H . ; Portlock, F.T. Flowering response of
western hemlock seedlings to gibberellin and
water-stress treatments. Can. J. For. Res. 12:
76-82; 1982.

Cecich, R. A. Flowering of jack pine seedlings
two years after application of GA^p. In:
Second N. Central Tree Improvement Conference:
proceedings; 1981; U.S. For. Service, Rocky
Mtn. For. and Range Expt. Sta.; 1982:
176-182.

Cecich, R. A. Flowering in a jackpine seedling
seed orchard increased by spraying with
gibberellin A, p. Can. J. For. Res. 13:
1056-1062; 1983.

Cecich, R. A. White spruce (Picea glauca)

flowering in response to spray application of
gibberellin A, p. Can. J. For. Res. 15:
170-174; 1985.

Chalupka, W. Effect of growth regulators on the
flowering of Scots pine ( Pinus sylvestris L.)
grafts. Silvae Genet. 27: 62-65; 1978.

Chalupka, W. Time of GA^p application may
effect the sex of Scots pine flowers
initiated. Silvae Genet. 33: 173-174; 1984.

Chalupka, W.; Giertych, M. The effect of

polyethylene covers on the flowering of Norway
spruce [Picea abies (L.) Karst.] graft. Arbor.
Kornickie. 22: 185-192; 1977.

177

Dunberg, A. Flower induction in Norway spruce.
In: IUFRO Meeting of Working Parties on Norway
Spruce Provenances and Norway Spruce Breeding:
proceedings; Escherode, FRG; 1979: 139-157.

Dunberg, A.; Oden, P. Gibberellins and conifers.
In: The Biochemistry and Physiology of
Gibberellins, Vol. 2, Crozier, A., ed . ,
Praeger, New York; 1983: 221-296.

Hall, P. Genetic improvement of black spruce in
Newfoundland. In: Yeatman, C.W., ed.
Nineteenth Mtg. Can. Tree Improvement Assoc.:
proceedings, Part 1; 1983 August 22-26;
Toronto Ontario. Ottawa, Ont.: Can. For.
Service; 1984: 38-41.

Hare, R. C. Application method and timing of
gibberellin A^/y treatments for
increasing pollen conebud production in
southern pines. Can. J. For. Res. 14: 128-131;
1984.

Harrison, D. School of Forestry, Austral. Natl.
Univ., Canberra, Australia; 1985. Unpublished
research results.

Hashizume, H. Experimental induction of female
flowers in young Japanese larch (Larix
leptolepis Gordon). J. Jpn. For. Soc. 49:
405-408; 1967.

Katsuta, M. Recent advances in the promotion of
strobilus production by gibberellins in
Japanese conifers. In: Symposium on Flowering
Physiology: proceedings; XVII IUFRO World
Congress, Kyoto, Japan. Krugman, S. and M.
Katsuta, eds., Japan, For. Tree Breeding
Assoc., Tokyo; 1981: 68-74.

Marquard, R. D.; Hanover, J. W. Sexual zonation
in the crown of Picea glauca and flowering
response to exogenous GA, i-j. Can. J. For.
Res. 1984a. 14: 27-30.

Marquard, R. D.; Hanover, J. W. Relationship
between GAwy concentration, time of
treatment, and crown position on strobilus
production of Picea glauca. Can. J. For. Res.
1984b. 14: 547-553.

Owens, J. N. ; Webber, J. E . ; Ross, S. D.; Pharis,
R. P. Interaction between gibberellin
^4/7 an<^ rootpruning on the reproductive
and vegetative processes in Douglas-fir. III.
Effects on anatomy of shoot elongation and
terminal bud development. Can. J. For. Res.
1985. 15: 354-364.

Pharis, R. P. Method and Composition For Treating
Trees, Patent 180378 (22 March, 1976; New
Zealand); 514194 (30 March, 1976; Australia);
4110102 (29 August, 1978; U.S.A.); 1547571 (22
August, 1979; U.K.); 1078203 (27 May 1980;
Canada) .

Pharis, R. P. Interaction of native or exogenous
plant hormones in the flowering of woody
plants. In: Regulation of Developmental
Processes in Plants, Schutte, H.R., and Gross,
D., eds., Acad. Sci., Halle, GDR. 1978:
343-360.

Pharis, R. P.; King, R. W. Gibberellins and
reproductive development in higher plants.
Ann. Rev. Plant Physiol. 36: 517-68; 1985.

Pharis, R. P.; Kuo, C. G. Physiology of

gibberellins in conifers. Can. J. For. Res.
7: 299-325; 1977.

Pharis, R.P.; Ross, S.D. Gibberellins: Their

potential uses in forestry. Outlook on Agric.
9: 82-87; 1976.

Pharis, R.P.; Ross, S.D.; McMullan, E.E.

Promotion of flowering in the Pinaceae by
gibberellins. III. Seedlings of Douglas-fir.
Physiol. Plant. 50: 119-126; 1980.

Pharis, R. P.; Ross, S. D. Progress in hormonal
cone induction in Pinaceae conifers. In:
Yeatman, C.W., ed. Nineteenth Mtg. Can. Tree
Improvement Assoc.: proceedings, Part 1; 1983
Aug. 22-26; Toronto Ont. Ottawa, Ont.: Can.
For. Service; 1984: 163-169.

Pharis, R.P.; Ross, S.D. Pinaceae Hormonal
Promotion of flowering. In: Handbook of
Flowering, Vol. 5, ed . Halevy, A., CRC Press,
Boca Raton, Fla. 1986: (in press).

Pharis. R. P.; Tomchuk, D.; Beall, F. D., Rauter,
R. M.; Kiss, G. 1985. Promotion of flowering
in white spruce (Picea glauca) by gibberellin
A^/y alone, and with auxin (naphthalene-
acetic acid), girdling, and Ca ^03)2 as
adjunct cultural treatments. Can J. For. Res.
1985. (accepted).

Philipson, J. J. Flowering of Sitka spruce, In:
Report on Forest Research, 1980, Her Majesty's
Stationary Office, London. 1981: 30-31.

Philipson, J.J. Flowering stimulated in Sitka-FC
research. Forestry and British Timber, Feb.
1983.

Philipson, J. J. The promotion of flowering in
large field-grown Sitka spruce by girdling and
stem injections of gibberellin A^y. Can.
J. For. Res. 1985. 15: 166-170.

Pollard, D. F. W.; Portlock, F. T. The effects of
photoperiod and temperature on gibberellin
A^/y-induced strobilus production of
western hemlock. Can. J. For. Res. 1984. 14:
291-294.

Ross, S.D. Differential flowering responses by
young Douglas-fir grafts and equisized
seedlings to gibberellins and auxin. Acta
Hortic. 56: 163-167; 1976.

178

Ross, S. D. Influences of glbberellins and
cultural practices on early flowering of
Douglas-fir seedlings and grafts. Forestry
Res. Tech. Rep., 042/3001/77/7. Centralia, WA:
Weyerhauser; 1977. 9p.

Ross, S.D. Influences of glbberellins and
cultural practices on early flowering of
Douglas-fir seedlings and grafts. In: Third
World Consult. For. Tree Breeding: proceed-
ings Vol. 2; 1977 March; CSIRO Canberra,
Australia; 1978: 997-1107.

Ross, S. D. Efficacies of glbberellins alone and
with auxins NAA and 2, 4, 5, -TP for promotion
of male and female flowering on Douglas-fir
Weyerhauser Forestry Res. Tech. Rep.,
042-3201/79/29. 1979a: 7p.

Ross, S. D. Evaluation of foliar spray formula-
tions for seed orchard application of glbber-
ellins, Weyerhaeuser Forestry Res. Tech.
Rep., 042-3001/79/10. 1979b: 12p.

Ross, S. D. Early flower induction in Douglas-
fir: 1976 trials at Sequim, Washington I.
Hormone treatments and application regimens,
Weyerhaeuser Forestry Res. Tech. Rep.,
042-3201/79/45. 1979c: 13p.

Ross, S. D. Enhancement of shoot elongation in
Douglas fir by gibberellin A^/y and its
relation to the hormonal promotion of flower-
ing. Can. J. For. Res. 1983. 13: 986-994.

Ross, S. D. Interactions between heat, drought,
gibberellin A^/y and their timing on
flowering and shoot elongation in potted Picea
engelmannii (Perry) grafts. Can. J. For. Res.
1985 (in press) .

Ross, S. D. Container seed orchard and crown
management research, In: Yeatman, C.W., ed .
19th Mtg. Can. Tree Improvement Assoc.: proc-
eedings, Part 1; 1983 Aug. 22-26; Toronto, ON.
Ottawa ON: Can. For. Service 1984: 192-195.

Ross, S. D.; Bollman, M. P.; Pharis, R. P.;
Sweet, C. B. Gibberellin Awy and the
promotion of flowering of Plnus radlata:
Effects of partitioning of photoassimilate
within the bud during primordia differenti-
ation. Plant Physiol. 76: 326-330; 1984.

Ross, S. D.; Pharis, R. P. Cibberellin-induced
flowering of Douglas-fir grafts, Plant
Physiol. 51 (Suppl.): 36 (Abstract); 1973.

Ross, S. D.; Pharis, R. P. Progress in the promo-
tion of early flowering in Douglas-fir by glb-
berellins. Forestry Research Tech. Rep.
50/3001/76/32. Centralia, WA: Weyerhauser;
1976: 9p.

Ross, S. D.; Pharis, R. P. Recent developments in
enhancement of seed production in conifers,
In: Edwards, D.G.W. and Yeatman, C.W., eds.
18th Mtg. Can Can. Tree Improvement Assoc.:
proceedings; Part 2; 1981; Ottawa, ON: Can.
For. Serv.; 1982: 26-38.

Ross, S. D . ; Pharis, R. P. Promotion of Flowering
in Crop Trees: Different Mechanisms and Tech-
niques with Special Reference to Conifers.
In: Attributes of Trees or Crop Plants,
Cannell, M.G.R., Jackson, F.E., and Gordon,
J.C., eds., Instit. Terrestrial Ecology, Monks
Wood Exper. Sta., Abbots Ripton, Huntingdon,
U.K.; 1986: 387-401.

Ross, S. D.; Pharis, R. P.; Binder, W. D. Growth
regulators and conifers: Their physiology and
potential uses in forestry. In: Plant Growth
Regulating Chemicals, Nickell, L.G., ed.,
Chemicals, CRC Press, Boca Raton, Florida.
1983: 35-78.

Ross, S.D.; Piesch, R.F.; Portlock, F.T.
Promotion of cone and seed production in
rooted ramets and seedlings of western hemlock
by gibberellins and adjunct cultural treat-
ments. Can. J. For. Res. 11: 90-98; 1980.

Ross, S. D.; Webber, J. E.; Pharis, R. P. Owens,
J.N. Interaction between gibberellin A, i-,
and rootpruning on the reproductive ana vege-
tative process in Douglas-fir. I. Effects of
flowering. Can. J. For. Res. 15: 341-347;
1985.

Tompsett, P.B. Studies of growth and flowering in
Picea sltchensis (Bong.), Carr. 1. Effect of
growth regulator applications to mature scions
on seedling rootstocks. Ann. Bot . 41:
1171-1178; 1977.

Webber, J. E. Tree Improvement physiology

research, B.C. Ministry of Forests 1981-1983.
In: Yeatman, C.W., ed. 19th Mtg. Can. Tree
Improvement Assoc.: proceedings, Part 1; 1983
Aug.; Toronto ON. Ottawa, ON: Can. For.
Service; 1984: 196-199.

Webber, J. E.; Ross, S. D.; Pharis, R. P.; Owens,
J.N. Interaction between gibberellin A, /-,
and rootpruning on the reproductive and vege-
tative process in Douglas-fir. II. Effects on
shoot growth. Can. J. For. Res. 15: 348-353;
1985.

Williams, D.J. Early evaluation of controlled

crosses of Black Spruce [Picea mariana (Mill.)
B.S.P.]. Edmonton, Alberta: University of
Alberta; 1985. Dissertation.

Williams, D. J.; Dancik, B. P.; Pharis, R. P.
Early progeny testing and evaluation of
controlled crosses of Black Spruce. Submitted
to Can. J. For. Res. 1985.

Zeevaart, J. A. D. Gibberellins and flowering.
In: The Biochemistry and Physiology of
Gibberellins. Vol. 2, Crozier, A., ed . ,
Praeger, New York; 1983: 333-374.

Zimmerman, R. H.; Hackett, W. P.; Pharis, R. P.
Hormonal aspects of phase change and
precocious flowering. In: Encyclopedia of
Plant Physiol. (N.S.), Vol. 11, Pharis, R. P.
and Reid, D. M., eds., Springer-Verlag,
Berlin/Heidelberg/New York; 1985: 79-115.

179

POTENTIAL FOR CONTAINER SEED ORCHARDS

Stephen D.^Ross, Andrea M.^JLastham and Ralph C.I Bower

ABSTRACT: Results are presented for Engelmann
spruce and western hemlock which demonstrate
the practical advantages of producing
genetically improved seeds on small potted
trees within a plastic-covered house. Relative
to conventional soil-based orchards, these
advantages include: earlier and more
consistently abundant flowering; improved
protection of cones and seeds; and strict
control of pollen parentage, together with
flexibility of clonal composition for maximum
genetic gains. Because of more efficient space
utilization and the greater ease and
flexibility of management, production costs
also promise to be lower.

(Sweet and Krugman 1978; Ross and Pharis
1982). These included insufficient initiation
of reproductive structures following a
generally long "juvenile" phase; problems of
pollination, including control of pollen
parentage, and subsequent cone and seed
development; and a host of management problems
associated with tree size.

In this paper we review the potential
advantages of indoor-potted seed orchards for
more rapidly, completely and economically
realizing the benefits of tree improvement.
For convenience of discussion these are grouped
into the three categories of production,
genetic and management efficiency.

INTRODUCTION

Potted trees, subject to strict control over
environmental and treatment factors, have long
been preferred by physiologists interested in
studying the mechanism and control of flowering
in conifers (e.g. Longman 1982; Ross 1985).
Tree breeders, too, have occasionally
capitalized on the relative ease with which
precocious flowering can be induced, and
controlled crosses subsequently made, using
potted trees (Greenwood and others 1979) . Until
recently, however, there has been little
serious interest in the management of small
potted trees for volume production of
genetically improved seed.

Since 1980, the British Columbia Ministry of
Forests has been researching the development
and evaluation of the indoor-potted seed
orchards for western hemlock ( Tsuga
heterophy 11a (Raf.) Sarg. ) and the interior
spruces, Engelmann ( Picea engelmanni i Perry)
and white (Picea glauca (Moench) Voss). With
these species we faced (or would soon face) the
same problems encountered with most conifers in
attempting to produce genetically improved
seeds in conventional soil-based orchards

Paper presented at the Conifer Tree Seed in
the Inland Mountain West Symposium, Missoula,
MT, August 5-6, 1985.

Stephen D. Ross and Andrea M. Eastham are,
respectively, Tree Physiologist and
Horticulturist, Research Laboratory, British
Columbia Ministry of Forests, Victoria, B.C.,
Canada; Ralph C. Bower is Forest Geneticist,
MacMillan Bloedel Ltd., Nanaimo, B.C., Canada

PRODUCTION EFFICIENCY

Initiation of Reproductive Structures

Many studies (Ross 1978, 1985; Tompsett and
Fletcher 1979; Greenwood 1981; Bonnet-Masimbert
and others 1982; Longman 1982; Philipson 1983)
have demonstrated the relative ease with which
potted trees can be induced to flower, given
an appropriate stimulation treatment (see
also Pharis and Ross in this volume) . We
illustrate this with two examples comparing,
for western hemlock (table 1) and Engelmann
spruce (table 2), the flowering response
to induction treatments for potted and field-
grown trees. Trees were vegetative propagules
of mature plus-trees, with different clones
and ramets of similar (spruce) or different
(hemlock) ages, represented in the two orchard
types. All treatment trees received spray
applications of the growth regulator gibberellin
A ,/7 (GA »/y)> calcium nitrate fertilizer and
moderate drought (by withholding irrigation)
appropriately timed for each species (Ross 1985;
unpublished results). Potted trees of each
species received these treatments both outdoors
and in an unheated plastic-covered house (hemlock)
or in a 30°C day:20°C night heated house
(spruce). Heat treatment was also attempted
for spruce in the soil-based orchard by
enclosing grafts in a polyethylene tent.

Without treatment, flowering by both species in
both orchards was rather poor. Treatment of
western hemlock in the soil-based orchard
increased from less than 5 to nearly 320 the
mean number of seed-cone buds initiated per
ramet, and from 27% to 100% the proportion of
clones that flowered (table 1). However, the

180

Table 1. — Comparison of 1985 flowering responses for 7-year-old potted and 12-year-old field-grown
western hemlock rooted cuttings at the MacMillan Bloedel Harmac Tree Improvement Centre,
B.C. 1.2

Potted seed orchard * soil-based

Untreated Treated seed orchard

Flowering response Outdoors Outdoors Indoors untrt. treated

Female flowering
Ramets (=clones) Flowering (%) 75 100 100

Females/tree (no. + s.e.) 52+22 257+65 427+63

27
3+2

100
319+131

Male flowering
Ramets (=clones) FLowering (%) 56
Mean Pollen Score-* 1

73
1.6

86
2.2

36
1

61
1

1 All treated trees received calcium nitrate fertilizer in spring and 6 weekly foliar sprays of 200 mg
L~l GA4 / 7 commencing at vegetative bud burst. Potted trees, additionally, were subjected to a
simultaneous 6 weeks of moderate drought

2 One ramet per clone per treatment, with a different group of 22-23 parent-tree clones represented
each orchard

3 Pollen cone production scored: 0 = none, 1 = light, 2 = moderate, 3 = heavy

Table 2. — Comparison of 1985 flowering responses to gibberellin A4/7, drought and heat treatments for
7-year-old potted and 8-year-old field-grown Engelmann spruce grafted ramets representing
different parent-tree clones (Ross, Birzins and Cox, unpublished results)

Potted orchard ( Vic tor ia , B . C . ) Soil-based orchard (Vernon, B.C.)
Flowering response Outdoor GA4/7 + drought Untrted GA4 / 7 + drought

Control Outdoors Indoors Control Alone Tented

Ramets ( c lones ) It rmt (no.) 16(16)

16( 16)

16(16)

1489(296) 193(193)

410(209)

Female flowering

Ramets producing (%) 12

Clones producing (%) 12

Cones/tree (no. + s.e.) 1

69
69
30+9

75
75
54+33

10
23
1+ 1

17
17
2+1

24
24
11+4

Male flowering

Ramets producing (%) 19

Clones producing (%) 19

Cones/tree (no. + s.e.) 1

81
81
44+10

88
88
3 3+9

4
19
2+2

16
16
2 + 2

15
23
2+2

181

much younger (7 vs. 12 years) potted trees
initiated in response to treatment indoors 33%
more seed-cone buds each than the treated
field-grown trees. Outdoors, the potted trees
were less responsive to treatment, although
here also 100% of the clones flowered.

The difference in treatment response was even
more dramatic for Engelmann spruce where the
grafts in the two orchards were of similar ages
(table 2). As found by Chalupka and Giertych
(1977) for Picea abies (L.) Karst., heat
treatment by means of tenting significantly
enhanced female flowering in the soil-based
orchard, although still only 24% each of the
ramets and clones produced seed-cone buds. In
contrast, nearly 90% of the clones whose potted
ramets received heat plus GA4/7 treatment
indoors flowered, and their mean production of
seed-cone buds was almost four times that of
the comparably treated field-grown trees.

As is typical for young soil-based orchards,
pollen production here by both species was very
sparse in relation to female flowering. In
contrast, the smaller potted trees produced
pollen in abundance. Furthermore, a
significantly greater proportion of the clones
contributed to this more profuse male flowering
by potted trees, this again being particularly
dramatic for interior spruce (table 2) .

Where potted trees are properly managed (see
below) , there is no reason to assume that
profuse flowering cannot be consistently
achieved on a biennial basis. The potted trees
in both studies reported here had flowered
nearly as profusely in response to essentially
the same treatments applied two years
previously. This, however, was a first-time
attempt at stimulation of the field-grown
trees. Retreatment of a soil-based orchard may
or may not be successful, depending on climatic
conditions in the year of stimulation (Wheeler
and others 1985). Adverse site and climatic
factors can negate even the most effective of
treatments, including applications of GA4/7
in conjunction with nitrogen fertilizer,
girdling and root-pruning (see Pharis and Ross
in this volume) .

High temperature and water stress generally
appear to be the most important requirements
for abundant flowering in many conifers (e.g.
Pollard and Portlock 1981; Ross 1985). Their
timing is also critical, as the results for
Engelmann spruce in figure 1 clearly show.
Cone-bud differentiation in this species is
known to occur during the late stage of slow
shoot elongation, and it is only at this time
that heat treatment promotes flowering.
Applied earlier, while shoots are rapidly
elongating, its effect on flowering is strongly
inhibitory, although water stress is highly
promotive at that time and only then. This
optimal sequence of early drought and late high
temperatures seldom occurs in nature but is
easily created in an indoor-potted orchard.

Treatment Period

(in heated shelterhouse)

Figure 1. — Effect of heat and drought treatment
timing on female flowering in young potted
Engelmann spruce grafts. Ramets were moved into
a 30° C day:20° C night heated shelterhouse
at different stages of lateral shoot elongation
(VBS and VBB refer to vegetative bud swelling
and burst, respectively), during which they
were either kept well watered or subjected to a
moderate or severe drought by withholding
irrigation until their midnight needle water
potential had decreased to -0.75 or -1.5 MPa
(adapted from Ross 1985).

Cone and Seed Efficiency

The ability to protect developing cones and
seeds from adverse climatic conditions (high as
well as low temperatures and wind) is another
important advantage of indoor-potted orchards
(see Bramlett in this volume). The ready
accessibility of cones on small potted trees
for inspection and spraying should also result
in more effective insect and disease control.

Table 3 compares cone and seed traits for
potted ramets of the same western hemlock
clones that were induced to flower and
subsequently maintained either outdoors or in a
plastic-covered house without supplemental
heat. Not only did the trees outdoors initiate
significantly fewer seed cones, but the
proportion of those that survived was also
significantly lower than indoors. This was the
result of several severe freezes during winter
and early spring. Outdoors, the surviving cones
contained fewer total (filled plus empty) seeds
as well, probably for the same reason.
Possibly the different conditions of heat and
water stress under which trees were induced to
flower indoors and outdoors the previous year
had a carry-over effect on subsequent cone and
seed development. However, our results suggest
that so long as trees are not overly stressed
during induction such carry-over effects are
relatively small.

In a soil-based orchard, especially a young
one, pollen is likely to be limiting for good
seed set (see tables 1 and 2). This was not a

182

Table 3. — Cone and seed traits compared for
potted western hemlock rooted
cuttings induced and maintained
outdoors or in an unheated greenhouse
(Eastham and Ross, unpublished
data ) ■ • 2

T n H nr* r

Trait

(X+s.e. )

(X+s.e. )

Cones init iated/ tree (no.)

320+23

633+30

Surviving cones/tree (no.)

198+13

529+26

Cone efficiency (X)

62

84

Total seed/cone (no.)

2 7+1

31+1

Filled seed/cone (no.)

10+1

14+1

Seed efficiency (%)

37

45

100-filled seed wt. (mg)

200+5

199+5

28-day germination (X)

79+5

82 +5

1 Rooted cuttings were 7-years old at time of
induction, with 3-8 ramets each of the same 8
plus tree clones

2 See footnote 1 of table 1 for description
of induction treatment.

consideration in the present comparison of
indoor and outdoor potted trees (table 3).
Unbagged female strobili on both groups of
trees were supplementally pollinated at least
three times when maximally receptive with the
same pollen lot (from potted ramets indoors).
Even then, however, the filled seed percentage
was significantly lower outdoors. Cool, moist
conditions prevailed outdoors throughout the
pollination period and this could have
adversely affected pollen germination and
subsequent fertilization of ovules.

Potted trees in this study were maintained in
the plastic-covered house following
pollination. However, this does not appear to
be necessary, or in the case of some species,
particularly desirable for good subsequent cone
and seed development. Where flowering
Engelmann spruce grafts were subject to
elevated temperatures within an unheated house,
the resulting cones were significantly smaller
relative to ramets whose cones matured outdoors
(Ross, in prep.). These cones also contained
significantly fewer total seed, of which a
smaller percentage was filled.

For this species, and presumably most conifers,
optimal conditions for cone and seed
development are the same as those that favor
vigorous vegetative growth — that is, abundant
water and nutrients along with moderate
temperatures. These are not the conditions
that promote flowering. Another advantage of
working with potted trees, therefore, is that
they may be managed separately for each
initiation and subsequent development of
reproduction structures to ensure maximum
production of high quality seed. This is not
generally possible in a soil-based orchard.

GENETIC EFFICIENCY

Control of Pollen Parentage

It is no secret that the wind-pollinated
orchard is not very efficient when it comes to
capturing the genetic benefits from tree
improvement (Sweet and Krugman 1978; Smith and
Adams 1983; El-Kassaby and others 1984).
Dilution of genetic gains from contaminating
foreign pollen is a serious problem in many
such orchards. Where orchards are located off
site, the seed produced may also be maladapted
to the planting region. Furthermore, only a
relatively small proportion of the orchard
parents may be contributing male and female
gametes. Of these an even smaller proportion
may have the opportunity to mate due to the
nonsynchrony in timing of pollen shedding and
female receptivity among clones. The situation
is thus ripe for selfing and other departures
from truly panmictic mating, with potentially
deleterious effects on realized genetic gains.

Indoor-potted orchards, on the other hand,
allow a degree of control over pollen parentage
not feasible in conventional wind-pollinated
orchards to achieve maximum-possible genetic
gains. The approach to pollen management will
depend to a large extent on the status of the
breeding and testing program.

Artificial pollination may not be warranted
when dealing with untested clones. But even
then positive steps can be taken to ensure more
nearly panmictic mating. At the very least,
the potted trees can be effectively isolated
indoors from foreign pollen. And, it may be
expected that a larger proportion of clones
will be contributing male and female gametes
than in the soil-based orchard (tables 1 and
2), which should reduce the potential for self
pollination. Reproductive bud development in
spring can be synchronized, either hastened or
slowed as required for late- and early flushing
clones, respectively, by moving their potted
ramets into a warmer or cooler house.

It is, however, where information on the mating
value of individual orchard clones is available
that the genetic benefits from pollen
management in indoor-potted orchards will be
greatest. This may involve supplemental mass
pollination using pollen collected from the
very best parents. Even more attractive is the
opportunity to mass produce progeny of elite
full-sib families through controlled crossing,
and thus capitalize on that portion of the
nonadditive genetic variance associated with
specific-combining effects.

Unlike the problems encountered with large
trees in soil-based orchards (see Sweet and
Krugman 1978), artificial pollination is
relatively easy with small potted trees.
Certain clones (and even ramets within clones)
have a tendency to initiate mainly seed cones
or pollen cones, and these may be managed
separately as female and male parents. In

183

early spring the female parents might be moved
outdoors to slow their reproductive bud
development relative to pollen parents left in
an unheated house. This will ensure the
necessary lead time for collection and
processing of pollen from all clones prior to
the onset of female receptivity (J.E. Webber,
pers . com.). The pollen is easily harvested,
either by picking individual pollen cones as
they mature, or shaking the tree and collecting
the shed pollen on a sheet of paper. We have
found the latter method to be particularly
convenient for species, such as western
hemlock, which have very small pollen cones.
The female parents are then moved back indoors
for artificial pollination at the optimal time
with high viability, fresh pollen.

Infusion of New Selections into Production

By the time a conventional orchard comes into
full production, 10 to 15 or more years after
establishment, its seed may already be
genetically obsolete. Even if progeny test
results are not available before then, most
orchards will have already required a
silvicultural thinning by that age. The
opportunity to use these results to rogue
inferior clones from the orchard therefore may
be limited as well. With an indoor-potted
orchard the lag time between selection and seed
production for new clones is considerably
shorter. Also, unlike conventional orchards,
whose clonal composition is determined at time
of establishment, the potted orchard may be
continually upgraded genetically as new
selections become available. It is not only
more efficient genetically, but far less
expensive as well to replace individual potted
ramets than an entire soil-based orchard.

We call this in situ advancing-front , and the
concept of orchard generations no longer
applies (Ross and Pharis 1982). In the case
of a new tree improvement program, the potted
orchard would probably contain ramets of all
plus-trees in sufficient number to meet
projected seed requirements over, say, the next
10 years. Breeding and seed production could
occur simultaneously, but with primary emphasis
given initially to the former so as to minimize
delays in progeny test establishment.
Progressively more severe roguing of inferior
clones would then occur over time as the
progeny tests become older and their results
more reliable. Initial roguing of parents
could perhaps begin on the basis of four year
or younger progeny test results, with selection
among progeny of the best families occurring
several years later (Lambeth 1980; Pharis and
Ross in this volume). The clones thus
eliminated might be replaced initially by extra
ramets of the proven parent clones (as required
for seed production), but ultimately by ramets
of new advanced-generation selections. Even
then, the potted orchard could still contain
some highly superior parents from previous
generation( s ) .

MANAGEMENT EFFICIENCY
Production Costs

Indoor-potted orchards may be more efficient
than soil-based orchards in terms of providing
earlier production of seed of higher genetic
quality, but we are frequently asked are they
practical and cost effective? We are only just
beginning to acquire comparative cost-benefit
data for the two types of orchards, but for
western hemlock and interior spruce the answer
does appear to be yes.

Consider the hypothetical case of two interior
spruce orchards, one potted and the other soil
based, each with a production capacity of five
million viable seeds per year. Using Birzin's
(this volume) expected yield of 4,000 viable
seeds per ramet per annum by age 15 years after
grafting, and a final spacing of 5m x 5m (1250
ramets rogued from an initial 2500), the
soil-based orchard would occupy nearly 3.5 ha,
exclusive of roads and support facilities. Our
results indicate that each interior spruce
ramet in an indoor-potted orchard is capable of
producing on average 3,000 viable seeds every
other year beginning at age 7 years in response
to biennial induction treatment. Thus, a
potted orchard with the same five-million
annual production capacity would contain about
3,300 ramets but only occupy 700m2 of covered
houses and a like area of outdoor container
yards. (Only half the ramets would be indoors
at any one time — for cone induction and
pollination — except possibly for freeze
protection over winter when they can be tightly
packed to conserve space.)

Even a very superficial analysis indicates that
the potted orchard will probably be less costly
to establish. One does not require a
sophisticated controlled-environment greenhouse
— a plastic-covered house with propane heater
and let-down sidewalls for natural ventilation
will suffice. In this example, the cost to
construct two such houses (10m x 35m) and a 700
m^ outdoor container yard, each equipped with
movable pallets and automated drip-irrigation
system, should not exceed $45,000 (costs in
1985 Canadian dollars). Assume another $8,000
to propagate the additional 800 ramets which
the potted orchard requires (3300 vs. 2500
initially for the soil-based orchard).

However, the potted orchard requires little
space. It can be sited anywhere, on relatively
inexpensive and otherwise nonproductive land —
perhaps in conjunction with a container nursery
where certain facilities and equipment can De
shared. On the other hand, proper site
selection is crucial to the success of the
soil-based orchard (Birzins, this volume). And,
sites conducive to abundant flowering tend to
be in short supply and high demand for
agricultural use. Such land in the Okanagan
Valley of interior B.C. currently goes for
about $17,000 per hectare (Birzins, pers.
comm.), for a total purchase price of nearly

184

£60,000 for the 3.5 ha orchard in this
example. That does not include site
preparation, road construction, fencing or
irrigation systems.

The indoor-potted orchard is equally attractive
from the standpoint of potentially lower
operating costs. Contrary to popular belief,
container culture for seed production need not
be highly labor intensive. Much of the work
(potting and plant handling, irrigation and
applications of fertilizers, pesticides and
CAs) can be automated to a large degree. There
are the additional expenses for pots, media and
heating, but these promise to be small in
relation to the large amounts of fertilizers
and other chemicals and water used in
9oil-based orchards. Finally, there are large
economies to be realized in the induction,
pollination, protection and harvesting of cones
when working with small, potted trees.

In comparing the two orchards one must also
consider the difference in time and value of
seed produced. There are two costs associated
with the longer nonproductive phase of the
soil-based orchard (15 vs. 7 years for the
potted orchard in this example). There is the
time-discount factor and the opportunity cost
of delayed realization of genetic benefits.
In this example, the indoor-potted orchard
could already have produced 40 million viable
seed by the time the soil-based orchard came
into commercial production. That relates to
approximately 16,000 extra hectares of new
interior spruce plantations containing the very
best possible genetically improved trees.

Production Flexibility

Indoor-potted orchards offer a unique
flexibility of management. Their production
capacity can be rapidly scaled upwards or
downwards in response to changing seed
requirements. The mobility of potted trees
also allows for most efficient site
utilization. Unlike soil-based orchards with
their fixed-tree arrangement, the potted
orchard can start off small, and then be
expanded as needed to accommodate the
increasing size of trees.

Another advantage of indoor-potted orchards
relates to a point made by Libby (1985). He
notes that "Seed-orchards are sensitive to
economies of scale, and one can rarely justify
a seed-orchard for local low volume demand."
Even with so-called 'major' species having a
large total annual seed requirement, this
production is often divided among many breeding
zones each represented by its own small
orchard. Several of these may be consolidated
on a single site for increased management
efficiency, but then there is the possibility
that seed will be maladapted to their site of
utilization due to cross-pollen contamination
between the different breeding zones. Trees
from many breeding zones may be managed as a

single group in a potted orchard, except when
separated for pollen management.

In a potted orchard one can also conceive of
utilizing controlled pollinations to capitalize
on those unique characteristics of individual
clones (such as disease or drought resistance
or special wood properties), of the same or
different breeding programs, to custom produce
progeny that are specifically adapted to
certain problem sites or for minor, but
valuable end products.

Finally, potted orchards are also most
attractive for those minor species whose annual
seed requirements hardly justify a seed-orchard
program, but which are poor seed producers in
nature. Chamaecypar is nootkensis (D. Don)
Spach, a species not widely planted but still
highly prized for its quality wood, is an
excellent example. In containers this species
flowers profusely at a very young age in
response to GA3 treatment (Bower, unpub.
data). A desirable strategy for such a species
might be to rapidly build up a number of years'
supply of seed in a small orchard, and then
discard the potted ramets which could easily be
repropagated when the seed supply again runs
low.

CONCLUSIONS

Compared to conventional orchards,
indoor-potted orchards offer the potential for
attaining earlier, more reliable seed
production and maximum-possible genetic gains
through sexual reproduction, all with a much
greater efficiency of management. The
biological feasibility of producing genetically
improved seed on small, potted trees is now
well established for many conifers. And,
further research will lead to improved
efficiencies of production and management.
However, for western hemlock and the interior
spruces this research has progressed to the
point where we are now ready to begin pilot
testing the approach for cost effectiveness on
a semi-operational scale.

REFERENCES

Uonnet-Masimbert , M.P. Delany, G. Chanteloup,
and J. Cupaye. 1982. Influence de l'etat

d'activite des racines sur la floraison

induite par des gibbere 1 1 ines 4 et 7 chez

Pseudotsuga menziesii (Mirb.) Franco.
Silvae Genet. 31:178-183.

Chalupka, W. and M. Giertych. 1977. The
effects of polythene covers on the
flowering of Norway spruce ( Picea abies
(L.) Karst.) grafts. Arbor. Kornickie
22: 185-192.

El-Kassaby, Y.A., A.M.K. Fashler. and 0.

Sziklai. 1984. Reproductive phenology and
its impact on genetically improved seed

185

production in a Douglas-fir seed Orchard.
Silvae Genet. 33:120-125.

Greenwood, M.S., C.H. O'Gwynn and P.G. Wallace.
1979. Management of an indoor, potted
loblolly pine breeding orchard. In: Proc.
15the South. For. Tree Impr. Conf.
Mississippi State Univ., MS. pp. 94-98.

Greenwood, M.S. 1981. Reproductive development
in loblolly pine. II. The effect of age,
gibberellin plus water stress and
out-of-phase dormancy on long shoot growth
behaviour. Amer. J. Bot. 68:1184-1190.

Lambeth, C.C. 1980. Juvenile-mature

correlations in Pinaceae and implications
for early selection. For. Sci. 26:71-580.

Libby, W.J. 1985. Potential of clonal

forestry. In: L. Zsuffa, R.M. Rauter, and
C.W. Yeatman (eds.) Proc. 19th Meet. Can.
Tree Impr. Assoc. Part 2. pp. 1-1.

Tompsett, P.B., and A.M. Fletcher. 1979.
Promotion offlowering on mature Picea
sitchensis by gibberellins and
environmental treatments. The influence
timing and hormonal concentration.
Physiol. Plant. 45:112-116.

Wheeler, N.C. , C.J. Masters, S.C. Cade, S.D.
Ross, J.W. Keeley, and L.Y. Hsin. 1985.
Girdling: a effective and practical
treatment for enhancing seed yields in
Douglas-fir seed orchards. Can. J. For.
Res. 15:505-510.

Longman, K.A. 1982. Effects of gibberellin,
clone and environment on cone initiation,
shoot growth and branching in P inus
contorta. Ann. Bot. 50-247-257.

Philipson, J.J. 1983. The role of gibberellin

A4/7, heat and drought in the induction

of flowering in Sitka spruce. J. Exp. Bot.
34:291-302.

Pollard, D.F.W. and F.T. Portlock. 1981.
Effect of temperature on strobilus
production in gibberel lin-treated seedlings
of western hemlock. Can. For. Serv. Res.
Notes. 1:21-22.

Ross, S.D. 1978. Influences of gibberellins and
cultural practices on early flowering of
Douglas-fir seedlings and grafts. In: Proc.
3rd Consult. For. Tree Breeding. Vol. 2,
CSIRO, Canberra, Australia, pp. 997-1107.

Ross, S.D. 1985. Promotion of flowering in
potted Picea engelmannii (Perry) grafts:
effects of heat, drought, gibberellin
A4/7, and their timing. Can. J. For. Res.
15: In press.

Ross, S.D. ad R.P. Pharis. 1982. Recent
developments in enhancement of seed
production in conifers. In: D.F.W. Pollard,
D.G. Edwards, and C.W. Yeatman (eds.) Proc.
18th Meet. Can. Tree Impr. Assoc., Part 2.
p. 26-38.

Smith, D.B and D.T. Adams. 1983. Measuring
pollen contamination in clonal seed
orchards with the aid of genetic markers.
In: Proc. South. For. Tree Improv. Conf.
17:69-77.

Sweet, G.B. and S.L. Krugman. 1978. Flowering
and seed production problems - and a new
concept of seed orchards. In: Proc. 3rd
World Consult. For. Tree Breeding. Vol. 2.,
CSIRO Canberra, Australia, p. 749-759.

186

MANAGING BLACK SPRUCE SEEDLING SEED

Ronald

ABSTRACT: The use of fertilizers to stimulate
cone production in black spruce ( Picea mar iana
(Mill.) B.S.P.) seedling seed orchards is dis-
cussed. The types, rates, and time of fertilizer
application are reviewed, with emphasis on the
causes of inconsistencies in responses to ferti-
lizer treatments. The interactions between ini-
tial tree spacing and the time and degree of
rogueing and tree response to fertilizing are
briefly reviewed.

INTRODUCTION

ORCHARDS FOR CONE AND SEED PRODUCTION

0

F. I Smith

FACTORS AFFECTING RESPONSE TO FERTILIZING
Tree Size

Female and male cone production were positively
correlated with both tree height and diameter,
but the correlations were higher for diameter
(Smith 1983). Small trees did not respond as well
to fertilizer treatments as did large trees
(fig. 1). In some instances, small trees did not
respond at all to treatment regardless of the
type or rate of fertilizer applied.

In New Brunswick, seedling seed orchards are
established on 'typical* planting sites. Although
seedling orchards are not managed as intensively
as clonal orchards, cultural practices such as
fertilizing to enhance cone production will be
necessary for these orchards to be fully
productive .

Applying fertilizer is the most common technique
currently employed for enhancing cone production
in seed orchards. However, no one type of ferti-
lizer or rate of application has been successful
in all instances (Sweet and Hong 1978). The
effectiveness of any fertilizer treatment in
enhancing cone production varies with tree size,
tree spacing, the type of fertilizer used, the
rate, method, and timing of application, the
weather immediately following fertilizer applica-
tion, and the genetic predisposition of the trees
to produce cones.

Fertilizer and spacing experiments were estab-
lished in 8-to 10-year old black spruce (Picea
mariana (Mill.) B.S.P.) plantations from 1 980-
1982 (Smith 1983). This paper summarizes some of
the factors that influenced the effectiveness of
applying fertilizers to enhance cone production
and discusses the results as they apply to manag-
ing seedling seed orchards.

Paper presented at the Conifer Tree Seed in
the Inland Mountain West Symposium, Missoula,
MT, August 5-6, 1985. Paper presented by
Dr. G. Edwards, Canadian Forestry Service,
Pacific Forest Research Centre, Victoria,
BC.

til

150-
125
100
75
50

O 25

120

90

60

30

0 100 200

RATE (kgN/ha)

TREE HEIGHT (m) -o . 2 — ■— 2.5

•-•▼--■3.5 - - 4

300

■•— 3

Figure 1. — Relation between a) female and b)
male cone production in black spruce (1982) and
tree height and the amount of ammonium nitrate
fertilizer applied in 1981 (May - June).

Ronald F. Smith is a Tree Improvement Forester,
Canadian Forestry Service, Maritimes Forest
Research Centre, Fredericton, NB.

187

Tree Spacing

Tree spacing affects the capacity of seed orchard
trees to respond to fertilizer treatments in that
trees growing at wider spacings have fuller
crowns with more potential flowering sites
(greater numbers of buds). Fertilizing did not
increase the number of buds, but rather the pro-
portion of buds which developed reproducti vely
(Smith 1985) .

The degree to which spacing will limit response
to fertilizing differs between species. In red
pine, ( Pinus resinosa Ait.) female cones can be
borne on a large portion of the live crown. The
number of cones per tree may increase with
increased spacing until the total number of cones
per unit area decreases due to the reduced number
of trees (Stiell 1971). In young black spruce
however, female cones were generally borne in
only the top three or four whorls. The number of
potential flowering sites in the upper crown
limited the degree to which female cone produc-
tion increased in response to wider spacings.
However, trees at wider spacings did produce more
cones of both sexes (fig. 2).

a)

(/>

LU

o

o

Z 250

b)

must be removed before growth and, concomitantly,
seed production within the orchard is affected.
Orchards located on very fertile sites may re-
quire rogueing earlier and to a heavier degree
than might be justified from the family test
measurements alone.

Type of Fertilizer

Nitrogenous fertilizers, particularly those cont-
aining the nitrate (N03) form of nitrogen, e.g.,
ammonium nitrate and potassium nitrate, have been
most successful in stimulating cone crops
(Puritch 1977). Applying ammonium nitrate in-
creased both male and female cone production in
black spruce whereas applying urea did not (Smith
1983) .

Rate of Application

The rate at which fertilizer is applied deter-
mines whether or not flowering will be affected,
and if so, positively or negatively. Fertilizing
black spruce at rates greater than 300 kg N/ha
decreased both female and male cone production,
but especially the latter (fig. 3). Both the
amount of fertilizer required per tree to elicit
the maximum cone production response, and the
rate at which overf ertilization occurs, increases
with tree size (Ebell 1972a).

120

CO

a.

o

600

-_,._1981

SPACING (m)

— 1982 --- 0---1983

Figure 2. — Effects of spacing on a) female and
b) male cone production in black spruce.

RATE (kg N/ha)

Figure 3. — Relation between rate of ammonium
nitrate fertilizer application and male cone
production in 1982 for black spruce trees ferti-
lized in 1980 (from Smith 1983).

Seedling seed orchards are often planted at close
spacing, e.g., in New Brunswick, 1 X 2 m (3 X
6 ft) or 1.5X 2m (1.5 X 6 ft). Rogueing removes
genetically 'inferior' families (trees) thus
allowing the remaining trees to develop full,
vigorous crowns capable of supporting large cone
crops. However, growth rate of the trees affects
the time and intensity of the rogueing, as trees

Method of Application

The amount of fertilizer applied is usually
expressed as the total amount of fertilizer or
nutrient per unit area. However, the effective-
ness of fertilization in stimulating cone produc-

188

tion is determined by how much of the applied
fertilizer each tree receives. Fertilizer should
be placed under the crown dripline because the
bulk of a tree's root system lies within this
area. Band-type spreaders are better than
cyclone types because there is better control of
fertilizer placement.

Broadcast applications are more effective for
large than for small trees as their root systems
occupy more of the site. If a broadcast method is
used, more fertilizer per hectare may be needed
to elicit a flowering response, i.e., the amount
of fertilizer lost to uptake by other vegetation
will be more than if the same total amount of
fertilizer was applied, but placed in a band
around each tree. When accessibility within the
orchard precludes using broadcast methods, hand
applications are necessary. Hand applications
allow for rates to be increased for large, and
decreased for small trees, respectively, which
should optimize the flowering response.

Timing

The time that trees are fertilized will determine
whether vegetative growth, reproductive growth,
or both, will be affected. Fertilizer must be
applied before the reproductive buds for the next
year are formed (primordia differentiation). In
Douglas-fir, April and May fertilizer applica-
tions significantly increased cone production
whereas June applications did not (Ebell 1972b),
because primordia differentiation in Douglas-fir
occurs in late May to June (Owens 1969). Owens
and Molder (1979) give growth and development
patterns for the major North American genera.
These general guidelines could be used by orchard
managers to determine approximately when diffei —
entiation occurs, enabling them to establish
approximate ' not-later-than' dates for fertilizer
applications when the exact phenology for their
species and orchard sites is not known.

In black spruce, applying fertilizers up to two
months prior to the approximate time of primordia
differentiation still significantly increased
cone production the following year (Smith 1983).
In southern pine orchards, fertilizer can be
applied as much as one month earlier than the
'recommended' times for female cone induction
(Schmidtling 1983) conferring the advantage of
coordinating fertilizer applications to periods
of reduced workloads.

Effects of Weather

Response to fertilizing is usually best when the
weather conditions during the year of application
are conducive to the initiation of strobili,
e.g., the response year is a 'good' cone crop
year. In black spruce, trees did not respond to
fertilizing as well in naturally 'poor* flowering
years as in 'good' years; only the larger trees
showed an increase in cone production in these
'poor' years.

Weather Influences the rate of fertilizer uptake
and the total amount of nutrient which actually
becomes available to the tree roots. Heavy rain-
fall immediately after applying fertilizer can
cause a high proportion of the fertilizer to be
leached through the soil. When this occurs, the
cone- induct ion effect of the applied fertilizer
can be lost. The cone-induction effect of ferti-
lizing can also be lost under warm, dry condi-
tions, i.e., fertilizer will deliquesce more
slowly, perhaps remaining In the soil longer.
When this occurs it also takes longer for the
fertilizer to reach the root surfaces. If ferti-
lizer Is applied close to the time of primordia
aif ferentiation, the proportion of the fertilizer
which actually reaches the roots in time to
affect cone production could be reduced, thus
reducing its effectiveness in stimulating flower-
ing. When this occurs, the rate at which ferti-
lizer should be applied to maximize cone produc-
tion must be increased (Ebell 1972b).

Genetic Effects

Probably the most important factor influencing
cone production in seed orchards is the genetic
predisposition of the trees to flower. Differ-
ences between clones account for a large propor-
tion of the total variation in flowering in seed
orchards (Sprague and others 1979; Schmidtling
1983). Consequently, genetic effects can mask the
effects of fertilizer treatments. Adding nitrogen
can increase the numbers of flowers produced, but
will generally not induce those trees to flower
which were not 'predisposed' to do so (Robinson
1979; Barnes and Bengston 1968).

Genotype-fertilizer Interactions at the family
level have been reported for conifers (Jahromi
and others 1976; Maliondo and Krause 1985). The
effectiveness of fertilizer treatments in induc-
ing cone production for different families is not
well documented. The strong fertilizer-clone
interactions in reponse to applying nitrogen
fertilizers (Robinson 1979; Sweet and Krugman
1977) would indicate that family differences can
be expected. It is, however, Impractical on an
operational scale to formulate fertilizer
recommendations on a family basis. Consequently
the fertilizer and rate which is effective for
'most' families should be used.

The only means of managing the genetic component
within seedling seed orchards is through select-
ing which families or individuals are to be
removed. Fecund individuals within the best fam-
ilies can be favored over nonbearing trees given
that these trees also exhibit good growth charac-
teristics. Similarly, trees which respond well to
cone-induction treatments may be favored over
those which do not respond well. Flowering
records, by family, should be kept to evaluate
accurately fecund families and trees within the
orchard .

189

CONCLUSIONS

The following practices should be followed if the
maximum benefit from applying nitrogen fertiliz-
ers to stimulate cone production in black spruce
seedling seed orchards is to be realized.

1 . Orchard rogueing must be scheduled such that
tree spacing never limits crown development and,
concomitantly, the capacity of the trees to res-
pond to cone-induction treatments.

2. The type of fertilizer used, and the rate,
method, and timing of application must be consid-
ered in planning fertilizer schedules.

3. Records of flowering in the orchard should be
kept so that the final rogueings favor not only
the best families (based on family test measure-
ments), but also the more fecund trees within
these best familes.

REFERENCES

Barnes, R. L. ; Bengston , G. W. Effect of ferti-
lization, irrigation and cover cropping on
flowering and on N and soluble sugar composi-
tion of slash pine. For. Sci . 1 72-180;
1 969.

Ebell, L. F. Cone-induction response of Douglas-
fir to form of nitrogen fertilizer and time of
treatment. Can. J. For. Res. 2: 317-326;
1 972a

Ebell, L. F. Cone-production and stem-growth

response of Douglas-fir to rate and frequency
of N fertilizer. Can. J. For. Res. 2: 327-338;
1972b.

Jahromi, S.T.; Goddard, R.E.; Smith, W.H. Geno-
type x fertilizer interactions in slash pine:
Growth and nutrient relations. For. Sci.
22: 211-219; 1976.

Maliondo, S.M.; Krause, H.H. Genotype and soil
fertility interaction in the growth of black
spruce progeny from a central New Brunswick
population. Can. J. For. Res. 15: 410-416;
1985.

Owens, J. N. The relative importance of initia-
tion and early development on cone production
in Douglas-fir. Can. J. Bot. H7: 1039-1019;
1969.

Owens, J. N . ; Molder, M. Times and patterns of
cone differentiation in western North American
conifers. In: Bonner, F . , ed. Flowering and
seed development in trees symposium: proceed-
ings; 1978 May 15-18; Mississippi State
Univ.; 1979: 25~32.

Puritch, G. S. Cone production in conifers. Can.
For. Serv . , Pac . For. Res. Cent., Inf. Rep.
BC-X-65; 1977. 56 p .

Robinson, J. F. Response to nitrate and ammonium
fertilizers-flowers, cones and seed in a lob-
lolly pine seed orchard. In: Proc. 15th South.
Tree Improv. Conf. 1979; 78-85.

Schmidtling, R. C. Timing of fertilizer applica-
tion important for management of southern pine
seed orchards. South. J. Appl . For. 7: 76-81;
1 983.

Smith, R. F. Effects of fertilizers and spacing
of trees on cone production in young black
spruce ( Picea mar iana (Mill.) B.S.P.) planta-
tions. MSc Thesis, Univ. Wisconsin, Madison;
1983. 101 p.

Smith, R.F. Data on file, Canadian Forestry Ser-
vice, Maritimes Forest Research Centre,
Fredericton, New Brunswick; 1985.

Sprague, J.; Jett, J. B. ; Zobel , B. The manage-
ment of southern pine seed orchards to
increase seed production. In: Bonner, F., ed.
Flowering and seed development in trees sym-
posium: proceedings; 1978 May 15-18;
Mississippi Sta . Univ.; 1979; 145-162.

Stiell, W. M. Comparative cone production in
young red pine planted at different spacings.
Can. For. Serv., Petawawa For. Exp. Sta.,
Publ. No. 1306; 1971 . 8 p.

Sweet, G. B. ; Hong, S. 0. The role of nitrogen in
relation to cone production in Pinus radiata.
New Zeal. J. For. Sci., 8: 225-238; 1978.

Sweet, G. B., Krugman, S. L. Flowering and seed
production problems-and a new concept of seed
orchards. In: Proc. 3rd World Consultation
Forest Tree Breeding, Canberra, Australia,
March 21-26; 1977; 7^9-759.

190

0 VARIATION IN CONE AND SEED PRODUCTION FROM NATURAL STANDS,
PLANTATIONS, AND CLONAL-BANKS OF LODGEPOLE PINE,

C. j^Yir

L1

ABSTRACT: Information reported in this paper is
based on: 1) 780 parent trees from 53 provenances
of the inland form of lodgepole pine (Pinus contor-
ta ssp. latifolia) ; 2) a wind-pollinated progeny
plantation of 778 families derived from (1),
planted in 1973, and in which conelet production
was recorded in 1979, 1980, and 1981 (age 10,
11, and 12 from seed); and 3) 160 clones (800
grafts) in two clone banks, which were planted
in 1972, and in which conelet production was
recorded annually since establishment. Both (2)
and (3) are located at Red Rock, south of
Prince George, British Columbia.

The results suggest that:

L« a managed clonal seed orchard can
produce commercial quantities of seeds (over 200
cones per graft) in 10 years;

2. a clonal orchard may produce better
quality seeds than natural stands (grafts
produced 44% more seeds per cone and seeds 48%
heavier than did natural stands);

3. large clonal variation in both seed- and
pollen-cone production capability will not only
affect the genetic composition of seed orchard
seeds, but also the balance between maximizing
genetic gain and maintaining seed production
capacity at the time of seed orchard roguing;

4. differential cone production capability
among provenances is inherited rather than
environmentally induced, and that provenances of
northern latitude origin are more precocious but
less prolific than the southern latitude ones;

5. the current strategy of concentrating
seed orchards for the central Interior at Red
Rock and for the southern Interior at Vernon
appears to be appropriate biologically and
economically, in view of the strong geographic
trend among provenances in cone production
capability, particularly of pollen cones;

6. seed orchards for northern British
Columbia and the Yukon probably should not be
located south of latitude 56°N in view of their
poor growth at many sites in the central and
southern Interior British Columbia.

Paper presented at the Conifer Tree Seed in the
Inland Mountain West Symposium, Missoula, MT,
August 5-6, 1985.

Cheng C. Ying is Provenance Forester, Research
Branch, and Keith Illingworth is Manager, Tree
Improvement Section and Research Stations,
British Columbia Ministry of Forests, Victoria,
AB.

INTRODUCTION

This paper reports geographic and year to year
variation in conelet production and seed yield
of lodgepole pine in natural stands, in a
wind-pollinated (w.p.) provenance-progeny test
plantation, and in two clone banks, emphasizing
their practical implications for seed orchard
planning.

Lodgepole pine has become a major planting
species in the Pacific Northwest. In the
interior region of British Columbia alone,
planting of lodgepole pine reached 15 million
seedlings in 1983 (i.e., 20% of the total
seedlings planted), and is projected to reach 73
million seedlings (40% of the Interior program)
at the turn of the century. The goal of
planting only genetically improved trees by the
year 2000 (B.C. Ministry of Forests 1980) has
led to the rapid expansion of a tree improvement
and seed orchard program in the British Columbia
Interior. Seed orchards are the most expensive
phase of a tree improvement program and their
efficiency determines, to a large extent, the
cost-benefit ratio of such a program.
Information reported in this paper provides a
basis for improving the efficiency of lodgepole
pine seed orchard planning in terms of their
location, size, and projected seed yields.

SOURCE MATERIAL AND DATA COLLECTION

Cone production and seed yield data were
gathered from three sources:

La about 780 parent trees from 53
provenances of the inland form of lodgepole pine
( Pinus contor ta ssp. latifolia) (Critchfield
1980). The majority of the provenances
comprised 15 parent trees. The number of seeds
per cone and 1000-seed weight were determined
for each parent tree;

2. a wind-pollinated progeny plantation of
778 families derived from (1). The plantation
was established in 1973 with 2+1+1 stock, and
was laid out according to a compact-family
(split plot) design (provenance designated as
main plot and family the sub-plot with three
replications of 6-tree family plot). Spacing
was 3.7 x 3.7 m. Seed- and pollen-cone
production were recorded by counting the number
of conelets and pollen clusters in the spring of
1979, 1980, and 1981 (age 10, 11, and 12 from
seed) as follows :

191

Number of

Year Prov. W.P. fmlies Trees

1979 53 778 8888 (in 2 reps)

1980 53 778 8835 (in 2 reps)

1981 10 14*3 828 (in 1 rep)

Various numbers of pollen clusters were sampled
from many flowering trees to determine the
average number of microstrobili per pollen
cluster for each family and provenance.

3. one hundred and sixty clones (800
grafts) in two clone banks, named SCA and SK.
Grafting was done in 1971 and 1972, and both
clone banks were established in the fall of 1972
with five ramets per clone in a single row,
spaced 3.7 x 3.7 m. The two clone banks were
planted side by side. SCA contains clones from
the north, and SK from the south, of latitude
56°N approximately (fig. 1), and

335 of the 375 grafts in SCA (75 clones) and 337
of the 425 grafts (83 clones) were living in
1985. Seed- and pollen-cone production were
recorded every year since planting, and in 1981
seed yield and seed weight were determined for a
sample of 20 clones representing their
geographic distribution (Ying and others 1985b).

Both the w.p. progeny plantation and the clone
banks are located at Red Rock (lat. 53° 46'N,
long. 122° 42'W, elev. 580 m) , south of Prince
George. The climate at both sites is similar,
but the soils are very different: the clone
banks are located on a deep, freely drained
silty sand, the progeny plantation is on a stony
glaciof luvial terrace.

Geographic locations (fig. 1) of parent trees in
(1) and (2) and ortets of 3) are summarized
below:

Source Lat.(N) Long.(W) Elev. (m)

(1 & 2) 49o04,-63o18' 114°25' -136° 18' 455-1815
(3) 49°18' -63°22' 120°10' -136°17' 250-1478

Data were subjected to variance analyses, and
correlation and multiple regression were
employed to elucidate the relationship of cone
production at different ages and with the
geographic origin of provenances.

SEED YIELD AND SEED WEIGHT

Both seed yield (i.e., number of seeds per cone)
and seed weight showed large variation from

Figure 1. — Geographic origin of the 53 provenances and the ortets
in SK and SCA clone banks.

192

Table 1. — Comparison in number of seeds per cone and 1000-seed weight between
natural and grafted lodgepole pine

Geographic
region

No. seeds per cone

Natura 1
s tands

Grafts

1000-seed weight (mg)

Na tura 1
s tands

Grafts

Yukon and Northern
B.C.

Southeastern B.C.
and Alberta foothills
Central and Southern
B.C.

Coast-interior
t rans i t ion

22
12
16
11

24

22

2.6
3. 2
2.8
2.5

4. 1

3.8

Mean 16 23 2.8 4.0

Range (provenance means) 9-31 2.2-3.9

Range (tree or clone means) 1-64 5-39 1.4-4.6 3.3-4.9

provenance to provenance, tree to tree, or clone
to clone (table 1). Variance analyses indicated
40% among- versus 60% within-stand variation.
Grafts produced 44% more seeds per cone and
these were 48% heavier than trees in natural
stands (table 1).

Significant and positive correlations between
number of seeds per cone and latitude and
longitude (table 2) suggest a northwest-
southeast geographic trend: trees from the
Yukon Territory and northern British Columbia
produced the highest number of seeds per cone,
and those from Alberta, southeastern British
Columbia, and near the coast-interior transition
zone were among the lowest (table 1). An
opposite geographic trend was evident in seed
weight (table 2) : trees from Alberta produced
the heaviest seed and those from the Yukon and
near the coast-interior transition, the lightest
(table 1). However, the range of variation in
seed weight was rather small (table 1). This
confirms the conclusions of Birot (1978) and
Critchfield (1980) that variation in seed weight
is largely a character of the subspecies.
Latitudinal and longitudinal trends were not as
obvious in the clone banks probably due to the
small number of clones (20) studied and the
large within-provenance variation (Ying and
others 1985b) . Results from both the clone
banks and natural stands indicated poor seed
yield of high-elevation trees.

Table 2.-- Correlation of number of seeds per

cone and 1000-seed weight of natural
lodgepole pine trees with latitude,
longitude and elevation of their
origin (based on provenance means,
n = 53)

No. seed per cone 1000-seed weight

Latitude .66 a -.32 b

Longitude .50 a -.60 a

Elevation -. 38 a . 18

a Significant at 0.01 level.
b Significant at 0.05 level;

SEED- AND POLLEN-CONE PRODUCTION
Grafts

Seed- and pollen-cone production were recorded
in the clone banks over a 12-year period
(fig. 2); both seed- and pollen-cone production
increased steadily after planting except for the
unexpected decline in 1983. Most grafts
produced a commercial quantity of cones (over
200 conelets per graft) 10 years after
planting. Pollen-cone production, although
about 5 years behind initially, reached the
level of commercial quantities at about the same
age as seed-cone production, and therefore would
not affect the commercial production of a clonal
seed orchard (fig. 2) .

Grafts in the SCA clone bank (northern latitude
origin) started to produce both seed and pollen
cones earlier, but were eventually out-produced
by those in the SK clone bank (southern origins)
(fig. 3).

Seedl ings

Seed- and pollen-cone production of the 10
provenances (148 families) counted in all 3
years (1979-81) are presented in table 3.
Pollen-cone production showed a sharp increase
from 1979 (age 10) to 1980 (age 11), and the
increase from 1980 to 1981, though not as
spectacular, was still substantial, with the
exception of #31 from the Yukon. Average number
of pollen clusters per living tree in 1981 was
25 times that in 1979 (table 3). Compared with
pollen-cone production, changes in seed-cone
production from 1979 to 1981 were relatively
small: most provenances showed a moderate
increase from 1979 to 1980, but decreased from
1980 to 1981 (table 3). The range of provenance
difference in pollen-cone production was much
larger than that for seed cones. For example,
in 1981 the most prolific provenance produced
almost 10 times more pollen cones than did the
least prolific one, excluding the atypical
provenance #31, but the ratio of difference

193

500-

400-

■ 300-

r 200-

100-

| 1 NORTHERN

1 1 CLONES

SEED CONE

SOUTHERN
CLONES

POLLEN CONE

10 II 12 7 8

NUMBER OF YEARS AFTER ESTABLISHMENT

Figure 3. — Comparison in seed- and pollen-cone production
between clones of northern latitude (SCA) and southern
latitude (SK) origin.

between the most and the least productive
provenances was only about 2 to 1 for seed-cone
production (table 3). The change of the pollen
cone - seed cone ratio from 0.9 at age 10 to
10.5 at age 12 (table 3) also reflected the
differential pollen- and seed-cone production.
Mean number of microstrobili per pollen cluster
varied very little among provenances or from
year to year (table 3) .

Differences among provenances and among families
within provenances were all statistically
significant, but over 60% of the variation was
associated with the tree-to-tree differences
within w.p. families.

Grafts Versus Seedlings

Comparison in conelet production between the
grafts in the clone banks and the seedlings in

the w.p. progeny plantation is not, strictly
speaking, valid because they were neither from
the same geographic locations, nor planted at
the same site in the same year, receiving the
same cultural treatments. However, a comparison
of mean production at different ages should be
reasonably valid because the numbers of grafts
and seedlings investigated were large and they
were from similar geographic areas (fig. 1).
This comparison is given in table 4. The grafts
produced 5 to 10 times more seed cones than the
seedlings, and showed similar superiority in
pollen-cone production at age 10 and 11. At age
12, however, the seedlings produced as many
pollen cones as the grafts. The percent of
individuals bearing seed cones or pollen cones
was higher in grafts in the clone banks than in
seedlings in the w.p. progeny plantation,
particularly in the earlier ages.

194

Table 3. — Average number of seed cones and pollen clusters, their ratio, and average number of

microstrobilus of the 10 provenances which were counted from 1979 (age 10) to 1981 (age 12),
provenances arranged from north to south

Pollen cluster

Provenances

Seed cone

Pollen cluster

/seed cone Microstrobilus

No.

Loca t ion La t .

Long Elev (m)

10 11 12

10

1 1

12

10

1 1

12 10

1 1

12

31

Frances L. , Yukon 61° 10'

129° 20' 884

18 26 3

0

I

2

0

0. 1

1.2 19

20

17

66

Stone Mt. , B.C. 58° 39'

124° 46' 1173

22 30 18

0

6

23

0

0. 3

1.9 21

22

24

27

Pink Mt. , B.C. 57° 00'

122° 24" 1113

23 30 32

1

16

53

0. 1

0.8

2. 3 29

21

26

60

Mt. Lemory, B.C. 55° 33'

122° 33' 732

13 29 27

2

22

82

0. 3

1.1

7.0 25

30

27

63

Albreda, B.C. 52° 35'

119° 10' 975

14 46 36

11

114

170

1.8

2.7

12.8 31

28

30

17

Oie L. , B.C. 52° 00'

121° 12' 991

10 21 21

6

35

181

1.7

2.9

22.1 27

27

31

14

Wentworth Cr. , B.C. 50° 58'

120° 20' 1059

8 30 16

3

34

111

1.0

1.3

18.0 25

28

29

72

Larch Hills, B.C. 50° 42'

119° LI1 777

6 33 20

7

106

189

2. 1

3.6

19.2 28

22

32

68

Kananaskis, Alta. 51° 01 '

115° 02' 1501

12 26 22

4

55

112

0. 7

2.6

6.8 28

28

27

45

Settlers Rd , B.C. 50° 31'

115° 44' 1036

28 82 39

13

199

254

0.8

3. 1

11.3 27

32

35

Mean

15 35 24

5

61

123

0.9

1.9

10.5 27

28

29

LSD. 05

6 10 8

6

36

39

1.8

1.5

8.2 6

4

6

Table 4 .--Comparison in seed- and

pol len-cone

production

between grafted and

seedling lodgepole pines

Seed cone bearing

Pollen

cone

bear

Lng

Graft Seedlings

Graft

Seed 1 ings

Age %a

Mean0 %a

Mean0

Mean0

%a

Mean0

10 98

113 84

15

68

28

21

5

11 100

237 96

35

89

185

62

61

12 99

128 90

24

80

103

78

123

a X of grafts or

seedlings bearing seed or pollen cone.

D Mean number of

conelet or pollen cluster.

Table 5. — Correlation coefficients of mean

number of

seed

cones and

pol len

clusters, and pollen

cluster - seed cone ratio

with

the latitude,

longitude, and elevation of the

jrovenance

origin

Seed cone

Pol len

cluster

Rat io

Origin/age

10 11 12

10

L L

12

10

I I

12

Latitude

.40 -.19 -.50 -.12

.37

-.89

04

-.03

-.78

Longitude

.23 -.24 -.67 -.32

.51

-.80

10

-. 29

-.46

Elevation

-.21 -.19 .09 -.06

. 12

-.07

10

-. 33

-.26

Notes: 1. Correlations were calculated from provenance means.

2. Age 10 and 11 data are based on 53 provenances; age 12 data
are based on 10 provenances.

3. Correlation coefficients significant at 0.05 level are underlined.

Geographic Trend

Correlations of cone production with latitude
and longitude of the provenances were relatively
high in the w.p. progeny test (table 5). The
correlation coefficient increased with the age
of the trees for pollen-cone production, but

changed from positive to negative for seed-cone
production. This indicates that provenances
from the northwest (i.e., the Yukon) were more
precocious but less prolific than those from
southeastern British Columbia (i.e., East
Kootenay). A similar trend was also evident
with the grafts (fig. 3) .

195

Table 6. — Regression analyses with seed cone, pollen cluster, and pollen
cluster - seed cone ratio as dependent variables and latitude,
longitude and elevation of the provenance origin as independent
variables using maximum method

Number of
variables
in model

Intercept

Latitude

Longitude

Elevat ion

One
Two
Three

226.6
300.9
466.1

1.64N
3.30N

Seed cone
-1.683
-3.02N
-4.92a

■0.03N

.45
.52
.68

One
Two
Three

1130.9
1242.5
1976. 7

•18. 72b
•19.35b
•10.13N

Pollen cluster

-9.75N

-0.07N
-0. 13N

79
.83
86

One
Two
Three

94.6
-69.3
-63. 1

Pollen cluster
-1.56b

seed cone ratio

■3.97b
3.90°

2.43b
2. 34b

•0.00

,61
95
95

N The regression coefficient
a The regression coefficient
b The regression coefficient

not significant,
significant at 0.05 level,
significant at 0.01 level.

Table 7. — Age/age correlation (r-value) in seed- and pollen-cone production,
and their ratio (pollen cluster:seed cone) estimated by covariance
analyses based on the 10 provenances counted from age 10 (1979) to
12 (1981) in w.p. progeny plantation

Seed cone Pollen cone Ratio

A*

>e pairs

A

ge pairs

A

ge pairs

Source

10/11

10/12

11/12

10/11

10/12

11/12

10/11

10/12 11/ i:

Provenance

.62

.39

.66

.95

.92

.87

.85

.88 .74

Families/Prov.

.40

.48

.49

.68

.41

.62

. 34

.32 .20

Trees/Fam. /Prov.

.57

.46

.56

.56

.30

.53

.31

.17 .20

Regression analyses (table 6) of w.p. progenies
revealed that latitude, longitude, and elevation
of the provenances together accounted for 68,
86, and 95% of the observed variation for
seed-cone production, pollen-cone production,
and their ratio respectively, but the effect of
elevation was not significant in any case.
Geographic trends in seed- and pollen-cone
production of grafts in the clone banks were not
as obvious (Ying and others 1985b). This can be
expected in view of the large within-provenance
variation mentioned before.

Age/Age Correlation

High age/age correlation in cone production was
observed for both seedlings (table 7) and grafts
(table 8). This was true for provenance,
family, and individual trees (table 7) . These
results suggest that a few precocious and
prolific clones or seedlings could contribute a
large proportion to the genetic composition of
the seeds, particularly those harvested from the

early years of a seed orchard (Jonsson and others
1976; O'Reilly and others 1982; Todhunter and Polk
1982) .

Table 8. — Age/age correlation (r-value) in seed-
cone and pollen-cone production based
on clonal means. All correlation
coefficients are significant at 0.05
leve 1 .

Year

(age )

1980

1981 1982

1982

Seed cone

1979

(8)

.78

.48 .44

.33

1980

(9)

.71 .66

.52

1981

(10)

.68

.52

1982

(11)

.50

Pollen cone

1979

(8)

.46

.81 .41

.43

1980

(9)

.67 .51

.57

1981

(10)

.56

.54

1982

(11)

.80

196

Cone Production and Crowth

Correlation between cone production and height
growth were all positive at the level of
provenance, family, and individual tree
(table 9), indicating that tall trees were also
heavy cone producers. In an earlier study, Lee
found an invariable positive correlation between
cone production and various growth
characteristics in these two clone banks.
There is no indication of negative effects of
flowering on vegetative growth in either grafts
or seedlings of lodgepole pine (Nilsson 1981).

Table 9. — Correlation of total height at age 13
with seed-cone and pollen-cone
production at age 12 in the w.p.
progeny plantation, using covariance
analysis technique

Correlation between height and

Source

Seed cone

Pollen cone

Provenances

.61

.86

Fami 1 ies/ P

.11

.35

Trees/F/P

. 15

. 27

DISCUSSION

AND IMPLICATIONS

The superiority of grafts in seed yield and seed
weight (table 1) suggests that seeds from a
clonal seed orchard will be of better quality
than the seeds from natural stands, and thus
should produce better quality seedlings and
improve the nursery recovery factor (the ratio
of number of plantable seedlings over number of
seeds sown). The nursery recovery factor for
seed from clonal seed orchards is predicted to
be 0.6 as compared to 0.4 for seed from natural
stands (Hewson and others 1984) . The low seed-
cone production by the seedlings (table 4)
suggests that clonal orchards would be superior to
seedling orchards in terms of seed yield.
However, the lack of pollen-cone production by
the grafts during the first 7-8 years (fig. 2)
can hinder the rapid advance of a breeding
program. Field observation indicated more
balanced seed- and pollen-cone production by the
seedlings at the early ages. Moreover, the
response of grafts to cone-enhancement treatment
in lodgepole pine has not been encouraging
(Wheeler and others 1982).

It is interesting and useful to compare the
seed- and pollen-cone production in the w.p.
progeny plantation at Red Rock with that of a
similar trial at Ange, Sweden (lat. 62° 32'N,
long. 15°42'W, elev. 200 m) , which contained
provenances from similar geographic origins
(Nilsson 1983):

( year )

Seed cone

Pol len

cone

Red Rock

An .: •

Red Rock Ange

Red Rock

Ang

L0( 1979)

15(1978)

13 8

5

26

11(1980)

16(1979)

25 10

61

35

12( 1981)

17(1980)

24 13

123

45

Although the Ange plantation was 4 years older,
the amount of cone production was only about
half of that at Red Rock. Apparently,
plantation location has a major impact on cone
production.

The pattern of geographic variation observed at
Red Rock was also very different from that at
Ange: positive correlations between seed-cone
production and latitude of the provenance
origins were found at Ange (Nilsson 1981), but
at Red Rock this correlation was positive only
at age 10 (table 5). In other words, at Ange
more northern provenances showed more abundant
seed-cone production, and this pattern changed
little from year to year. At Red Rock,
however, this north-south pattern was observed
only in 1979, and in 1981 provenances of
southern latitude out-produced the northern
ones (table 3). No geographic trend in
pollen-cone production was apparent at Ange,
but a declining northwest to southeast trend
was obvious at Red Rock (tables 3, 5, and 6).

Differential adaptation among provenances to
respective test sites is apparently the main
reason for this disparity in the geographic
variation pattern of cone production between
Ange and Red Rock. Our experience indicates
that vigorous vegetative growth is a
prerequisite for abundant cone production in
lodgepole pine (Wheeler and others 1982); high
correlation between growth and cone production
(table 91 suggests that vigorous trees
produced more cones (Nilsson 1981). Trees of
the Yukon and northern British Columbia seed
sources grew poorly in the Red Rock
plantation, and at most sites in Interior
British Columbia, indicating their poor
adaptation (Ying and others 1985a). This
apparently also affected their cone production
capacity. On the other hand, these northern
provenances have been the major sources of
lodgepole pine seeds for reforestation and
were found to grow well at most sites in
Sweden (Lindgren 1983). Poor cone production
of the southern provenances at Ange was
evidently related to their low vigor. These
results suggest that the selection of seed
orchard location should follow the similar
general guidelines used for seed transfer
(Ying and others 1985a, 1985b).

At Red Rock, the prolific cone production of
seedlings and grafts originating from southern
latitudes (fig. 3, table 3) does not seem to
support the common belief that the southward
transfer of seed orchards will benefit the
cone production (Gansel 1973; Werner 1975;
Schmidtling 1983). Until more is known about
the effect of site and cultural treatment on
cone production of lodgepole pine, the current
British Columbia strategy of concentrating
seed orchards for the central Interior at Red
Rock and for the southern Interior at Vernon
appears to be appropriate biologically and
economically. The results also suggest that
the seed orchard for lodgepole pine from
northern British Columbia and the Yukon
probably should not be moved to south of
latitude 56°.

197

Both seed- and pollen-cone production showed
similar patterns of geographic variation, but
the magnitude of differences among provenances
was much larger for pollen than for seed cones
as reflected in their ratio (table 3). Does
this differentiation in sexual ratio among
provenances represent the strategy of parental
resource allocation, reflecting evolutionary
fitness (Doust and Doust 1983)? Or is it
environmentally induced due to geographic
displacement (i.e., long day length conducive
to pollen-cone production as a result of
northern displacement) ( Larson . 1961 ; Giertych
1967)? Daylength is the prevailing factor
affecting cone initiation following geographic
transfer. In our case, the maximum northern
transfer was about 3° of latitude, equivalent
to the maximum increase of about 40 minutes in
daylength occurring in June. It is unlikely
that this relatively small change in daylength
would have such a drastic effect on
pollen-cone initiation (Mirov 1956).

The pollen-ovule ratio plays a central role in
natural selection associated with sexual
reproduction. It is well established that the
pollen-ovule ratio increases with the
increasing rate of outcrossing (Willson 1979;
Doust and Doust 1983); with wind-pollination
species, a large quantity of pollen production
is probably the most effective way to ensure
outcrossing (Willson 1979). That the
differential pollen-seed-cone ratio among
lodgepole provenances is related to their
outcrossing rate appears to be in line with
the results of Yeh and Layton (1979). They
found the geographic trend of decreasing
genetic variability among lodgepole pine
populations extending from the southern
Interior of British Columbia toward the Yukon
Territory. However, it is still not clear why
natural selection should favor high
outcrossing rates in the southern Interior
populations.

Differential seed- and pollen-cone production
capacity among clones and geographic sources
is of practical significance. First, a few
prolific producers (particularly pollen cones)
can affect the genetic composition of the
seeds from seed orchards (O'Reilly and others
1982). Second, large clonal variation in cone
production makes it difficult to achieve the
goal of maximizing genetic gain while
maintaining the cone production capacity at
the time of seed orchard roguing. Third, the
risk of pollen contamination increases if seed
orchards for different geographic regions are
proximally located. In the latter case,
pollen contamination could consequently weaken
the genetic adaptive capability of seed
produced. For example, excessive
contamination of the seed orchard for the
northern and central Interior of British
Columbia by the southern Interior sources can
severely impair winter hardiness of progeny
seed.

Information reported here is useful in guiding
the selection of seed orchard location,
projecting seed yield, and alleviating
potential problems in lodgepole pine seed
orchard management related to pollen
contamination and clonal disparity in cone
production. However, since the clonal banks
and the w.p. progeny plantation are not
planted at the same site and not specifically
designed to address seed orchard management,
questions such as the effect of site and
cultural treatments on cone production and
seed yield of lodgepole pine grafts and
seedlings remain inadequately answered. A
study designed to answer the above questions
is now in progress. Information generated
from this study will further our understanding
about the flowering response of lodgepole pine
grafts and seedlings under different
environments .

REFERENCES

Birot, Y. 1978. Variability geographique du
poids de la graine de Pinus contorta.
Silvae Genet. 27(0:32-40.

British Columbia Ministry of Forests. 1980.
Seed orchard and tree improvement program
for interior region of British Columbia.
Victoria, B.C. 36 p.

Critchfield, W.B. 1980. Genetics of

lodgepole pine. U.S. Dep. Agric. For.
Serv. Res. Pap. WO-37. 57 p.

Doust, J.L. and L.L. Doust. 1983. Parental
strategy: gender and maternity in higher
plants. Bioscience 33:180-185.

Gansel, C.R. 1973. Should slash pine seed
orchard be moved south for early
flowering? In: 12th Southern Forest Tree
Improv. Conf. Proc. 310-316.

Giertych, M.M. 1967. Analogy of the
differences between male and female
strobilus in Pinus to the differences
between long- and short-day plants. Can.
J. Botany. 45-1907:1910.

Hewson, C.A. and J. Konishi. 1984. Seed

orchard planning and development for the
central interior region. B.C. Min. For.
Victoria, B.C. 31 p.

Jonsson, A., I. Ekberg , and G. Eriksson.
1976. Flowering in a seed orchard of
Pinus s ilves tris L. Studia Forestalia
Suecica No. 135. Swedish College of
Forestry, Dep. For. Genetics. Uppsala,
Sweden. 38 p.

Larson, P.R. 1961. Influence of date of

flushing on flowering in P inus banksiana.
Nature. 192:82-83.

198

Lee, C. E. Vegetative and reproductive growth in
lodgepole pine. Vancouver, BC: University of
British Columbia. Undergraduate thesis.

Lindgren, K. 1983. Provenances of Pinus
contor ta in northern Sweden. Swedish
Univ. of Agricultural Sciences, Dep. of
For. Genetics and Plant Physiol. Umea,
Sweden. 91 p.

Mirov, N.T. 1956. Photoperiod and flowering
of pines. For. Sci. 2:328-332.

Nilsson, J-E. 1981. Flowering in Pinus

contor ta : a study of flowering abundance
and flowering phenology of provenances
from latitude 48°N to 63°N in a Swedish
field trial. Rep. No. 2. Swedish Univ.
of Agricultural Sciences, Dep. For. Gen.
and Plant Physiol. Umea, Sweden: 128 p.

O'Reilly, C. , W.H. Parker, and J.E. Banker.
1982. Effect of pollination period and
strobili number on random mating in a
clonal seed orchard of Picea mar iana.
Silvae Genet. 31 :90-94.

Schmidtling, R.C. 1983. Geographic location
affect flowering of loblolly pine.
In: 17th Southern Forest Tree Improv. Conf.
Proc. pp. 138-143.

Todhunter, M.N. and R.B. Polk. 1982. Seed

and cone production in a clonal orchard of
jack pine (_P. banks iana) . Can. J. For.
Res. 12:512-516.

Werner, M. 1975. Location, establishment and
management of seed orchards. In: Seed
orchards: Forest Commission Bull. 54.
C.R. Faulkner (editor). Her Majesty's
Stationery Office, London, pp. 49-57.

Wheeler, N.C. , C. C. Ying, and J.C. Murphy.
1982. Effect of accelerating growth on
flowering in lodgepole pine seedlings and
grafts. Can. J. For. Res. 12:533-537.

Willson, M.F. 1979. Sexual selection in
plants. The American Naturalist
113:777-790.

Yeh, F.C. and C. Layton. 1979. The

organization of genetic variability in
central and marginal populations of
lodgepole pine Pinus contorta spp.
Latifolia. Can. J. Genet. Cytol.
24:487-503.

Ying, C.C. , K. Illingworth, and M. Carlson.
1985a. Geographic variation in lodgepole
pine and its implications for tree
improvement in British Columbia. In: The
Lodgepole Pine Symp. : the species and its
management. Proc, May 8-10, 1984;
Spokane, Wash, and May 14-16, Vancouver,
B.C., pp. 45-53.

Ying, C.C, J.C. Murphy, and S. Anderson.

1985b. Cone production and seed yield of
lodgepole pine grafts. For. Chron.
61:223-228.

199

Section 4. Effects of Biological Factors On Seed
Production

200

INFLUENCE OF VERTEBRATES ON CONIFER SEED PRODUCTION

y/

Curtis H. Halvorson

ABSTRACT : Many birds and mammals utilize
conifer tree seed but few have feeding
strategies or physical capabilities that allow
for cone predation prior to seed dispersal.

The American red squirrel is the primary and
most effective vertebrate cone predator,
although chipmunks also forage in trees for
food. Grizzly and black bears use conifer
cones in limited fashion. Clark's nutcrackers
can have a major impact on cone crops of
large-seeded conifers. Jays, crossbills,
siskins, and finches are other common bird
predators but for most birds our knowledge of
their impact is more anecdotal than measured.
Low densities and erratic movements suggest
most birds can only exert sporadic local
pressure on cone crops but could be damaging to
high-value seed orchards and production areas.

Red squirrel feeding on ponderosa pine and
Douglas-fir cones begins in June and continues
through summer, but clipping for storage is
timed to seed maturation and starts in early
August. Squirrel feeding removed 17 to 100
percent of the annual crop on an island, but a
consistent correlation between crop removal and
crop production was not found. Seedfall after
squirrels' harvest ranged from 15,000 to
172,000 sound seed per acre in good years.
Cone clipping by squirrels is often accompanied
by loss of branch and terminal buds and
first-year conelets. Vegetative parts of
conifers may be used when crops fail.

Two highly adapted seed predators, the Clark's
nutcracker and the pinyon jay, return some
value by being effective seed predators. The
cone caches of the most effective predator, the
red squirrel, serve as a source of high quality
seed for nurseries. The impact of most birds
cannot be fully determined from the present,
mostly descriptive, literature. And the red
squirrel's inconsistent harvest of cone crops
appears to be more influenced by weather, seed
quality, and the availability of other food
rather than simply by cone abundance or
squirrel population density.

Paper presented at the Conifer Tree Seed in
the Inland Mountain West Symposium, Missoula,
MT, August 5-6, 1985.

Curtis H. Halvorson is Wildlife Biologist/
Technical Editor, Division of Wildlife
Research, Office of Information Transfer,
U.S. Fish and Wildlife Service, USDI , Fort
Collins, CO.

INTRODUCTION

Many vertebrates are specialized predators on
conifer seeds. Some are more dependent than
others. Red squirrels (Tamiasciurus
hudsonicus) cache cones in fall, prior to seed
dispersal, so that they are assured of a
dependable food supply through long boreal
winters. The Clark's nutcracker (Nucif raga
columbiana) gathers limber pine (Pinus
f lexilis) , whitebark pine (P. albicaulis, and
pinyon (P. edulis) seeds into throat pouches
during the fall harvest and transports these as
far as 14 mi (22 km) (Tomback 1977; Vander Wall
and Balda 1977; Lanner and Vander Wall 1980) to
ground caches. These become the winter and
spring food supply for adult and juvenile
nutcrackers. Vertebrate interest in conifers
thus produces seed loss if the animal's role is
primarily that of a predator, or seed
distribution and tree regeneration if the role
is primarily that of a disperser. These roles
have been studied and will be included in my
discussion of the influence of vertebrates on
seed production.

Man's interest in seed includes an economic
dependency that seeks to enhance seed
production to produce wood commodities. We are
interested in seed losses and have catalogued
vertebrate influences and their control in the
past. By 1915, according to Pank's (1974)
bibliography on animals and tree regeneration,
there were eight government and forestry
publications discussing reforestation and
vertebrate agent problems. Several hundred
reports and numerous conferences have followed
(Smith and Aldous 1947; Tevis 1953; Radwan
1963; Black 1969, 1974; Janzen 1971; Pank
1974). Most focus on post-dispersal seed loss
and subsequent stand establishment, damage, and
protection. The usual scenario is: cones open
releasing seeds, which are gathered and
consumed by rodents.

I will confine my review to those animals that
reach the seed before it leaves the tree. This
is predispersal impact and is far less
quantified than that for shed seed. Those
animals that have specialized adaptations for
harvesting predispersed seed are primary
influences. There are secondary influences on
seed production — bears stripping cambium,
porcupines (Erethizon dorsatum) girdling
conifers — but I will confine descriptions to
the primary agents. I will describe these
adaptations and their effects and implications
on seed reproduction. The birds and mammals
discussed herein (table 1) are mostly residents

201

Table 1* — Vertebrates of the Inland Mountain West that use conifer seed prior to seed dispersal (Table modified from Smith and Balda 1979)

Taxonomlc Group

Genera of trees

Method of using seed

BIRDS:

Family:Picidae
Hairy Woodpecker
Picoides villosus

White-headed woodpecker
Picoides albolarvatus

Extracted seed from closed and open

cones are eaten by adults and juveniles

Extracted seed from closed and open

cones are eaten by adults and juveniles

Llgon 1973, Smith and Balda 1979; Stallcup
1968, 1969; Tomback 1977

Curtis 1948; Ligon 1973; Smith and Balda
1979; Tevis 1953; Tomback 1977

Family : Co rvidae

Gray Jay (questionable)
Perisoreus canadensis

Abies , Picea

"May extract seeds from open cones

or retrieve from ground to eat and store
as boll"

Smith and Balda 1979 list without evidence;
see Dow 1965 for feeding test; Turcek
and Kelso 1968 list European jay P.
inf austus

Steller's Jay
Cyanocitta stelleri

Pinyon Jay
Gymnorhinus cyanocephalus

Clark's Nutcracker
Nucif raga Columbiana

Black-billed Magpie
Pica pica

Family: Pa ridae

Black-capped Chickadee
Par us atricapillus

Mountain Chickadee
Parus gambeli

Boreal Chickadee
Parus hudsonicus

Pinus , Pseudotsuga

Abies, Picea , Pseudotsuga
Pinus

Abies , Picea , Pseudotsuga
Pinus

Abies, Picea, Pseudotsuga

Cache seed from open cones to
feed adults

Cache seed extracted from closed

and open cones to feed adults and young

Cache seed extracted from closed

and open cones to feed adults and young

Cache seed from open cones and
ground to feed adults

Cache seeds from open cones and
ground to feed adults

Cache seeds from open cones and
ground to feed adults

Cache seeds from open cones and
ground to feed adults

Benkman et al. 1984; Curtis 1948; Smith
and Balda 1979; Tevis 1953

Ligon 1978; Smith and Balda 1979; Turcek
and Kelso 1968

Benkman et al. 1984; Curtis 1948; Guintoli
and Mewaldt 1978; Lanner and Vanderwall
1980; Smith and Balda 1979; Tomback 1977,
1980, 1981, 1982; Vandervall et al. 1981

Smith and Balda 1979 (Vanderwall and
Balda , unpublished)

Bent 1946; Curtis 1948

Benkman et al. 1984; Haftorn 1974; Smith
and Balda 1979; Stallcup 1968, 1969; Tevis
1953; Tomback 1977

Haftorn 1974; Smith and Balda 1979

Family :Sittidae

Red-breasted Nuthatch
Sltta canadensis

White-breasted Nuthatch
Sitta ca rolinensis

Pygmy Nuthatch
Sitta pygmaea

Family :Fringillidae
House Finch
Carpodacus mexicanus

Red Crossbill, White-
winged Crossbill
Loxia curvirostra, L.
leucoptera

Common Redpoll
Carduelis f lammea

Abies , Picea , Pinus ,
Pseudotsuga , Tsuga

Abies , Picea, Pseudotsuga ,
Pinus

Abies, Larlx , Picea , Pinus
Pseudotsuga , Tsuga

Seeds from open cones cached
and eaten by adults

Uses and makes cache of seeds
and finds seed on ground

Seeds from open cones cached and
eaten by adults

Feeding on opening cones and
seed fallen on ground

Seeds from closed and open cones
fed to adults and young

Feeding on opening cones

Benkman et al. 1984; Curtis 1948; Smith
and Balda 1979; Stallcup 1969; Tevis 1953

Curtis 1948; Smith and Balda 1979;

Stallcup 1969; Tevis 1953; Tomback 1977

Ligon 1973; Smith and Balda 1979; Stallcup
1969

Curtis 1948

Bock and Lepthien 1976; Curtis 1948;
Kemper 1959; Ligon 1973; Newton 1973;
Smith and Balda 1979; Stallcup 1969;
Stickney (1966 pers. comm.); Tomback 1977

Bent 1968; Curtis 1948; Newton 1973

Pine Siskin
Carduelis spinus

MAMMALS

Abies , Larlx, Picea, Pinus
Pseudotsuga , Tsuga

Seeds from open cones and the
ground fed to young adults

Benkman (pers. comm. ); Bent 1968;
Smith and Balda 1979

Order :Rodentia
Red Squirrel
Tamiasciurus hudsonicus

Abies , Larlx , Picea, Pinus ,
Pseudotsuga , Tsuga

Cache and feed from whole closed
cones

Curtis 1948; Duncan 1954; Finley 1969;

Gurnell 1984; Kemp and Keith 1970; Schmidt
and Shearer 1971; Shaw 1936; Smith 1970;
Smith and Balda 1979

Chipmunk
Eutamius

Abies, Larix, Picea, Pinus
Pseudotsuga , Tsuga

Cache and eat seeds from closed
or open cones

Benkman et al. 1984; Broadbrooks 1958;

Schmitz (1966 pers. comm.); Tomback 1977,
1981

Golden-mantled Ground
Squirrel

Spermophllus lateralis

Ord er : Ca mivo ra

Grizzly and Black Bear
Ursus arctos; U. americanus

Pinus albicaulls

Forage for single seeds

Major fall food from squirrel caches
some tree climbing by black bear

Benkman et al. 1984; Smith and Balda 1979;
Tomback 1977, 1981

Kendall 1983; Mealey 1980; Murle 1981;
Tlsch 1961

202

of the geographic area prescribed for this
symposium, the Inland Mountain West, lying
between the east slopes of the Cascades and the
western edge of the mixed-grass prairie in
Alberta and Montana. Important knowledge of
some bird species has come from adjacent areas;
these and pertinent foreign literature are
included where applicable.

Background References

Background information on animals and their use
of tree mast has been reviewed by Janzen (1969,
1971-insects, birds, and mammals), Smith and
Balda ( 1979-compet i tion between animals for
conifer seed), and Smith and Reichmann
(1984-food caching behavior). Boreal forest
birds, their Holarctic distribution and
adaptations to coniferous forests, have been
covered by Udvardy (1969), with emphasis on the
Northern Rocky Mountains. The Corvidae (crows
and jays), have special behavioral adaptations
for food movement and storage, particularly
conifer seeds, and these habits are reviewed by
Turcek and Kelso (1968) for 22 European and
North American species. Tree squirrels are the
only mammals that relate Importantly to tree
mast. Gurnell (1983) has provided a major
review of Palearctic and Nearctic tree squirrel
behaviors and population responses to seed
crops; Heaney (1984) included tree squirrels in
summarizing climatic and latitudinal
relationships to life histories of the
Sciuridae. The nature of red squirrel cone
caching, energy supplies, and some forest
regeneration implications are described by
Finley (1969). The evolutionary history of
squirrels is thoroughly discussed by Black
(1972), Emry and Thorington (1984), and Hafner
(1984); Brodkorb (1971) describes the fossil
record on birds. These reviews are useful
references for readers seeking perspective on
vertebrates and predispersed seed crops.

Some History

Modern conifers were geologically
well-established in the Cenozoic Era about 70
million years ago (Egler (1977). Ancestors of
our present birds and squirrels appeared much
later, in the mid-Miocene, about 18 million
years before present (ybp; Brodkorb 1971, Black
1972, Hafner 1984); by the Pleistocene, modern
forms used conifer seeds much as we see them
now. The adaptations between vertebrates and
conifer seeds in the Inland Mountain West were
already well established genetically when the
late Pleistocene glacial ice sheets excluded
most forests from the Northern Rocky Mountains,
beginning 42000 ybp, and were maintained after
re-establishment of forests as recently as
12000 or fewer years ago in some areas (Heusser
1969) .

The adaptive behaviors of animals of the
montane West stem from the seasonality of the
taiga — the extremes in seasonal temperatures
and moisture. Most birds migrate alti tudinally
at least, or lat i tud i nally at most, in response

to annual food shortages. Invertebrates
disappear, most seeds and berries fall, plants
dry and go dormant. A few, chiefly the corvids
(crows, jays, nutcrackers), store food over
winter whereas nuthatches (Sitta spp.) and
chickadees (Parus spp.) do so only for days or
weeks (Udvardy 1969). Red squirrels cache and
defend their winter supply of cones (Smith
1968; Gurnell 1983) but chipmunks (Eutamias
spp.) take their seed hoards to bed, packing
several thousand at a time under and around
their underground nests (Broadbrooks 1958).
Bears (Ursus a rc tos , U. ameri canus ) also take
their food supplies into hibernation but as
energy already assimilated as fat tissue.
Man's economic concerns about vertebrate use of
seed seldom reflect on the fact that animal use
of seed is an evolutionary phenomenon enabling
species to occur in certain geographic areas.
It is difficult for us to comprehend the pace
of adjustments that occurred during the
coevolution of birds, mammals, and trees.
Animals must adapt to external conditions of
food abundance and scarcity or die, while man
can create and, to large extent control, his
own food supply and living environment.

Interactions: Mutualism and Defense

In recent years there has been renewed interest
in vertebrate-seed relationships that goes
beyond the economic aspects of seed-eating
related to losses In tree regeneration. This
interest has looked at seeds and their
vertebrate predators as biological interactions
between two organisms — the plant seeking
effective mechanisms for reproducing and the
vertebrate using the plant for survival.
Negative interactions as coevolved defensive
mechanisms were initially emphasized in this
"new look" (Smith 1968, 1981; Janzen 1969,
1971; Elliot 1974). Janzen (1971) considered
that trees are not entirely helpless and over
time have envolved mechanisms to defend
themselves. These mechanisms include satiating
the predators and minimizing the feeding
intensity (Janzen 1971). The entire seed crop
could ripen synchronously and overwhelm the
harvesting ability of predators, as Benkman and
others (1984) concluded for southwestern white
pine (Pi nus strobi f ormis) and the red
squirrels. If seed crop abundance occurs only
at long intervals, and predator peaks lag the
seed crop peaks, then predators whose
population build-ups are related to seed crop
abundance have reduced populations and less
predation capability.

Another defense mechanism may be present in
lodgepole pine (Pi nus contorta) . Cone serotiny
and smaller, fewer seeds in the harder
serotinous cones have been theorized as tree
defenses against red squirrels (Elliott 1974;
Smith and Balda 1979). The squirrels, in turn,
develop heavier jaw musculature, according to
Smith (1970, 1981). But if hard nuts and cones
predate the origin of squirrels, a suggestion
by Emry and Thorington (1984), then plants
might first have developed some of their

203

defenses against geologically older factors
such as insects or fire.

Odum recently (1985) advised against population
ecology focusing on the negative inter-
actions— competition and predation — and
predicted more study of positiveness —
cooperation and mutualism. Mutualism requires
that the association between different
organisms increase the fitness of both, not
just one. The seed dispersal activities of
fruit-eating birds and mammals examined by
Krefting and Roe (1949) have been recognized as
mutualism, as have pollinating insect
interactions reviewed by Owen (1980). An
argument has been made by Owen and Wiegert
(1981) for the stimulating effects of grazer
saliva and feeding on grasses. Owen (1980)
further theorizes that removal of plant growing
points, apical shoots and buds, is favorable to
the plant because it induces branching, to the
extent that a plant is better supported,
photosynthetic parts are more favorably
exposed, and seed production is increased by
profuse branching. Although he does not
mention squirrel clipping of terminal buds, it
is an activity I have observed and will allude
to later, as a factor in ponderosa pine (Pinus
ponderosa) branch morphology. A recent paper
by Tomback (1982) portrays this increasing
interest in mutually beneficial coevolved
relationships and provides substantial evidence
that the seed dispersal activities of the
Clark's nutcracker are more effective than that
of rodents and other animal agents (see also
Hutchins and Lanner 1982). The nutcracker
caches whitebark pine seed in conditions
particularly suited for germination and
seedling survival and most likely to provide
the best recruitment to the pine population.
Favorable aspects of nutcracker caching include
moving this heavy wingless seed away from under
the immediate tree canopy, often into
pioneering habitats such as burns, thereby
reducing concentrations for rodent predators.
In Utah a high elevation burn is being
restocked with limber pine by the seed caching
activity of Clark's nutcracker (Lanner and
Vander Wall 1980). Similar beneficial
relationships have been suggested between the
wingless pinyon pine (Pinus edulis) seed and
the pinyon jay (Gymnorhinus cyanocephalus )
(Ligon 1978) but hard evidence to date shows
the benefit only to the bird.

SEED AS FOOD

Seeds are a particularly efficient food
resource for animals. They are a sessile prey
that periodically occurs in large abundance,
are durable to store, and have a dormancy that
improves storage survival, particularly in the
cooler temperate zones where caching behavior
among birds and mammals seems more common
(Smith and Reichman 1984). Viable conifer
seeds, the condition animals normally select
for, are uniform in size and have a clumped
distribution when in the cone, making them

highly visible and accessible to those animals
having the physical abilities and feeding
strategies adapted to using this resource in
the tree. Conifer seeds have higher energy
values when compared to seeds of most
angiosperms; deciduous tree exceptions being
shagbark hickory nuts (Carya ovata) and black
walnut ( Juglans nigra) whose 6302 and 7221
cal/g dry weight (dw) of endosperm and embryo
(Smith 1970) is in the mean range (6781+ 760
cal/gm dw) of 19 conifer species reviewed by
Grodzinski and Sawicka-Kapusta (1970). Smith
(1968) determined the average caloric value of
eight conifer species to be about 7000 cal/g
dw; ponderosa pina having the highest at 7558
(SD 132) cal/g dw and lodgepole pine the lowest
at 6827 (SD 152) cal/g dw. The average value
of 400 plant species seeds tested by Golley
(1961) was 5065 (CV 219) g cal/g dw, although
he included seed head and coat material. In
conifer seeds the shell can constitute about 20
percent of a seed's caloric value (Smith
1970). The superior nutritive quality of
conifer seeds comes from their high lipid
content (Smith 1970) and lipids have the
highest caloric content (about 9000 cal/g dw)
of basic nutrition elements. Proteins average
about 5000 and carbohydrates about 4000 cal/g
dw (Smith and Follmer 1972). Smith (1970)
recorded the digestive efficiency of two red
squirrels at 90 percent on nut kernels and
reported the digestive ability was about in
proportion to the caloric value or lipid
content. Fat is the principal food storage
form in conifer seed (Rediske 1961) and a most
durable form in cool storage such as a squirrel
midden.

The importance of conifer seeds in the diets of
their animal users is incompletely known. It
has been explored for the red squirrel by Smith
(1968) and Finley (1969) in speculative
discourse, and for the Clark's nutcracker by
Tomback (1982) more precisely. It is difficult
to measure the total energy needs, use, and
contribution of conifer seeds to free-living
birds or squirrels and at present the
relationships can only be crudely estimated.
Anyone who has watched a red squirrel in daily
activities can see they are extremely energetic
creatures. Their basal metabolic rate is 1.76
times that expected for an animal its size
(Irving and others 1955). The daily energy
consumption for an adult male red squirrel in
summer averaged 117 kg cal and for a lactating
female 322 kg cal (Smith 1968). Five
half-grown juveniles consumed about 25 percent
(80.5 kg cal) of their mother's daily
requirement during nesting. Smith calculated
that a squirrel's territory contained nearly
the average total estimated food energy
required (42,700 kg cal-Finley 1969) by a
squirrel. These figures are to be accepted
with fair caution, as mentioned earlier. On
the basis of a squirrel perhaps getting half
its subsistence from other foods such as fungi,
fruit, and flower heads, Finley (1969) provides
estimates of the amount of seed needed from
five conifer species for the other half of a
squirrel's energy needs (table 2).

204

Tabic 2. — btlutcd aoounta1 of aeed needed to provide 21,330 kg cal, half
of the annual energy requirement of an adult Taalaac lurua

Douglaa-

r. fin** 1 3rtn n

Blue

Ponde roaa

Lodge pole

f lr

spruce

spruce

pi ne

pine

Tn ouaand f re a h

whole aeeda

400

12U0

910

170

1300

Kg of freah

whole aeeda

4.4

3.9

3.9

5.2

S.8

Kg of dry

aeed kernela

3.0

3.0

3.0

2.8

3.1

Nuaber of

cone a

9200

2400

62,000

Buahela of

conea

13

17

8.6

8.7

24

Nuaber of good

aeed treea

8.3

1*

17

6

47

Ac re n of

f oreat

3.2

2.0

1.7

0.8

4.1

'The values In this tsble are Internally Inconsistent because they are
derived frost seversl Independent sources. (Frost Flnley 1969). Figures
represent a broad linking of averages and areas to derive relations between
quantities of seed, cones, trees, and area. Figures could be refined
considerably through further study.

The Clark's nutcrackers studied by Tomback
(1982) in the eastern Sierra Nevada recovered
stored whitebark pine seed from about April to
July and the total energy requirements were
estimated at 16000-25500 kJ (min-max) or 133 to
213 kJ per day. These caloric requirements
convert to 50 to 84 seeds per day, amounts a
nutcracker can readily carry in its throat
pouch. The red squirrel and Clark's nutcracker
energy requirements have been more intensively
evaluated than any other species I will
discuss. Because conifer cone crops fluctuate
considerably, the effects of the varying,
highly nutritious food supply will be discussed
later.

PREDATOR OR SEED DISPERSAL AGENT

Predispersal seed removal by vertebrates
represents an interception in the normal cycle
of a tree. A superficial judgment would class
the act as predation and the agent as a
predator. A distinction exists between seed
predators and seed dispersers (Smith and Aldous
1947; Turcek and Kelso 1968); predators being
those whose seed use results in seed
destruction, as opposed to the seed
dissemination effect of dispersers. Janzen
(1971) judges the two roles according to the
precision with which seed processing (i.e.,
completeness of assimilation) occurs, and
whether the agent seeks the fruit or the seed.
After passage through the gut of six bird and
four mammal species, the fruit of 25 different
woody plant species mostly showed either no
viability loss or improved germination
(Krefting and Roe 1949). Such processing
inefficiency would class the vertebrate
consumers as dispersal agents. Krefting and
Roe (1949) also found that the dispersal role
of many animals was enhanced because passage
through the gut helped overcome seed-coat

dormancy. In most cases the desired food item
was fleshy fruit material rather than seed.
With conifer cones, the seed is the
sought-after food material and most vertebrate
agents that gather conifer seed act primarly as
predators.

Of the 44 mammals and 37 bird species listed by
Smith and Aldous (1947) as North American
conifer seed eaters, the red squirrel can be
considered the primary predator because of its
efficiency in cutting cones from trees (Tevis
(1953; Smith 1970; Shearer and Schmidt 1970;
Schmidt and Shearer 1971; Harvey and others
1980; Gurnell 1983; C. H. Halvorson
unpublished), and deeply burying these under
damp conditions that preclude seed release
(Shaw 1936; Finley 1969). Other mammalian
gatherers and eaters of predispersed seed are
given lesser attention because either they do
not forage on commercially important conifers,
or they eat seed but also cache it in clusters
that often are not recovered, and later
germinate. Grizzly and black bears are in the
first category because the large seeds of the
non-commercial whitebark pine are eagerly
sought (Tisch 1961; Mealey 1980; Kendall
1983). Chipmunks and golden-mantled ground
squirrels ( Spermophi lus lateralis) eat but also
pouch seeds and cache them in shallow pits,
failing to recover some portions of those
(Tevis 1952, 1953; Broadbrooks 1958; West
1968). Red squirrels do not have cheek
pouches, and thus store whole cones that do not
germinate, therefore their storage act is not
that of dispersal.

Most notable in the role of a dispersing agent
are some members of the crow family (Corvidae),
chiefly the pinyon jay and the Clark's
nutcracker (table 1), who distribute seeds away
from the trees and cache more than they
recover. Of course they also consume seed but
when the gathering is done in years of seed
abundance the birds bury more seed than they
consume (Tomback 1982). The other bird
families (table 1) recognized as conifer seed
gathers and feeders from cones are finch-types
(Fringillidae) , nuthatches (Sittidae),
chickadees (Paridae), and a few woodpeckers
(Picidae). Some of these groups may cache
seeds for short periods of days or weeks, as
with chickadees and nuthatches (Kilham 1974;
Sherry 1984), but for the most part the
families other than corvids can be considered
predators on predispersed seed.

SPECIES THAT FORAGE ON PREDISPERSED SEED

The manner in which vertebrates forage on
predispersed seed has been described by Smith
and Balda (1979) as: removing the entire cones
and caching (squirrels); extracting seed from
green or ripe closed cones, with some birds
performing storage (woodpeckers, crossbills,
jays); searching and extracting seed from
opening or opened cones (chickadees,
nuthatches, small finches).

205

The physical and behavioral adaptations to
these foraging methods are the subjects of this
section. None of the species discussed
(table 1) are such extreme specialists that
they subsists only on conifer seeds. The
sporadic nature of seed crops would not have
allowed survival. The crossbills (Loxia spp.)
are most closely linked to conifer seeds, in
feeding, reproduction and movement, but in
common with the other birds listed, they also
utilize buds and insects. Fungi are important
to rodents (McKeever 1964; Smith 1968; Sanders
1983; personal observation) and bears eat
carrion and forage on herbaceous vegetation,
roots and berries (Tisch 1961; Mealey 1980;
Murie 1981). Despite certain foods being
especially sought, usually large-seeded pines,
none of the birds or mammals have distributions
confined to the range of any one tree species.

I have tried to limit the vertebrate list
(table 1) to those whose primary feeding
behavior is that of taking cones or seed from
the tree. That does not preclude their also
foraging on shed seed or clipped cones as
secondary efforts. The five bird and two
mammal families that forage on predispersed
seed include at least 22 species of vertebrates
(table 1). All those shown occur across the
northern Rocky Mountains. The listing is based
on Smith and Balda (1979) but omits species out
of the range in question, such as the scrub and
gray-breasted jays (Aphelocoma coerulescens , A.
ultramarina) , plain titmouse (Parus inornatus),
and shrews (Soricidae) and mice (Cricetidae)
that only use dispersed seed. Added are two
bears — the grizzly and black; and three
birds — the black-capped chickadee ( Parus
atricapillus) , house finch ( Carpodacus
mexicanus) , and common redpoll ( Ca rduelis
f lammea) . The additions are based on
references not available or not included by
Smith and Balda (1979). To the bird listings
could be added Williamson's sapsucker
(Sphyrapicus thy roideus) , reported by Tomback
(1977) to occasionally take whitebark and
Jeffrey pine seed (Pinus jeffreyi) from cones,
or forage on Arizona limber and southwestern
white pine cones during opening (Benkman and
others 1984). The pine grosbeak ( Pinicola
enucleator) was observed by Tomback (1977) as a
subalpine transient in the Sierra Nevada,
taking whitebark and Jeffrey pine seed from
cones. This grosbeak has been recorded
primarily as a bud, fruit, and catkin feeder
(Bent 1968; Newton 1973). My inclusion of the
gray jay ( Perisoreus canadensis) and
black-billed magpie (Pica pica) is based on
Smith and Balda (1979) although the former
species remains undocumented as a conifer seed
feeder. The authors cite unpublished data that
the magpie harvests pinyon pine in Utah where
the Steller's jay (Cyanocitta stelleri) is
absent. In feeding tests (Dow 1965) that
deliberately exposed gray jays to conifer cones
( Abies lasiocarpa , Psuedotsuga menziesii) the
birds showed a weak but negative interest in
cones compared to strong search effort on bark
and rotted logs. The test timing (March and

April) would have been at the extreme end of
seed dispersal time (Schopmeyer 1974) for these
conifers in the western ends of their range
(British Columbia Cascades, Vancouver) and
might have contained unsound seed, normally the
last to shed. When later tested with lodgepole
pine that had seeds visible in partly opened
cones the birds pecked but did not attempt to
extract them, despite the birds being without
food several hours prior (Dow 1965). The
European jay (Perisoreus inf austus) is cited in
many accounts listed in Turcek and Kelso
(1968:282) as taking and caching Siberian stone
pine (Pinus siberica), Norway spruce (Picea
abies ) and other conifer seeds. Further study
of the gray jay relationship with conifer seeds
is desirable because it is a Nearctic winter
resident throughout its range in the boreal
forest (Udvardy 1969; Welty 19 75).

In the inland West all species in table 1 are
year-long residents in a portion of those
states bordering Canada and southward. It is
characteristic of resident species of the taiga
or northern coniferous forest that they are
mostly seed-and nut-eaters and insect feeders
(Welty 1975). In Montana only the white-headed
woodpecker (Picoides albolarvatus) is not a
resident. However, many of the birds show
altitudinal migratory behavior after breeding,
moving between forest communities with the
onset of harsh weather, or latitudinally from
one region to another, sometimes showing
irruptive behavior in response to food
scarcity. True migratory birds of our northern
coniferous forests outnumber the residents.
Udvardy (1969) attributes this paucity of
complete adaptation to the sharp seasonal
changes and relatively recent and interrupted
expansion of boreal forests during interglacial
periods. The adaptations birds especially have
made is to migrate, to shift their food habits
between summer and winter, and to prolong food
availability by storing it. Mammals also show
these adaptations, but rather than migrate some
are dormant during winter.

The species descriptions that follow are
patterned after Smith and Balda' s (1979)
succinct presentation, which should be read for
additional insight on competition between seed
eaters .

Birds

Woodpeckers . — Only recently was the hairy
woodpecker (Picoides villosus) described
feeding on intact conifer cones in the tree.
However, an early record from 1928 is cited by
Bent (1939) that seed, mostly conifer, averaged
12 percent in the diet (not identified as being
volume or occurrence) of a humid-forest west
coast subspecies of the hairy woodpecker (P^.v^.
harrisi). The origin of the seed was not
disclosed and could have been from ground
foraging which is out~of-character for this
bird. Stallcup (1968, 1969) provided the first
accounts after observing hairy woodpeckers
perched and feeding on pine cone clusters.

206

Attacking only the upper surface of ponderosa
pine cones with their bill, they hammered and
pecked, sometimes twisting bracts loose from
the seeds which were also pecked open instead
of cracked as finches do. The feeding occupied
about 65 percent of their foraging time from
mid-October through February in this Colorado
study. Early spring foraging on ponderosa pine
was also observed by Ligon (1973), and
occasional feeding on whitebark and Jeffrey
pine occurs in the Sierra Nevada (Tomback 1977).

The white-headed woodpecker was recognized as a
pine seed feeder in 1911 (Bent 1939). All
reports (Curtis 1948; Tevis 1953; Ligon 1973;
Tomback 1977) refer to feeding on the large
seeded species: ponderosa pine, Jeffrey pine,
sugar pine (P. lambertiana) , and whitebark
pine. Spring (April) foraging on ponderosa
pine cones in Idaho consisted of a few birds
foraging steadily through open second-growth
pine, chipping cones open for seed while
clinging to them. In early August the birds
attacked the hard green cones and extracted
unripe seed (Ligon 1973). Three females that
Ligon collected in October had their stomachs
60-70 percent filled with pine seeds. Tevis
(1953) describes late-August feeding on
California sugar pine cones two weeks prior to
ripening. Groups of four to six birds "slashed
open every cone on which they alighted and ate
the seeds, leaving deep vertical

trenches " He reported 34 percent (559)

cones destroyed. They concentrated on certain
trees, thus one pine lost 85 percent (252) of
its cones. White-headed woodpeckers reach the
northern and eastern limits of their range in
Idaho (Burleigh 1972) and are not common in
Washington, although they are residents there
also (Jewett and others, 1953). The
Williamson's sapsucker may be a conifer cone
feeder (Tomback 1977; Benkman and others 1984),
but this is not substantiated enough to assess
the importance to regeneration. Of the
woodpeckers listed, the white-headed presently
appears to be the species most interested in
cones and able to damage them in quantity.

Jay s . — Despite not having real evidence that
the gray jay is a conifer seed consumer, the
"Whiskey Jack" is a ubiquicous and territorial
inhabitant of northern coniferous forest. Yet
their use of plant material as food is hardly
known (Rutter 1969). They are well-known
omnivores especially attracted to meat and camp
scraps. Their particular attribute is the
ability to store food amidst conifer needles
and in and on tree trunks and branches (Dow
1965). This storage is achieved by orally
manipulating food until it is completely coated
with thick saliva to become a mucous pellet or
bolus. The saliva allows adhesion to a
substrate (Dow 1965) and would seem ideally
suited for caching seed, but there are no
records to that effect. Captive jays were
unable to open sunflower seeds and gray jays
seemed committed to a single bill-shoving
motion when burying (in captivity only) or
storing food. The hammering motion that is

used by other seed feeding corvids to store or
extract conifer seeds was never observed by Dow
(1965). Smith and Balda's (1979) inclusion of
the gray jay as a conifer seed eater seems
unwarranted at present and I have included it
only because of its congener, the European jay,
a known seed user (Turcek and Kelso 1968).

Three other jays, Steller's, pinyon, and
Clark's nutcracker, share typical jay
characteristics of strong bills and feet,
social system, and a food storage instinct.
The Steller's jay remains the least studied of
the three. Tevis (1953) described the
Steller's foraging in sugar pine when cones
opened, "launching" from a branch to strike a
cone and dislodge seed which was seized in
mid-air, carried to a limb, and the kernel
pounded open. This jay travels in raucous
groups, concentrating on trees with opening
cones. Only Smith and Balda (1979) mention
storage, and large flocking movements do not
appear in the reviews previously cited. The
Steller's jay seems to be an inconsequential
seed predator from knowledge presently
available .

Pinyon jays live and feed in close association
with pinyon pine - juniper ( Juni perus spp.)
communities but also occur with ponderosa pine,
a seed readily accepted in lieu of the
infrequent pinyon seed crops (Bent 1946; Balda
and others 1972; Ligon 1978; Schopmeyer 1974).
They readily extract conifer seed (table 1)
from green or opening cones by hammering in jay
fashion; they bury seed clusters in pockets,
and practice communal caching, usually on open,
warm-aspect slopes. From 30-50 seeds at a time
can be transported in a distensible esophagus.
The food is used later in winter and as food
for new hatchlings in early spring (Ligon
1978). These strong fliers search widely for
food, up to 13 miles (21 km) daily (Balda and
others 1972). These behavioral characteristics
are quite similar to those in the Clark's
nutcracker, but the pinyon jay shows a unique
physiological adaptation to cone abundance that
is shared with the red crossbill (Loxia
curvi rost ra ) , a bird more distinctly associated
with montane and boreal forest. Pinyon jays
can be stimulated to breeding readiness and
actual breeding by the presence of an abundant
pine crop. Thus, Ligon (1978) found breeding
readiness (gonadal development) in early
December, February breeding after a large
pinyon seed crop, and August breeding in the
presence of a large maturing seed crop that was
used to feed young. He reported this only for
southwestern New Mexico and similar behavior in
northern parts of the jay's range is unknown.
Pinyon seed volume in stomachs reached 95
percent in December to 50 percent in May as
birds continued to use caches, then tapered to
20 percent in summer until new seed became
available .

The Clark's nutcracker is the most thoroughly
studied of the predispersal seed-eating jays,
partly because its eruptive migrations are

207

well-known (Davis and Williams 1964; Turcek and
Kelso 1968; Bock and Lepthien 1976) and its
seed storage behavior has been thoroughly
documented, but only in the last 15 years
(table 1). In contrast, its Palearctic
counterpart, the Eurasian Nutcracker (Nucif raga
caryocatactes) , has a long record of
observation on feeding behavior and population
migration (Formosof 1933; Turcek and Kelso
1968) in conjunction with fluctuations in its
primary food, the Siberian stone pine (P.
cembra) . The Clark's nutcracker is superbly
adapted to feeding on conifer seeds year round
and, like the Eurasian nutcracker, possesses a
sublingual throat pouch (Bock and others 1973)
to store and transport food. This is more
highly specialized than the pinyon jay's
expandable esophagus and holds about 20 ml (125
limber pine) of seed (Lanner and Vander Wall
1980) . The nutcracker depends heavily on
whitebark, limber, and ponderosa pine (and
related Jeffrey) pine extensively (Curtis 1948;
Tomback 1977, 1981; Lanner and Vander Wall
1980) but readily takes Douglas-fir
( Pseudotsuga menziesii) in the absence of the
larger-seeded species (Giuntoli and Mewaldt
1978; Fisher and Myres 1980; Vander Wall and
others 1981) . The typical foraging pattern was
first delineated by Tomback (1977) in the
Sierra Nevada. The adult birds began harvest
from unripe closed whitebark pine cones in
mid-July. Seeds were exposed by forcefully
repeated stabs to loosen and tear off scales.
Soft seeds were removed piecemeal and eaten or
moved into the sublingual pouch by a backward
toss of the head. Hardened ripe seeds were
extracted whole at a rate of about one per 7
sec. after scales were twisted and broken off.
Seed soundness was detected by "bill-clicking"
or rattling the seed within the mandibles;
unsound seed being discarded. Pouch contents
averaged 77 whitebark seeds. Storage occurred
initially in late August in high elevation
subalpine habitat but flights were also made at
this time to lower elevations in the Jeffrey
and ponderosa pine zones, to store whitebark
seed for winter use, and also to "test" lower
elevation seed. Harvest and storage of
subalpine whitebark seed ended about
mid-October but many birds had already moved to
lower elevation wintering areas to harvest
ponderosa or Jeffrey pine. Because whitebark
pine cones are indehiscent, the birds continued
to forage on any available in winter. The seed
recovery period at lower elevations was from
late December to late March with young being
fed cached seed starting very early in spring.
Upon returning to higher elevations in early
summer the birds used subalpine caches.
Tomback (1977) concluded that these altitudinal
movement patterns were comparable to those
reported from Montana (Giuntoli and Mewaldt
1978). The onset of harvest on limber pine in
Utah was also similar (Lanner and Vander Wall
1980) .

From a full pouch of whitebark pine seed a
nutcracker will successively distribute 3.7+2.9
seed per cache, burying them at a mean depth of

2.0+0.8 cm (Tomback 1982); caches of limber
pine were similar (4.2+3.1 seeds per cache at
2-3 cm depth (Lanner and Vander Wall 1980) .
Seed cache recovery apparently relies heavily
on memory, supplemented by visual cues (see
Tomback 1980) . Eurasian nutcrackers can find
stored seed under snow less than 2 m deep (in
Tomback 1980). Olfactory detection was
recently tested in another corvid, the magpie,
and reported to show at least close distance
effectiveness (less than 1 m) on cod liver oil
(Buitron and Nuechterlein 1985). Corvid
olfactory bulbs are quite small compared to
birds with known olfactory capabilities.

The seed use pattern of Clark's nutcracker (and
pinyon jay) is to store seed when it is
seasonally plentiful (fall), with subsequent
recovery when it is scarce but needed by
nestlings (winter-spring). Conifer seed was
the most important year-long food in 426
nutcracker stomachs examined from Montana
(Giuntoli and Mewaldt 1978). Average
importance of weighted samples from four years
showed conifer seeds constituted 83 percent of
stomach volume balance but occurred 59 percent
of the time, mostly in summer. Ponderosa seed
volume and frequency (52 percent, 80 percent
respectively) were greater than whitebark (19
percent, 42 percent respectively) in this
Montana study. The collection elevations
between 3280-8500 ft (1000-2500 m) spanned the
distribution of the pines in western Montana.
Seed consumption was strongly influenced by
availability. A bumper Douglas-fir seed crop
in 1946 provided most of the fall-to-spring
diet in the presence of only a moderate
ponderosa cone crop. In 1948, a very heavy
ponderosa crop was reflected by only those
seeds occurring in stomachs between September
1948 through last samples from May 1949; this
occurring with "moderate crops" of whitebark
and fir present.

The black-billed magpie (Pica pica) is not
cited as a conifer seed eater except by Smith
and Balda (1979) who refer to unpublished data
from Utah (Vander Wall and Balda) that the
magpies take pinyon pine seed in the absence of
Steller's jays. The magpie in Europe stores
angiosperm nuts as does the yellow-billed
magpie (P. nuttali) in California (Turcek and
Kelso 1968)"

Chickadees . — Chickadees (Parus spp.) are small
tame birds resident to the coniferous forest of
Canada and the U.S. They forage industriously
on tree trunks, branches, foliage, and the
ground, using their short, strong acute bills
to extract and store arthropods, spiders,
insects, and seed in bark crevices (Haftorn
1974). In searching open conifer cones they
typically cling to the underside or hang upside
down, grasping seed wings and tugging the
kernel from between the cones scales (Bent
1946). All conifer species are used, including
the larger seeds like ponderosa, limber,
whitebark, and Jeffrey pine (Curtis 1948;
Tomback 1977; Benkman and others 1984). The

208

black-capped and the mountain (P_. gambe 1 i )
chickadee occur with conifer distribution in
the West Coast Corderilleran and Rocky Mountain
Ranges. The boreal chickadee (P_. hudsonicus)
is most common in Canada and Alaska but dips
down into Washington and Montana. At least
short-term (24 hour) memory has been
demonstrated for the black-capped chickadee in
relocating stored seed (Sherry 1984) and food
storage allows prolonged predation. Although
no evaluation of impact occurs in the
literature, the chickadee, like other small
seed-foragers (nuthatches, small finches)
cannot engage closed cones and must concentrate
seed gathering for the brief period between
cone opening and seed shedding. It therefore
seems unlikely that chickadees have much impact
on predispersed conifer seed.

Nuthatches. — The nuthatches ( Si t ta spp.)
(table 1) are behaviorally related to
chickadees in many seed foraging mannerisms.
Although the nuthatches are particularly adept
at working up and down vertical surfaces in
their search for insects they also mount open
cones from all angles to pull seed out, eating
it or storing by poking or hammering them under
bark or into crevices (Bent 1948; Ligon 1968;
Kilham 1974; Smith and Balda 1979). Species of
both families search assiduously among foliage
and branches but nuthatches feed on the ground
less than chickadees and only occasionally
probe fallen cones (Stallcup 1968; Smith and
Balda 1979) . Nuthatches and chickadees are
winter residents in western coniferous forest
and often form loose flocks with other seed
eaters during the fall cone-opening period and
during winter (Stallcup 1968, 1969; Ligon
1973). Foraging has been reported on the
squirrel-dropped cones of sugar pine (Tevis
1953), and in the tree on ponderosa pine
(Curtis 1948; Stallcup 1968; Ligon 1973; Smith
and Balda 1979).

A food habits study of three nuthatch species
(table 1) in western Oregon showed the bulk of
their diet from May 1969 through February 1970
to be insects (Anderson 1976). Birds were
collected from Douglas-fir and ponderosa pine
forests but conifer seeds were not listed in
their diet, or cone crops mentioned. Because
seed-eaters are opportunists adapted to
irregular supplies of tree seed it may not be
evident in their diet some years. However, the
red-breasted nuthatch (Sitta canadensi s) was
reported taking ponderosa pine seed in Idaho
(Curtis 1948), spruce ( Picea spp.), fir (Abies
spp.), and pine in Arizona (Smith and Balda
1979), limber pine and southwestern white pine
in New Mexico (Benkman and others 1984), and is
the only nuthatch observed to hammer open
pinyon pine seeds with its bill (Smith and
Balda 1979). It is also the only nuthatch that
is in the group of nine North American boreal
tree-seed eating birds showing eruptive
migrations (Bock and Lepthien 1976). The
white-breasted nuthatch (S_. caroli nensis) ,
although presented by Smith and Balda (1979) as
having the least specialized diet, appeared

with the red-breasted taking ponderosa seeds
(Curtis 1948) and foraging on whitebark and
Jeffrey pine cones in California (Tomback
1977). They also feed from fallen cones of
sugar pine (Tevis 1953) and ponderosa pine
(Stallcup 1968; Smith and Balda 1979).

The smallest nuthatch, the pygmy (J3. pygmaea) ,
occurs west of the Continental Divide up into
southern British Columbia. Like the others, it
has been observed storing ponderosa pine seed
taken from cones on the ground and from the
tree in late fall and through winter (Stallcup
1968; Ligon 1973). Smith and Balda (1979)
describe its feeding rate on a ponderosa pine
cone as one seed extracted per second and the
entire cone searched in under one minute.

Nuthatches exhibit intense foraging behavior,
like all bark-searching birds who probe for
insects, and their strong, narrow, relatively
long bills are quite suited to probe the parted
scales of conifer cones. But their feeding
capability is limited to open cones in the
period after squirrels, jays, and crossbills
harvest, and in company with chickadees and
finches. They also appear to store seeds for
relatively short periods (days) , cannot carry
any quantity beyond one or two seeds, and have
an evolutionary orientation for storing in
rather limiting locations (bark or tree-trunk
crevices) (Smith and Reichman 1984). These
characteristics suggest far less predator
efficiency than that associated with jays or
squirrels. Without more thorough evaluation of
seed predation nuthatches probably can't be
rated a serious factor in seed loss. The
red-breasted nuthatch appears to be more
dependent on conifer seed then the pygmy or
white-breasted nuthatch because it periodically
moves great distances beyond its normal timber
habitat, as in 1969-70, when Bock and Lepthien
(1976) saw red-breasted nuthatches working on
cemetery fenceposts in eastern Colorado prairie
far from any tree.

Fl nches . — Finches (Fringillidae) are agile
birds who coordinate their feet and bills,
parrots-like, to clamber over conifer cones and
search for seed. A finch characteristic is a
specialized bill that is internally grooved
along each side of the upper mandible. The
sharp rim of the bottom jaw fits into this
groove to hold and crack seed, a process in
which the tongue manipulates and holds the seed
in position. Bills vary in groove width, jaw
strength, and external shape, thus limiting the
size, hardness, and location of seed that can
be used by a particular species. The pine
siskin (Carduelis spinus) uses its long narrow
bill as tweezers to insert between and even pry
open cone scales to pull out seed, an
adaptation allowing it to exploit many conifer
species (table 1; Bent 1968; Newton 1973) but
the stubby-billed common redpoll and house
finch are probably not as adept or efficient at
cone-f eeding .

209

The hooked, crossed tips of crossbill (Loxia
spp.) mandibles are uniquely formed to extract
seed from hard, closed cones of most conifers,
but the shape precludes ground foraging since
the tips do not converge enough to grasp small
seed. For closed cones the bill tip is
inserted behind a scale and the lower mandible
moved sideways by powerful muscles to separate
the scales and release the seed. Then the
specially adapted tongue (longer than other
finches with an extra piece of cartilage on the
end) scoops the seed out. Closed cones are
usually wrenched off the branch and fed on
while perched, using one foot as a clamp.
According to Newton (1973) crossbills usually
take only a few seeds from closed cones then
drop the cones. These show frayed and split
scales. At seed-dispersal time, crossbills
feed while clinging to the cone, similar to
other birds.

None of the finches store food, unlike all but
the woodpeckers in table 1. This means they do
not have a "banking" mechanism to prolong seed
predation beyond the dispersal period. With
few exceptions the Inland West conifers
disperse most of their seed between late August
and November. Trees that do extend seed
dispersal into winter or longer, and can
provide winter food, include western larch
(Larix occidentali s) , blue spruce ( Picea
pungens ) , west coast Douglas-fir ( Pseudotsuga
m. var. menziesii) , western hemlock (Tsuga
heterophylla) , and the _p. scopulorum variety
of ponderosa pine in Colorado (Schopmeyer
1974). The indehiscent whitebark pine cones
can retain seed for months if squirrel or jay
predation is not complete. Serotinous
lodgepole pine is seldom included as seed prey
for birds because cones would either be
impenetrable for most birds or an
energy-inefficient food source. Lodgepole pine
seeds were only 2 percent of the cone weights
measured by Smith (1970), compared to 7.4
percent (Douglas-fir), 11.1 percent (ponderosa
pine), 14.5 (Engelmann spruce), and 25.9
percent (subalpine fir — Abies lasiocarpa).
Animals obviously benefit more by feeding
efforts that secure large seed.

Red crossbills were observed near Missoula,
Montana, fluttering around ponderosa pine cones
and extracting seed from them on 9 May 1966.
About the same period, farther west, chipmunks
( Eutamias spp.) were seen up in ponderosa pine,
clipping cones that retained seed from the 1965
crop (P. F. Stickney and R. F. Schmitz, pers.
comm.). Ponderosa pine normally sheds 90-95
percent of its seed between September and the
end of October in Idaho and western Montana
(Fowells 1965; Shearer and Schmidt 1970). The
year 1965 was particularly cold and wet from
March through October, with the coldest
September in 86 years. The remaining winter
was one of the mildest on record. Dispersal of
the 1965 seed crop apparently was unusually
delayed by these climatic conditions, allowing
extended foraging.

Seed can be exposed longer to predispersal
predators when a wet fall inhibits cone
opening. Normally, seed is released by a
drying process where the scale fibrils shrink
and contract, to draw the scales apart (Harlow
and others 1964) . Scale spreading takes place
gradually over days and weeks and varies within
and among individual trees. Cones also open
and close, depending on humidity, causing
discontinuous or delayed shedding beyond the
normal dispersal period, at least for spruce,
pine, and Douglas-fir (Shearer and Schmidt
1970; Allen and Owen 1972; Schopmeyer 1974;
Hoff and Coffen 1982) .

Irregular seed production and aberrant
dispersal patterns are the norm however, and
the foraging pattern for finches is most nearly
a catch-as-catch-can matter. Such is reflected
both in the intensity of seed-foraging and the
restless movements of particularly the
crossbills and pine siskin. Curtis (1948)
singled red crossbills out as the most
"voracious" feeders, present in the largest
numbers, among nine bird species that descended
on a medium-heavy ponderosa cone crop near
Idaho City, Idaho, starting about 10 September
1947. Feeding birds persisted until early
November in groups up to 40, although numbers
peaked in mid-October. A rapid buildup of red
crossbills also occurred at Yellow Bay
Biological Station in western Montana between
Aug. 1-15, 1954 (Kemper 1959). The population
showed both physiological (gonadal) and
morphological (brood patch, enlarged cloaca)
breeding signs. These changes were attributed
to excellent cone crops among Douglas-fir,
grand fir, and Engelman spruce, and a moderate
ponderosa pine crop.

Mammals

The Order Mammalia contributes relatively few
vertebrates to the group who take seed in the
tree, if compared to birds (table 1). The
flying squirrel of the west (Glaucomys
sabrinus) is endemic to coniferous forests but
is omitted from my listing because I found no
firm data that established its conifer seed
eating activity, much less using cones still in
the tree. Four flying squirrels nearly starved
on an ad libitum diet of white spruce (P.
glauca) seed, preferring laboratory chow and
many other foods instead (Brink and Dean
1966). By contrast, red squirrels preferred
spruce cones over the chow, taking 2-3 weeks to
adapt to the processed food. Stomach contents
of flying squirrels collected each month in
California conifer stands showed year-long use
of chiefly fungi and lichens but no seed,
despite conifer seed being easily available in
fall (McKeever 1960). Tests of flying squirrel
food preference by Laurance and Reynolds (1984)
did find they readily accepted pinyon nuts.
Additional study is required because flying
squirrels have been popularly accused of
damaging conifers by clipping branches and
terminals if not cones (Tom Lawrence, pers.
c omm . ) .

210

Red Squirrel. — As an acknowledged seed predator
wherever It lives in North America (Murie 1927;
Hatt 1929, 1943; Tevis 1953; McKeever 1964;
Brink and Dean 1966; Smith 1970; Harvey and
others 1980; Benkman and others 1984; Smith and
Reichman 1984), the red or pine squirrel
(table 1) cuts cones of all conifer species and
stores them in winter larders or middens (Murie
1927; Finley 1969; Smith and Reichman 1984).
These middens provide necessary energy for
over-winter survival. They also supply
sustenance needed during the later winter-early
spring breeding period when other food is
scarce. Each squirrel normally has at least
one primary and several secondary middens (C.
H. Halvorson, unpublished) which, In abundant
crop years, are stocked with as many cones as
can be secured prior to seed shedding.

In heavy seed years squirrel caches can supply
food well into the second year beyond the crop
harvest (C. H. Halvorson, unpublished), thus,
the effect of a crop failure is not immediate.
With such a food supply concentrated at great
effort, it is understandable why squirrels call
aggressively if intruders pass through the area
of harvest. Red squirrel aggressiveness has
long been considered territorial defense.
Smith (1968), in southern British Columbia, has
described territory size as varying inversely
with food abundance but defended all year.
Unchanging territorial boundaries were depicted
by Rusch and Reeder (1978) in Alberta spruce
and pine. Territoriality in a Colorado
lodgepole pine forest was not dependent on
large central stores because middens were
highly variable in size and cone content
(Gurnell 1984). Lodgepole pine stands differ
from most conifer forests because a seed supply
is always available in serotinous cones on
trees, as well as in squirrel caches. My
studies (C. H. Halvorson, unpublished) suggest
that territorialism in the rigid sense of Smith
(1968) may not be a constant nor perhaps
appropriate concept to describe red squirrel
behavior, although it certainly can be a
perception. I have found food supplies to be
defended and competition existing for areas in
which middens were prominent features.
However, the degree to which defense took place
was more dependent on season, on population
density, and on the year of the conifer seed
supply than on a sustained and perpetual
react iveness . Their territorial calling (Smith
1978) is never a seasonal or yearly constant.
However, red squirrels are solitary year-long,
except for an apparent one-day mating period
(Smith 1968). Therefore exclusiveness, how and
whenever achieved, is distinctive to the
species.

Red squirrel tree nests (less commonly
underground) are globular and composed of
grass, moss, needles, and twigs. The nest is
usually one of several in the same or nearby
tree. Nests and primary middens are always in
proximity, usually 50 ft or less.

Published figures of red squirrel densities
from across its geographic range are reviewed

and tabulated in Rusch and Reeder (1978).
Spring densities varied from about 1 acre (0.4
ha) per adult squirrel in spruce to 19.8 acres
(8.1 ha) per squirrel in eastern hardwoods. In
my 12-year study (unpublished data) minimum and
maximum densities in Douglas-f ir/ponderosa pine
were 4.5 acres (1.8 ha) to 0.8 acres (0.3 ha)
per adult in spring and slightly higher in
fall, on the average. This variation
illustrates the nature of squirrel
populations — they can fluctuate greatly between
years. This variation seems universal in tree
squirrel populations (Gurnell 1983; Heaney
1984), and is often associated with tree seed
abundance (e.g., Sciurus vulgaris in Finland -
Rajala and Lampio 1963; T. hudsonicus in
Alberta - Kemp and Keith 1970) in theory but
not by proven hypotheses.

The red squirrel's characteristic cone clipping
on ponderosa pine and Douglas-fir starts almost
tentatively in late June and early July after
seed fertilization. Douglas-fir and ponderosa
pine cones are cut and peeled, usually in the
tree or on a favorite feeding log or stump.
The time at which red squirrel primary use of
cones shifts from feeding to caching coincides
with seed maturation, as it does with the
Clark's nutcracker (C. H. Halvorson,
unpublished; Tomback 1977). Seed maturation is
marked by biochemical changes, the more
important being a sugar content decrease and a
fat increase in the endosperm (Rediske 1961).
As mentioned earlier, fat resists spoilage,
especially in the cool sites that squirrels
select for middens (Shaw 1936; Hatt 1929;
Finley 1969). Adult squirrels usually sever
and drop individual cones from the branch, but
ponderosa pine cones, being tightly appressed
to each other and the shoot, are often dropped
by cutting the branch just behind the cluster.
This causes loss of cone primordia and any
first year cones, and could remove three years
of fruit growth. If the cluster happens to be
on the terminal shoot, the trees' s primary
elongation point is also lost. Cone clipping
rates can be rapid, almost frantic, and
squirrels ignore their early morning and late
afternoon bimodal daily activity pattern to
work through the day. I have seen squirrels
cutting Douglas-fir cones at rates of 4 to
8/min, and eating one cone each 3 minutes for
90 minutes. Giant sequoia cones were cut at
rates from three/min to 538 in 30 minutes, or
18/min (Harvey and others 1980). Ponderosa
pine cones are often cut and stored
individually while fir usually are dropped, to
be stored later. In one instance it took about
6.5 min per ponderosa pine cone from cutting to
burying .

The quality of cones cached by squirrels is
sometimes questioned. White spruce and
Douglas-fir cones gathered from stores during
the entire caching period were found to be ripe
and have better viability than seed collected
directly from trees in the same periods; there
was no significant increase in viability with
later dates of cutting, compared to collections

211

begun at the onset of major caching (Lavender
and Engstrom 1956; Wagg 1964). Similar testing
apparently has not been done for ponderosa pine
or other conifers. Where cones ripen
asynchronously between trees a question arises
whether squirrels concentrate first on those
trees whose mast ripens first. Olson and Silen
(1975) found that coastal Douglas-fir seed
showed steady weight increases and improved
germination even during the final two weeks
prior to seedfall.

Considerable data on cutting variability were
gathered during a study of squirrels and cone
crops on an island in Flathead Lake in western
Montana (Halvorson and Engeman 1983). The
years 1962 and 1965 had similarly dense fall
squirrel populations, averaging >l/acre
(>2.5/ha, fig. 1) between August and November.
Roughly 65 percent of the cones counted on
marked sample limbs were cut each year,
although the number of cones on the limbs
differed considerably (fig. 1). A heavy seed
crop was produced in 1962, measuring about 11
lb/acre (12 kg/ha) of high quality (61 percent
sound) seeds of Douglas-fir and ponderosa pine
combined. 1965 had a very light crop (0.30
lb/acre, 0.34 kg/ha), consisting almost
entirely of low quality (22 percent sound)
ponderosa pine seed. In 1962, the Douglas-fir
cutting rate was 8-10 days earlier than for
pine, a typical pattern because ponderosa pine
cones opened later than fir. By September 10,
1962, squirrels had gathered 179 pine and 636
fir samples cones, or over three-quarters of
the sample ultimately removed that year. In
contrast, in 1965, only 78 pine cones were cut
by September 20 and only 90 total removed, at a
desultory rate (fig. 1), by a larger population
than present in 1962. Weather in 1962 was a
"normal" warm but somewhat dry fall, and 1960
and 1961 had been poor crop years. 1965 had
record September cold temperatures and a wetter
than average entire year, causing seed to be
held in cones (see earlier). In 1965, middens
(caches) were still full from a 1964 seedfall
of 28 lb/acre (31 kg/ha), the largest in 12

HQ 10°

z£ o

LU <
CL LU

PONDEROSA PINE
DOUGLAS FIR

1962

SQUIRREL/ACRE: 1.10
CONE SAMPLE CUT: 234/328 P
686/1277 F

1965

10 20
JULY

SQUIRRELS/ ACRE: 1.33
CONE SAMPLE CUT: 90/136 P
3/7 F

10 20
AUGUST

10 20
SEPTEMBER

10 20
OCTOBER

MONTH

Figure 1. — Variability in cone-clipping rates
on tagged-limb samples of ponderosa pine (P)
and Douglas-fir (F) cones in a year of heavy
(11 lb/acre; 12 kg/ha) cone crops and warm,
dry, fall weather (1962), and poor (0.3
lb/acre; 0.34 kg/ha) cone crops and very wet,
cool fall weather (1965). Similarly high
squirrel densities occurred both years. Cedar
Island, Montana (C. H. Halvorson, unpublished).

years. That seed was 60 to 70 percent sound.
This example depicts the influence of weather,
seed quality, and the availabilty and influence
of previously stored food. To 1965 could be
added the greatest abundance of fungi found on
the island to that time — an effect from the
"unusual" weather that year. Squirrels are
known for their catholic food habits (Hatt
1929), especially utilizing the seasonal
abundance of fungi and hanging it in trees to
dry (Hatt 1929; McKeever 1964; Fogel and Trappe
1978) .

Other Squirrels. — Chipmunks and golden-mantled
ground squirrels are terrestrial rodents who
forage extensively on the ground for seed,
fungi, fruit, herbaceous material, and insects
(Tevis 1952, 1953; Broadbrooks 1958). Both
genera have internal cheek pouches that are
effectively used to transport food.
Over-wintering strategies differ in that the
ground squirrel converts its fall feeding into
fat layers metabolized during true
hibernation. Chipmunks do not accumulate fat
or gain weight before winter but go into torpor
that may be interrupted by mild periods during
winter. Their strategy of lining their
underground nests with seeds allows them ready
and safe access to nutritious food during
winter awakenings. Broadbrooks (1958)
described finding 1080 ponderosa pine seeds in
one nest and 5360 Douglas-fir plus about 10,000
herbaceous seeds in another. One cache
contained 170 g of seed, about 50 g being pine.

All studies of seed eating in these two genera
have dealt with ground foraging for seed and
those impacts on tree regeneration (Pank
1974). Seed predation in the tree canopy has
been observational and anecdotal, thus we have
no measure of its importance, and only sparse
data that such foraging occurs. Spermophilus
lateralis and Eutamius spp. have been seen
foraging up in conifers. The chipmunks are
more agile and seed constitutes a larger part
of their diet than it does for the ground
squirrel (Ingles 1965). E. amoenus has been
seen feeding at heights to 100 ft (30 m) in
ponderosa pine and Douglas-fir, pulling seed
out as cones opened. Broadbrooks (1958)
described two chipmunks feeding side by side
from the same cone 50 ft (15 m) up in a tree.
Whitebark and Jeffrey pine seed were gnawed
from cones by golden-mantled ground squirrels
and chipmunks in trees amidst seed-gathering by
Clark's nutcrackers (Tomback 1977). Benkman
and others (1984) in Arizona reported similarly
for the two sciurids feeding on limber pine.
These reports all describe fall feeding. Dick
Schmitz's observation on ponderosa pine seed
retained into at least early May and being used
by chipmunks was mentioned earlier. None of
the accounts I found mention any in-tree
behavior other than pulling seeds from opening
cones or gnawing cones. Some seed is probably
pouched and carried to storage but
cone-clipping and caching, a much more
efficient form of foraging behavior (Janzen
1971; Smith and Balda 1979; Smith and Reichman

212

1984) Is not discussed. The habit of seeking
and pouching individual seed for storage is an
evolved feeding characteristic, therefore the
most efficient practice for terrestrial
sciurids. Cone storing or clipping has not
been the adaptive path taken by the chipmunk
and ground squirrel, therefore is probably
energy-inefficient for them and they would seem
to impose minor impact on predispersed seed
compared to arboreal squirrels.

Bears. — What is written on bear use of conifer
seed in the Inland West comes mainly from
Yellowstone National Park (Mealey 1980; Murie
1981; Hutchins and Lanner 1982; Kendall 1983)
and northern Montana (Tisch 1961). Both the
grizzly and black bear (table 1) eat whitebark
pine seeds and berries as the main fall food,
when available. Spring use of pine seeds is
also a matter of availability if the previous
fall had a heavy cone crop (Kendall 1983).
Limber pine seeds are mentioned as grizzly food
only by Craighead and Craighead (1972) and C.
Jonkel (pers. comm.). The seeds are obtained
by robbing red squirrel cone caches or rodent
seed stores (Mealey 1980; Hutchins and Lanner
1982; Kendall 1983) but relatively little by
climbing trees (black bear only — Tisch 1961).
Therefore, bears would be secondary predators
on cones and only on two noncommercial pines.
Murie (1981) in Alaska suggested that spruce
cones were not palatable, otherwise they would
be expected to appear in grizzly droppings
during heavy crop years when squirrels pile
cones above and underground. Both bear species
are able to extract seed without ingesting
cones scales, therefore scales don't appear in
scat. Bears feed by breaking cones up in their
mouth or underfoot, then licking the seed up
and expelling scales from sides of the mouth
(Kendall (1983). In summary, bear feeding
concentrates on noncommercial pines, depends
mostly on squirrels storing cones, and is tied
to periodicity of good cone crops. At least
the two large-seeded pines are recognized as
very important fall foods in the northern Rocky
Mountains, and are sought after in spring prior
to herbaceous vegetation growth (Kendall 1983).

VERTEBRATE IMPACT ON PREDISPERSED SEED

The known impact that vertebrates have on
reducing potential seed crops depends on
factors of economics, species characteristics
that contribute to feeding efficiency, amount
of study, and the response or dependencies of
animals to crops. To portray relative
importance, I have distilled from the foregoing
accounts those qualifications that potentially
affect or relate to species relative impact
(table 3).

Judging Importance

The categories of table 3 list: 1) whether the
species has been studied with the purpose and
detail needed to quantify their effect on
seedfall, has been only described, or has not

been observed in measurable terms; 2) the
economic status of food trees; 3) specialized
and efficient feeding behaviors (beyond simply
extracting seed from cones in trees or moving
normal distances to exploit a food source, once
located); 4) unique morphological adaptations
that increase foraging effectiveness (beyond
standard family attributes like gnawing
incisors or grooved bills). The fifth category
concerns responses to seed crops rather than
impacts. It infers the importance of cones to
animals by listing some responses that occur
from cone scarcity or abundance, but also
suggests the extent of evolutionary
adaptiveness to cone crops. Winter residency
is not shown in table 3, but would qualify a
bird as a potentially more effective predator
than transients. The white-headed woodpecker
is the only bird listed whose eastern winter
range stops at western Montana.

Species importance can be interpreted from
table 3 by tabulating the number of categories
having a positive (yes) statement and how many
"Special" factors are assigned. Animals
feeding on noncommercial whitebark, limber, or
even the commercial seed of pinyon pine
(Fowells 1965) would be of little concern to
the timber industry regardless of how well
studied the animals were. Thus, bear use of
whitebark pine cones would be of major interest
only to agencies managing these omnivores,
especially since bears are secondary predators
on cone crops. Siskins and redpolls would be
of less relative importance even though they
double their breeding period in response to
unusually abundant spruce seed (Newton 1973),
but they only feed exploi tatively without
caching when a crop appears. However, both
species show irruptive migration and their
impacts have not been measured. Benkman (pers.
comm.) classed the pine siskin as a major but
temporary predator on tamarack (Larix larcina),
its competitor being the white-winged
crossbill. Nuthatches and chickadees are birds
that simply pull seed out and cache them
individually in bark crevices, as they do other
prey. The pygmy nuthatch is reported to
extract one ponderosa pine seed/sec. and work
one cone/minute (Smith and Balda 1979). Their
impact is far less than the crossbills' who can
attack closed cones effectively with their
laterally abducting mandibles, irrupt, and
breed as a result of stimulus effect from
abundant cones (Kemper 1959; Newton 1973; Bock
and Lepthien 1976). Although finches exploit
but do not store food, the crossbills
especially are extremely effective predators
because they range widely in large flocks and
can open cones. In high-value sites, such as
seed orchards and nurseries, they have
potential to consume large amounts of seed on a
local basis (Curtis 1948). Their feeding
potential is unmeasured.

The Important Vertebrates

Any bird or mammal that feeds on open and
closed cones of commercial tree species, has

213

Table 3. — Status^ and ordering of factors that relate predispersal seed users to conifer cone crops

Impact & Feeding Commercial Special feeding Special Special Response

Species Quantified conifers behaviors morphology to cone crops

Hairy Woodpecker

No

Yes

Closed and open cones

Strong bill, feet

Unknown

W. -headed Woodpecker

Yes-limited

Yes

Slash green cones

Strong bill, feet

Unknown

Gray Jay

No

Unverified

Not conifer seed

Salivary glands

2

Steller's Jay

No

Yes

No

No

Unknown

Pinon Jay

Yes .

Yes-limited

Disperse caches

Esophagus expands

Breeding

[SW only)2

Clark's Nutcracker

Yes

Yes

Disperse caches

Sublingual pouch

Breeding ,

weight2

Blk. -billed Magpie

No

Unverified

No

Buccal cavity

Unknown

Blk.-cap. Chickadee

No

Yes

Cache seed

No

Unknown

Mtn. Chickadee

Descriptive

Yes

Cache seed

No

Unknown

Boreal Chickadee

No

Yes

Cache seed

No

Unknown

Red-brst. Nuthatch

No

Yes

Cache seed

Strong feet, bill

2

W.-brst. Nuthatch

No

Yes

Cache seed

Strong feet, bill

Unknown

Pygmy Nuthatch

Descriptive

Yes

Cache seed

Strong feet, bill

Unknown

House Finch

No

Yes

No

No

Red, W. Crossbills

Descriptive

Yes

Closed and open cones

Prying bill

Breeding2

Common Red Poll

No

Yes

No

No

Breeding2

Pine Siskin

Descriptive

Yes

No

No

Breeding2

Red Squirrel

Yes

Yes

Cache cones

Wrist joint

Breeding ,

longevity

Chipmunk

Descriptive

Yes

Cache seed

Cheek pouch

Unknown

G.M. Grd. Squirrel

No

Yes

No

Cheek pouch

Unknown

Grizzly & Black Bear

No

No

Rob squirrel caches

No

Concentrate feeding

1 For European corvids (jays and crows) see Turcek and Kelso 1968

2 Periodic irruptive movement and (generally) southward invasions

particularly effective feeding behaviors and
morphology, and shows a distinctive response
that suggests strong dependency on cone crops
would qualify as the potentially most effective
predator on predispersed seed. These criteria
quickly narrow the list to four vertebrates:
the white-headed woodpecker, pinyon jay, Clark's
nutcracker, and red squirrel (table 3). The
white-headed woodpecker's propensity and
efficiency in slashing open green cones (Tevis
1953) suggest they be especially identified in
seed orchard management, the same as
crossbills. White-headed woodpeckers have been
studied little and their importance to cone crop
impact is undetermined.

For the most part bird foraging on conifer cones
is scattered and sporadic throughout forest
stands during seed ripening, and is not
conspicuous like that seen in cultivated crops
(ducks and grain, blackbirds on rice and
sunflower seeds) where loss is assessed in
bushels or tons. Even concerted effort by
researchers often achieves only approximations
of agricultural seed loss (J. Besser, pers.
comm.). However, some quantitative estimates of
seed removal by Clark's nutcrackers and pinyon
jays have been possible because the seed
capacities of the nutcracker's sublingual pouch
and the jay's distensible esophagus are known.
Swallowing a seed completely or pouching it is
distinguishable by the slight backward head toss
which accompanies pouching (Tomback 1977). By
counting the number of daily trips an individual

(Bock and Lepthien 1976).

nutcracker made with filled pouches, Tomback
(1982) was able to estimate seed take. A single
bird might store about 32,000 whitebark pine
seeds in a good crop year, or 3-5 times its
energy requirements, thus leaving seed for
germination, rodent food, and other fates.
Twenty-five birds foraging their average 4.5
hr/day for the approximate 42 day storage season
were estimated as caching about 800,000 seeds
(18,000 cones) in a 125 acre (50 ha) area. This
was about 200,000 seed caches. From 75-100
percent of the whitebark pine cones were
destroyed or partly eaten by nutcrackers and
chipmunks in 1979 on Tomback' s (1981) tree
transects. Similar high losses of limber pine
cones occurred in Utah where 90 percent (1322)
were removed in 1979 and 70-90 percent even in
the preceding bumper crop year (Lanner and
Vander Wall 1980). The major result of
unrecovered seed caches that germinate is the
frequently seen multiple stemmed occurrence of
whitebark pine. Tomback (1977) found 1 to 4
(mean 1.7+1.0) seedlings per cluster in a Sierra
Nevada burn where nutcrackers were storing
seed. This corresponded to the average of
3.2+2.9 (minimum 1 to maximum 15) seeds per
cache recorded. Multiple-stems are also
frequent on a Utah burn where nutcrackers are
similarly reestablishing limber pine (Lanner and
Vander Wall 1980) .

Ligon (1978) observed flocks of 200 to 300
pinyon jays and calculated conservatively that
they stored 30,000 seeds/day and perhaps 4.5

214

million seeds In the September-January storage
period. He noted they frequently select very
open areas (e.g., areas cleared of
plnyon-junlper) to store seed next to shaded,
moisture-favorable sites such as bushes or
downed trees and away from competition with the
parent tree.

Cone ripening phenology is a factor In seed loss
--the "satiation" principle mentioned earlier.
Benkman and others (1984) speculated that
nutcrackers could take 6.6 percent of a limber
pine seed crop because the cones appeared to
ripen simultaneously on individual trees but
asynchronously among trees. But if cones on all
trees ripened together, which they interpreted
for southwestern white pine, foraging would not
encompass the entire crop before seed shed and
only 3.7 percent would be harvested. Because
seed-eating birds tend to concentrate their
harvesting in flocks, seed loss would depend on
seed-ripening patterns of conifers. Total loss
contribution from birds is also related to the
presence or absence of tree squirrels. Red
squirrels took about 80 percent of an Arizona
limber pine crop and nutcrackers harvested about
1.6 percent, whereas the birds gathered 12-22
percent when squirrels were absent (Benkman and
others 1984). In assessing the impact of
vertebrates on predispersed seed, the cone
ripening pattern of conifer species should be
identified and the contribution of seed-cache
germination to re-establishing tree stands
should be considered.

The Red Squirrel As A Cone Predator

The red squirrel's superior efficiency as a
predator stems from its ability to rapidly
collect large quantities of seed by cutting
cones. The Clark's nutcracker and pinyon jay
also gather seed in groups but their seed
extraction and storage does not appear as
efficient. I timed a red squirrel cutting
ponderosa pine cones and carrying them from the
tree to a cache and back at 6.5 min/ cone (6
cones). This equaled one seed stored each 6.5
sec, using a figure of 60 sound seed/ cone
(Schmidt and Shearer 1971). It took nutcrackers
an average of 24 (11-66) min to harvest one
pouch load (77 seeds) of whitebark pine and
return from caching them; an average of 19
sec/seed (Tomback 1982). However, whitebark
pine seed is 4.6 times heavier than ponderosa
seed (Schopmeyer 1974), reducing the weight
equivalent carried to about 4 sec. But
nutcrackers took 31 sec to extract seed from
closed cones (7 sec for open ones) where the
cutting rate for squirrels is unaffected by cone
stage and squirrels harvest continually through
the day in the presence of cone abundance.
Nutcracker harvesting conservatively takes 4.5
hr/day (Tomback 1982). Guintoli and Mewaldt
(1978) only recorded 25 ponderosa seeds per
pouch (n=20) in Montana, suggesting far less
efficiency than Tomback measured. A
radio-collared squirrel (C. H. Halvorson,
unpublished) was tracked for 90 minutes during
which time it cut, peeled, and consumed the seed

of 30 mature but closed Douglas-fir cones, i.e.
1 seed/4 sec or 3 min/cone, including some
movement to reach cones in the trees. In other
observations squirrels cutting an average of 6
(4-8) fir cones/min were removing 270 seed/min
or 1 seed/1.3 sec. (about 45 seeds/cone using
Finley 1969). The fact that squirrels guard
their caches and caches are available only to
bears increases greatly the squirrel efficiency.

The proportion of a cone crop harvested by red
squirrels varies considerably and the effect on
seedfall is not stated simply. Many harvest
figures can be quoted: sugar pine 54 percent
(Tevis 1953); limber and southwestern white pine
75-83 percent (Benkman and others 1984);
ponderosa pine 19-82 percent (Schmidt and
Shearer 1971); Douglas-fir 17-100 percent (C. H.
Halvorson unpublished). Tevis (1953) noted that
the harvest on individual trees also varied from
11-100 percent. In a study of ponderosa pine
cone loss factors in Montana (Schmidt and
Shearer 1971), the amount of cutting by
squirrels could not be related directly to crop
size but may have been affected by both the
abundance of the companion conifer, Douglas-fir,
and the weather. The pine cone harvest was
inverse to the size of the fir crop. With a
1954 heavy fir crop (229,000 sound seeds/acre)
only 19 percent (24 of 125 sample cones) of the
pine cones were cut. This was a cool, moist
fall. When the fir crop failed in 1956, during
a warm, dry fall, 82 percent (40 of 49 sample
cones) of a visually estimated light pine crop
was removed. Intermediate to these rates was a
54 percent removal (69 of 128 samples cones) of
a light (7,100 seeds/acre) pine and fir crop
during average fall weather.

Many combinations of cone abundance, harvest,
and squirrel population occurred between 1962
and 1972 (table 4) on the Flathead Lake study
island (C. H. Halvorson, unpublished annual
rept.). A medium-density (0.96 squirrels/acre)
population took about half of a fair pine and
fir crop (about 4 lb/acre total) in 1967 but the
same density cut only about 18 percent of sample
cones in the very heavy crop year of 1971
(21.7 lb total seed/acre). The seed-fall
amounts (table 4) would all be adequate for
regeneration, except for the 6000 fir seed/acre,
but varying amounts are also taken by
ground-dwelling rodents and would further reduce
potential germinants. Seed abundance and the
proportion of cones taken by squirrels determine
how many seed reach the ground and that is
variable but important information to a forest
manager.

Previous studies have only reported cone crop
abundance and loss without determining squirrel
population densities. But even with densities
accurately known it has not been possible to
establish a simple linear relationship between
squirrel numbers and cone loss (C. H. Halvorson
unpublished). The correlations between cone
crops of either the same or previous year and
August and November squirrel densities have been
of low order (r=0.581 to 0.381), suggesting that

215

factors other than crop levels determine
squirrel densities and the proportion of a cone
crop taken — factors such as other food
availability, cone quality, predation,
competition, breeding-age ratios, and weather
(C. H. Halvorson unpublished). The study by
Sanders (1983) on foraging by the Douglas
squirrel T. douglasi in California similarly
reported a lack of correlation between crop size
and the proportion of white fir and Jeffrey pine
cones taken. She concluded that squirrels
foraged according to food encounter or energy
intake rates rather than to an expectation of
the cones quantity to be removed. Sanders also
found evidence that squirrels with more cones on
their home ranges took a significantly lower
proportion of available cones than squirrels on
cone-poor areas.

Seed quality is an important factor that enters
into our perception and examination of cone
abundance, the cone-cut, and squirrel population
relationship. Douglas-fir apparently can
develop full size cones without pollination or
fertilization (Allen and Owens 1972) . Ponderosa
pine and other conifer cones also can develop
normally with a low level of fertilized seed
(Geo. Howe pers. comm.). Viewed from the ground
in early summer, a good cone crop can be
forthcoming but the promise of a crop is a
familiar one to foresters. In this northern
region, the year 1978 looked good for lodgepole
and ponderosa pine, western larch, spruce, and
fir but, except for yellow pine, the seed
quality was poor in the presence of a very wet
spring (John McBride pers. comm) , suggesting
possible pollination disruption because most
species, except pine, have a one-year
development cycle. Cone insects, especially
Hemiptera (e.g., Leptoglossus occidentalis) are
especially adept at inconspicuously damaging
maturing cones that otherwise appear normal.
Red squirrels, as with other seed predators,
readily detect and reject poor quality food,
often without having to sample it.

The total effect of
ovulate bud formati
been measured for p
Schmidt and Shearer
of cone development
percent, the female
with mature cone-be
squirrels are prone
of mature cones wer
figure varying year
percent loss of the
combined effect of
two-year cycle.

squirrel clipping, from
on to seed germination, has
onderosa pine in Montana by

(1971). In the first year
, squirrels removed only 2

strobili being associated
aring shoots ends that

to clip. About 44 percent
e lost the second year, the
ly as outlined earlier. A 14

potential cone crop was the
squirrel impact during the

Squirrel Clipping of Buds and Shoots. — Red
squirrels also indirectly affect cone production
by their feeding on buds, clipping branch ends,
and girdling in the tree crown. There are many
such references for all tree species squirrels
associate with (Hosley 1928; Hart 1936;
Shantz-Hansen 1945; Rowe 1952; Cook 1954; Duncan
1954; Adams 1955; Lutz 1958; Walters and Soos
1961; Pulliainen and Salonen 1965).

In my experience, reproductive buds of
Douglas-fir were often incorporated as food, but
only when squirrels were hard-pressed for cones
did they concentrate over-winter feeding on
buds. In the spring of 1967 on Cedar Island, I
found Douglas-fir twigs with primarily
pistillate but also staminate buds eaten. Twig
density over most of the island was 20-25 per
3 ft2 (0.84 m2) sample (C. H. Halvorson
unpublished). Foresters elsewhere in the
northern Rockies reported peeled basal ends of
ponderosa pine shoots (see Adams 1955), needles
fed on, and pine shoots blanketing the ground
where squirrels lived. Specimens I examined
showed typical squirrel tooth patterns. Cone
crops in the region for two years prior to 1967
had been poor, and I attribute the feeding
directly to lack of cones.

Terminal and lateral shoot clipping causes
concern for tree form and lumber quality but

Table 4. — Cone crop harvest rate and residual seedfall in different crop years
and levels of fall (Aug. -Nov.) squirrel densities. Cedar Island, Montana (C.
H. Halvorson, unpublished)

Squirrels

Potential

cropl

Cone harvest (%)

Residual

seed2

Year

per acre (ha)

P. pine

D.-fir

P. pine

D.-fir

P. pine

D.-fir

1962

1.1 (2.7)

5.4

5.4

71

54

15

106

1964

1.8 (4.5)

7.8

20.2

83

80

19

172

1967

0.9 (2.2)

3.6

0.3

55

53

19

6

1971

0.9 (2.2)

12.8

8.9

17

20

127

300

^-Pounds per acre (kg/ha = 16/acre x 2.2046 t 2.47),
2Thousands per acre (seed per ha = 1000/s t 2.47).

216

coastal Douglas-fir saplings showed no height
loss after three growing seasons where terminals
had been clipped off by squirrels (Fisch and
Dimock 1978). The injury was considered minimal
and temporary by the authors, similar to Rowe's
(1952) opinion for white spruce. Lutz (1958)
observed that the club-top or tufted appearance
of black spruce could be ascribed to squirrels
cutting twigs in the apical aggregate of cones
that characterizes the species. The bare bole
section below the top functioned as a fire-break
and prevented destruction of the cone crop. The
effect of terminal clipping in conifers can be
multiple- or fork-tops, but Hoff and Coffen
(1982) saw little height difference between
terminally pruned compared to unpruned
single-stemmed western white pine in a seed-tree
plantation. Because multiple-topped trees
produced more cones and pollen, the authors
recommended that leaders and terminal buds of
plantation trees be pruned to maintain 1-3 stems
from the 10 ft (ca. 3 m) height upwards.

Although 1 believe from field observation that
the stag-horn branching appearance of mature and
"overmature" ponderosa pine is a direct result
of red squirrel feeding activity, whether for
cones or survival in coneless years, it remains
to be proven that shoot pruning (with
accompanying apical bud removal) increases
pollen and cone production in wild trees as it
does for cultivated western white pine (Hoff and
Coffen 1982). Pruning buds in higher plants
stimulates branching and there would seem little
argument that squirrels are stimulating pine
branching. If more cones are produced, the
squirrel benefits.

In summary, red squirrel total impact can be 14
percent of a potential ponderosa pine crop
(Schmidt and Shearer 1971) but is unknown for
other conifer species. Cone harvest is highly
variable and crop size is not directly related
to harvest rate. However, expressing the impact
of squirrel cutting simply as the proportion of
cones cut has little meaning to forest
management without knowing crop quality and size
of both the potential and residual seedfall.
Even heavy loss of a bumper cone crop is likely
to produce adequate seedfall for germination,
but a light harvest of a poor or fair crop
probably will give an inadequate seed supply.
Unmeasured but real value can be assigned to
squirrel cone harvest because some of the 14
percent total impact on potential crop, or 66
percent on immediate cone crop, is recoverable
from caches for nursery planting (Finley 1969).
Value can also be postulated for bud and shoot
injury by squirrels, at least in terms of
enhancing fruit production and food production
structure of some tree species, but this too is
unmeasured. While birds are primarily
exploitive on cone crops, at least the Clark's
nutcracker and pinyon jay have been shown to
benefit the distribution of noncommercial,
large-seeded pines in the absence of any active
management programs by agencies for those
species .

ANIMAL RESPONSES TO IRREGULAR CONE CROPS

Animal response to changes in prey density has
been described as either numerical (Solomon
1949), or functional (Morris and others 1958).
If the prey (cones) density changes and
predators show increased numbers the response is
termed numerical. It could include breeding
effect but also the irruption phenomenon
described by Bock and Lepthien (1976). Red
squirrels also appear to respond numerically
with increased breeding effort. A functional
response occurs where access to increased prey
(cones, seeds) tends to concentrate or shift
foraging activities onto that prey (Morris and
others 1958). The sense of "functional" has
been typified mostly in forest insect outbreaks
where birds actually shift from their usual
foods. All seed predators show a functional
shift by including more seed in their diet when
it is available; some birds going back to buds
and insects when the seed supply fades, or
emphasizing an alternative conifer species if
available. A physical response was seen in
Clark's nutcrackers who weighed significantly
less the winter after cone crops failed to
appear in western Montana (Guintoli and Mewaldt
1978). Underweight nutcrackers in Utah in 1977
(Vander Wall and others 1981), were thought to
be recent immigrants because resident birds were
feeding normally on an abundant pine crop. In
1977, all cone crops in the U.S. Forest Service
Northern Region were reported poor; also very
spotty in 1976 (U.S.F.S. Reg. 1 Nursery Reports
and R. Shearer, pers. comm.).

The stimulus effect of food (Kemper 1959) and
the exploitive flocking to cone crops as they
ripen is typical of crossbills (Newton 1973;
Bock and Lepthien 1976); this was similarly
recognized in the pinyon jay by Ligon (1978).
Crossbills are also known to breed in any month,
in response to cone crop abundance, but chiefly
in late winter and then again late summer.
Pinyon jay's second breeding in June and July
(Ligon 1978) is also ascribed to food stimulus,
in this case when abundant green pinyon cones
are maturing. Even pine siskins and redpolls
will double their breeding period in the
presence of unusually abundant spruce seed crops
(Newton 1973) .

In summary, Nearctic redpolls, pine siskins, and
crossbills all move seasonally in response to
seed-ripening or in response to availability,
wherever cones still hold seed. When fall
conifer crops are found, the birds settle on
those until the seeds are mostly eaten up or
shed, then travel to the next supply.
Crossbills will go into a breeding condition
anytime the new supply is plentiful (e.g.,
Kemper 1959). The periodic mass movements
("irruptions") are also food-related but only
the crossbills depend on conifer seed in cones
throughout the year. When mass migrations
occur, and six of the irruptive species are
finches (Bock and Lepthien 1976), a combination
of food shortage and large populations is
believed responsible (Newton 1973). However,

217

Tomback (1977) suggests that a bird such as
Clark's nutcracker that is intricately tied to
conifer seed may only need total failure of all
its conifer seed resources to irrupt, in spite
of population density. The crossbill is similar
enough to the nutcracker in tactical respects
that it could be convergently responding the
same way, or else has evolved to move and
compensate for local crop failures.

Squirrel life history also contains response to
cone crop abundance that may serve to drive the
mechanism postulated earlier in this section,
but the way this happens is still being examined
(C. H. Halvorson, unpublished data). Gurnell
(1983) summarized responses from the European
and American literature on all tree squirrels
species: "...seed availability can affect the
length of the breeding season, the number of
adults which produce two litters, the number of
adults and yearlings which breed, and the mean
litter size at birth and weaning." Not
mentioned is evidence that adult squirrels, if
born in high seed crop years (5 lb/acre;
5.6 kg/ha), show a median longevity of 17 months
compared to 22 months for births in a poor crop
year (C. H. Halvorson and Engeman 1983). I also
found the breeding incidence among yearling red
squirrels on Cedar Island to be 88 percent in
high crop years and 51 percent in low years, a
not quite significant difference and low
correlation (p=0.052, r=0.628) with crop years,
but crop years may not be the largest source of
variation in the repeated measure ANOVA used
(Halvorson unpublished; also Gurnell 1983
citation). However, an analysis of litter size
difference between high and low seed abundance
years showed that production of young was
predictably and significantly greater if seed
was plentiful (p<0. 00003, r=0.95 for number of
young per female). Finally, I found that in the
12-yr study some females had second litters only
in the three heaviest seed crop years, 1962,
1964, and 1971 (C. H. Halvorson unpublished
data) , thus providing further direct evidence
that squirrel populations respond to conifer
seed abundance. This association has often been
inferred but until now has not been previously
established on a basis of both measured seed
crops and accurately counted squirrel
populations — a need expressed by Gurnell (1983).

SUMMARY

I have presented the names and nature of
seed-eating vertebrates common to the North
American West but particularly to the Inland
Mountain West where most are resident species.
At least 21 of the birds and mammals listed
(table 1) make important use of conifer seed
before it is shed from the cone. A few other
bird species do so to lesser extent. The most
effective predators are those who can feed
habitually on closed and green cones. These
include one mammal — the red squirrel; one
woodpecker — the white-headed; two corvids —
the pinyon jay and Clark's nutcracker; and two
finches — the red- and white-winged crossbills.

Except for the white-headed woodpecker whose
habits are not well known, the other five have
unique behaviors, morphology, and strategies
that enable them to effectively exploit cone
crops whenever they occur, especially in
abundance. The crossbills do not store food,
hence must search continuously for cones, but
once found, the birds quickly transform the food
energy into a reproductive thrust. The squirrel
and corvids intensively gather and store food
when it becomes available, using it in
maintenance and later reproduction, thereby
extending their predation period on seed.
Corvids somewhat control competition for seeds
by burying them. The squirrel eliminates most
of its competitors by burying the entire closed
cone and making the seed unavailable to all
except other squirrels, insects, and bear. The
grizzly and black bear apparently use squirrel
caches when they contain whitebark and limber
pine cones. The remaining predators on
predispersed seed must compete for seed that is
more accessible in opening cones, and these
animals basically exploit an immediate food
source or store portions for brief periods of
days or weeks.

Those predispersal seed predators most highly
adapted and effective on cone crops also return
some value by being effective seed dispersers —
the case of the nutcracker and pinyon jay — or,
if a squirrel, by serving as a source of high
quality seed for nurseries (but only if the seed
didn't originate from costly seed-tree
orchards). Whether benefits to trees accrue
from squirrel feeding is speculative, but
answerable by further study. The impact of most
birds on predispersal seed is not fully assessed
from the present primarily descriptive accounts,
especially for the commercial conifers growing
in this Inland Mountain West.

ACKNOWLEDGMENTS

I thank Michael Bogan, Karen Manci, and Diana
Tomback for critical comments and helpful
suggestions on the manuscript. Craig Benkman
generously provided unpublished material that
was useful in preparing this paper.

REFERENCES :

Adams, L. Pine squirrels reduce future crops
of ponderosa cones. J. For. 53:55; 1955.

Allen, G. S.; Owen, J. N. The life history of
Douglas-fir. Can. For. Serv. , Environment
Canada. Ottawa: 1972. 139 p.

Anderson, S. H. Comparative food habits of
Oregon nuthatches. Northwest Sci. 50:
213-221; 1976.

Balda, R. P.; Bateman, G. C; Foster, G.:

Flocking associates of the pinon jay. Wilson
Bull. 84: 60-76; 1972.

218

Benkman, D. W.; Balda, R. P.; Smith, C. C.
Adaptations for seed dispersal and the
compromises due to seed predation in limber
pine. Ecology 65: 632-642; 1984.

Bent, A. C. Life histories of North American
woodpeckers. Washington, D.C.: U.S. Natl.
Mus. Bull. 174; 1939. 334 p. (Reprinted by
Dover Publ., Inc., New York, 1964).

Bent, A. C. Life histories of North American
jays, crows, and titmice. Washington, D.C.:
U.S. Natl. Mus. Bull. 191; 1946. 495 p.
(Reprinted by Dover Publ. Inc., New York,
1964).

Bent, A. C. Life histories of North American
nuthatches, wrens, thrashers, and their
allies. Washington, D.C. : U.S. Natl. Mus.
Bull. 195; 1948. 475 p. (Reprinted by Dover
Publ., Inc., New York, 1964).

Bent, A. C. and collaborators. Life histories
of North American cardinals, grosbeaks,
buntings, towhees, finches, sparrows, and
allies. 0. L. Austin, Jr. (ed.) Washington,
D.C: U.S. Natl. Mus. Bull. 237; 1968.
602 p. (Reprinted by Dover Publ., Inc., New
York, 1968).

Black, C. C. Holarctic evolution and dispersal
of squirrels (Rodentia: Sciuridae). Evol.
Biol. 6: 305-332; 1972.

Black, H. C. (ed.). Wildlife and reforestation
in the Pacific Northwest. Corvallis, OR:
Oregon State University School of Forestry;
1969. 92 p.

Black, H. C. (ed.) Wildlife and forest
management in the Pacific Northwest.
Corvallis, OR: Oregon State University
School of Forestry; 1974. 236 p.

Bock, C. E.; and Lepthien, L. W. Synchronous
eruptions of boreal seed-eating birds. Am.
Nat. 110: 559-571; 1976.

Bock, W. J.; Balda, R. P.; Vander Wall, S. B.
Morphology of the sublingual pouch and
tongue musculature in Clark's nutcracker.
Auk 90: 491-519; 1973.

Brink, C. H.; Dean, F. C. Spruce seed as a
food of red squirrels and flying squirrels
In interior Alaska. J. Wildl. Manage. 30:
503-512; 1966.

Broadbrooks, H. E. Life history and ecology of
the chipmunk, Eutamias amoenus, in eastern
Washington. Univ. of Mich., Misc. Publ.,
Mus. Zoo. No. 103; 1958. 48 p.

Brodkorb, P. Origin and evolution of birds.
In: Farner, D. S.; King, J. R.; Parkers, K.
D.; eds. Avian Biology, Vol. 1. New York:
Academic Press; 1971. 586 p.

Buitron, D.; Nuechterleln, G. Experiments on
olfactory detection of feed caches by
black-billed magpies. Condor 87: 92-95; 1985.

Burleigh, T. D. Birds of Idaho. Caldwell,
ID: Caxton Printers, Ltd.; 1972. 467 p.

Cook, D . B. Susceptibility of larch to red

squirrel damage. J. For. 52: 491-492; 1954.

Craighead, F. C, Jr.; Craighead, J. J.
Grizzly bear pre hi be mat ion and denning
activities as determined by radio tracking.
Wildl. Monogr. 32: 1972. 35 p.

Curtis, J. D. Animals that eat ponderosa pine
seed. J. Wildl. Manage. 12: 327-328; 1948.

Davis, J.; Williams, L. The 1961 irruption of
the Clark's Nutcracker in California. Wilson
Bull. 76: 10-18; 1964.

Dow, D. D. The role of saliva in food storage
by the gray jay. Auk 82: 139-154; 1965.

Duncan, D. P. A study of some of the factors
affecting the natural regeneration of
Tamarack (Larix laricina) in Minnesota.
Ecology 35: 498-521; 1954.

Egler, F. The nature of vegetation. Aton
Forest, CT 06058; 1977. 527 p.

Elliott, P. F. Evolutionary responses of
plants to seed-eaters: pine squirrel
predation on lodgepole pine. Evolution 28:
221-231; 1974.

Emry, R. J.; Thorlngton, R. W., Jr. The tree
squirrel Sciurus (Sciuridae, Rodentia) as a
living fossil. In: Eldredge, N.; Stanley, S.
M. eds. Living fossils. New York:
Springer-Verlag; 1984: 23-31.

Finley, R. B., Jr. Cone caches and middens
of Tamiasciurus in the Rocky Mountain
region. Lawrence, KS : Univ. of Kans. Publ.
Mus. Nat. Hist., Misc. Publ. No. 51:
233-273; 1969.

Fisch, C. G . ; Dimock, E. J., II. Shoot
clipping by Douglas squirrels in
regenerating Douglas-fir. J. Wildl. Manage.
42: 415-418; 1978.

Fisher, R. M.J Myres, M. T. A review of

factors influencing extralimital occurrences
of Clark's Nutcrackers in Canada. Can.
Field-Nat. 94: 43-51; 1980.

Fogel, R.; Trappe, J. M. Fungus consumption

(mycophagy) by small mammals. Northwest Sci.
52: 1-31; 1978.

Formosof, A. N. The crop of cedar nuts,
invasions into Europe of the Siberian
nutcracker (Nucifraga caryocatactes
mac rorhynchus Brehm) and fluctuations in
numbers of squirrel (Sciurus vulga ri s L.).
J. Anim. Ecol. 2: 70-81; 1933~

219

Fowells, H. A., ed. Silvics of forest trees of
the United States. U.S. Department of
Agriculture, Agriculture Handbook 271; 1965.
762 p.

Giuntoli, M.; Mewaldt, L. R. Stomach contents
of Clark's Nutcrackers collected in western
Montana. Auk 95: 595-598; 1978.

Golley, F. B. Energy values of biological
materials. Ecology 42: 581-584; 1961.

Grodzinski, W.; Sawicka-Kapusta, K. Energy

values of tree-seeds eaten by small mammals.
Oikos 21: 52-58; 1970.

Gurnell, J. Squirrel numbers and the abundance
of tree seeds. Mammal. Rev. 13: 133-148;
1983.

Gurnell, J. Home range, territoriality,

caching behavior, and food supply of the red
squirrel ( Tamiasciurus hudsonicus Fremonti)
in a subalpine lodgepole pine forest. Anim.
Behav. 32: 1119-1131; 1984.

Hafner, D. J. Evolutionary relationships of
the Nearctic Sciuridae. In: Murie, J. 0.;
Michener, G. R. , eds. The biology of
ground-dwelling squirrels, Chapter 1, 3-23;
Lincoln, NE : Univ. Nebraska Press; 1984.
464 p.

Haftorn, S. Storage of surplus food by the

boreal chickadee Parus hudsonicus in Alaska,
with some records on the mountain chickadee
Parus gambeli in Colorado. Ornis Scand. 5:
145-161; 1974.

Halvorson, C. H. Red squirrel ecology in the
montane forest of the northern Rocky
Mountains. Missoula, MT: U.S. Department of
Interior, Fish and Wildlife Service, Denver
Wildlife Research Center. Annual reports for
1962 through 1970. Unpublished.

Halvorson, C. H.; Engeman, R. M. Survival

analysis for a red squirrel population. J.
Mammal. 64: 332-336. 1983.

Harlow, W. M.; Cote, W. A., Jr.; Day, A. C.
The opening mechanism of pine cone scales.
J. For. 62: 538-540. 1964.

Hart, A. C. Red squirrel damage to pine and
spruce plantations. J. For. 34: 729-730;
1936.

Harvey, H. T.; Shellhammer, H. S.; Stecker, R.
E. Giant Sequoia ecology. Washington, D.C.:
U.S. Department of Interior, Natl. Park
Serv. Sci. Monogr. Ser. No. 12; 1980. 182 p.

Hatt, R. T. The pine squirrel in Colorado. J.
Mammal. 24: 311-345; 1943.

Hatt, R. T. The red squirrel: its life

history and habits, with special reference
to the Adirondacks of New York and the
Harvard forest. Roosevelt Wildlife Annals
2: 11-146; 1929.

Heaney, L. R. Climatic influences on

life-history tactics and behavior of North
American tree squirrels. In: Murie, J. 0.;
Michener, G. R., eds. The biology of
ground-dwelling squirrels. Lincoln, NE:
University of Nebraska Press; 1984. 43-78.

Heusser, C. J. Late-Pleistocene coniferous

forests of the Northern Rocky Mountains. In:
Center for Natural Resources, ed. Coniferous
forests of the Northern Rocky Mountains.
Missoula, MT : University of Montana
Foundation; 1969. 1-23.

Hoff, R. J.; Coffen, D. 0. Recommendations for
selection and management of seed orchards of
western white pine Res. Note INT-325. Ogden,
UT: U.S. Department of Agriculture, Forest
Service, Intermountain Forest and Range
Experiment Station; 1982. 7 p.

Hosley, N. W. Red squirrel damage to

coniferous plantations and its relation to
changing food habits. Ecology 9: 43-48; 1928

Hut chins, H. E.; Lanner, R. M. The central

role of Clark's nutcracker in the dispersal
and establishment of whitebark pine.
Oecologia 55: 192-201; 1982.

Ingles, L. G. Mammals of the Pacific States.
Stanford Univ. Press, Stanford, CA: 1965.
506 p.

Irving, L; Krog, H.; Monson, M. The metabolism
of some Alaskan animals in winter and
summer. Physiol. Zool. 28: 173-185; 1955.

Janzen, D. H. Seed-eaters versus seed size,
number, toxicity and dispersal. Evolution
23: 1-27; 1969.

Janzen, D. H. Seed predation by animals.

Annu. Rev. Ecol. Syst., 2: 465-492; 1971.

Jewett, S. G.; Taylor, W. P.; Shaw, W. T.;
Aldrich, J. W. Birds of Washington State.
Seattle, WA: University of Washington Press;
1953. 767 p.

Kemp, G. A. Keith, L. B. Dynamics and

regulation of red squirrel (Tamiasciurus
hudsonicus) populations. Ecology 51:
763-779; 1970.

Kemper, T. Notes on the breeding cycle of the
red crossbill (Loxia curvirostra) in
Montana. Auk 76: 181-189; 1959.

Kendall, K. C. Use of pine nuts by grizzly and
black bears in the Yellowstone area. In:
Meslow, E. C, ed. Fifth International
Conference Bear Research and Management,
Glacier National Park, MT 59936; 1983. 336 p

220

Kilham, L. Covering of stores by

white-breasted and red-breasted nuthatches.
Condor 76: 108-109; 1974.

Krefting, L. W.; Roe, E. I. The role of some
birds and mammals in seed germination.
Ecological Monogr. 19: 269-286. 1949.

Lanner, R . M . ; Vander Wall, S. B. Dispersal of
limber pine seed by Clark's Nutcracker. J.
For. 78: 637-639; 1980.

Laurance, W. F . ; Reynolds, T. D. Winter food
preferences of captive-reared northern
flying squirrels. Murrelet 65: 20-22; 1984.

Lavender, D. P.; Engstrom, W. H. Viability of
seeds from squirrel-cut Douglas fir cones.
Res. Note /27. Oregon State Board of
Forestry. Salem, OR; 1956.

Ligon, J. D. Foraging behavior of the

white-headed woodpeckers in Idaho. Auk 90:
862-869; 1973.

Ligon, J. D. Reproductive interdependence of
pinon jays and pinon pines. Ecological
Monogr. 48: 111-126; 1978.

Lutz, H. J. Effect of red squirrels on crown
form of black spruce in Alaska. U.S.
Department of Agriculture, Forest Service,
Alaska Forest Research Center Technical Note
42; 1958. 3 p.

McKeever, S. Food of the northern flying
squirrel in northeastern California. J.
Mammal. 41: 270-271; 1960.

McKeever, S. The biology of the golden-mantled
ground squirrel Ci te llus laterali s .
Ecological Monogr. 34: 383-401; 1964.

Mealey, S. P. The natural food habits of

grizzly bears in Yellowstone National Park,
1973-74. In: Martinka, C. J.; McArthur, K.
L., eds. Fourth International Conference
Bear Research and Management, Glacier
National Park, MT 59936; 1980. 375 p.

Morris, R. F.; Cheshire, W. F; Miller, C. A.;
Mott, D. G. The numerical response of avian
and mammalian predators during a gradation
of the spruce budworm. Ecology 38: 487-494.
1958.

Murie, 0. J. The Alaska red squirrel providing
for winter. J. Mammal. 8: 37-40; 1927.

Murie, A. The grizzlies of Mount

McKinley. Washington, D.C. : U.S. Department
of Interior, Natl. Park Ser. Sci., Monogr.
Ser. No. 14; 1981. 251 p.

Newton, Ian. Finches. New York: Taplinger
Publishing Co.; 1973. 288 p.

Odum, E. P. Ecology: the state of our

science. Ecol. Soc . Amer. Bull. 66: 14-15;
1985.

Olson, D. L.; Silen, R. R. Influence of date
of cone collection on Douglas-fir seed
processing and germination: a case history.
Res. Pap. PNW-190. Portland, OR: U.S.
Department of Agriculture, Forest Service,
Pacific Northwest Forest and Range
Experiment Station; 1975. 10 p.

Owen, D. F. How plants may benefit from the
animals that eat them. Oikos 85: 230-235;
198U.

Owen, D. F.; Wiegert, R. G. Mutualism between
grasses and grazers; an evolutionary
hypothesis. Oikos 36: 376-378; 1981.

Pank, L. G. A bibliography on seed-eating
mammals and birds that affect forest
regeneration. U.S. Department of Interior,
Fish and Wildl. Ser., Spec. Sci. Rep.,
Wildl. No. 74; 1974. 28 p.

Pulliainen, E.; Salonen, K. On eating of
pine-buds by the squirrel ( Sci urus
vulgaris) . Silva Fenn. 117: 1-22; 1965.

Radwan, M. A. Protecting forest trees and
their seed from wild mammals. Res. Pap.
PNW-6. Portland, OR: U.S. Department of
Agriculture, Forest Service, Northwest
Forest and Range Experiment Station; 1963.
28 p.

Rajala, P.; Lampio, T. The food of the

squirrel (Sciurus vulga ri s ) in Finland in
1945-1961. Abstract pp. 8-9 of Suomen Riista
16: 155-185. 1963.

Rediske, J. H. Maturation of Douglas-fir seed;
a biochemical study. For. Sci. 7: 204-213;
1961.

Rowe, J. S. Squirrel damage of white spruce.
Ottawa, Ontario: Can. Dep. Resour. and
Develop., Silvic. Leaf. No. 61; 1952. 2 p.

Rusch, D. A.; Reeder, W. G. Population ecology
of Alberta red squirrels. Ecology 59:
400-420; 1978.

Rutter, R. J. A contribution to the biology of
the gray jay (Peri soreus canadensis) . The
Can. Field-Nat. 83: 300-316; 1969.

Sanders, S. D. Foraging by Douglas tree

squirrels (Tamiasciurus douglasii; Rodentia)
for conifer seeds and fungi. Davis, CA:
University of California; 1983. 107 p.
Unpublished diss.

Shantz-Hansen, J. Red squirrel damage to

mature red pine. J. Forestry 43: 604-605;
1945.

Schmidt, W. C; Shearer, R. C. Ponderosa pine
seed — for animals or trees? Res. Paper
INT-112. Ogden, UT: U.S. Department of
Agriculture, Forest Service, Intermountain
Forest and Range Experiment Station; 1971.
14 p.

221

Schopmeyer, C. S. Seeds of woody plants in the
United States. Agriculture Handbook 450.
Washington, D. C. : U.S. Department of
Agriculture, Forest Service; 1974. 883 p.

Shaw, W. T. Moisture and its relation to the
cone-storing habit of the western pine
squirrel. J. Mammal. 17: 337-349; 1936.

Shearer, R. C; Schmidt, W. D. Natural

regeneration in ponderosa pine forests of
western Montana. Res. Paper INT-86. Ogden,
UT: U.S. Department • of Agriculture, Forest
Service, Intermountain Forest and Range
Experiment Station; 1970. 19 p. '

Sherry, D. Food storage by black-capped
chickadees: memory for the location and
contents of caches. Anim. Behav. 32:
451-464; 1984.

Smith, C. C. Structure and function of the
vocalizations of tree squirrels
(Tamiasciurus) . J. Mammal. 59: 793-808. 1978.

Smith, C. C. The adaptive nature of social

organization in the genus of tree squirrels
Tamiasciurus . Ecol. Monogr. 38: 31-63; 1968.

Smith, C. C. The coevolution of pine squirrels
(Tamiasciurus) and conifers. Ecol. Monogr.
40: 349-371; 1970.

Smith, C. C. The indivisible niche of

Tamiasciurus : an example of nonpartitioning
of resources. Ecol. Monogr. 51: 343-363;
1981.

Smith, C. C; Follmer, D. Food preferences of
squirrels. Ecology 53: 82-91; 1972.

Smith, C. C; Reichman, 0. J. The evolution of
food caching by birds and mammals. Annu.
Rev. Ecol. Syst. 15: 329-351; 1984.

Smith, C. C; Balda, R. P. Competition among
insects, birds, and mammals for conifer
seed. Amer. Zool. 19: 1065-1083; 1979.

Smith, C. F.; Aldous, S. E . The influence of
mammals and birds in retarding artificial
and natural reseeding of coniferous forests
in the United States. J. For. 45: 361-369;
1947.

Solomon, M. E. The natural control of animal
populations. J. Anim. Ecol. 18: 1-35; 1949.

Stallcup, P. L. Spatio-temporal relationships
of nuthatches and woodpeckers in ponderosa
pine forest of Colorado. Ecology 49:
831-843; 1968.

Stallcup, P. L. Hairy woodpeckers feeding on
pine seeds. Auk 86: 134-135; 1969.

Tevis, L., Jr. Autumn foods of chipmunks and
golden-mantled ground squirrels in the
northern Sierra Nevada. J. Mammal. 33:
198-205; 1952.

Tevis, L., Jr. Effect of vertebrate animals on
seed crop of sugar pine. J. Wildl. Manage.
17: 128-131; 1953.

Tisch, E. L. Seasonal food habits of the black
bear in the Whitefish Range of northwestern
Montana. Missoula, MT : Montana State
University; 1961. 100 p. M.S. Thesis.

Tomback, D. F. Foraging strategies of Clark's
Nutcracker. Living Bird 16: 123-161; 1977.

Tomback, D. F. How nutcrackers find their seed
stores. Condor 82: 10-19; 1980.

Tomback, D. Notes on cones and

vertebrate-mediated seed dispersal of Pinus
albicaulis (Pinacae) . Madrono 28: 91-94;
1981.

Tomback, D. F. Dispersal of whitebark pine
seeds by Clark's Nutcracker: a mutualism
hypothesis. J. Anim. Ecol. 51: 451-467; 1982.

Turcek, F. J.; Kelso, L. Ecological aspects of
food transportation and storage in the
Corvidae. Commun. Behav. Biol. 1: 277-297;
1968.

Udvardy, M. D. F. Birds of the coniferous
forest. In: Center for Natural Resources,
ed . Coniferous forests of the Northern Rocky
Mountains. Missoula, MT : University of
Montana Foundation; 1969. 99-130.

Vander Wall, S. B.; Balda, R. P. Co-adaptations
of the Clark's nutcracker and the pinon pine
for efficient seed harvest and dispersal.
Ecol. Monogr. 47: 89-111; 1977.

Vander Wall, S. B.; Hoffman, S. W.; Potts, W.
K. Emigration behavior of Clark's
nutcracker. Condor 83: 162-170; 1981.

Wagg, J. W. B. Viability of white spruce seed
from squirrel cut cones. For. Chron. 40:
98-110; 1964.

Walters, J.; Soos, J. Douglas squirrel damage
to Douglas fir. Res. Notes. Faculty of
Forestry, No. 32. Vancouver, Canada:
University of British Columbia; 1961. 2 p.

Welty, J. C. The life of birds, 2nd. ed.

Philadelphia, PA; W. B. Saunders Co.; 1975.
623 p.

West, N. E. Rodent-influenced establishment of
ponderosa pine and bitterbrush seedlings in
central Oregon. Ecology 49: 1009-1011; 1968.

222

COLLE

Jeanne L.J^Pedro White and Monte D.J^White

SQUIRREL BEHAVIOR INFLUENCES QUALITY OF CONES AND SEEDS
COLLECTED FROM SQUIRREL CACHES—FIELD OBSERVATIONS^

ABSTRACT: Field observations were made in north-
central Idaho during the 1980 through 1984 seed
procurement seasons. Squirrel behavior affects
the quality, location, species, maturity, and
viability of seeds collected. This paper discusses
some aspects of collecting seed cones for
reforestation from squirrel caches. With proper
care in collecting and handling cones, good quality
seed can be obtained.

INTRODUCTION

The authors have observed the behavior of pine
squirrels during cutting and caching of coniferous
seed cones since 1980. Observations were made on
the Pierce Ranger District of the Clearwater
National Forest and surrounding private forest
lands. This paper discusses conventional opinions
about squirrel cache collection of seed cones
for reforestation. Cache location, cone handling,
and relative collection costs are also discussed.

CONVENTIONAL OPINION

Opinions differ on the acceptability of seed
collected from squirrel caches. The most prevalent
of these and our responses to them are:

Seed Maturity

Squirrels tend to harvest cones before they are
mature. Response: Seeds collected from squirrel
caches on the Pierce Ranger District had viability
ranging from "acceptable" to "excellent", as shown

Paper presented at the Conifer Tree Seed in the
Inland Mountain West Symposium, Missoula, MT,
August 5-6, 1985.

Jeanne L. Pedro White is a Forestry Technician,
Pierce Ranger District, Clearwater National
Forest, USDA Forest Servcie, Kamiah, ID.; Monte D.
White is a Forester, Clearwater County, Orofino, ID.

in table 1. These data were obtained from germina-
tion test records of the Coeur d'Alene Nursery
located in Coeur d'Alene, ID.

Seed Sources

Seed sources are uncertain. Response: Individual
source trees cannot be determined with certainty
when collecting cones from squirrel caches. However,
while checking cone crops of potential collection
sites, the authors observed that the majority of
cones are to be found in the dominant and codominant
trees in even-aged stands. When collecting cones,
most squirrel activity was found in those trees.
This is probably due to advanced maturity and relative
abundance of cones found in the upper canopy. One can
be relatively certain that most of the cones found in
caches located within the stand were borne by those
trees, although some could have been imported from
adjacent stands. To ensure genetic diversity, we
collected from at least seven separate cache sites
within stands. Collection site desirability can thus
be based on phenotypic characteristics of the
larger stand components.

Squirrels Winter Food Supply

Collection from caches deprives squirrels of
their winter food supply. Reponse: Squirrels
collect far more cones than they can eat. If
available, they will cache seeds from a variety
of plants, then fail to find all of the caches.
Individuals dying before winter obviously do not
return to consume their caches. The authors
have found old caches which have evidently remained
untouched for years.

Only a portion of the squirrel caches will be
found by human collectors. An example of this is
that on one occasion cone pickers assigned to
collect cones from the tops of felled trees didn't
collect all the cones before leaving for the
evening. Returning the following day, they found
all the cones missing from the trees with no nearby
caches to account for them.

223

Table 1. — Viability of seeds collected from squirrel caches on the Pierce Ranger District, Clearwater
National Forest, ID, 1980-1983

No. of Seedlots

Species

Average
Viability

Range of
Viability

Seedlots Below
Minimum
Acceptable
Viability^

1 QftO

US!

7Q _ Ql"/

U

4

ES

91%

77 - 97%

0

7

GF

77%

56 - 88%

0

1981

: 6

LP

77%

42 - 90%

1

1982

: 5

DF

95%

92 - 98%

0

1983

: 1

LP

88%

0

lAc

ceptable minimums are:

Douglas

Fir (DF) = 70%

Grand Fir (GF) = 50%

Engelman

Spruce (ES) = 60%

Lodgepole Pine (LP) = 70%

FINDING THE CACHES

The art of locating squirrel caches is acquired
primarily through experience. Favorite cache
sites include: small ground depressions
(including animal tracks); cavities in and around
logs, stumps, roots, and rocks; moist seeps;
along the banks of small creeks; in and around
structures such as fences, outbuildings, and
foundations. Squirrels may enlarge the storage
locations. Cones may be concealed beneath wet moss,
cone scales, needles, or other debris and there
may be layers of this type material between pockets
of cones within the storage locations. Cache
volume will range from a few cones to more than a
bushel. One will not always be able to find specific
types of caches or species of cones in a given area.
Squirrels seem quite opportunistic in caching
behavior, taking advantage of whatever storage
areas are available.

Squirrels will sometimes carry cones amazing
distances to be cached. On one occasion, a cone
cache was found along a stream bank and the
nearest seed source was 10 to 12 chains (200-240
meters) away. Many times squirrels can also
be seen carrying cones across roads and skid
trails. In most cases, though, cones are cached
near or below the source trees.

OTHER CONSIDERATIONS

1) During years of abundance, squirrels seem to
prefer true firs and Douglas-fir over pines in
mixed stands. White pine cones seem to be less
desirable when other species are available in
collectible quantities.

2) Cone collection from squirrel caches does not
result in damage to the source trees . Human
collection results in tree damage. Damage ranges
from incidental bark gouging and branch breakage,
when using "cherry pickers" or climbing spurs, to
seed tree destruction, when trees are felled prior
to cone picking.

3) Cone collection costs will usually be lower if
cones are obtained from caches than by more labor-
and equipment- intensive means.

PROPER CONE HANDLING IS IMPORTANT

Cached cones are often wet, dirty, and/or insect-
damaged. With proper on-site sorting, the collector
can dispose of most undesirable cones. If cones are
wet, storage of fewer cones per bag will speed
drying time and prevent molding. If sound collection
procedures are followed, good quality cones will
usually be received at the collection station.

I

CONCLUSIONS

Squirrel caches are now a major source of reforesta-
tion seed on the Pierce Ranger District. Viability
tests from the Coeur d'Alene Nursery on squirrel-
cached seed show that mature, viable seed can be
obtained in this area using the cache collection
method. Caches can be found in diverse locations
although, at times, they may be difficult to locate.
Cache location experience combined with proper cone
handling during and after collection will usually
result in high-quality reforestation seed in storage.

224

^ INSECTS AND CONIFER SEED PRODUCTION

Gordon

ABSTRACT: Hundreds of insects are associates of
cones and seeds in the Inland Mountain West, but
only a few cause economic damage. Damage by
insects is one of the major impediments to seed
production in Douglas-fir, spruces, ponderosa
pine, western white pine, and western larch,
based on the few published damage surveys from
the region. Each tree species has its own
complex of associated insects. Salient features
of the bionomics to development of pest
management systems are discussed. Major
short-comings in the management of insects
affecting seed production are the lack of
methods for monitoring pest populations and *he
limited options in pest control.

INTRODUCTION

Cone and seed insects are major impediments to
seed production by many conifers. Hundreds of
insect species are known associates of cones in
the Inland Mountain West (Keen 1958; Hedlin
1974; Kulhavy and others 1975; Hedlin and others
1980) . These include insects that feed on cones
and seeds, parasitoids and predators, and
species of unknown feeding habit. For example,
at least 67 insect species were reared from
ponderosa pine cones in New Mexico in 1964-67
(Kinzer and others 1972a). Of these, 33 species
fed on cones and seeds, 43 were parasitoids, 10
were predators, and 21 were of unknown feeding
habit.

There have been 22 published reports of damage
by cone and seed insects in the Inland Mountain
West (table 1), of which 13 were general surveys
for damage to various conifers while the others
were concerned with specific insects. Several
surveys have reported damage as the percentage
of cones attacked rather than the percentage of
seeds damaged. Such surveys are difficult to
interpret with regard to importance of insect
pests because some insect groups, e.g., scale
midges, may infest a large proportion of a cone
crop but, unless large numbers are present and
cones or scales are killed, cause little or no
seed damage. Some of the surveys have covered
many sites for only one year (Allen and Ruth
1969; Ruth and others 1980), while others

Paper presented at the Symposium on Conifer Tree
Seed in the Inland Mountain West, Missoula, MT .
August 5-6, 1985.

G. E. Miller is Research Scientist, Pacific
Forestry Centre, Canadian Forestry Service,
Victoria, B.C.

IN THE INLAND MOUNTAIN WEST: A REVIEW

* 8 ~ 7

E. jMiller

have covered only one site but for several years
(Barnes and others 1963; Schmid and others 1981;
Jenkins 1983) .

The importance of cone and seed insects in the
region varies among conifers (table 1) . Heavy
seed losses (up to 100%) have been reported in
the literature for most of the tree species for
which damage surveys have been conducted, i.e.,
Douglas-fir, ponderosa pine, western white pine,
and western larch. To date losses reported in
true firs and western red cedar have been less
than 30%, and lodgepole pine and western hemlock
have suffered very little seed loss. However,
the number of site-years for which seed losses
have been reported is very limited (table 1) .
Even in Douglas-fir, the most intensively
sampled species, 81 of the 108 site-years
reported were carried out in the interior of
British Columbia in 1980. The lack of adequate
survey makes assessment of the general
importance of cone and seed insects, as well as
the relative importance of individual species,
in the seed production by some conifers
difficult. The seed losses (table 1) may not
reflect maximum potential losses for all
conifers. For example, Dewey and Jenkins (1980)
reported that at one site 35% of lodgepole pine
cones were infested by coneworms and scale
midges. Seed losses undoubtedly occurred, but
were not measured, so the loss value of 0
reported in table 1 is not a true indication of
potential damage. Unpublished data from the
interior of British Columbia showed that more
than 75% seed loss can occur in subalpine fir.
Due to difficulties in damage identification for
some pests, most notably seed bug (Krugman and
Koerber 1967), estimates of damage by these
pests are probably conservative. The need for
further surveys is obvious, especially for true
firs, lodgepole pine, western larch, hemlocks,
and cedars .

Seed losses vary dramatically among sites and
years (table 1), For example, the percentage of
seed damaged in the Cariboo Region of British
Columbia in 1979 varied from less than 1% to 75%
for spruce at 23 sites and from 13 to 100% for
Douglas-fir at 19 sites (Allen and Ruth 1979).
Variation at individual sites is largely due to
the irregular fluctuations in cone crop size
(Mattson 1971; Forcella 1978, 1980; Miller and
others 1984) . When large cone crops are
preceded by light or nil crops, the size of
insect populations is limited so that large
crops usually suffer low levels of insect
damage. Where large cone crops occur closely
together losses are more severe. For example, a

225

Table 1. — Cone and seed insect damage reported from the Inland Mountain West and
estimated seed losses

Range

Number of seed loss

Tree species Surveys site-years (%)

Douglas-fir
(PseudotsuRa menziesii
(Mirb . ) Franco)

Clark and others 1963*; Dewey 1972*; Allen and
Ruth 1979; Dewey and Jenkins 1979* , 1980* , 1982* ;
Ruth and others 1980; Shearer 1984

108

0-100

Grand fir

(Abies srandis (Dougl.)
Lindl. )

Pfister and Woolwine 1963*; Kulhavy and Schenk
1976a*. b; Dewey and Jenkins 1980*, 1982*

30

0-19

Subalpine fir

(Abies lasiocarpa (Hook.)

Nutt.)

Kulhavy and others 1976; Dewey and Jenkins 1982*

29

Western larch

(Larix occidentalis Nutt.)

Dewey and Jenkins 1980*, 1982*, Ruth and others
1980; Shearer 1984

10

0-40

Spruces

(Picea enselmanni Parry)
P. Rlauca (Moench) Voss)

Hedlin 1973; Allen and Ruth 1979; Ruth and others
1980; Schmid and others 1981; Dewey and Jenkins
1982*

40

4-100

Lodgepole pine

(Pinus contorta Dougl.)

Piny on pine
(Pinus edulis Engelm.
P. monophylla Torr. &
Frem. )

Dewey and Jenkins 1979*; 1980*; Ruth and others
1980

Forcella 1978+, 1979+

14

<5-90

Ponderosa pine

(Pinus ponderosa Laws)

Kinzer and others 1972a; Dale and Schenk 1978+;
Dewey and Jenkins 1979*, 1980*. 1982*; Ruth and
others 1980

40

<1-100

Western white pine
(Pinus monticola Dougl . )

Barnes and others 1962+; Williamson and others
others 1966+; Schenk and Goyer 1967; Dewey and
Jenkins 1980*, 1982*; Jenkins 1983+

16

5-98

Western red cedar
(Tsuga heterophylla (Raf.)
Sarg. )

Ruth and others 1980

0-20

Western hemlock
(Thuja plicata Donn)

Ruth and others 1980

Damage reported as percentage of cones infested only, seed losses not reported.

+ Seed loss can be estimated from the percentage of cones infested for pests that kill whole
cones, e.g., Conophthorus spp .

very heavy crop of Douglas-fir cones occurred
around Keremeos, British Columbia, In 1983. Most
of the Douglas-fir cone moth (major pest)
population remained in diapause in 1984 when a
very light crop was produced. The moderate 1985
cone crop was heavily infested by the moth.

Stand density may affect seed loss and pest
complex. Seed losses are higher in less dense
stands in western white pine (Barnes and others
1961; Schenk and Goyer 1967). In ponderosa
pine, cone beetle is more damaging in open
stands whereas seed moth is more prevalent in
dense stands (Dale and Schenk 1978) .

226

Table 2. — Most damaging cone and seed insects associated with conifers in the Inland Mountain West

Pest species no . 1

Conifer

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29

Douglas- fir
Grand fir
Subalpine fir
White fir

*

Western larch

Spruces

Pinyon pine
Ponderosa pine
Western white pine

Western red cedar

* *
*

1 See table 3 for species identification.

THE PESTS
Pest Species

Only a few insect species that feed on cones and
seeds of each conifer are significant pests.
Based on the damage surveys, as well as other
sources (Keen 1958; Hedlin 1974; Hedlin and
others 1980) , the important pests on conifers in
the Inland Mountain West are listed in tables 2
and 3. Cone and seed insect pests represent a
wide variety of insect groups, each with its own
biological characteristics. Not all pests have
been identified, e.g. the most damaging insects
on western larch, a wooly aphid and a scale
midge (Shearer 1984).

Since most pest species are host species — or
genus- specif ic — each conifer has its own
specific complex of pests. Notable exceptions
are the fir coneworm, western conifer seed bug,
and western budworm. The relative importance of
each pest varies throughout host range. For
example, spruce seed moth and spruce cone maggot
occur throughout the range of spruces in the
region but the moth is most damaging in the
south and the maggot in the north. Ponderosa
pine cone weevil occurs sporadically within the
range of its host but has done significant
damage at some sites (Bodenham and others
1976). The complex of pests attacking a conifer
should therefore be determined at each cone and
seed production site.

Insects also damage pollen cones (Dale and
Schenk 1979; Hedlin and others 1980) but
apparently do not reduce pollen availability.

Pest Biology

The bionomics of most of the economically
important cone and seed injects are known to

some extent (table 4) and are summarized
elsewhere (Hedlin and others 1980) . All
species, except the seed bug, complete their
feeding over one summer. The seed bugs feed as
nymphs and young adults over one summer and as
mature adults over the next summer. Aspects of
biology that are significant considerations in
the development of pest management systems
(table 5) include stage(s) of cone attacked and
damaged, methods of mate and host location,
rates of survival, amount of damage per insect,
location of insects at cone harvest, and
capacity for prolonged diapause.

Stage(s) of cone attacked and damaged determine
the timing of insecticide applications. Female
cone beetles girdle and kill conelets prior to
oviposition so an insecticide should be applied
when females are active. Damage by seed
chalcids is done by larvae but, because chalcid
larvae are protected inside conelets, an
insecticide should be applied at the time of
oviposition to kill adult wasps. Many pests
such as cone moths, cone maggots, and some seed
moths, oviposit on open strobili but damage
occurs after egg hatch. Against these insects,
insecticides may be applied as larvicides when
the conelets have closed and are turning down or
as adulticides when strobili are open to receive
po 1 len .

Knowledge of survival rates and amount of damage
per insect allows for development of damage
prediction systems based on counts of an insect
stage, notably eggs, that occurs prior to
damage. This has been done for Douglas-fir cone
gall midge (Miller 1984).

Identification of cues used in mate and host
location may provide tools useful in population
monitoring. The utilization of pheromones in

227

Table 3. — Insect pests of cones and seeds in the Inland Mountain West, their common
names and species number in table 2

Insect

Coleoptera: Curculionidae

Conotrachelus neomexicanus Fall

Scolytidae
Conophthorus edulis Hopkins
C. ponderosae Hopkins

Diptera: Anthomyiidae

Lasiomma anthracina (Czerny)
L. abietis (Huckett)

Cecidomyiidae
Dasyneura abiesemia Foote
Mayetiola thu.i ae (Hedlin)

Lonchaeidae
Earomyia abietum Mc Alpine

Hemiptera: Coreidae

Leptoglossus occidentalis Heidemann

Hymenoptera: Torymidae

Megastismus albif rons Walker

M. lasiocarpae Crosby

M. pinus Parfitt

M. rafni Hoffmeyer

M. spermotrophus Wachtl

Lepidoptera: Olethreutidae

Barbara coif axiana (Kearfott)
Barbara sp .

Cydia bracteatana (Fernald)
C. miscitata (Heinrich)
C. piperana (Kearfott)

C. strobilella (L.)
Eucosma bobana Kearfott
E. rescissoriana Heinrich

Pyralidae
Dioryctria abietivorella (Grote)

D. albovittella (Hulst)
D. aurenticella (Grote)
D. pseudotsugae Munroe

D. reniculelloides Mutuura
& Munroe

Tortricidae
Choristoneura occidentalis Freeman

Unidentified species

mating behavior occurs in many cone and seed
insects. Sex attractants have been identified
for several Inland Mountain West insects,
namely: Cydia piperana, Eucosma bobana, E.
ponderosae , E. recissoriana , Dioryctria
pseudotsugella and Barbara ulteriorana (Sartwell
and others 1984), Barbara coif axiana (Hedlin and

Species

Common name number

pine cone weevil 1

pinyon cone beetle 2
ponderosa pine cone beetle 3

spruce cone maggot 4
fir cone maggot 5

fir seed midge 6
western redcedar cone midge 7

fir seed maggot 8

western conifer seed bug 0

ponderosa pine seed chalcid 10

11

fir seed chalcid 12

13

Douglas-fir seed chalcid 14

Douglas-fir cone moth 15

fir cone moth 16

fir seed moth 17

18

ponderosa pine seedworm 19

spruce seed moth 20

pinyon cone borer 21

lodgepole pine cone borer 22

fir coneworm 23

24

ponderosa pine coneworm 25

26

spruce coneworm 27
western spruce budworm 28

29

others 1983), and Cydia strobilella (Roelofs and
Brown 1982; Booij and Voerman 1984). Pests
known to or suspected of using pheromones but
for which sex attractants have not yet been
identified include Conophthorus ponderosae and
Leptoglossus occidentalis (Kinzer and others
1972b; Dale and Schenk 1979).

228

Table 4. — Literature on bionomics of important cone and seed insects in the Inland
Mountain West

Insect

Rererence

Coleoptera:

cone weevil
cone beetles

Bodenham and others 19 76

Williamson and others 1966; Kinzer and others 1970, 1972a;
Dale and Schenk 19 79

Diptera:

seed midge 2
seed maggots 2
cone maggots
cone midge

Hemiptera:

seed bug2

Hymenoptera:

seed chalcids

Lepidoptera:

cone moths
budworm
seed moths

coneworms^
cone borer

Hedlin 1967b
Hedlin 1967b
Hedlin 1973
Hedlin 1964b

Koerber 1963

Hussey 1955; Hedlin 1967b; Kinzer and others 1972a

Hedlin 1960; Clark and others 1963; Hedlin and Ruth 1974
McKnight 1968

Tripp 1954; Hedlin 1967a; Kinzer and others 1972; Hedlin
1973; Dale and Schenk 1979; Schmid and others 1981
McLeod and Daviault 1963; Dale and Schenk 1979
Ollieu and Goyer 1966

1 In addition to the detailed accounts listed, Keen (1958) gives some information
on most groups .

2 The bionomics are well known for only a few groups, but information on these
groups is particularly lacking.

Insects may be attracted to cones through
host-produced volatiles (olfaction), vision, or
most likely both. Pine, spruce, and larch
strobili produce unique blends of volatiles
(Borg-Karlson and others 1985) which insects
could use in finding their host. Kinzer and
others (1972b) reported that presence of
ponderosa pine conelet increased attraction to
the opposite sex in both male and female
Conophthorus ponderosae and that females were
attracted to host tree resins.

However, Mattson
that branch sele
cone location wa
resinosae Hopkin
cone let-produced
significance in
cone maggot uses
locate strobili
host- associated
and seed insects

and others (1934) indicated
ction within a host tree during
s random in Conophthorus
s, suggesting that

volatiles are of little
these scolytids. The spruce

vision, at least in part, to
(Roques 1984). The role of
stimuli in host location by cone

has yet to be determined.

Insect location at cone harvest and capacity for
prolonged diapause (table 5) influence size of
infestation. In seed orchards, insects which
are in cones at the time of harvest are removed
with the cones so that the insects must invade

orchards each year for infestations to occur.
Insects which are not in the cones at harvest
will continually be on site. Damage will
probably be greater where resident populations
occur. Methods of monitoring, e.g. trap
location, may differ between migrating and
resident populations. Adult emergence can occur
after diapause for two or more winters. This
emergence can augment emergence of adults from
the preceeding year's crop, resulting in greater
damage (Hedlin 1964a) .

The bionomics of some cone and seed pests are
not well known (tables 4 and 5) . The bionomics
of all pests at a site must be known if pest
management systems are to be effective.

PEST MANAGEMENT

Monitoring Pest Populations

Schedules for applying insecticides can be fixed
(i.e., by calendar date or host phenology),
flexible (i.e., when an important biological
event, such as peak of oviposition is predicted
to occur) or on a treat-as-necessary basis.
Pest monitoring has no application in
f ixed- schedule spraying but is an important

229

Table 5. — Aspects of bionomics with implications for pest management and control

Average

Host Damage number seeds Location at Prolonged

Insect stage attacked by damaged/ insect cone harvest diapause

Coleoptera:
cone weevil
cone beetles

Diptera:

seed midge
seed maggots
cone maggots
cone midge

Hemiptera:
seed bug

2nd yr conelets
2nd yr conelets

open strobili
open strobili
open strobili
open strobili

all

larva
adult

larva
larva
larva
larva

adult

1

?

30
?

duff
cones on ground

cone
cone
duff
cone

?

yes

yes
yes
yes
yes

tree

Hymenoptera:

seed chalcids

2nd yr conelets (pines) larva
conelets^ (others)

cone

yes

Lepidoptera:
cone moths
budworm
seed moths

cone worms

cone borers

open strobili larva

all larva

2nd yr conelets (pines) larva

open strobili (others) larva

all (pines) larva

o

conelets"1 (others) larva

2nd yr conelets larva

15

?

12
14

Ll
?

9

cone
tree
cone
cone
cone/?
?

cone

yes

no

yes

yes

no

no

?

Number cones/insect
* 1 month after pollination

component in the latter two schedules . Cameron
(1984) reviews the experiences with each
approach in southern pine seed orchards. The
main purpose of monitoring insect populations is
the achievement of maximum control of the target
pest with least amount of insecticide. Fixed
schedule spraying is the least efficient way of
utilizing insecticides and treating when
necessary the most efficient because cone and
seed insect numbers and damage vary dramatically
among years and sites and control is not always
warranted. Therefore, monitoring pest
populations is desirable in most situations
where control actions are contemplated.

Monitoring may consist of simply detecting pest
presence through to quantification of pest
populations and predicting damage. An effective
monitoring system should allow for proper timing
of insecticide applications and, when possible,
for prediction of damage from the observed level
of infestation.

Adult trapping with pheromone/sex attractant
lures holds considerable potential for
monitoring cone and seed insects. To date, none
of the known attractants has been developed into
a monitoring scheme. However, sex

attractant-based monitoring systems have been
developed for insect pests in deciduous fruit
orchards (Madsen and others 1975; Madsen and
Peters 1976; Vakenti and Madsen 1976) and it
should be possible to develop similar systems
for cone and seed insects, especially for use in
seed orchards and seed production areas.

A potential problem in using sex attractant
lures in seed orchards is trap placement.
Because the insects must migrate into the
orchards every year and mating probably takes
place prior to migration, traps may need to be
placed within or on the edge of nearby forest
stands, which is sometimes a problem. Also, it
is possible that not all cone and seed insects
utilize pheromones in mating behavior. For
example, the courtship behavior of Megastigmus
spp. has been observed (Hussey 1955; Orr and
Borden 1983) but no mention was made of
pheromones. In 1984, attempts to trap M. pinus
with virgin insects near Victoria, British
Columbia failed.

Host attractants hold similar potential for
monitoring adult populations. Host attractants
have the following advantages over sex
attractants: i) they should be attractive to

230

several species rather than being
species-specific; ii) they should attract
females whereas in most species males are
attracted to sex attractants, and counts of
females should provide a more direct estimate of
potential damage; and iii) they could be used in
seed orchards, thus avoiding the potential
problem of trap location associated with sex
attractant traps. Visual traps which simulate
hosts and which also emit host volatiles are
used to monitor populations of some pests in
deciduous fruit orchards (Prokopy and Hauschid
1979; Reissig and Tette 1979) and such traps may
be best for cone and seed insects.

Blacklight traps have been used to monitor
coneworm populations in southern pine seed
orchards (McLeod and Yearian 1979, 1982). They
are most effective for moths; many other
insects, including seed bugs (Yates 1973), are
not attracted. Since these traps are not
selective they can catch large numbers of many
non-target lepidopteran species, resulting in
difficulties in specimen identification. They
are not considered practical for operational use
in southern pine seed orchards (Cameron 1984).

Sampling schemes for quantifying Douglas-fir
cone moth egg populations and determining the
need for applications of systemic insecticide
are currently being developed in British
Columbia. Similar schemes are needed for
populations of other pests, especially for
insects which can be monitored prior to
occurrence of damage, such as those which
oviposit on open strobili. The current lack of
monitoring schemes is a major gap in the
development of pest management systems.

Control Techniques

Currently, insecticides are the only practical
technique for controlling cone and seed insect
populations or damage. However, in seed
orchards, certain cultural practices can reduce
seed losses. For instance, annual removal of
insect populations in harvested cones, including
unwanted cones on rootstocks or crops too small
to manage for seed production, should reduce
insect damage. Leaving cones in orchards allows
resident insect populations to build up and
results in heavier seed losses. This has
happened in coastal Douglas-fir seed orchards
(Miller 1984). Similarly, cone-beetle infested
cones lying on the ground should be removed from
orchards. Delaying reproductive bud burst
through overhead misting with cold water for
reducing pollen contamination may also reduce
insect damage in some years (Miller 1983b) .
However, the effects of this technique on damage
by insect pests in the Inland Mountain West have
not been evaluated.

Insecticides . - Many contact and systemic
insecticides have been tested for cone and seed
insect control but only a few are effective.
For example, of 26 insecticides tested on
Douglas-fir cones only 11 reduced insect
populations or damage by 90% or more (Miller

1980). Because of timing of sprays, contact
insecticides can only be used as preventatives
whereas systemic insecticides can be used after
the need for insecticide application has been
established, at least for insects that oviposit
on open strobili. Systemic insectides have a
further advantage in the number of ways that
they can be applied. Systemics can be applied
as sprays, injections, paint- ons, or incorporated
into the soil whereas contacts can only be
applied as sprays.

Insecticides effective against inportant cone
and seed insects in the Inland Mountain West are
listed in table 6. Significant increases in
seed production after insecticide application
have also been reported by Stipe and Green
(1981) and Reardon and others (1984). There are
few reports of insecticide efficacy against
pests on conifers other than Douglas- fir, and
even against some pests on Douglas- fir, such as
seed bug. In Canada, azinphosmethyl and
dimethoate are registered for use on Douglas fir
(Agriculture Canada 1982) and dimethoate is
currently being registered for spruce cone and
seed insects. In the United States,
azinphosmethyl, dimethoate, Pydrin® and
oxydemetonmethyl were registered for specific
uses against Douglas- fir cone and seed insects
(Overhulser and Sandquist 1985) and fenvalerate
is registered for use on western white pine.
Obviously options in pest control are limited.

Research elsewhere on cone and seed pest control
may be applicable in the Inland Mountain West.
For example, insecticides have been evaluated
against seed bug and coneworms of southern pines
(DeBarr and Nord 1978; DeBarr and Fedde 1980;
Nord and DeBarr 1983; Nord and others 1984)
which are related to the species in the West.
These evaluations in the south provide a list of
most suitable candidate insecticides for trial
against western species.

Efficacy of various available application
methods has not always been consistent,
especially for aerial sprays and injections.
Aerial sprays have sometimes been more than 90%
effective (Miller and Hutcheson 1981) but have
also been totally ineffective (Johnson and
Winjum 1960; Johnson 1963). These "one- shot"
trials have not been attempts at systematic
technique development. In southern pine seed
orchards, frequency of aerial applications is
increasing (Barber 1984), suggesting
satisfactory results are being obtained.
Injections have been tested for effectiveness in
reducing insect damage in Douglas- fir (Koerber
and Markin 1984; Reardon and others 1984),
spruce (Fogal and Lopushanski 1984) and southern
pines (Merkel and DeBarr 1971), but again the
results have ranged from excellent to poor. Two
causes of the variation are poor uptake of the
insecticide by individual trees and uneven
distribution of insecticides within the tree
crown (Koerber and Markin 1984). With proper
timing, ground- based sprays have generally
resulted in consistent effective control,
provided sprays are not affected by rain (Miller

231

1983a; Nord and others 1984). However,
ground-based sprays are limited by tree height
and size of the area being treated.
Incorporation of insecticides into soil has been
effective in southern pine seed orchards (DeBarr
1978) and requires only one application per
year, compared to several sprays for coneworm
and seed bug control (Neel 1980) . As with
injections, level of control has varied (Nord
and others 1979). Soil incorporation trials
have not been reported from the West. More
technique evaluation and development, as
occurred in southern pine seed orchards
(reviewed by Barber (1984)), is needed in our
region if consistent and effective control of
cone and seed insects is to be achieved in the
future.

Not all application methods are equally
effective against all pests. Soil incorporation
of carbofuran controlled several pests on
eastern white pine but not seed chalcid (DeBarr
and others 1982). Likewise, timing of
application for one pest may not control
another. For example, injecting or spraying
Douglas-fir when strobili were open or turning
down controlled Douglas-fir cone gall midge and

cone moth but not seed chalcid (Koerber and
Markin 1984; Miller unpubl . ) . These points
should be remembered when developing pest
management systems .

The most practical method of insecticide
application will vary with the type of stand to
be treated. Protection of crops in forests is
only practical through aerial applications
because of tree size and variable terrain.
Protection of seed crops on plus trees, being
individual trees within forest stands, is
possible through injecting insecticide and
aerial spraying. In seed production areas
(depending on terrain) and especially in seed
orchards, all methods of application are
possible. Provided the trees are not too tall
and the area being treated is small,
ground-based sprays or soil incorporations are
likely the most cost effective. Otherwise, only
aerial applications are practical. Injections
are too expensive and time-consuming to be used
in the treatment of whole orchards or seed
production areas.

In some conifers, pest control can be relatively
simple; a single application of systemic

Table 6. — Insecticides demonstrated to be 90% effective against insect pests of cones and seeds in
the Inland Mountain West

Insecticide

Tree

type

Application
method

Pest

Reference

Douglas-fir

contact

systemic

spruce

azinphosmethyl

spray

cone

moth

Johnson and Winjum (1960)

spray

coneworm

Cade (1977)

DDT

spray

cone

moth

Rudinsky (1955)

spray

seed

chalcid

Rudinsky (1955)

lindane

spray

coneworm

Cade (1977)

malathion

spray

seed

chalcid

Stoakley (1973)

acephate

spray

coneworm

Cade (1977)

dicrotophos

spray

cone

moth

Meso (1975)

injection

cone

moth

Meso (1975)

carbofuran

spray

coneworm

Cade (1977)

dimethoate

spray

cone

moth

Hedlin (1966)

spray

coneworm

Cade (1977)

spray

seed

chalcid

Hedlin (1966)

injection

cone

moth

Meso (1975)

injection

seed

chalcid

Meso (1975)

f enitrothion

spray

seed

chalcid

Hedlin (1966)

oxydemetonmethyl

spray

cone

moth

Meso (1975)

injection

cone

moth

Schenk and others (196 7)

injection

cone

moth

Koerber and Markin (1984)

phorate

spray

cone

moth

Johnson and Winjum (1960)

dimethoate

spray

cone

maggot

Hedlin (1973)

oxydemetonmethyl

spray

cone

maggot

Hedlin (1973)

spray

seed

moth

Hedlin (1973)

injection

seed

moth

Fogal and Lopushanski (1984)

injection

cone

maggot

Fogal and Lopushanski (1984)

dicrotophos

injection

seed

moth

Fogal and Lopushanski (1984)

injection

cone

maggot

Fogal and Lopushanski (1984)

western
white pine

contact permethrin

spray

cone beetle

Shea and others (1984)

232

insecticide controls both major seed pests (seed
moth and cone maggot) of spruce (Hedlin 1973).
In conifers attacked by only one pest species,
such as western red cedar and western white pine
at some sites, control should be relatively
easy. In other conifers which are attacked by
different pests at different stages of cone
development, e.g. Douglas-fir and ponderosa
pine, or by pests that Initiate attacks on cones
and seeds over several weeks or months, e.g.
coneworms and seed bug, control of damage may be
more difficult to achieve consistently and may
require multiple insecticide applications.

Biological control. — The prospects for
biological control appear limited, even though
some insects, such as Douglas-fir cone moth
(Hedlin 1960), suffer heavy losses to natural
enemies because pest damage is usually completed
prior to pest death caused by these mortality
factors. Entomophagous fungi may provide an
alternative to chemical insecticides (Timonin
and others 1980) but further research is needed
in evaluating these pathogens and in application
techniques. Research in the southern U.S. has
shown that Bacillus thuringiensis is another
potential control agent for coneworms (McLeod
and others 1982) .

In summary, seed crops protected from one pest
may suffer damage by other pests, or damage may
be caused by a less important pest because no
insecticide applications were made due to low
levels of infestation by key pests (Cameron
1984; Miller 1984). Obviously pest management
systems should be developed to include all pests
associated with a conifer. Considerably more
research is needed before pest management
systems will be available for cone and seed
pests in the Inland Mountain West.

REFERENCES

Agriculture Canada. Compendium of pesticides

registered for use in Canada. Registered pest
control products. Publ. 1954 RP/82. Ottawa,
ON: Agriculture Canada, Plant Products and
Quarantine Directorate, Pesticides Division;
1982 .

Allen, S. J.; Ruth D. S. Cone and seed insects,
Cariboo Forest Region. Pest Report. Victoria,
BC: Canadian Forestry Service, Pacific Forest
Research Centre. 1979. 3 pp.

Barber, L. R. Insecticide application methods
for southern pine seed orchards. In: Yates,
H. 0. Ill, ed. Proceedings of the cone and
seed insects working party conference,
Working Party S2. 07-01, International Union
of Forestry Research Organizations; 1983 July
31-August 6, Athens, GA. Asheville, NC: U.S.
Department of Agriculture, Forest Service,
Southeastern Forest Experiment Station; 1984:
130-136.

Barnes, B. V.; Bingham, R. T. ; Schenk, J. A.
Insect- caused loss to western white pine

cones. Res. Note No. 102. Ogden, UT: U.S.
Department of Agriculture, Forest Service,
Intermountain Forest and Range Experiment
Station; 1961. 7 pp.

Bodenham, J.; Stevens, R. E.; Thatcher, T. 0.
A cone weevil, Conotrachelus neomexicanus , on
ponderosa pine in Colorado: Life history,
habits, and ecological relationships
(Coleoptera: Curculionidae) . Can. Entomol.
108: 693-699; 1976.

Borg-Karlson, A. K. ; Eidmann, H. H.; Lindstrom,
M. ; Norin, T.; Wiersma N. Odiferous compounds
from the flowers of the conifers Picea abies,
Pinus sylvestris and Larix sibirica .
Phytochemistry 24: 455-456; 1985.

Cade, S. C. Personal communication. Hot Springs,
AR: Weyerhaeuser Co., Tree Improvement and
Propagation; 1977.

Cameron, R. S. Timing of insecticide applica-
tions in southern pine seed orchards. In:
Yates, H.O. III. ed. Proceedings of the cone
and seed insects working party conference,
Working Party S2. 07-01, International Union
of Forestry Research Organizations; 1983 July
31-August 6, Athens, GA. Asheville, NC: U.S.
Department of Agriculture, Forest Service,
Southeastern Forest Experiment Station; 1984.
p 147-156.

Clark, E. C; Schenk, J. A.; Williamson, D. L.
The cone- infesting moth Barbara coif axiana as
a pest of Douglas-fir in northern Idaho.
Ann. Entomol. Soc. Am. 56: 246-250; 1963.

Dale, J. W.; Schenk, J. A. Cone production and
insect-caused seed loss of ponderosa pine in
Idaho and adjacent Washington and Montana.
Bull. No. 24. Moscow, ID: University of
Idaho, Forest Wildlife and Range Experiment
Station. 1978. 16 pp.

Dale, J. W. ; Schenk, J. A. Bionomics and natural
control of the cone and seed insects of
ponderosa pine in Idaho and adjacent
Washington and Montana. Bull. No. 29. Moscow,
ID: University of Idaho, Forest Wildlife and
Range Experiment Station; 1979. 24 pp.

DeBarr, G. L. Southwide tests of carbofuran for
seedbug control in pine seed orchards. Res.
Pap. SE-185. Asheville, NC: U.S. Department
of Agriculture, Forest Service, Southeastern
Forest Experiment Station; 1978. 24 pp.

DeBarr, G. L. ; Berisford, C. W. Use of sex
pheromones for seed and cone insect pest
management in conifer seed orchards. In:
Yates, H.O. III. ed. Proceedings of the cone
and seed insects working party conference,
Working Party S2. 07-01, International Union
of Forestry Research Organizations; 1983 July
31-August 6, Athens, GA. Asheville, NC: U.S.
Department of Agriculture, Forest Service,
Southeastern Forest Experiment Station; 1984:
194-203.

233

DeBarr, G. L. ; Fedde, V. H. Contact toxicity of
17 insecticides to larvae of Dioryctria
amatella (Lepidoptera: Pyralidae) . Can.
Entomol. 112: 521-523; 1980.

DeBarr, G. L. ; Nord, J. C. Contact toxicity of
34 insecticides to second-stage nymphs of
LeptoRlossus corculus (Hemiptera: Coreidae) .
Can. Entomol. 110: 901-906; 1978.

DeBarr, G. L. ; Barber, L. R. ; Maxwell, A. H.
Use of carbofuran for control of eastern
white pine cone and seed insects. For. Ecol.
Manage. 4: 1-18; 1982.

Dewey, J. E. Damage to Douglas-fir cones by

Choristoneura occidentalis . J. Econ. Entomol.
63: 1804-1806; 1970.

Dewey, J. E. A three year evaluation of

Douglas-fir cone and seed insects in Montana
and Yellowstone National Park. Rep. No. 72-1.
Missoula, MT: U.S. Department of Agriculture,
Forest Service, Northern Region; 1972. 19 pp.

Dewey, J. E. ; Jenkins, M. J. An evaluation of
cone and seed insects in selected seed
production areas in Region 1. Rep. No. 79-16.
Missoula, MT: U.S. Department of Agriculture,
Forest Service, Northern Region, Forest Insect
and Disease Management; 1979. 18 pp.

Dewey, J. E. , Jenkins, M. J. An evaluation of
cone and seed insects in selected seed
production areas in Region 1 (progress report
II). Rep. No. 80-18. Missoula, MT: U.S.
Department of Agriculture, Forest Service,
Northern Region, Forest Insect and Disease
Management; 1980. 11 pp.

Dewey, J. E.; Jenkins, M. J. An evaluation of
cone and seed insects in selected seed
production areas in Region one (final
report). Rep. No. 82-5. Missoula, MT: U.S.
Department of Agriculture, Forest Service,
Northern Region, Forest Pest Management;
1982. 22 pp.

Fogal, W. H. ; Lopushanski , S. M. Stem injection
of insecticides for control of white spruce
seed and cone insects. In: Yates, H.O. III.
ed. Proceedings of the cone and seed insects
working party conference, Working Party
S2. 07-01, International Union of Forestry
Research Organizations; 1983 July 31-August
6, Athens, GA. Asheville, NC: U.S. Department
of Agriculture, Forest Service, Southeastern
Forest Experiment Station; 1984. p. 157-167.

Forcella, F. Irregularity of pinyon cone

production and its relation to pinyon cone
moth predation. Madrono 25: 170-172; 1978.

Forcella, F. Cone predation by pinyon cone
beetle ( Conophthorus edulis ; Scolytidae) :
Dependence on frequency and magnitude of cone
production. Am. Nat. 116: 594-598; 1980.

Hedlin, A. F. On the life history of the

Douglas-fir cone moth, Barbara coif axiana
(Kft.) (Lepidoptera: Olethreutidae) , and one
of its parasites, Glypta evetriae Cush.
(Hymenoptera: Ichneumonidae) . Can. Entomol.
92: 826-834; 1960.

Hedlin, A. F. Results of a six-year plot study
on Douglas-fir cone insect population
fluctuations. For. Sci. 10: 124-128; 1964a.

Hedlin, A. F. Life history and habits of a
midge, PhytophaRa thu.j ae Hedlin (Diptera:
Cecidomyiidae) in western red cedar cones.
Can. Entomol. 96: 950-957; 1964b.

Hedlin, A. F. Prevention of insect-caused seed
loss in Douglas-fir with systemic
insecticides. For. Chron. 42: 76-82; 1966.

Hedlin, A. F. The pine seedworm, Laspeyresia
piperana (Lepidoptera: Olethreutidae) , in
cones of ponderosa pine. Can. Entomol. 99:
264-267; 1967a.

Hedlin, A. F. Cone insects of grand fir, Abies
Krandis (Douglas) Lindley, in British
Columbia. J. Entomol. Soc . B.C. 64: 40-44;
1967b.

Hedlin, A. F. Spruce cone insects in British
Columbia and their control. Can. Entomol.
105: 113-122; 1973.

Hedlin, A. F. Cone and seed insects in British
Columbia. Info. Rep. BC-X-90. Victoria, BC:
Canadian Forestry Service, Pacific Forest
Research Centre; 1974. 63 p.

Hedlin, A. F. ; Ruth, D.S. Barbara colfaxiana
siskiyouana (Kft.), a pest in cones of Abies
grandis. J. Entomol. Soc. B.C. 71: 13; 1974.

Hedlin, A. F. ; Weatherston J.; Ruth D. S.;
Miller, G.E. Chemical lure for Douglas-fir
cone moth, Barbara colfaxiana (Lepidoptera:
Olethreutidae). Environ. Entomol. 12:
1751-1753; 1983.

Hedlin, A. F. ; Yates, H. 0. Ill; Cibrian-Tovar ,
D.; Ebel, B. H.; Koerberl, T. W. ; Merkel , E.
P. Cone and seed insects of North American
conifers. Victoria, BC: Canadian Forestry
Service/ U.S. Department of Agriculture,
Forest Service/ Secretariade de Agricultura y
Recursos Hidraulicos, Mexico; 1980. 122 pp.

Hussey, N. W. The life histories of MesastiRmus
spermotrophus Wachtl (Hymenoptera:
Chalcidoidea) and its principle parasite,
with descriptions of the developmental
stages. Trans. R. Entomol. Soc. Lond. 106:
133-151; 1955.

234

Jenkins, M. J. Relationship between attacks by
the mountain pine cone beetle (Coleoptera:
Scolytidae) to clone and cone color in
western white pine. Environ. Entomol. 12:
1289-1292; 1983.

Johnson, N. E. Helicopter application of Guthion
for control of the Douglas-fir cone midge. J.
Econ. Entomol. 56: 600-603; 1963.

Johnson, N. E.; Winjum, J. K. Douglas-fir cone
and seed insect biological and control
studies: Progress in 1958, 1959. For. Res.
Note No. 22. Centralia, WA: Weyerhaeuser
Company, Forestry Research Center; 1960. 23 p.

Keen, F. P. Cone and seed insects of western
forest trees. Tech. Bull. 1169. Washington,
DC: U.S. Department of Agriculture, Forest
Service, California Forest and Range
Experiment Station; 1958. 168 p.

Kinzer, H. G. ; Ridgill, B. J.; Watts, J. G.
Biology and cone attack behavior of
Conophthorus ponderosae in southern New
Mexico (Coleoptera: Scolytidae). Ann.
Entomol. Soc. Am. 63: 795-798; 1970.

Kinzer, H. G . ; Ridgill, B. J.; Watts, J.G.
Seed and cone insects of ponderosa pine.
Bull. 594. Las Cruces, NM: New Mexico State
University, Agriculture Experiment Station;
1972a. 36 p.

Kinzer, H. G.; Ridgill, B. J.; Reeves, J. M.
Response of walking Conophthorus ponderosae
to volatile attractants. J. Econ. Entomol.
65: 726-729; 1972b.

Koerber, T. W. Leptoglossus occidentalis

(Hemiptera, Coreidae) , a newly discovered
pest on coniferous seed. Ann. Entomol. Soc.
Am. 56: 229-234; 1963.

Koerber, T. W. ; Markin, G. P. Metasystox-R®

injections increase seed yield of Douglas-fir
in California, Oregon, and Washington. In:
Yates, H. 0. Ill, ed . Proceedings of the cone
and seed insects working party conference,
Working Party S2. 07-01, International Union
of Forestry Research Organizations; 1983 July
31-August 6, Athens, GA. Asheville, NC: U.S.
Department of Agriculture, Forest Service,
Southeastern Forest Experiment Station; 1984:
137-146 .

Krugman, S. L . ; Koerber, T.W. Effect of cone
feeding by Leptoglossus occ idental is on
ponderosa pine seed development. For. Sci.
15: 104-111; 1969.

Kulhavy, D. L . ; Schenk, J. A. An evaluation of
damage by cone and seed insects of grand fir
in northern Idaho. Anniv. Publ. No. 5.
Moscow, ID: University of Idaho, Deptartment
of Entomology; 1976a: 27-35

Kulhavy, D. L. ; Schenk, J. A. Cone and seed
insect damage and prediction of cone

production in grand fir in the Potlatch area
of northern Idaho. Note No. 23. Moscow, ID:
University of Idaho, Forest, Wildlife and
Range Experiment Station; 1976b. 6 p.

Kulhavy, D. L. ; Dale, J. W. ; Schenk. J. A.

A checklist of the cone and seed insects of
Idaho. Cur. Info. Ser. No. 6. Moscow, ID:
University of Idaho, College of Forestry,
Wildlife and Range Sciences; 1975. 28 p.

Kulhavy, D. L. ; Schenk, J. A.; Hudson, T. D.

Cone and seed insects of subalpine fir during
a year of low cone production in northern
Idaho. J. Entomol. Soc. B.C. 73: 25-28;
1976 .

Madsen, H. F.; Peters, F. E. Pest management:
monitoring populations of Archips
argyrosphilus and Archips rosanus
(Lepidoptera: Tortricidae) with sex pheromone
traps. Can. Entomol. 108: 1281-1284; 1976.

Madsen, H. F.; Peters, H. F.; Vakenti, J. M.
Pest management: Experience in six British
Columbia apple orchards. Can. Entomol. 107:
873-877; 1975.

Matson, W. J., Jr. Relationship between cone
crop size and cone damage by insects in red
pine seed production-areas. Can. Entomol.
103: 617-621; 1971.

Mattson, W. J.; Tabashnik, B. E . ; Miller, J. R.
Developing a conceptual model of cone- finding
behavior by the red pine rone beetle,
Conophthorus resinosae Hopkins (Coleoptera:
Scolytidae). In: Yates, H. 0. Ill, ed.
Proceedings of the cone and seed insects
working party conference, Working Party
S2. 07-01, International Union of Forestry
Research Organizations; 1983 July 31-August
6, Athens, GA. Asheville, NC: U.S. Department
of Agriculture, Forest Service, Southeastern
Forest Experiment Station; 1984: 65-76.

McKnight, M. E. A literature review of the
spruce, western and 2- year cycle budworms
Choristoneura fumif erana , C. occidentalis and
C. biennis (Lepidoptera: Tortricidae). Res.
Pap. RM-44. Ft. Collins, CO: U.S. Department
of Agriculture, Forest Service, Rocky
Mountain Forest and Range Experiment Station;
1968. 35 p.

McLeod, J. M. ; Daviault, L. Notes on the life
history and habits of the spruce cone worm,
Dioryctria reniculella (Grt.) (Lepidoptera:
Pyralidae). Can. Entomol. 95: 309-316; 1963.

McLeod, P. J.; Yearian, W. C. Insect control
tactics in loblolly pine seed orchards.
Arkansas Farm Res. 28: 4; 1979.

McLeod, P. J.; Yearian, W. C. Insect control
tactics evaluated in loblolly pine seed
orchards. Arkansas Farm Res. 31: 10; 1982.

235

McLeod, P. J.; Yearian, W. C; Young, S. Y.
Effectiveness of Bacillus thurinRiensis
against the southern pine coneworm,
Dioryctria amatella (Lepidoptera: Pyralidae) .
Environ. Entomol. 11: 1305-1306; 1982.

Merkel, E. P.; DeBarr, G. L. Trunk implantations
of dicrotophos for cone-insect control in
slash pine seed production stands. J. Econ.
Entomol. 64: 1295-1298; 1971.

Meso, S. W. Douglas-fir cone and seed insect
impact and control. Portland, OR: U.S.
Department of Agriculture, Forest Service,
Region 6; 1975. 19 p. Unpublished report.

Miller, G. E. Pest management in Douglas-fir
seed orchards: A problem analysis. Pest
Management Pap. No. 22. Burnaby, BC: Simon
Fraser University, Pestology Centre; 1980.
138 p.

Miller, G. E. Biology, sampling and control of
the Douglas-fir cone gall midge, Contarinia
oreRonensis Foote (Diptera: Cecidomyiidae) ,
in Douglas-fir seed orchards in British
Columbia. Burnaby, BC: Simon Fraser
University; 1983a. 192 p. Dissertation.

Miller, G. E. Evaluation of the effectiveness of
cold water misting of trees in seed orchards
for control of Douglas-fir cone gall midge
(Diptera: Cecidomyiidae). J. Econ. Entomol.
76: 916-919; 1983b.

Miller, G. E. Pest management in Douglas-fir
seed orchards in British Columbia. In: Yates,
H. 0. III. ed. Proceedings of the cone and
seed insects working party conference,
Working Party S2. 07-01, International Union
of Forestry Research Organizations; 1983 July
31-August 6, Athens, GA. Asheville, NC: U.S.
Department of Agriculture, Forest Service,
Southeastern Forest Experiment Station; 1984:
179-185.

Miller, G. E.; Hutcheson, D. W. Aerial spraying
for control of the spiral spruce-cone borer,
Hylemya anthracina (Diptera: Anthomyiidae) .
J. Entomol. Soc. B.C. 78: 3-6; 1981.

Miller, G. E. ; Hedlin, A. F.; Ruth, D. S. Damage
by two Douglas-fir cone and seed insects:
Correlation with cone crop size. J. Entomol.
Soc. B.C. 81: 46-50; 1984.

Neel, W. W. An assessment of cone and seed

losses in a slash pine seed orchard following
two types of insecticide application. Tech.
Bull. 102. Mississippi State, MS: Mississippi
State University, Agriculture and Forest
Experiment Station; 1980. 4 p.

Neel, W. W. ; DeBarr, G. L. ; Lambert, W. E.
Variability of seed-bug damage to pine
following different methods of applying
carbofuran granules to a slash pine seed
orchard. In: Bonner, F. , ed. Proceedings of a
symposium on flowering and seed development

in trees, Working Parties S2. 01-05 and
S2. 07-01, International Union of Forestry
Research Organizations; 1978 May 15-18,
Starkville, MS. Starkville, MS: U.S.
Department of Agriculture, Forest Service,
Southeastern Forest Experiment Station; 1979:
314-318.

Nord, J. C; DeBarr, G. L. Contact toxicity of
12 insecticides to LeptoRlossus corculus
adults. Can. Entomol. 115: 211-214; 1983.

Nord, J. C; DeBarr, G. L. ; Overgaard, N. A.;
Neel, W. W. ; Cameron, R. S.; Godbee, J.F.
High-volume applications of azinphosmethyl ,
fenvalerate, permethrin, and phosmet for
control of coneworms (Lepidoptera: Pyralidae)
and seed bugs (Hemiptera: Coreidae and
Pentatomidae) in southern pine seed orchards.
J. Econ. Entomol. 77: 1589-1595; 1984.

Ollieu, M. M.; Schenk, J. A. The biology of
Eucosma rescissoriana Heinrich in western
white pine in Idaho (Lepidoptera:
Olethreutidae) . Can. Entomol. 98: 268-274;
1966.

Orr, D. B.; Borden, J. H. Courtship and mating
behavior of MeRastiRmus pinus Parfitt
(Hymenoptera: Torymidae) . J. Entomol. Soc.
B.C. 80: 20-24; 1983.

Overhulser, D. L. , and R. E. Sandquist.

Protecting Douglas-fir seed orchard crops
from insect damage. Tree Improvement News No.
47. Olympia, WA: Industrial Forestry
Association, Tree Improvement Laboratory;
1985: 1-7.

Pfister, R. D.; Woolwine, P. C. Insect damage in
grand fir cones. Res. Note. INT-8. Ogden, UT:
U.S. Department of Agriculture, Forest
Service, Intermountain Forest and Range
Experiment Station; 1963. 4 p.

Prokopy, R. J.; Hauschild, K. I. Comparative
effectiveness of sticky red spheres and
Pherocon AM standard traps for monitoring
apple maggot flies in commercial orchards.
Environ. Entomol. 8: 696-700; 1979.

Reardon, R. C; Stipe, L. E.; Dewey, J. E.

Systemic insecticides implanted and injected
in Douglas-fir in Montana to increase seed
yield. In: Yates, H. O. Ill, ed. Proceedings
of the cone and seed insects working party
conference, Working Party S2. 07-01,
International Union of Forestry Research
Organizations; 1983 July 31-August 6, Athens,
GA. Asheville, NC: U.S. Department of
Agriculture, Forest Service, Southeastern
Forest Experiment Station; 1984: 186-191.

Reissig, W. H. ; Tette, J. P. Apple maggot

(Diptera: Tephritidae) monitoring in an apple
pest management program. Can. Entomol. Ill:
621-625; 1979.

236

Roques , A. Etudes sur le role des stimuli

visuels dans la reconnaissance des cones de
meleze par les adultes de Lasiomma (Diptera:
Anthomyiidae) , p. 46-56. In: Yates, H. 0.
Ill, ed. Proceedings of the cone and seed
insects working party conference, Working
Party S2. 07-01, International Union of
Forestry Research Organizations; 1983 July
31-August 6, Athens, GA. Asheville, NC: U.S.
Department of Agriculture, Forest Service,
Southeastern Forest Experiment Station; 1984:
46-56 .

Rudinsky, J. A. Douglas-fir cone and seed

insects (progress report). For. Res. Note No.
13. Centralia, WA: Weyerhaeuser Company,
Forestry Research Center; 1955. 5 p.

Ruth, D. S.; Senecal, M. A.; Karlson, J. A.
Cone and seed pests 1980. Pest Report.
Victoria, BC: Canadian Forestry Service,
Pacific Forest Research Centre; 1980. 16 p.

Sartwell, C; Daterman, G. E. ; Sower, L. L.

Synthetic attractants for cone and seed moths
in the western United Stated (Summary) . In:
Yates, H. 0. Ill, ed. Proceedings of the cone
and seed insects working party conference,
Working Party S2. 07-01, International Union
of Forestry Research Organizations; 1983 July
31-August 6, Athens, GA. Asheville, NC: U.S.
Department of Agriculture, Forest Service,
Southeastern Forest Experiment Station; 1984:
204 .

Schenk, J. A.; Goyer, R. A. Cone and seed
insects of western white pine in northern
Idaho: distribution and seed losses in
relation to stand density. J. For. 65:
186-187; 1967.

Schenk, J. A.; Giles, R. H.; Johnson, F. D.

Effects of trunk-injected oxydemetonmethyl on
Douglas-fir cone and seed insects, seedling
production, and mice. Station Pap. No. 2
Moscow, ID: University of Idaho, Forestry,
Wildlife and Range Experiment Station; 1967.
20 p.

Schmid, J. M. ; Mitchell, J. C; Stevens, R. E.
Cydia youngana (Kearfott) (Lepidoptera :
Tortricidae) and associates in Engelmann
spruce cones, Fraser Experimental Forest,
Colorado 1974-1977. Res. Note RM -394. Ft.
Collins, CO: U.S. Department of Agriculture,
Forest Service, Rocky Mountain Forest and
Range Experiment Station; 1981. 5 p.

Shea, P. J.; Jenkins, M. J.; Haverty, M. I.

Cones of blister rust-resistant western white
pine protected from Conophthorus ponderosae
Hopkins (C. monticolae Hopkins). J. GA.
Entomol. Soc. 19: 129-138; 1984.

Shearer, R. C. Influence of insects on

Douglas-fir, Pseudotsuga menziesii (Mirb.)
Franco, and western larch, Larix occidentalis
Nutt., cone and seed production in western
Montana. In: Yates, H. 0. Ill, ed .
Proceedings of the cone and seed insects
conference, Working Party S2. 07-01,

Stipe, L. E. ; Green, A. K. A multiple ground
application of acephate and carbaryl for
protection of Douglas-fir cones from western
spruce budworm. Rep. 81-22. Missoula, MT:
U.S. Department of Agriculture, Forest
Service, Northern Region, Forest Pest
Management; 1981. 17 p.

Stoakley, J. T. Laboratory and field tests of
insecticides against the Douglas- fir seed
wasp (Megastigmus spermotrophus ) (Hym. ,
Torymidae). Plant Pathol. 22: 79-87; 1973.

Timonin, M. I.; Fogal , W. H.; Lopushanski, S. M.
Possibility of using white and green
muscardine fungi for control of cone and seed
insect pests. Can. Entomol. 112: 849-854;
1980.

Tripp, H. A. Description and habits of the

spruce seedworm, Laspeyresia youngana (Kft.)
(Lepidoptera: Olethreutidae) . Can. Entomol.
86: 385-402; 1954.

Vakenti, J. M. ; Madsen, H. F. Codling moth
(Lepidoptera: Olethreutidae): monitoring
populations in apple orchards with sex
pheromone traps. Can. Entomol. 108: 433-438;
1976.

Williamson, D. L. ; Schenk, J. A.; Barr, W. F.
The biology of Conophthorus monticolae in
northern Idaho. For. Sci. 12: 234-240; 1966.

Yates, H. 0. III. Light trapping in seed
orchards under a pest management system.
Twelfth Southern Forest Tree Improvement
Conference Proceedings 1973: 91-96.

237

J?

INSECTS DESTRUCTIVE TO PONDEROSA

PINE CONE CROPS IN
Elizabeth A.^Blake,
and Thomas

ABSTRACT: The insect damage to the 1984 ponderosa
pine, Pinus ponderosa, cone and seed crop is re-
ported. Three cone infesting insects ( Dioryctria
auranticel la , Conophthorus ponderosae, Conotrache-
lus neomexicana) and two seed infesting insects
(Cydia piperana , Megastigmus al bifrons) were re-
sponsible for as much as 100 percent reduction in
seed yield. Impact of seed and cone insect damage
in 1984 is compared to previous estimates of insect
damage. An economic comparison of the ponderosa
pine seed value with and without insect damage is
presented.

INTRODUCTION

Cone and seed insects are one of the most import-
ant biotic factors affecting seed production and
thereby regeneration of ponderosa pine, Pinus pon-
derosa Dougl . ex Laws., in the Southwest. Studies
in Cal ifornia (Koerber 1967), New Mexico (Kinzer
and others 1972), Colorado (Bodenham 1973),
and Arizona (Schmid and others 1984, unpublished
manuscripts) indicate four insect species as the
major damaging agents on ponderosa pine. These
include the ponderosa pine coneworm (Dioryctria
auranticella (Grote) (Lepidoptera : Pyral idae) ) ,
the ponderosa pine cone beetle (Conophthorus pon-
derosae Hopkins (Coleoptera: Scolytidae) ) , the
ponderosa pine seedworm (Cydia piperana (Kearfott)
(Lepidoptera: Olethreutidae) ) , and the ponderosa
pine seed chalcid (Megastigmus albifrons Walker
(Hymenoptera : Torymidae)). Other less prevalent
cone and seed insects are the cone midge (Thomas-
iniana sp.), the pine cone weevil (Conotrachel us
neomexicana Fall (Coleoptera: Curcul ionidae) ) , and
the western conifer seed bug (Leptoglossus occiden-
tal is Heidemann (Hemiptera: Coreidae) ) .

The reduction of seed yield by cone and seed in-
sects varies by location, tree, and year. Koerber

Paper presented at the Conifer Tree Seed in the
Inland Mountain West Symposium, Missoula, MT,
August 5-6, 1985.

Elizabeth A. Blake is Research Associate, Northern
Arizona Unversity, School of Forestry, Flagstaff,
AZ: Michael R. Wagner is Assistant Professor of
Forest Protection, Northern Arizona University
School of Forestry, Flagstaff, AZ: Thomas W.
Koerber is Research Entomologist, Pacific South-
west Forest and Range Experiment Station, Forest
Service-USDA, Berkeley, CA.

NORTHERN ARIZONA^
Michael R.^Wagner
Koerber

(1967) reports 'damage from C. pi perana of 37
percent in 1963 and only 9 percent the following
year. Kinzer and others (1972) report an overall
reduction in seed yield of 82 percent in New
Mexico, with individual trees exceeding 90 per-
cent. Schmid and others (1984) show damage by
Conophthorus ponderosae, M. albifrons , and Cydia
pi perana in northern Arizona varies significantly
between study areas and between trees within study
areas. They also report no significant differences
between tree crown levels. The year to year varia-
tion in ponderosa pine cone production coupled
with cone and seed insect damage makes estimates
of seed yield uncertain at best.

METHODS

Five study plots were selected in fall 1984 on the
Coconino and Kaibab National Forests. Each plot
included 10 ponderosa pine trees ranging in height
from 10 m to 15 m. Twenty live and 20 dead cones
were collected from each of two crown levels (upper
half and lower half). The entire live cone crop
was collected from trees which produced less than
20 live cones. The collected cones were placed in
labeled burlap bags and taken to the laboratory for
evaluation. The number of remaining live and dead
cones on each tree was counted and recorded.

Evaluation of cone and seed damage occurred in three
steps. Dead cones were visually examined and the
cause of mortality (Conophthorus ponderosae, Dioryc-
tria auranticel la, Conotrachel us neomexicana, or
aborted) was recorded. The second step was a scale
by scale dissection of all live cones. The number
of sound-appearing seeds, aborted seeds, and Cydia
piperana damaged seeds were recorded. Finally, all
sound-appearing seeds were x-rayed and categorized
as filled, unfilled, Megastigmus albifrons damaged,
and other-damaged seeds.

Cone and seed insect specimens found during the
study were reared to adults for species identifi-
cation. Damage was then associated with the reared
specimens .

All data were subjected to analysis of variance
(AN0VA, a=0.05) by crown level, plot, and trees
within a plot. Data were transformed where neces-
sary to meet the assumptions of homogeneity of
variance. The Student-Newman-Keul s and Least Sig-
nificant Difference multiple range tests were used
to determine where differences occurred.

238

RESULTS

Five insect species were found to cause significant
damage to the ponderosa pine cone and seed crop in
1984. A brief description of each insect species
and the damage it causes is given below.

Cone Insects

Damage to cones by the ponderosa pine coneworm,
Dioryctria auranticella , ranged from 19 percent to
71 percent of the total cones produced per tree.
This insect species caused the heaviest damage to
the cone crops in our study areas.

Dioryctria auranticella feeds on seeds and scale
tissue of ponderosa pine and knobcone pine during
the larval stage (Hedlin and others 1981). The
larvae enter the basal portion of the cones and
bore irreqular shaped feeding cavities within the
cone. Reddish-brown frass and webbinq fill the
large cavities (fig. 1). Pupation occurs within
the cavity; the pupal staqe lasts from 10 to 14
days. Adults emerge, mate, and lay eggs in July.
Little is known of the lifecycle from July until
the larvae appear in the cones the following
spring. Entire cones are usually killed by D.
auranticel la infestation; partially killed cones
become distorted and do not open.

Figure 1 . --Dioryctria auranticella larva feeding
on seeds and scales of ponderosa pine. Frass and
webbing fill the feeding cavities.

In this study damage to cones by the ponderosa
pine cone beetle, Conophthorus ponderosae , ranged
from 2 percent to 19 percent per tree. The
adult female beetle kills cones by severing the
conductive tissue of the cone stalk (Hedlin and
others 1981). A pitch tube usually marks the point
where the adult enters the cone in early summer to
construct a qallery and lay eggs (fig. 2). The
dead cones turn brown and may remain on the tree
or drop to the ground. Larvae feed randomly in
the cone where they complete development in approx-
imately one month, pupate, and emerge as adults.
The adults may overwinter in the dead cones, or
enter shoots or conelets where they feed and over-
winter. In Arizona very few adults have been
found overwintering in cones. No viable seed is
produced by cones infested with £. ponderosae.

Figure 2.--A pitch tube at the base of a ponderosa
pine cone marks the place where an adult Conoph-
thorus ponde rosae entered the cone.

Figure 3. --Larvae of Conotrachel us neomexicana
feeding on scales and seeds of a developing pon-
derosa cone.

Cone damage by the pine cone weevil, Conotrachelus
neomexicana, was relatively light at our study
plots when compared to JJ. auranticel la and Conoph-
thorus ponderosae. Damage ranged from 7 percent
to 14 percent of the cone crop per tree.

Larvae of £. neomexicana feed indiscriminately on
the scales and seeds of developing cones (Bodenham
1973) (fig. 3). The pinkish-white to yellowish-
white larvae feed for four to six weeks and, after
being stimulated by rain, mature larvae chew exit
holes through the outer shell of the cone and drop
to the ground. They then burrow into the soil and
pupate. Adults are formed in eight to 12 days but
remain in the pupal cell for a few days while they
darken and harden. The adults make their way to
the surface, feed on shoots for a brief period,
and burrow into the litter where they overwinter.

Seed Insects

Damage to seeds by the ponderosa pine seedworm,
Cydia piperana, ranged from 1 percent to 11
percent of seeds per cone in this study. The

239

larvae consume the entire contents of each seed as
they migrate from one seed to another within the
cone (Hedlin and others 1981). Damaged seeds are
filled with frass; seed pairs may be fused together
by silk-lined tunnels of frass. Larvae burrow
exit tunnels through mined seeds and then retreat
to the cone axis where they overwinter (fig. 4).

Figure 4.--A mature larva of Cydia piperana in the
cone axis. Larvae migrate from seed to seed, con-
suming the entire contents of each seed they en-
counter .

Pupation occurs in the spring and, after approxi-
mately two weeks, the pupae wriggle halfway through
the exit hole and the adult moth emerges. One to
three larvae may inhabit a single cone and may
destroy a significant amount of the seed produced.

An average 53 percent of seeds which appeared to
be normally developed were found to contain lar-
vae of the ponderosa pine seed chalcid, Megastig-
mus albifrons, when x-rayed. Overall, 7 per-
cent of the total seed production per cone was
damaged by M. albifrons in this study.

The larvae of M. albifrons completely consume the
contents of ponderosa pine seeds (Hedlin and others
1981). No external evidence of seed infestation
is apparent until adults emerge in spring. Female
adults oviposit through the cone scale into the
immature seed. Only one larva completes develop-
ment in each seed. The presence of larvae can be
detected by x-raying the seed (fig. 5). Pupation
occurs in the seed; adults emerge by chewing exit
holes through the seed coat.

Comparison of Two Studies in Northern Arizona

Two studies of cone and seed insects in northern
Arizona report the incidence of damage to ponderosa
pine cone and seed production. Schmid and others
(1984) reported findings for 10 locations; we report
findings for five locations, three of which were
studied in 1982 by Schmid and others (1984). These
three plots are used for comparison of cone and
seed insect damage.

Both studies report no statistically significant
difference in damage between crown levels (middle

Figure 5. --Larva of Megastigmus albifrons inside
a ponderosa pine seed. Seeds must be x-rayed to
detect the presence of M. albifrons because no
external evidence of infestation is apparent.

and lower, Schmid and others (1984), upper and lower,
this study). Differences in damage caused by C_.
pondersae, Cydia piperana, and M. albifrons weFe
significant between plots and among trees within
plots in both studies. However, Schmid and
others (1984) reported no difference in damage
by JJ. auranticella between plots or trees
within plots in 1982, while we did find these
differences in 1984. Table 1 compares the
incidence of D. auranticel la and Conophthorus
ponderosae at the US Highway 89, Deadman Flat,
and Williams 345 Road plots. The incidence
of D. auranticel la increased dramatically in all
ploFs from 1982 to 1984. Damage to cones by C. pond-
erosae decreased at the US Highway 89 and Deadman
Flat plots but increased substantially at the Wil-
liams 354 Road plot. The common factor between the
data is an apparent displacement of one species by
the other in each year. Schmid and others (1984)
did not report damage by Conotrachel us neomexicana .

A comparison of seed damage between 1982 and 1984
showed the relative effect of seed insects and en-
vironmental factors that influence viable seed pro-
duction (table 2). The percentage of sound and
hollow seed decreased substantially at the Deadman

Table 1. --Comparison of the percentage of dead

cones killed by Dioryctria auranticella
and Conophthorus ponderosae at three
plots in northern Arizona

FJ. auranticel la C. ponderosae

Plot

19821

19842

1982

1984

US Hwy 89

1

80

39

10

Deadman Flat

1

78

15

9

Williams 354

0

55

0

32

Road

X

<1

71

18

17

Data for 1982 from Schmid and others (1984).
Data for 1984 from this study.

240

Flat and Williams 354 Road plots; the US Highway
89 plot experienced a slight decline in these same
categories. Schmid and others (1984) did not
report aborted seed in their 1982 damage
estimates. Perhaps aborted seeds were counted as
hollow seeds or no obviously aborted seeds were
found. Our study showed aborted seeds constituted
75 percent or more of the seed crop at all three
study areas. If the percentage of aborted and
hollow seeds are combined, the impact on total
seed production in 1984 is substantial.

Comparison indicates a decrease in both £. piperana
and M. albifrons in 1984. This decrease is prob-
ably due to the large percentage of aborted seeds
which are unavailable for insect infestation.
Of the seeds which appeared to be normally devel-
oped at the time of dissection, 45 to 67 percent
were found to be infested with M. al bi f rons when
x-rayed .

CONCLUSIONS

Damage to ponderosa pine cone and seed crops by
insects is highly variable from year to year.

One cause of this variation appears to be a com-
bination of the population dynamics of the damaging
insects and the size of the previous and current
cone crops. If good cone crops are produced for
several years and then a poor crop is produced,
the insects will concentrate on the few cones and
damage much of the crop. However, if several poor
crops are followed by a moderate to good crop, the
relatively small insect population that was
supported by the poor cone crops will be dispersed
and cause little damage to the large crop.

Table 3 compares the seed yield and cost per pound
of seed, assuming an initial crop of 200 cones per
tree. The comparison is for no insect damage and
insect damage using estimates for 1982 (Schmid and
others 1984) and 1984. The value of the sound seed
produced per tree in 1984 is 83 times less than the
value per tree if no insect damage were to occur,
and 27 times less than the estimated value in 1982.
The variability of the damage and value estimates
between 1982 and 1984 point out the need for pre-
cise estimates of cone and seed damage prior to
cone collection each year.

Table 2. --Comparison of the percentage of sound, hollow, aborted, Cydia piperana damaged, Megastigmus
albifrons damaged, and other damaged seeds per cone at three locations in northern Arizona

Plot

Sound

Hoi 1 ow

Aborted

C. piperana

M. albifrons

Other

19821

1984

1982

1984

1982 1984

1982

1984

1982

1984

1982 1984

US Hwy

89

2

<12

8

5

88

77

2

16

5

<1

Deadman

Flat

38

<1

16

4

75

31

11

11

10

<1

Wi 1 1 iams

354 Road

60

<1

33

7

84

3

1

14

6

<1

X

33

<1

19

5

82

37

5

14

7

<1

iData for 1982 from Schmid and others (1984).
Percentages may not add to 100 due to rounding.

Table 3. --Comparison of the value of sound seed produced per tree for non-insect damaged and insect
damaged cone crops

i lnsecj- Insect Damage Insect Damage
Damage 19821 1984

Initial numoer
of cones/tree

200

200

200

% cone mortal ity

0

28

60

Surviving cones

200

144

80

Average number of
sound seed/cone

47

21

0.9

Sound seed/tree

9.4002

3,024

72

Seeds/pound

11 ,4003

11 ,400

11,400

Pounds of sound
seed/tree

0.82

0.27

0.01

Value/pound of seed

$38. 004

$38.00

$38.00

Value of seed/tree

$31.60

$10.26

$ 0.38

Data from Schmid and others (1984).

3ased on U.S. Forest Service estimate of 75 percent sound seed per tree.
Estimate from Schopmeyer (1974).
Estimate from U.S. Forest Service.

241

REFERENCES

Bodenham, J. L.; Insects attacking ponderosa pine
cones in northern Colorado, with emphasis on
a cone weevil, Contrachelus neomexicanus Fall.
M.S. thesis, Colorado State University, Fort
Collins; 1973. 97 p.

Hedlin, A. F. ; Yates, H. 0. Ill; Tovar, D. C. ;
Ebel, B. H. ; Koerber, T. W. ; Merkel , E. P.
Cone and seed insects of North American
conifers. Joint publication of Canadian
Forest Service, U.S. Forest Service and
Secretaria de Agricultura y recursos Hidrau-
licos, New Mexico; 1980.

Kinzer, H. G. ; Ridgill, B. J., and Watts, J. G.
Seed and cone insects of ponderosa pine. Agri
cultural Experiment Station Bulletin 594,
New Mexico State University, Las Cruces; 1972.
36 p.

Koerber, T. W. Studies of the insect complex af-
fecting seed production of ponderosa pine in
California. Ph.D. dissertation. University of
California, Berkeley; 1967. 86 p.

Schmid, J. M. ; Mitchell, J. C; Carlin, K. D. ;
and Wagner, M. R. Insect damage, cone dimen-
sions, and seed production in crown levels of
ponderosa pines. Great Basin Naturalist 44(4)
575-578; 1984.

WESTERN SPRUCE BUDWORM IMPACT ON DOUGLAS-FIR CONE PRODUCTION

Jerald E. Dewey

ABSTRACT: The western spruce budworm is the most
destructive pest of Douglas-fir cones in much of
the Inland Mountain West. Budworm feed on
Douglas-fir flowers and cones throughout their
entire larval stage, sometimes destroying more
than one cone. This paper discusses the general
impact of budworm on cone production as deter-
mined from numerous surveys and evaluations, as
well as a specific study conducted in 1983. In
1983, 772 cones on 14 trees were tagged and
examined four times between early May and late
August. The number of cones remaining in late
August was approximately 60 percent of what was
counted in May, and 21.6 percent of those were
conspicuously injured by insects. This cone loss
was attributable to all factors; however, the
western spruce budworm was the single most
damaging agent.

INTRODUCTION

For many years forest managers have recognized
the western spruce budworm (Chor is toneura
occidentalis Freeman) as a severe defoliator of
Douglas-fir, true fir, and spruce forests.
Native to western forests, the budworm population
periodically becomes epidemic over very large
areas. Western spruce budworm defoliated
approximately 4.2 million hectares (10.38 million
acres) in the western U.S. in 1984, and another
62,000 hectares (153,202 acres) in British
Columbia (Kucera and Taylor 1985).

This voracious defoliator is not widely
recognized for its influence on cone and seed
production. Budworm affect seed production in a
number of ways. As a defoliator, they concen-
trate in the upper crowns of host trees. Several
consecutive years of feeding will leave very thin
crowns or dead tops.

Most Douglas-fir cones are produced in the upper
tree crown. By topkilling potential cone-bearing
trees, budworm reduce cone production. Severe
defoliation may trigger a stress cone crop
response in the trees, but there is uncertainty
regarding the viability of seed from these cones.

Budworm reduce cone production more directly by
feeding on developing flowers, cones, and

Paper presented at the Conifer Tree Seed in the
Inland Mountain West Symposium, Missoula, MT,
August 5-6, 1985.

Jerald E. Dewey, Supervisory Entomologist, USDA
Forest Service, Cooperative Forestry & Pest
Management, Missoula, MT.

reproductive buds during all stages of larval
development, from emergence in the spring until
pupation. Fellin (1985) reports budworm larvae
become active as early as the end of April and
that nearly all larvae are feeding by mid-June.
Vegetative buds are usually still very tight
through the month of May. But reproductive buds
are swollen or may have burst by the first of
May.

OBSERVATIONS

Newly emerged larvae have three food sources:
mining old needles; mining tight vegetative buds;
or feeding on reproductive buds or flowers. I
have observed that buds and flowers are a pre-
ferred food source. While they feed freely on
developing pollen buds and male flowers, the
impact of this feeding has not been well evalua-
ted. Because most conifers produce an excess of
pollen, I suspect the feeding on male buds and
flowers usually does insignificant damage to
Douglas-fir seed production.

Budworm larvae feeding on reproductive buds,
female flowers, or small conelets generally kill
them by severing the conductive tissue. This
results in rapid dessication and shedding,
requiring the larvae to seek new feeding sites.
As a result of this feeding characteristic, the
budworm is particularly damaging to cones. A
single larva can completely destroy one to
several cones. In contrast, many of the other
cone-feeding insects may destroy only a single
seed or portion of a cone.

Budworm larvae are only about 1 to 2 millimeters
(.04 to .08 inches) in length when newly emerged
from overwintering. As a result, budworm damage
is frequently overlooked or misdiagnosed. Cone-
lets killed by budworm are often wrongly attrib-
ubted to lack of pollination, abortion, or frost.

As cones grow in size, budworm feeding damage
symptoms change. Fewer cones are killed out-
right. But various degrees of cone mining and
surface feeding can be observed as the cones
develop. This mining and surface feeding results
in seeds being devoured, deformed cones that
reduce seed recovery, and perhaps reduced seed
viability.

PREVIOUS STUDIES

A number of studies provide information about
budworm damage to seed production. In a cone and
seed insect survey in Montana and Yellowstone
National Park, Dewey (1970) found that budworm

243

damaged or destroyed up to 71 percent of the
Douglas-fir cones. Stipe and Hard (1980), in a
light cone-crop year coupled with a large budworm
population, found a peak of 8.2 early instar
larvae per cone. The result was shedding of most
of the developing cones and loss of the entire
seed crop by the end of the season.

A survey of Idaho and Montana seed production
areas revealed budworm are Douglas-fir cones'
most damaging pest. Damage was greatest where
the defoliator was at epidemic levels (Dewey and
Jenkins 1982). Chrisman and others (1983),
examining Montana Douglas-fir cone production in
12 stands with varying levels of defoliation,
found that average cone production is usually
higher in stands with little or no defoliation.
That is, western spruce budworm infestations
reduce cone production.

1983 MONTANA STUDY

A study, designed to measure the amount of
Douglas-fir cone loss from flowering to cone
maturity, was conducted in 1983 in a budworm-
infested Douglas-fir stand near Frenchtown, MT .
Fourteen flowering Douglas-fir trees, ranging
from 8.5 (27.9 feet) to 13.7 meters (44.9 feet)
in height, were selected for monitoring. Four
cone-bearing upper crown branches were tagged on
each tree. The entire cone complement per branch
was counted four times between May 4 and August
31. Table 1 presents the results of cone counts
at various dates.

Table 1. — Cone counts in 14 Douglas-fir

trees near Frenchtown, MT, 1983

Tree Date Percent

number loss

5/4 6/1 7/25 8/31

Because the conelets were only about 10 to 20
millimeters (.4 to .8 inches) long at the time of
the first count, a number of them were not seen.
On one tree (#5) , 11 more cones were noted on the
second count than on the first. As reported by
Stipe and Hard (1980) , considerable cone loss can
occur to the small developing conelets in May.
Hence more cones were apparently missed on the
first count than indicated by the second count.

Agents other than budworm can cause a decline in
cone numbers between May and September. While
winds, hail, and rain can dislodge cones from
trees, there was no evidence that these elements
contributed to the reduced cone numbers during
the 1983 study. Squirrels often begin cone
cutting prior to August 31, but a significant
amount of cone cutting had not occurred at this
location prior to the final count. The cone
counts in table 1 do not reflect cone conditions,
just cone presence.

At the time of the last count, all cones on the
survey branches were collected, taken to the
laboratory, rated for insect damage, and
weighed. Cones weighed an average of 4.4 grams
(.16 ounces) per cone, compared to 8.8 grams (.31
ounces) for cones collected at the same time from
a nearby forest where an insecticide had been
used to control pests (Stipe and Dewey 1985) .
There was visual evidence of insect damage on
21.6 percent of the cones collected in the study.

SUMMARY AND CONCLUSIONS

Forty percent of the 772 Douglas-fir cones
present on June 1 had fallen from the trees by
August 31. Of the remaining 465 cones, 21.6
percent had obvious symptoms of the insect damage.
Although actual counts of the pest complex were
not made, the western spruce budworm was
responsible for most of the damage. The study
year, 1983, was a moderate to good cone-crop
year, and a year with a light (nondef oliating
level) budworm population.

1

102

108

97

98

9

2

75

69

52

20

73

3

48

54

39

7

87

4

61

62

47

35

44

5

84

95

77

77

19

6

57

55

51

42

26

7

66

62

42

35

47

8

38

40

36

31

22

9

43

40

34

27

37

10

44

43

17

16

64

11

21

21

15

12

43

12

43

47

34

31

34

13

37

34

28

25

32

14

39

42

16

9

79

Total or

Average

758

772

585

465

40

REFERENCES

Chrisman, Allen B.; Blake, George M. ; Shearer,
Raymond C. Effect of western spruce budworm
on Douglas-fir cone production in western
Montana. Res. Pap. INT-308. Ogden, UT: U.S.
Department of Agriculture, Forest Service,
Intermountain Forest and Range Experiment
Station; 1983. 7 p.

Dewey, Jerald E. Damage to Douglas-fir cones by
Choristoneura occidentalis . J. Econ. Entomol.
63: 1804-1806; 1970.

Dewey, Jerald E. ; Jenkins, Michael J. An
evaluation of cone and seed insects in
selected seed production areas in Region One.
Missoula, MT: U.S. Department of Agriculture,
Forest Service, Forest Pest Management; 1982.
22 p.

244

Fellin, David G. Personal communication.

Missoula, MT: U.S. Department of Agriculture,
Forest Service, Intermountain Research Station;
1985.

Kucera, Daniel R. ; Taylor, Robert G. Spruce
budworm situation in North America 1984.
Misc. Pub. 1448. U.S. Department of
Agriculture, Forest Service, Canadian
Forestry Service; 1985.

Stipe, Lawrence E. ; Hard, John S. An early single
tree application for Douglas-fir foliage and
cone protection from the western spruce
budworm. Report 80-3. Missoula, MT: U.S.
Department of Agriculture, Forest Service,
Forest Insect and Disease Management; 1980.
1 p.

Stipe, L. E.; Dewey, J. E. Protecting Douglas-fir
cones and foliage with a systemic insecticide.
Report 85-6. Missoula, MT: U.S. Department
or Agriculture, Forest Service, Forest Pest
Management; 1985. 21 p.

245

PROTECTION OF BLISTER RUST-RESISTANT WESTERN WHITE PINE CONES FROM

INSECT DAMAGE WITH PERMETHRIN AND FENVALERATE

4

Michael I. ' Haverty and Patrick J., Shea

ABSTRACT: Production of seed from blister rust-
resistant western white pine in northern Idaho has
been severely reduced by the mountain pine cone
beetle and the fir coneworm. In 1 98 1 and 1984,
insecticide treatments to control these insects to
maximize seed production were evaluated in Idaho.
At one test site, Sand point, 0.03 pet. permethrin
significantly reduced loss of cones to the moun-
tain pine cone beetle; however, 0.06 pet. perme-
thrin was more cost effective. At the other test
site, Moscow, the fir coneworm infested 46.6 pet.
of the untreated cones, significantly more than
13.6 pet. in the single application of 0.025 pet.
fenvalerate. A double application of fenvalerate
increased seed yield significantly from 31.3 to
56.0 seeds/ cone when compared to the untreated
check. A third application of fenvalerate was
apparently unnecessary.

INTRODUCTION

Western white pine (Pinus monticola Douglas) is
one of the more valuable species in the northern
Rocky Mountains. However, the introduction of the
pathogenic fungus Cronartium ribicola Fischer to
the United States from Europe in the early 1900's
nearly eliminated western white pine (Haig and
others 1941). To preserve this species for tim-
ber, the USDA Forest Service started a breeding
program to select for resistance to this disease
(Bingham 1983).

The life history of Cj_ ponderosae with respect to
the phenology of white pine cones has been des-
cribed (Williamson and others 19 66 ) . White pine
cones require about 15 months to reach maturity
(Owens and Molder 1977). In early spring, as
second-year cones begin to elongate, adult cone
beetles emerge from overwintering sites within old
cones. Female beetles initiate attack and bore
into the young cones. They rapidly girdle the
axis of the cone, thereby severing the conductive
tissue and killing the cone. Adult females may
attack up to four cones, depositing a maximum of
100 eggs. Normally, however, only 4 to 8 adults
emerge from an infested cone (Jenkins 1 982 ) . The
rapid attack and subsequent mortality before the
cones are half-grown make successful control
difficult (Furniss and Carolin 1977).

The fir coneworm is a transcontinental species
that attacks cones of many conifer species. It
also mines in the buds, shoots, and trunks of
conifers. Its life history is variable and not
well known. Apparently, in Moscow, larvae pupate
in cocoons on the ground in August and September
and emerge as moths in June. Eggs are laid soon
after emergence, and larvae feed from June to
August or September. Larvae mine the inside of
cones but also crawl on the outside throughout
cone development (Furniss and Carolin 1977; Hedlin
and others 1 980 ) . Feeding throughout the cone
development period could necessitate multiple
insecticide applications.

Recently, the production of blister rust-resistant
seed from western white pine in seed orchards has
been severely reduced in Idaho because of periodic
infestations of the mountain pine cone beetle
Conophthorus ponderosae Hopkins ( =C . monticolae)
at the Sandpoint (Idaho) Seed Orchard (Jenkins
1982; Shea and others 1984) and the fir coneworm
Diorvctria abietivorella (Grote) at the Moscow
(Idaho) Arboretum ( Haverty and others 1985; Shea
1985). The western conifer seed bug Leptoglossus
occidentalis Heidemann and the lodgepole cone moth
Eucosma rescissoriana Heinrich have been observed
in the Moscow Arboretum; however, we have not yet
associated any significant damage with these
insects.

Paper presented at the Conifer Tree Seed in the
Inland Mountain West Symposium, Missoula, MT,
August 5-6, 1985.

Michael I. Haverty and Patrick J. Shea are Prin-
cipal Research Entomologists, Pacific Southwest
Forest and Range Experiment Station, Forest Ser-
vice, U.S. Department of Agriculture, Berkeley,
CA.

This paper reports the results of experiments
conducted in Idaho in 1981 and 1984 to evaluate
single and multiple, high- volume ground applica-
tions of two insecticides — permethrin and fen-
valerate— for protection of cones of blister
rust-resistant western white pine. Our objective
was to protect the cone crop from the major pest
at two test sites (Sandpoint and Moscow) that
attack cones during the second year of cone
development.

MATERIALS AND METHODS

The insecticides selected were chosen on the basis
of human safety and efficacy against £_,_ ponderosae
(Haverty and Wood 1981 ) or Dioryctria spp. and
Leptoglossus corculus (Say)(Nord and others 1984).

Sand po in t- Conophthorus

The study site was the Sandpoint Seed Orchard. In
1981, the 17.3-acre (7.0-ha) orchard contained 800
grafts of 21 -year-old western white pine. Trees
ranged in height from approximately 15 to 45 ft (5

246

to 15 m). Pennethrln (Pounce) was sprayed in
April 1981. Ten trees between 20 and 45 ft (7 and
15 m) tall were randomly assigned to each of seven
treatments. Ramets were from three clones with
similar susceptibility to £. ponderosae attack
(Jenkins 1982). Trees next to a previously selec-
ted tree or in a previously designated buffer zone
were not used. The seven treatments consisted of
an untreated check, and 0.03t 0.06 and 0.12 pet.
permethrin in water applied once, the same concen-
trations applied again 14 days later. Trees were
sprayed between 0500 and 0930 within a 2-day
period with a Bean hydraulic sprayer mounted on a
trailer. The sprayer was calibrated to deliver
approximately 5 gallons (18.9 liters) per tree
within 48 seconds. Mixing was done Just before
application and all trees within a treatment were
treated consecutively. The order of treatment
within any morning was random.

Twenty-one days following the second application
and after all attacks had occurred, all cones on
all trees were inspected and counted as infested
vs. noninfested. Because timing is critical to
protection of the cone crop, five screened cages,
each with 50 infested cones collected in the
orchard or from the surrounding area, were placed
throughout the orchard. Spraying began 1 day
after the first beetle emerged. Treatments were
analyzed by pairwise tests of differences with a 2
x 2 contingency table (infested vs. noninfested
vs. treatment) and a chi-square statistic at a =
0.01.

Previous work in the orchard indicated a high
probability that the beetle population would be
low in 1981 (Jenkins 1982). Because of concern
that insufficient attacks would not allow an
adequate test of the insecticide, beetles were
collected elsewhere and placed throughout the
orchard. Infested cones containing overwintering
beetles were collected in and around Sandpoint,
and held until the trees were treated. Immedi-
ately after spraying, 15 infested cones were
placed under test trees (including checks) to
ensure adequate attack.

Mo s co w-Diorvctrla

The Moscow seed orchard is rectangular in shape
and covers approximately 12 acres (4.9 na) on tne
western edge of the University of Idaho campus.
Our study area comprised the northeast quarter of
the seed orchard and had approximately 340 £.
monticola of cone-bearing age. Only trees with an
initial cone crop of J> 20 second-year cones were
used. The remainder of the seed orchard was
separated from the study area by a draw at least
60 ft (20 m) wide. To protect the majority of the
cone crop in the seed orchard from seed-destroying
insects, this area was treated three times (once
each in May, June, and July) with fenvalerate
applied aerially at a rate of 0.75 lb Al/acre
(0.81* kg Al/ha) delivered in 10 gallons of water
(93.5 liter/ha). The draw served as a clearly
visible buffer to guide the helicopter pilot so
that the insecticide would not drift into the
study area. Insecticide drift was monitored
during each spray application with water-sensitive
spray deposit cards. No drift was detected, and

the study area was assumed to be free of insec-
ticidal contamination from the adjacent control
operation.

Fenvalerate (Pydrin Insecticide 2.4 EC) was
diluted in water to a concentration of 0.025 pet.
(wt/wt) and applied with a trailer-mounted
sprayer. Mixing was done just before applica-
tion. The tank mixture was applied to near the
point of runoff with an FMC Bean hydraulic pumper
using a hand-operated gun. Trees were sprayed in
the early morning and late evening when wind was
minimal to avoid contamination of adjacent trees.
Between applications, spray equipment was cleaned
and rinsed with Nutra Sol.

Treatments were done on three dates in 1984: 9
May, 13 June, and 18 July. Since exact phenolo-
gies of the three insect pests were unknown, the
application schedules were modified after the
procedure of Nord and others (1984) to be approx-
imately 30 days apart. Pheromone traps baited
with candidate E. recissorlana pheromone were
distributed diagonally across the orchard at about
6 ft (2m) above ground in trees, and inspected
every other day for the appearance of adult
males. The first application coincided with the
first E. recissorlana catch.

The experiment was executed with a completely
randomized design. Four treatments were com-
pared: an untreated check, a single application
(9 May), a double application (9 May and 13 June),
and a triple application (9 May, 13 June, and 18
July). Each insecticide treatment was randomly
assigned to 12 trees; 22 trees were randomly
assigned to the untreated check. We selected
trees spaced sufficiently apart to avoid con-
tamination. As a result, we occasionally replaced
randomly selected trees if they were too tall or
too close to adjacent trees.

Before the first insecticide application, all
cones on the treatment and check trees were
examined and counted. Before each subsequent
insecticide application, all cones were re-
examined for obvious insect damage or the presence
of occidentalls adults or nymphs. Cones with
insect damage or insects present were flagged,
numbered, and left on the tree. A final observa-
tion was made on 21 and 22 August. Previously
infested and newly infested cones were collected
and bagged separately and returned to the labora-
tory in Berkeley, California. The remaining cones
were picked from 22 to 25 August. Cones were
counted, put in separate burlap bags, and air
dried.

The seed was extracted at the USDA Forest Service
Coeur d'Alene Nursery, Idaho. Uncleaned seed lots
from each tree were put in plastic bags and mailed
to Berkeley. All seed lots were carefully clean-
ed. Eight groups of 100 seeds from each tree were
weighed and placed in envelopes. The remaining
seeds were also weighed. Seeds per tree were
estimated based on the mean weight of the 800
seeds for that tree. In lots with less than 800
seeds, all seeds were counted.

247

The eight envelopes with 100 seeds per envelope
were taped to an 8 by 10 inch (20 by 25 cm) sheet
of paper and radiographed to determine percentages
of empty seed, seed with a viable embryo, or seed
damaged by L.. occiden talis or some unknown cause.
Individually collected cones damaged by insects
were air dried and the seed were extracted by
shaking. Seeds from each cone were counted and
radiographed. Data from infested and noninfested
cones were combined and used to calculate cone and
seed yields for each tree.

The response variables were number of cones har-
vested per tree, proportion of cones infested, and
number of seed per cone. Analysis of variance and
analysis of covariance with the number of cones
per tree as the covariate were used to detect
differences between treatments at the a = 0.05
level. Bonferroni's t-statistic (Miller 1980) was
used to compare means and to maintain a a = 0.05
for all comparisons (Jones 1981). Percentages of
filled, empty, or damaged seed were also computed.

RESULTS AND DISCUSSION
Sandpoint-Conophthorus

Beetles began to emerge on 18 April, 1981, and
trees were first sprayed on 19 and 20 April.
Trees were resprayed on 2 May. No statistically
significant difference occurred in the mean number
of cones per tree by treatment (table 1). All
permethrin treatments had levels of infested cones
that were significantly different from the un-
treated controls. Additional pairwise comparisons
revealed the following statistically significant
relationships in percent loss of cones: 0.03 pet.
once < 0.06 pet. once < 0.12 pet. once = 0.03 pet.
twice < 0.06 pet. twice < 0.12 pet. twice (table
1). Percent loss of cones ranged from 75.8 pet.
in the untreated controls to 1.7 pet. in the 0.12
pet. double treatment (table 1).

Mo s co w-Diorvctria

No statistically significant differences occurred
between treatments in the number of cones har-
vested per tree (table 2). However, the final
number of cones varied considerably between trees,
and ranged from 11 to 1 86 . Diorvctria abieti-
vorella infested 46.6 pet. of the cones in the
untreated check. This was significantly more than
in any of the insecticide treatments. The propor-

tion of infested cones among insecticide treat-
ments did not differ significantly (table 2).
Furthermore, we found no statistically significant
relationship between the number of cones on a tree
and the proportion of the cones which were
infested .

Observations in the seed orchard in previous years
indicated the presence of three potential pests of
cones and seeds. Although E. recissoriana was
present and captured in pheromone traps during
1984, we saw little evidence of damage by this
species. Leptoglossus occidentalis also had been
abundant during past years, but very few insects
were observed during our test. Little, if any,
cone damage was observed until 16 July 1984 (fig.
1). All damage was apparently caused by p_.
abietivorella. In July, less than 4.0 pet. of the
cones receiving the single application of 0.025
pet. fenvalerate were damaged by coneworms,
whereas ca. 25 pet. of the untreated cones were
infested. By August, the proportion of damaged
cones on untreated trees increased to 46.6 pet.
while only 13.6 pet. of the cones on trees sprayed
once and 4.1 and 5.1 pet. of the cones on trees
receiving two and three sprays, respectively, were
damaged (fig. 1; table 2).

Double or triple applications of 0.025 pet. fen-
valerate increased seed yield significantly com-
pared to the untreated check, that is, from 31.3
to 56.0 or 50.0 seeds/cone. The 95 pet. confi-
dence interval for the difference in mean seeds/
cone between two applications of fenvalerate and
the untreated check is 24.7+.15.3- In other words,
we are 95 pet. confident that this treatment
increased seed production by at least 9.4 seeds/
cone (an increase of 30.0 pet.), and possibly as
much as 40.0 seeds/cone (an increase of 127.8
pet. ).

Analysis of covariance showed no effect of cone
crop size on number of seed per cone during 1984.
All treatments also had approximately the same
proportion of filled, empty, and damaged seed
(table 2). This undoubtedly results from little
or no feeding by L. occidentalis and random
oviposition behavior of D. abietivorella.

The high cost of establishing seed orchards and
the fact that these orchards are the primary, if
not sole, source of resistant western white pine
seed make the develoment of effective insecticide
treatments an important research effort. The

Table 1. — Infested and noninfested western white pine cones, by permethrin treatment

Cones Cones Mean ^ ? loss

Treatment infested noninfested cones/tree of cones

0.03? once
0.06? once
0.12? once
0.03? twice
0.06? twice
0.12? twice
Untreated

254
203
105
131
66
8

547

234
379
353
478
600
488
174

48.8a
58.2a
45.8a
60.9a
66.6a
45.6a
72. 1a

52.0b
34.8c
22. 9d
21 .5d
9.9e
1 .7f
75.8a

Means in a column followed by the same letter are not significantly different at the a = 0.01 level.

248

50

45

35

!25

u 15

5

Application
I Cnsck

Two

□ tiw#

June

July

August

May

Figure 1. — Percent of western white pine cones
damaged by Diorvctria abietivorella. Single
applications of 0.025$ fenvalerate were applied on
9 May, 13 June and 18 July.

pet. pennethrin nearly eliminated cone losses.
However, 0.06 pet. permethrin applied once was the
most cost effective (Shea and others 1 98^4 ) . The
fir coneworm P_. abietivorella was the only insect
species to cause noticeable damage in the Moscow
Seed Orchard in 1 9 84 . This insect damaged almost
half the cone crop on unprotected trees and re-
duced seed yield by about 44 pet. Two applica-
tions of 0.025 pet. fenvalerate, once in May and
once in June, significantly increased seed yield.
A single application in mid-June might have been
sufficient to prevent coneworm damage, but was not
tested. A third application in July was apparent-
ly unnecessary.

ACKNOWLEDGMENTS

We thank Michael Jenkins, Larry Stipe, and Rodney
Nakamoto for managing various field aspects of
these experiments; Rodney Nakamoto for supervising
the preparation of the seed samples and processing
and analyzing the data; Allen Robertson, Julie
Duff in, and Lori Nelson for helping to extract and
clean the seed; Thomas Koerber for obtaining
radiographs of the seed; and James Baldwin for
clarifying our thinking in the evaluation of the
results.

REFERENCES

Bingham, Richard T. Blister rust resistant

western white pine for the Inland Empire: The
story of the first 25 years of the research and
development program. Gen. Tech. Rep. INT-146,
Ogden, UT: U.S. Department of Agriculture,
Forest Service, Intermountaln Forest and Range
Experiment Station; 1983. 45 p.

mountain pine cone beetle C.. ponderosae was the
only insect of concern in the Sandpoint Seed
Orchard in 1981. This insect has destroyed up to
75 pet. of the cone crop in this seed orchard. A
single 0.03 pet. application of permethrin signi-
ficantly reduced losses of the cone crop when
compared to the untreated check (from 75.8 pet.
loss to 52.0 pet. loss). Two applications of 0.12

Furniss, R. L. ; Carolin, V. M. Western forest
insects. U.S. Dep. Agric. Misc. Publ. No.
1339; 1977. 654 p.

Haig, Irvine T.J Davis, Kenneth P.; Weidman,

Robert H. Natural regeneration in the western
white pine type. U.S. Dep. Agric. Tech. Bull.
767; 1941. 98 p.

Table 2. — Cones harvested per tree, percentage of coneworm- infested
treated once, twice, or three times with 0.025? fenvalerate

cones, and seed per cone for trees

2

Treatments

Cones/tree^

Infested
Cones

3

Seed/cone

Filled
seed

Empty
seed

Damag
seed

Percent

- Percent -

Once

59.8 (8.8)a

13.6 (9.7)b

43.4 (9.0)ab

82.4a

17.1a

0.5a

Twice

62.1 (8.8)a

4.1 (9.7)b

56.0 (9.0)b

84.9a

14.8a

0.3a

Three times

65.8 (8.8)a

5.1 (9.7)b

50.0 (9-0)b

85.5a

13.8a

0.7a

Check

56.8 (6.5)a

46.6 (6.7)a

31.3 (6.6)a

85.9a

13.5a

0.6a

Infested by Dioryctria abietivorella (Grote)

Twenty-two trees were untreated and 12 trees each received either one, two, or three applications of
fenvalerate .

Mean (+95$ confidence limit) . Means in a column followed by the same letter are not significantly different
by Bonferroni's t-statistic at the -i = 0.05 level (Miller 1980).

249

Haverty, Michael I.; Wood, John R. Residual

toxicity of eleven insecticide formulations to
the mountain pine cone beetle, Conophthorus
monticolae Hopkins. J. Ga. Entomol. Soc.
16(1) :77-83; 1981.

Haverty, Michael I. ; Shea, Patrick J. ; Stipe,
Lawrence E. Single and multiple applications
of fenvalerate to protect western white pine
from Dioryctria abietivorella (Grote)
(Lepidoptera: Pyralidae) . J. Econ. Entomol.
(in review); 1985.

Hedlin, Alan F. ; Yates, Harry 0., Ill; Tovar,

David C. ; Ebel, Bernard H. ; Koerber, Thomas W. ;
Merkel, Edward P. Cone and seed insects of
North American conifers. Environment Canada,
Canadian Forestry Service, Ottawa, Ontario,
Canada; 1980. 122 p.

Jenkins, Michael J. Western white pine: the
effect of clone and cone color on attacks by
the mountain pine cone beetle. Ph.D. disser-
tation. Utah State Univ.; Logan, Utah; 1982.
91 P.

Jones, Davey. Use, misuse, and role of multiple-
comparison procedures in ecological and agri-
cultural entomology. Environ. Entomol. 13:
635-649; 1984.

Miller, Rupert G. , Jr. Simultaneous statistical
inference. 2nd ed. New York: Springer-Verlag;
1980. 299 P.

Nord, J. C. ; DeBarr, G. L.; Overgaard, N. A.;
Neel, W. W.; Cameron, R. S.; Godbee, J. F.
High-volume applications of azinphosmethyl,
fenvalerate, permethrin and phosmet for control
of coneworms (Lepidoptera: Pyralidae) and seed
bugs (Hemiptera: Coreidae and Pentatomidae) in
southern pine seed orchards. J. Econ. Entomol.
77:1589-1595; 1984.

Owens, John W.; Molder, Marte. Seed and cone
differentiation and sexual reproduction in
western white pine ( Pinus monticola) . Can. J.
Bot. 55:2574-2590; 1977.

Shea, Patrick J. Impact of insects on cone
production in two blister rust-resistant
western white pine seed orchards. Proc.
symposium conifer tree seed in the inland
mountain west, Missoula, MT. Aug. 5-7; 1985.

Shea, Patrick J.; Jenkins, Michael J.; Haverty,
Michael I. Cones of blister rust-resistant
western white pine protected from Conophthorus
ponderosae Hopkins (=C. monticolae) . J. Ga.
Entomol. Soc. 19( 1 ): 1 29-138 ; 1984.

Williamson, Darrell L.; Schenck, John A.; Ban,
William F. The biology of Conophthorus
monticolae in northern Idaho. For. Sci.
12:234-240; 1966.

INSECT PROBLEMS AND CONTROL EFFORTS ASSOCIATED WITH
SELECTED J)OUGLAS-FIR PLUS TREES^
William F.) Johns

ABSTRACT: The Beaverhead National Forest has ground-
sprayed and implanted their Douglas-fir (Pseudotsuga
menziesii) plus trees in an effort to overcome
effects of the western spruce budworm (Choristoneura
occidentalis) on cone crops. Both methods were
successful in protecting cones, but implants have
several advantages over spraying.

INTRODUCTION

The tree improvement program for Douglas-fir
(Pseudotsuga menziesii) calls for cones to be col-
lected from each of the plus tree stands assigned
to the Beaverhead National Forest. The seedlings
from these cones will be grown in a progeny test
plantation. This stage of the tree improvement
program has been held up for several years due to
the absence of cones in our Douglas-fir. Even
without the effects of insects, good cone crops in
Douglas-fir are infrequent on the Beaverhead
National Forest.

Added to the natural phenology of the plant is the
fact that Douglas-fir on the Forest has been
heavily infested with the western spruce budworm
(Choristoneura occidentalis) for a number of years.
The effect of the budworm on Douglas-fir cone
production is twofold: first, the budworm feeds on
new foliage, lowering the tree's vigor and thereby
lowering the tree's ability to produce cones.
Second, the larvae feed on conelets causing cone
mortality.

CONTROL EFFORTS

Around January 9, 1982, bud conditions indicated
that 1982 would be a good cone year. Shortly there-
after, I contacted Jed Dewey, Regional Entomologist,
Forest Service Northern Region, about what the
Forest could do to overcome the negative effects of
the budworm on cone production. With the help of
Jed and Larry Stipe, a program was established to
ground-spray our individual plus trees with a mix-
ture of carbaryl and water.

Between June 24 and July 15, 33 stands were
sprayed. The upper limit of the spray machine
was about 50 feet. At the time I felt that upward
drift would give us protection on to the top of
the tree. I later came to feel that this was not

the case. We did feel that the spray program was
successful in that it allowed us to collect cones
from many of our stands.

In 1984 we decided to try Acecap implants. In
retrospect, that was probably a poor decision due
to the poor cone crop. We had not done our home-
work on cone crop forecasting and were able to
collect from only five of the 25 stands that were
implanted. Neither method controlled cone midges.

PROJECT COSTS

Very poor cost records were kept on the spray
project, but it took about 20 person-days to
spray 33 stands. It took more time than desirable
because our stands are scattered and I was
unfamiliar with their locations.

Accurate cost records were kept on the implant
operation. Fifty trees in 25 stands were im-
planted at a cost of $2,485. At $50 per tree, the
cost seems to be excessive, but we had already
collected from easy-to-reach stands with the
spray operation. The biggest factor that drove
up the costs was snowmobiling into the stands.

SUMMARY

To summarize, I feel that both operations were
worthwhile, but the implants have considerable
advantages over spraying. They are:

1. Access with a vehicle is not critical.

2. Height of the tree is not a factor.

3. Weather is not a factor.

4. Individuals do not come in contact with
the chemical.

5. There is virtually no chance of
environmental contamination.

6. Coverage is more consistent.

Probably the weakest part of our program during
this time has been the ability to make an early
estimate of our upcoming cone crop. The know-
ledge exists to make such an estimate; we just
need to concentrate more on getting it done.
In spite of our poor performance in cone fore-
casting, however, the use of chemicals has allowed
us to reach a point where our collections are
almost completed.

Paper presented at the Conifer Tree Seed in the

Inland Mountain West Symposium, Missoula, MT, This spring indications were that a fair cone crop

August 5-6, 1985. was coming so we implanted our remaining

uncollected stands (18).

William F. Johns is Forest Silviculturist ,
Beaverhead national Forest, Forest Service, U.S.
Department of Agriculture, Dillon, MT.

251

^ IMPLANTATION AND INJECTION OF SYSTEMICS TO INCREASE

SEED YIELD IN

Richard C. jReardon

ABSTRACT: The western spruce budworm,
Choristoneura occidentalis Freeman, and spruce
coneworm, Dioryctria reniculelloides Mutuura and
Munroe, cause widespread damage to cones of
Douglas-fir, Pseudotsuga menziesii var. glauca
(Beissn.) Franco. The systemic insecticide
acephate injected at 4-inch (10-cm) and 6-inch
(15-cm) spacings, and implanted at 4-inch spacing
in Douglas-fir, increased the yield of filled
seeds when compared to the checks.

INTRODUCTION

Douglas-fir, Pseudotsuga menziesii var. glauca
(Beissn.) Franco, seed production areas as well
as natural stands are subject to unpredictable
year-to-year variation in cone crops. In the
northern Rocky Mountains, western spruce budworm,
Choristoneura occidentalis Freeman, is often the
most serious insect pest affecting Douglas-fir
cones (Dewey 1970). Other major insect pests
are the Douglas-fir cone moth, Barbara
coif axiana (Kearfott); the Douglas-fir scale
midge, Contarinia washingtonensis Johnson; the
Douglas-fir seed chalcid, Megastigmus
spermotrophus Wachtl (Dewey 1968, 1969) ;
and the spruce coneworm, Dioryctria
reniculelloides Mutuura and Munroe.

Historically, attempts to suppress populations
of seed and cone insects have relied on chemical
insecticides applied as foliar sprays (Dewey and
others 1975; Stipe and Hard 1980; Stipe and Green
1981). This strategy is not practical for widely
scattered trees in rough terrain or where
adjacent areas might be sensitive to
contamination. Implantation and injection
of systemic chemical insecticides offer
promise for individual tree protection.

Medicaps are the most widely used implantation
method. Medicaps containing powdered acephate
and designated as ACECAPS have been used to
protect foliage and reduce western spruce
budworm larval populations on Douglas-fir and
grand fir, Abies grandis (Dougl.) (Markin 1979;
Reardon and Haskett 1981; Reardon 1984a; Reardon
and Barrett 1984). ACECAPS implanted at 4-inch

Paper presented at the Conifer Tree Seed in
the Inland Mountain West Symposium, Missoula,
MT, August 5-6, 1985.

Richard C. Reardon is Entomologist, Northeastern
Area Forest Pest Management, U.S. Department of
Agriculture, Forest Service, Morgantown, WV;
Larry E. Stipe is Entomologist, Northern Region,
U.S. Department of Agriculture, Forest Service,
Missoula, MT.

DOUGLAS-FIR,
>/

id Larry E.j Stipe

(10-cm) spacing are registered fcr control of
spruce budworms in the United States (Reardon
1984b).

Mauget systemic injector units containing
oxydime ton -methyl and designated Inj ect-A-Cide
are effective in reducing insect populations and
increasing seed yield on Douglas-fir in
California (Koerber 1978; Dale and Frank 1981).
At present, mauget units with acephate are not
registered for control of spruce budworms.

This paper reports a study to determine the
effectiveness of Medicaps containing acephate or
dimethoate and Mauget systemic injector units
containing oxydemeton-methyl or acephate at two
spacings in increasing the seed yield of
Douglas-fir. Nutrients were also injected at
6-inch (15-cm) spacing.

MATERIALS AND METHODS

The study area of 500 acres (200 ha) , about 10 miles
(16 km) west of Whitehall, MT, contained open-grown
Douglas-fir at ca. 5906 feet (1800 m) elevation.
The trees ranged in height from 33 feet (10 m) to
64 feet (20 m) and averaged 37.1 ± 2.3 cm (x ± SD)
in diameter at 4 inches (10 cm) from the soil
surface. Most of the sample trees were separated
by a distance of at least 20 m, numbered and
randomly assigned one of eight treatments:
oxydemeton-methyl injected at 10- or 15-cm
intervals, nutrients injected at 15 cm, acephate
injected at 10 or 15-cm, dimethoate or acephate
implanted at 10 cm, and untreated checks. There
were 20 sample trees per treatment except for
acephate injected at the 10- (8 trees) and 15-cm
(10 trees) spacings, due to a limited number of
injector units.

Trees were treated on 6 and 7 April 1982, when
cone buds were swollen and vegetative buds were
still tight.

The powdered formulations of acephate (97 percent
Orthene) and dimethoate (95 percent Dimethoate) were
introduced directly into the xylem by using
Medicap plastic cartridges (1 cm x 3 cm). Each
acephate cartridge contained 0.9 gm active
ingredient (AI) and each dimethoate cartridge
contained 0.6 gm AI.

Liquid formulations of oxydemeton-methyl, as 50
percent Metasystox-RR; acephate, as 35 percent
Orthene; and nutrients, as 1 percent iron and 1
percent zinc, were injected into the xylem by
using Mauget injection units. Each unit con-
tained 1.5 g AI of oxydemeton-methyl or 1.8 g
AI of acephate. The Mauget injectors were re-
moved after 12 days. ,

252

Four methods were used to process mature cones —
air drying and shaking in bags, axial slicing,
dismantling, and nursery processing; thereby,
four estimates of the proportion of filled seeds
per cone were obtained per treatment. Fifty
mature cones were collected from the upper half
of the crown from each sample tree on 30 August
1982, Twenty-five of these cones were placed in
a paper bag, allowed to air dry, shaken, and the
proportion of opened cones, external cone
damage, and total dislodged seeds per tree were
recorded. One hundred seeds from each tree were
randomly selected and X-rayed to determine the
proportions that were filled, hollow, and damaged
by insects.

The other 25 cones from each tree were bisected
longitudinally along the cone axis with cone
cutters. One cut surface of each cone was
examined for the number of seeds filled, hollow,
and damaged by insects (Dobbs and others 1976).
A subsample of 10 sliced cones per tree was
dismantled by removing one scale at a time from
the axis. The average number of seeds damaged by
lepidopterous larvae, and numbers of scale midge
and Douglas-fir cone-gall midge, Contarinia
oregonensis Foote, per cone were recorded.

Three 1-bushel (35.2-L) samples of mature cones
were collected per treatment on 1-3 September
1982, with pole pruners and by climbing the trees.
Each 1 bushel of cones was collected from at
least three sample trees for a treatment. At
Luck Peak Nursery in Boise, Idaho, the cones were
air and kiln dried, followed by tumble extraction
of the seeds, seed dewinging, and air separation
of filled and hollow seeds. A subsample of 300
seeds taken from the seeds determined as "filled"

by air separation that were recovered from each
bushel were X-rayed to determine the proportion
filled, hollow, and damaged by insects. A
subsample of 200 sound seeds recovered from each
bushel was forwarded to the Idaho State Seed
Laboratory to assess germination.

All data were analyzed by using the Games and
Howell T modification for paired multiple
comparisons with unequal variances (Keselman and
Rogan 1978).

RESULTS

Damage by insects, as measured for mature cones
processed by shaking, by axial slicing, and by
dismantling, was significantly less in some
treatments (table 1) . The average proportion of
external cone damage for each treatment, except
nutrients and dimethoate, was significantly less
than that for the checks. The average proportion
of opened cones for each treatment provided an
additional estimate of external cone damage;
significant differences between treatments and
checks were similar to those for external cone
damage.

For the mature cones bisected longitudinally, the
average insect-damaged seed per cone surface was
significantly less for acephate injected at 10-cm
intervals than that for each of the other
treatments and checks.

For mature cones that were dismantled, the average
number of seeds per cone damaged by lepidopterous
larvae was significantly less for each acephate
treatment than for the checks.

lahle I. — Insect damage to Douglas-fir cones on trees Injected or Implanted with systemic Insecticides or nutrients, Montane 1982

Shaking

Cones External
processed cone damage

per (I)'
t reatment

X ± SE

Cones opened

(X)1

Slicing

Dismantling

Cones
processed

per
treatment

Insect-damaged

seeds per
cone surface

.1

Cones
processed

per
treatment

Insect-damaged seeds per cone1

C.

C. washlngton-
oregonensls cnsls Lepldopteran

SE

SE

Injected

Oxydemeton-me thy 1

15 cm 500 29. 7l ♦ 3.1 b 63 t 1.0 b 494 1.8 t 0.4 a 200 0.44 ± 0.16 0.07 ± 0.04 9.9 ♦ 1.2 a

10 cm 475 33.2 ♦ 2.9 b 54 ♦ 1.1 b 475 3.0 ♦ 0.4 a 200 0 - 0.42 ♦ 0.18 9.5 ♦ 1.4 a
Acephate

15 cm 250 17.9 ♦ 4.2 b 76 ♦ 1.4 c 250 1.8 ♦ 0.3 a 100 0 - 0.65 i 0.31 8.5 i 1.3 b

10 cm 160 10.0 1 3.5 c 88 ; 0.0 c 198 0.7 ♦ 0.2 b 80 0.90 i 0.38 0.11 ♦ 0.10 3.1 1 0.6 b

Nutrients— 15 cm 500 43.2 ♦ 3.4 a 40 1 1.1 a 492 2. 1 1 0.3 a 200 0.67 1 0.23 0 - 13.3 ♦ 1.4 a
I mp 1 anted

Acephate— 10 cm 500 20.7 i 3.6 b 64 1 1.3 b 498 1.9 ♦ 0.3 a 200 0.15 ♦ 0.01 0.29 ♦ 0.11 4.5 ♦ 1.0 b

Dimethoate— 10 cm 450 41.5 i 4.1 a «8 ! 1.4 a 488 2.5 ♦ 0.3 a 200 0.06 i 0.04 0 - 10.5 ± 1.1 a

Checks 500 50.7 ♦ 3.9 a 43 i 11.3 a 493 3.0 1 0.4 a 200 0.16 i 0.08 0 - 14.6 ± 1.2 a

Means In the sane column followed by the same letter do not differ significantly (P>0.05).

253

Most damage to seeds and cones was caused by the
western spruce budworm and spruce coneworm.
Insects found in low numbers (avg. < 1 per cone)
were the Douglas-fir seed chalcid, cone-scale
midge and cone-gall midge.

Trees treated with dimethoate or nutrients
consistently yielded greater numbers of filled
seeds per cone than did the checks, but the
difference was not significant (table 2) .
Significantly and consistently greater numbers of
filled seeds per cone were detected, except for
the nursery process, for oxydemeton-methyl at
15-cm intervals, and for acephate implanted and
injected at 10 and 15-cm intervals, compared with
the checks. Oxydemeton-methyl at the 10-cm
spacing was not as consistently effective at 15-cm
even though more insecticide was injected into
each tree.

DISCUSSION

Oxydemeton-methyl at 15-cm intervals, acephate
injected at 10- and 15-cm intervals and implanted
acephate consistently and significantly increased
the number of filled seeds per cone when compared

with the checks. Acephate injected at 10-cm
intervals was more effective than the
oxydemeton-methyl treatments in protecting cones,
as determined by external cone damage and
percentage of opened cones, and seeds from damage
by lepidopterans .

Air drying and shaking was the least time consum-
ing of the four methods used to process mature
cones. Numbers of filled seed per cone determined
by this method were higher than determined by
axial slicing, lower than by dismantling, and
higher than by processing at the nursery.

Medicaps, with implanted acephate at 10-cm
intervals and Maugets, with injected acephate at
10-cm intervals, are both effective in reducing
larval densities of western spruce budworm and
spruce coneworm and in increasing the number of
filled seed per cone for Douglas-fir. Medicaps
are registered for use against spruce budworms and
do not require removal of the empty cartridge from
the drilled hole. Maugets, with acephate, are not
registered for use against spruce budworms,
although the company has petitioned for
registration. Both Medicaps and Maugets
containing acephate are available at the same cost
per cartridge or unit.

Table 2. — Filled Douglas-fir seed per cone (means ± SE) by treatments and cone-processing methods,
Montana, 1982

Treatment

Shaking

Slicing
(per cone surface)

Dismantling

Nursery
Processing

Inj ected

Oxydemeton-methyl

15 cm

10 cm
Acephate

15 cm

10 cm
Nutrients — 15 cm

13.8 - ± 1.6 b

9.4 ± 1.1 a

13.6 + 1.3 b

16.0 + 2.6 b

8.0 ± 1.2 a

4.1 ± 0.5 b

3.2 + 0.4 b

4.3 ± 0.7 b
3.3 ± 0.4 b
2.3 ± 0.3 a

15.3 ± 2.0 b

12.0 ± 1.3 b

11.0 ± 1.2 b

17.6 ± 2.9 b

9.5 + 1.3 a

6.6 ± 1.0

7.6 ± 1.6

9.5 ± 1.6

8.4 ± 0.8

8.5 ± 0.8

Implanted

Acephate — 10 cm
Dimethoate — 10 cm

14.9 ± 1.9 b
7.0 ± 1.3 a

5.0 ± 0.5 b
2.8 ± 0.4 a

16.9 ± 1.6 b
10.3 ± 1.5 a

13.9 ± 2.9
5.7 ± 1.0

Checks

5.5 + 0.9 a

1.6 ± 0.2 a

6.0 ± 0.7 a

5.6 ± 0.7

1/ Means in the same column followed by the same letter do not differ significantly (P > 0.05).

254

REFERENCES

Dale, J.W.; Frank, C.L. Injecting

metasystox-R increases seed yields of large
superior Douglas-firs by reducing losses caused
by cone insects. U.S. Department of
Agriculture, Forest Service, FPM Rep. 81-33;
1981. 10 p.

Dewey, J.E. A study of insects attacking
Douglas-fir cones in eastern Montana and
Yellowstone National Park, 1967. U.S.
Department of Agriculture, Forest Service;

1968. 15 p.

Dewey, J.E. Results of a Douglas-fir cone
and seed insect study in Montana and
Yellowstone National Park, 1968. U.S.
Department of Agriculture, Forest Service;

1969. 9 p.

Dewey, J.E. Damage to Douglas-fir cones by
Choristoneura occidentalis . J. Econ. Ent.
63:1804-1806; 1970.

Dewey, J.E.; Meyer, H.E.; Parker, D. ; Hayes, F.
Ground application of dimethoate (cygon) for
the control of cone and seed insects of
Douglas-fir and grand fir. U.S. Department of
Agriculture, Forest Service, FPM Rep. 75-13;
1975. 6 p.

Keselman, H.J.; Rogan, J.C., A comparison of the
modified Tukey and Schaffe' methods of multiple
comparisons for pairwise contracts. J. Am.
Statist. Assoc. 73:47-52; 1978.

Koerber, T.W. Bole injected systemic insecticides
for control of Douglas-fir cone insects. In:
Proceedings Symposium on Systemic Chemical
Treatments in Tree Culture, Michigan State
University; 1978: 295-302.

Markin, G.P. Western spruce budworm control with
Orthene implants, 1978. Insecticide and
Acaracide Tests. 4:175; 1979.

Reardon, R.C. Two consecutive yearly applications
of Orthene medicaps increase protection of
grand fir against western spruce budworm.
Forest Ecology and Management. 7:183-190; 1984.

Reardon, R.C. How to protect individual trees from
western spruce budworm by implants and
injections. CANUSA Spruce Budworm Program.
Agriculture Handbook No. 625; 1984b. 15 p.

Reardon, R.C; Barrett, L. Effects of treating

western spruce budworm populations on grand fir
and Douglas-fir with acephate, carbofuran,
dimethoate, oxydemeton-methyl , and
methamidophos. Forest Ecology and Management .
8:1-10; 1984.

Reardon, R.C; Haskett, M.J. Effect of

Orthene medicaps on populations of western
spruce budworm on grand fir and Douglas-fir.
J. Econ. Ent. 74:266-270; 1981.

Stipe, L.E.; Hard, J.S., An early single tree
application for Douglas-fir foliage and cone
protection from the western spruce budworm.
U.S. Department of Agriculture, Forest Service,
FI&DM Rep. 80-3; 1980. 19 p.

Stipe, L.E.; Green, A.K. A multiple ground
application of acephate and carbaryl for
protection of Douglas-fir cones from western
spruce budworm. U.S. Department of
Agriculture, Forest Service, FPM Rep. 81-22;
1981. 17 p.

255

IMPACT OF INSECTS ON CONE/SEED PRODUCTION IN THREE

BLISTER RUST-RESISTANT WESTERN WHITE PINE SEED ORCHARDS

ABSTRACT: Little is known about the impact of
insects on production of blister rust-resistant
western white pine seed. Results from sampling
the only three producing seed orchards in Idaho
indicate insects can severely reduce production.
Further, the insect species responsible for damage
differ among the various orchards. A recommen-
dation for locating future orchards in white pine
type is made based on the potential for minimizing
insect-caused seed losses.

Y/

Patrick J.j Shea

coneworms have been reared from cones. If we are
to develop pest management systems for these
orchards, it is critical that we accurately assess
the impact of insects on BRR/WWP seed, and attempt
to rank or quantify the amount of seed damaged by
each insect species. The objective of this paper
is to present the results of the 1984 sampling
efforts.

MATERIAL AND METHODS

INTRODUCTION

A significant accomplishment of the breeding pro-
gram for blister rust-resistant western white pine
(BRR/WWP) is the establishment of three producing
seed orchards in Sandpoint, Coeur d'Alene and
Moscow, Idaho (Bingham 1983). Western white pine,
Pinus monticola Douglas, is valued for its clean-
boled form, and soft, white, easily-processed
lumber. It regenerates naturally and is charac-
terized by relatively rapid initial and continued
growth. The effect and history of the introduced
fungus, Cronartium ribicola Fisher, on WPP, and
attempts by federal and state agencies to combat
the spread of blister rust is well known (Haig and
others 1941; Hepting 1971). Because of its econ-
omic importance, forest land managers of the
northern Rocky Mountains have made the reestab-
lishment of WWP to its former habitat a priority
management objective.

Specific data are lacking on the effects of
insects on production of BRR/WWP seed. However,
insects are suspected of being responsible for
substantial seed losses in all three orchards.
Previous research in the Sandpoint orchard
strongly indicates that Conophthorus ponderosae
( =C. monticola Hopkins), the mountain pine cone
beetle, is the insect primarily responsible for
losses of up to 90 percent of the cones (Jenkins
1982; Bingham 1983). The situation in Moscow and
Coeur d'Alene is not quite as clear. Observations
by Forest Pest Management, R-1, and orchard per-
sonnel indicate that at least three insect genera
are present and probably cause substantial loses.
Species such as Leptoglossus occidentalis
(Heidemann), western conifer seed bug, Eucosma
recissoriana Heinrich, lodgepole pine cone borer,
and one or more species of Dioryctria, the fir

Paper presented at the Conifer Tree Seed in the
Inland Mountain West Symposium, Missoula, MT,
August 5-6, 1985.

Patrick J. Shea is Principal Research Entomolo-
gist, Pacific Southwest Forest and Range Experi-
ment Station, U.S. Department of Agriculture,
Forest Service, Berkeley, CA.

The three producing orchards were periodically
sampled for insect damage throughout the late
spring and summer of 1984. The northernmost
orchard is located in Sandpoint, Idaho. It was
established in 1960, and is about 17 acres (7 ha)
in size. It contains 800 grafts from 13 clones
and has been producing harvestable quantities of
cones since 1978. In this orchard 24 trees,
ranging from 20-45 ft (9 to 15 m) in height, were
randomly selected for study.

The Coeur d'Alene orchard is located approximately
42 miles (67.5 km) south of Sandpoint. This
orchard was established in 1960 and is about 13
acres (5 ha) in size. It is considered a low-
elevation orchard and is stocked by the same
families that occur in the Moscow Arboretum. This
orchard is just beginning to produce harvestable
quantities of cones but still requires artificial
pollination. It had very few cone bearing trees
in 1984 so that it was possible to examine the
cones on all trees (68) that had at least three
cones per tree.

The Moscow Arboretum is located about 70 miles
(112.7 km) south of Coeur D'Alene and beginning in
1957 was planted with WWP seedlings that survived
intense, artificial inoculation with blister rust
(Hoff and Coffen 1982). This orchard covers about
23 acres (8 ha), and 22 trees 20-45 ft (9-15 m)
in height were randomly selected for study. The
seed produced by this orchard is shared among_the
11 cooperatives in the Inland Empire Cooperative
Tree Improvement Program. Only trees in the NW
quadrant of the orchard were used. It is noted
that this orchard is on the edge of the Palouse
and is quite some distance (> 50 miles [129 km])
from any natural stands of WWP.

A hydraulic manlift was used to sample all trees.
The first sample (May 18, 1984) was started one
week after the second-year cones began to elon-
gate. All cones on all sample trees were examined
for insect entrance holes, presence of L.. occiden-
talis (nymph or adult), or any other external
evidence of damage. Mean numbers of cones per
tree were calculated for all orchards and ANOVA
(alpha =.05) was used to detect significant

256

differences between orchards. Mean numbers of
seeds per cone were calculated for the Moscow and
Coeur d'Alene orchards only. Damaged cones were
flagged and coded so that individual cones could
be re-examined throughout the summer. At the end
of the summer all damaged cones were removed and
taken to the laboratory where they were dissec-
ted. The remaining cones on all sample trees in
Coeur d'Alene and Moscow were picked, counted, put
in separate labeled burlap bags, and air dried.
Seeds were extracted at the USDA Forest Service
Coeur d'Alene Nursery. These uncleaned seeds were
put in plastic bags and shipped to Berkeley, CA
for analysis. Seed lots were kept separate by
tree. Eight envelopes per tree with 100 seed per
envelope were radiographed to determine percentage
of filled seed with viable embryo and empty seed,
or seed damage by L_. occidental is. or some other
unknown cause. This provided the estimate for
seeds per cone and damage by L_. occidentalis.

RESULTS Afro DISCUSSION

In 1984, the Sandpoint Orchard had significantly
more cones per tree (88.8 cones per tree) than
either Moscow (56.9 cones per tree) or Coeur
d'Alene (5.2 cones per tree)(alpha = .05, table
1). The small cone crop in Coeur d'Alene compared
to either Moscow or Sandpoint was not unexpected
since this orchard has just recently begun to
produce cones. However, cone production would
have been much higher than 5.2 cones per tree
except that there was a high rate of cone abortion
in the fall of 1 983 . The yield of 31 .3 seeds per
cone in the Moscow Arboretum is not much different
than the 39 seeds per cone reported by Hoff and
Coffen (1982) for this same orchard. The very low
seed per cone (10.2) yield at Coeur D'Alene is
thought to be due to poor pollination.

Five species of insects from three orders and four
genera cause damage to BRR/WVP seeds and cones in
the three orchards. When cone beetles emerge from
overwintering diapause, the adult females attack
maturing cones and kill them in the process. Each
attacking female is later joined by an adult
male. After mating she lays eggs as she feeds
down the axis of the cone. One female may attack
several cones (Jenkins 1982; Williamson and others
1966). Upon hatching, the larvae begin feeding
throughout the cone. Both adult and immature
forms of the western conifer seed bug feed on the

Table 1 . — Number of trees sampled in each of three
western white pine seed orchards, 1984

Mean Mean
Orchard N Cones/ Tree Seeds/Cone

( + SE) (+SE)

Moscow 22 56.9 (±6.1) 31.3 (3.16)

Coeur d'Alene 68 5.2 (+2.2) 10.2 (0.68)

Sandpoint 24 88.8 (+12.1) NA

NA = Not assessed.

individual maturing seeds of a wide range of coni-
fers. The life history and habits of the seedbug
on WWP are not completely known, but in natural
Douglas- fir stands, there is one generation per
year with ovipositing adults present from May to
July and immatures from June to September (Hedlin
and others 1 980 ) . This generally coincides with
our observations in 1984. The fir coneworms and
the lodgepole pine cone borer both attack the
cones of several species of conifers. Adults
oviposit on the surface of developing cones and
the resulting larvae mine throughout the cone and
destroy much of the seed. Preliminary results of
pheromone trapping in 1984 and 1985 indicate that
J3. recissoriana adults begin to appear in the
orchard in early June. In 1985, Diorvctria spp.
were not captured in pheromone traps until two
weeks after the first £. recissoriana were caught.

The Moscow and Coeur d'Alene orchards experience
considerably more cone damage, 46.6 percent and
47.3 percent respectively, than the Sandpoint
orchard (10.4 percent) (fig. 1). In Moscow vir-
tually all the damage was attributed to the fir
coneworms and perhaps the lodgepole pine cone
borer (fig. 2). We have been unable to assess
the amount of damage by species in this orchard.
Very few seed bugs were observed in the orchard
during sampling. In Coeur d'Alene, D. abietivorella
was the only lepidopteran species causing damage;
37 percent of the cones were infested with larvae
of this species. Cones killed by C. ponderosae were
also collected from this orchard but in relatively

70-

65-

60-
. 55-
£50-

o

I45'

|40-

<D 35"

'5 30-
o

S 25-

o

5 20-
* .5-

10- I
5-

0J 1 1 1 1 1

Moscow CDA Sandpoint

Orchard

Figure 1 .--Total percentage of infested cones in
western white pine seed orchards, 1984.

257

70 -j

65-
60-
? 55-

<D

^50-

Q.

HI Lepidoptera

ZZ1 Mountain pine cone beetle

I I Unknown damage

Moscow
CD'A

Moscow CDA Sandpoint

Orchard

Figure 2. — Percentage of infested cones by insect
group for western white pine seed orchards, 1984.

small numbers (fig. 2). L. occidentalis popu-
lations were heaviest at Coeur d'Alene and results
of radiographed seed reveal that 12 percent of the
seed were destroyed by adults and immatures of
this species. The full impact of L.. occidentalis
on production of BRR/WWP is not fully captured by
x-ray analysis of seed. There is considerable
suspicion that this insect may be responsible for
a large share of the conelet abortion experienced
in Coeur d'Alene during the fall of 1983. Studies
are in place to validate whether L. occidentalis
is causing conelet abortion and, if so, to what
extent. C. ponderosae was the only insect species
causing damage in the Sandpoint orchard (fig. 2).
None of the lepidopteran species were recovered in
Sandpoint and only an occassional L.. occidentalis
was observed.

Earliest damage occurs in the Sandpoint orchard
during late April to mid-May, with the emergence
of adult female cone beetles (fig. 3), but the
attack period is completed by mid-June (Jenkins
1982; Shea and others 1984). At Coeur d'Alene and
Moscow, visible damage begins to appear in mid-
June and continues to accumulate through August
(fig. 3).

Several tentative observations can be made from
the 1984 impact study, all of which relate to the
development and implementation of an IPM system.
First, there appear to be distinct differences
in the insect complex associated with the three
BRR/WWP seed orchards. Second, there may be an
association between the amount of damage exper-
ienced and the number of species causing damage.
The more pest species in the orchard the greater

Figure
sampli
1984.

June July
Assessment
Period

3. — Percentage of infested cones by
ng date in western white pine seed orchards,

the damage, e.g. Moscow vs. Sandpoint. Thirdly,
in those orchards with multiple pest species the
period of cone and seed vulnerability or insect
attack period is longer than it is in the orchards
with a single pest species such as Moscow vs.
Sandpoint. If these initial observations are
confirmed during the remaining two years of study,
development and implementation of an IPM system in
each orchard could be profoundly affected. For
instance, decisions regarding management strate-
gies to protect cones from damage in orchards with
multiple pests may be quite different than in
orchards with a single pest, i.e. multiple insec-
ticide applications vs. a single application; or
multiple monitoring systems vs. single monitoring
system.

Lastly, note that the Sandpoint orchard is the
only one of the three orchards studied that is
located in white pine type; whereas, both the
Moscow and Coeur d'Alene sites are located out of
type. In addition, note that the three lepidop-
teran species and L. occidentalis can all be con-
sidered generalists as defined by Fox and Morrow
(1981), that is they utilize a more diverse array
of host plants than does C. ponderosae (Hedlin
1981). This suggests that contrary to the recom-
mendation of Hoff and Coffen (1982) and Bingham
(1983), it may be advantageous to locate future
BRR/WWP seed orchards in white pine type. In
doing so the orchard manager may only be con-
cerned with management of a single pest as
compared to a multiple pest situation and its
attendant complexity.

258

ACKNOWLEDGMENTS

I extend thanks to Larry Stipe, Rodney Nakamoto,
and Mel Morton for management and assistance in
various field and laboratory aspects of this
study. Special thanks to Rodney Nakamoto for
supervising the preparation of the seed samples,
and processing and analyzing the data. I thank
Allen Robertson, Julie Duffin, and Lori Nelson for
helping with the miserable task of extracting and
cleaning the seed, and Tom Koerber for obtaining
radiographs of the seed. Lastly I thank Mike
Haver ty, John Hard, Jed Dewey, and Marion Page for
their helpful comments in the preparation of this
paper.

Williamson, Darrell L. ; Schenck, John A. ; Ban,
William F. The biology of Conophthorus
monticolae in northern Idaho. For. Sci. 12
234-240; 1966.

REFERENCES

Bingham, Pichard T. Blister rust resistant western
white pine for the Inland Empire: The story of
the first 25 years of the research and develop-
ment program. Gen. Tech. Rep. INT-146, Ogden,
UT: U.S. Department of Agriculture, Forest
Service, Intermountain Forest and Range Experi-
ment Station; 1 983 . 45 p.

Fox, L. R.; Morrow, P. A. Specialization; species
property or local phenomenon? Science. 211: 887-
893; 1981.

Haig, Irvine T. ; Davis, Kenneth P.; Weidman, Robert
H. Natural regeneration in the western white
pine type. U.S. Department of Agriculture Tech-
nical Bulletin 767; 1941 . 98 p.

Hedlin, Allen F.; Yates, Harry 0. Ill; Tovar, David
C. ; Ebel, Bernard H. ; Koerber, Thomas V/.;
Merkel, Edward P. Cone and seed insects of
North American conifers. Environment Canada,
Canadian Forestry Service, Ottawa, Ontario,
Canada; 1980. 122 p.

Hepting, George H. Diseases of forest and shade
trees of the Onlted States. U.S. Department of
Agriculture, Agric. Handb. 396; 1971. 659 P.

Hoff, P. J; Coffen, D. 0. recommendations for
selection and management of seed orchards of
western white pine. Research Note INT-325,
Ogden, Utah: U.S. Department of Agriculture,
Forest Service, Intermountain Forest and Range
Experiment Station; 1982. 7 p.

Jenkins, Michael J. Western white pine: the

effect of clone and cone color on attacks by the
mountain pine cone beetle. Ph.D. dissertation.
Utah State Univ.; Logan, Utah; 1982. 91 p.

Shea, Patrick J.; Jenkins, Michael J.; Haverty,
Michael I. Cones of blister rust-resistant
western white pine protected from Conophthorus
ponderosae Hopkins (=C. monticolae) . J. Ga.
Entoool. Soc. 19(1): 129-138; 1984.

259

i?

?) INFLUENCE OF DISEASES ON SEED PRODUCTION

R.'^Si

Jack R.'i Sutherland

ABSTRACT: In this, the first of three papers on
the influence of diseases on seed production,
hosts, life cycles and damage, and management
recommendations are given for important diseases
of Inland Mountain West seed orchard trees ,
cones and seeds. Highlighted are Armillaria
root rot, needle diseases, Inland spruce cone
rust, the seed or cold fungus and Sirococcus
blight.

INTRODUCTION

Although foresters and pathologists have always
assumed that diseases affect seed production, it
is only with the recent development of seed
orchards and container nurseries that the
importance of diseases of seed-producing trees
and their crop have begun to be clearly
defined. The high value and feasibility of
protecting seed orchard trees and their cones
and seeds has both necessitated and justified
the development of management strategies for
these diseases while container seedling
production has demonstrated the importance of
seed-borne pathogens in seedling disease
incidence and losses. Many seed-borne problems
either did not develop, were not evident, or
were attributed to other causes in bareroot
nurseries. Because of the increasing demand for
high quality seeds we need to know how to
recognize and manage diseases that affect seed
orchard trees, cones and seeds and seedling
pathogens such as Fusarium that are seed-borne.
Not only have technological advancements
increased our awareness of these various
diseases, but new techniques such as the use of
monoclonal antibodies have changed and improved
our methods of assaying for pathogens. My goal
and that of Drs . James and Mitchell in this
portion of the symposium is to update and
synthesize what is known about diseases
affecting conifer seed production in the Inland
Mountain West.

Review paper presented at the Conifer Tree Seed
in the Inland Mountain West Symposium, Missoula,
MT. August 5-6, 1985.

Jack R. Sutherland is a Senior Research
Scientist and Project Leader (Regeneration
Pests), Pacific Forestry Centre, Canadian
Forestry Service, 506 W. Burnside Rd. , Victoria,
B.C. V8Z 1M5, Canada.

DISEASES OF SEED ORCHARD TREES

Any tree disease likely harms seed production or
quality (e.g. Schaffer and others 1983). This
is true even when the disease results in
increased cone production such as happens with
distress cone crops on root-rot diseased trees
because probably such cones yield inferior
seeds. To date there is no evidence that it is
worthwhile controlling diseases of seed
producing trees in natural stands, including
those where some silvicultural treatment has
been used to stimulate or facilitate cone
production, thus the following deals only with
diseases of seed orchard trees. In seed
orchards, tree and crop values plus easy access
usually make disease management practical.
Potentially, any species of seed orchard tree is
susceptible to all the diseases affecting it in
nature. However, the special environment of
seed orchards created by fertilizing, watering,
growing trees out of their natural geographic
locality and other practices, may increase the
risk and impact of diseases or alter their host
range .

Experience indicates diseases encountered on
orchard trees varies with the orchard's age.
Usually the first to appear are diseases
indigenous to the site or those acquired by the
seedlings prior to outplanting in the orchard.
For example, Armillaria root rot, Armillaria
me Ilea (Vahl.: Fr.) Quel.), has killed 1-2% of
the white spruce, Picea glauca (Moench) Voss, in
an Interior B.C. orchard in the 5 years
following planting, but losses are expected to
decrease as inoculum, which was present in the
forest that was cleared for the orchard, dies
out. Western gall rust, Endocronartium
harknessii (J. P. Moore) Y. Hirat. , of lodgepole
pine, Pinus contorta Dougl. exemplifies a
disease that can be brought into a seed orchard
on diseased stock. Such seedlings not only
introduce the disease into the orchard, but as
infection is usually on the bole, the trees are
subsequently subject to wind breakage at the
infection site or killing of weakened trees by
secondary organisms. Presence of indigenous
diseases should be one of the major selection
criteria in choosing an orchard site. Although
presence of a disease usually will not eliminate
a potential site, it is best to reduce disease
risk as much as possible before orchard
establishment. When forest sites are to be used
for orchards, survey and mark the root rot
centers before clearing the trees, then repeatedly

260

root pick the area and especially the disease
centers to remove pathogen- infested roots
(inoculum). Alternate hosts such as those of
lodgepole pine stem rusts (e.g., Indian
paintbrush (Castille.ja miniata Dougl. ex Hook)
should be eliminated from within and adjacent to
any proposed pine orchard. Alternate hosts can
be prevented from re- invading the site by sowing
competing grasses or other cover crops,
especially perennials which can withstand
repeated mowing which is detrimental to most
alternate hosts. Herbicides should be used to
kill alternate hosts that re-invade the seed
orchard and nearby areas .

As the orchard ages, previously unencountered
diseases may appear. These include foliage
pathogens which are often favored by the greater
volume of needles on larger trees, particularly
dead and senescent needles, which may enhance
pathogen buildup or survival. Undoubtedly,
microclimate within the crown changes as trees
become larger and this leads to other changes
such as an increased volume of shade and
senescent needles. Changes that may favor
needle pathogens include higher humidity, longer
moisture retention on needles, lower
temperatures and decreased exposure to
sunlight. At a Douglas-fir, Pseudotsuga
menziesii (Mirb.) Franco, orchard on Vancouver
Island, B.C. Meria laricis Vuill. and a newly
discovered fungus, Hormonema merioides Funk,
Woods & Hopkinson (Funk and others 1985), both
on needles, only became evident in older
orchards where cold water misting was used to
retard flower development for preventing pollen
contamination from outside the orchard.
Defoliation was most severe on trees nearest
misting stand pipes. As trees reach
cone-bearing age the cultural practices change.
Crown and sometimes root pruning can cause both
wounds (infection courts) and stress. Cone
picking can cause wounds too, especially on
species such as lodgepole pine. With larger
orchard trees there are the increased
difficulties of detecting pathogens and applying
fungicides .

Little information is available on
pathogen- insect associations in Inland Mountain
West seed orchards, but recently in coastal B.C.
the needle cast fungus Meria has been found in
wounds of a needle midge (Contarinia sp . ) . on
Douglas-fir. This resembles a similar situation
in loblolly pine, Pinus taeda L., orchards in
the southern U.S.A. where the pitch canker
Fusarium colonizes needle wounds made by
Contarinia (Dwinell and others 1981) . Pine wood
nematode, Bursaphelenchus xy lophi lus (Steiner &
Buhrer) Nickle, which is both beetle transmitted
and host stress related, could occur in orchards
as the result of cultural practices such as root
pruning (Dwinell and Barrows-Broaddus 1983).

CONE DISEASES

Inland Spruce Cone Rust

This rust, Chrysomyxa pirolata Wint., is the

most damaging cone disease in the Inland
Mountain West. Cone losses of up to 60% and 6 7%
have been recorded for natural stands in western
Canada (Ziller 1974) and Utah (Nelson and
Krebill 1982), respectively. Seed orchard cones
appear to be equally affected as evidenced by
annual losses of up to 60% in the first few
years of cone production in an interior B.C.
spruce orchard. Cones of all indigenous spruces
are susceptible. The alternate (non-conifer)
hosts are Pyrola spp . and Monesis unif lora
(L.)A. Gray which are relatively small,
herbaceous to woody plants. Pyrola spp. are the
predominant component of the ground cover in
some spruce forests. Although the fungus is
holartic in distribution and occurs on alternate
hosts as far south as Guatemala (Cummings 1943),
cones are only affected in the northern part of
its range. Nelson and Krebill (1982) give some
of the possible reasons for this. In the Inland
Mountain West this disease is likely to be more
of a problem in the north than in the south.

The states and spores (in parentheses) in the
life history of C. pirolata are spermagonial
(spermatia), aecial (aeciospores) , uredinial
(urediniospores) and telial (teliospores) . The
first two occur on spruce cone scales; spermatia
as a yellow-orange honeydew-type exudate in
early summer and as soon as 2 weeks later, dry,
yellow-orange aeciospores. Aeciospores are
produced in profusion and shed from diseased
cones which desiccate and open prematurely.
Both the uredinial and telial states occur as
yellow-orange sori on the under surface of
alternate host leaves. Urediniospores spread
the fungus to other alternate host plants and
occur, depending upon the species of alternate
host and perhaps locality, from late spring
through to snowfall, but they are usually most
abundant about the time of spruce cone
pollination. At this time too, telia, similar
in appearance to uredinia, form on the
undersurface of alternate host leaves. A leaf
may bear mostly uredinia, mostly telia, or both
in about equal numbers. Telia produce
teliospores which germinate , giving rise to
basidiospores which lead to cone infection about
pollination time. Information on these various
states was recently published (Sutherland and
others 1984) .

Chrysomyxa pirolata usually becomes systemic in
cones which dry out, initially becoming brownish
green and later tan-brown, and open
prematurely. Resinosis often accompanies these
symptoms as does twisting and malformation of
cone scales. Sometimes cone rust appears to be
restricted to one side of the cone causing it to
be slightly convex. Diseased cones should not
be collected because they yield few seeds which
may germinate poorly or abnormally (Nelson and
Krebill 1982; Sutherland 1981A) .

Cone rust loss estimates have traditionally been
made by counting diseased, aeciospore-bearing
cones at collection time; however, such counts
underestimate cone rust losses because about
one-third of the cones with spermatia in early

261

summer do not produce aeciospores (Sutherland
and others 1984). Instead, shortly after
spermatia production the cones cease
development, dry out, and are often attacked by
insects. In orchards where cones are readily
accessible, cone rust appraisals should include
comparative counts of spermatia-bearing versus
aeciospore-bearing cones. Under forest
conditions cone rust severity tends to be
sporadic and localized. The disease depends on
numerous factors including presence and
phenology of the cone crop and disease on
alternate hosts , weather conducive to
basidiospore production, dissemination and cone
infection plus synchronization of these climatic
and biological phenomena. As cultural practices
such as root pruning induce more regular cone
crops, cone rust losses in seed orchards could
exceed those in forests. Effects of other seed
orchard practices are unknown, e.g., cold water
misting to retard cone development might favor
the disease in certain years by synchronizing
cone and rust phenology. Conversely, misting
could reduce losses by preventing spores from
reaching susceptible cones.

Prevention of cone rust is easy, provided a seed
orchard site is selected that is free of
alternate hosts. The B.C. situation
demonstrates this with a spruce seed orchard
near Salmon Arm, where Pyrolas are abundant,
regularly experiencing 15-60% cone losses while
spruce orchards about 50 km south in a drier,
Pyrola-f ree. area are free of cone rust.
Although it is not known how far viable cone
rust basidiospores are carried, it is suspected
that disease intensity is directly proportional
to closeness of the non-conifer host. Thus,
practices that reduce the abundance of alternate
hosts immediately around the orchard should be
helpful. For example, when sufficient ground
fuel was available at Salmon Arm, summer burning
eliminated the Pyrolas. Summer application of
paraquat or mineral oil (agricultural weed
killer) was also effective against Pyrolas.
Nitrogen fertilizers (NH4S04 or urea)
applied at forest fertilization rates in the
spring and fall had no detrimental effects on P.
asarif olia Michx. or P. (Orthillia) secunda L.
in the second growth forest surrounding the seed
orchard. Two years after fertilizer
application, neither Pyrola numbers nor the
ratio of healthy to diseased plants were
changed, but growth of grasses and other
understory plants increased dramatically. This
suggested that fertilizers might be used to
increase understory fuel for burning to
eliminate Pyrolas. One area adjacent to the
Salmon Arm orchard is pastured, but cattle avoid
the abundant P. asarifolia.

Ferbam fungicide, applied to cones during the
period beginning 1 week before through
pollination, reduces cone rust incidence 7-10
fold (Summers and others 1985) . Two sprays are
recommended to compensate for differences in
cone phenology. Since germination of seeds from
treated cones is reduced slightly, an extremely
important consideration for container nurseries,

other fungicides need testing, especially
systemics such as triademefon.

Other Potential Cone Diseases

Examples of other rusts that could damage cones
locally include western gall rust (Byler and
Piatt 1972 ) of hard pines, spruce bud rust
(McBeth 1984), and American spruce- raspberry
rust (Ziller 1974) . Since these pathogens are
not confined to cones the potential damage is
very unpredictable, especially to seed orchard
cones .

SEED DISEASES

Seed or Cold Fungus

This fungus, Caloscypha fulsens (Pers.) Boud. ,
imperfect state = Geniculodendron pyrif orme Salt
(Paden and others 1978) is the best known
conifer seed pathogen in this area, having been
isolated from seeds from Oregon and Washington
(Harvey 1980), Idaho (Wicklow-Howard and Skujins
1980) and British Columbia (Sutherland 1979;
Sutherland and Woods 1978). Seeds of numerous
conifers are susceptible (Salt 1970; Salt and
Brown 1969). The fungus occurs naturally on
seeds of species with non-serotinous cones that
are collected from the ground (Sutherland and
Woods 1978) , especially from squirrel caches
(Sutherland 1979). About one-third of all
spruce (Picea spp.) seedlots in B.C. are
infested (Sutherland 1979). Other common hosts
are seeds of Douglas-fir and true firs, Abies
spp., the latter mainly because mature cones
disintegrate and the pieces are collected from
the forest floor. Seedlots of species such as
lodgepole pine that usually have serotinous
cones are disease free. Within most infested
seedlots 1-5% of the seeds are diseased;
however, seedlots with up to 60% affected seeds
have been found. Even low numbers of diseased
seeds are important because the pathogen spreads
from diseased to healthy seeds at low
temperatures (Epners 1964), the origin of the
cold fungus name (Salt 1974), such as during
seed stratification, pre-sowing storage or after
sowing in cool, wet seedbeds or container
cavities (Thomson and others 1983) .

There are both qualitative and quantitative
indicators of seed fungus infestation. Seedlots
with much poorer germination when stratified are
prime suspects. Losses increase as exposure to
cool, wet conditions lengthens (Salt 1974). The
rate of germination is important since only
ungerminated seeds are susceptible, i.e., seeds
are immune once germination begins (Epners 1964;
Salt 1974). Seeds affected by C. fulgens are
mumified and not rotted as are seeds killed by
pre-emergence , damping-off fungi (Epners 1964).
The seed fungus may also produce patches of
indigo pigment in or around the embryo (Salt
1974; Woods and others 1982). White to whitish
blue infection cushions of C. fulgens, often
most abundant at the seed's distal end, may
cover up to 80% of the seed coat surface (Woods

262

and others 1982). Quantitative determinations
can be made by isolating C. f ulgens from a
500-seed sample of surface sterilized (30%
H202 for 30 min) seeds which are
incubated at 15°C on 2X water agar (Sutherland
and others 1978). The distinctive mycelium
(Salt 1974), often indigo, grows from seeds
within 3 weeks. Another technique (Sutherland
and others 1981B) that provides either
qualitative or quantitative assays for C.
fulgens is based on the presence and amounts of
alkaline phosphatase in infested seedlots. A
major advantage of isozyme assays is that they
allow use of much larger sample sizes than are
physically possible with isolation-by-plating
procedures .

As stated earlier, seeds acquire C. fulgens when
cones contact infested forest duff. Hence,
disease incidence increases with exposure time
(Sutherland 1981B) and can increase further if
contaminated cones are improperly stored, e.g.,
under cool, wet, poorly ventilated conditions.
However, spread ceases if stored cones are
properly air-dried (Sutherland 1981B) . Seedlots
originating from squirrel-cache collected cones
are most likely to contain C. fulgens
(Sutherland 1979) . Squirrels disseminate
infested cones and along with other rodents
consume diseased seeds (Sullivan and others
1984). Other than disseminating the fungus,
conidiospores and ascospores apparently play no
obligatory role in disease development because
the pathogen penetrates seeds following
infection cushion formation by vegetative
mycelium (Woods and others 1982) . Ascospores
are produced in cup-shaped fruit bodies, with a
dull to bright orange hymenial layer, which
occur on forest duff in early spring (Ginns
1975), often soon after snow melt.

High incidence of C. fulgens is an excellent
indicator of improper cone collection or
storage, or frequently both. For example, (i)
cones were collected that had been on the ground
for a long time, (ii) collections were from
squirrel caches, and (iii) such cones were
improperly stored (cool, wet, poor
ventilation) . The worst possible situation
results from a combination of all three. When
collecting cones from natural stands, hand
picking them from standing trees or from slash
plus proper handling afterward should greatly
reduce seed fungus incidence. Since seed
orchard cones are usually hand picked and
properly handled afterward, the pathogen should
not occur in these seeds. Cultural practices
that will reduce losses when sowing infested
seedlots include: (i) not stratifying the seeds
or stratify them for the shortest possible
period, (ii) sowing seeds quickly after
stratification to avoid pre- sowing storage where
the pathogen spreads, (iii) delay sowing until
temperatures are warm enough to promote rapid
germination and (iv) sowing as few seeds as
possible per container cavity or as far apart as
practical in bareroot drills. Probably the most
practical recommendation is to add a suitable
fungicide to the stratification water or dust

the seeds with it before sowing (Gordon and
others 1976; Salt 1974).

Sirococcus Blight

This disease, caused by the fungus Sirococcus
strobilinus Preuss, affects conifer regeneration
and forest nursery seedlings throughout the
North Temperate Zone (Sutherland and others
1981A; Wall and Magasi 1976 and references cited
therein; Wicker and others 1978 and references
cited therein). Occasionally, minor damage
occurs on older forest trees, e.g. western
hemlock, Tsuga heterophylla (Raf.) Sarg.,(Funk
1972). Sirococcus blight is especially
troublesome on both regeneration and nursery
seedlings along those portions of the North
American west coast where cooler, wet and often
overcast weather favors the disease. Local
hosts are spruces, e.g., white, Engelmann, Picea
engelmannii Parry, Sitka, P. sitchensis (Bong.)
Carr. , pines such as lodgepole and yellow, Pinus
ponderosa Laws., Douglas-fir and western
hemlock.

In 1981 it was shown that S. strobilinus is
seed-borne on spruces (Sutherland and others
1981A) ; subsequent observations (J.R. Sutherland
and W. Lock, unpublished) indicate that it also
may be seed-borne on western hemlock and yellow
pine, but not lodgepole pine (Sutherland and
others 1982) . While the pathogen has been found
on seedcoats , it normally penetrates seeds and
ramifies throughout the contents (Sutherland and
others 1981A) so that diseased seeds do not
germinate. However, in container nurseries
these dead seeds serve as inoculum for adjacent
seeds which apparently acquire the fungus,
germinate and emerge, then become diseased.
These diseased seedlings are foci for subsequent
spread of the fungus via splashing irrigation
water. The situation seemingly is different in
bareroot nurseries where the fungus appears to
be unable to bridge the gap between diseased and
healthy seeds, i.e., the latter do not acquire
the pathogen before germinating. At least two
factors are thought to allow the bridging in
containers. Firstly the medium is probably more
conducive (less microbial competition?) to
Sirococcus and secondly, container seeds are
sown more densely with frequently two or more
seeds being sown per cavity. The latter
practice also increases the likelihood of
placing a diseased seed in a cavity. Sirococcus
blight of bareroot seedlings can originate from
inoculum in wind-blown rain (Riffle and Smith
1979), such as from nearby windbreak trees, or
from inoculum on cones which fall into seedbeds
(Srago 1978). Observations leading to the
suspicion that S. strobilinus was seed-borne
were that the disease first appeared on very
young germinants of specific seedlots over
several years and that the fungus is common on
cones, particularly spruce.

Sirococcus blight damage and symptoms are well
known on seedlings (Smith 1975; Sutherland and
Van Eerden 1980) and regeneration (Wicker and

263

others 1978). Funk (1981) summarizes
information on the pathogen such as spore size
and shape. These references will be useful to
persons collecting cones from natural stands.
Seed orchard managers likely will not have to
deal with the disease on either cones or trees,
mainly because orchard cones are collected each
year which removes an important source of
inoculum. In nature the fungus is thought to
build up on old cones that remain on trees and
when these are inadvertently included in
collections they are a major source of
Sirococcus- infested seeds (J.R. Sutherland and
T.A.D. Woods, unpublished). Another reason
Sirococcus blight is unlikely to be important in
seed orchards is that prolonged light stress, a
major factor predisposing hosts to the pathogen
(Wall and Magasi 1976), seldom occurs in
orchards .

Seed-borne Sirococcus is another example of a
problem created by poor cone collection
practices, i.e. by including diseased cones in
collections. Old cones are most likely to be
diseased. This problem is difficult to overcome
because late in the collection season it is
extremely difficult to distinguish current year
from old cones. Presence of S. strobilinus
fruit bodies on cones indicates that the seeds
will be infested, but laboratory confirmation is
necessary because other similar-appearing fungi
also fruit on cones. Attempts to remove
Sirococcus -diseased seeds by repeatedly cleaning
infested seedlots with air or by passage over a
gravity table have been unsuccessful (J.R.
Sutherland and W. Lock, unpublished). In
container nurseries, benomyl or daconil
fungicide drenches after sowing Sirococcus-
infested spruce seedlots has produced
conflicting results for disease control and the
trials need repeating (G. Matthews, personal
communication). In B.C., recommendations for
container nurseries are to warn managers when
infested seedlots are being sown so that
fungicide spraying, and when practical, roguing
diseased seedlings, can begin when the disease
appears. Withholding watering or watering in
the morning so that seedlings dry off quickly
also helps alleviate damage. A disadvantage of
fungicides is that they must be applied
frequently to protect rapidly expanding tissues
and this leads to fungicide accumulation on
seedlings which concerns nursery workers and
tree planters. Also, fungicides applied early
in the season against Sirococcus may lead to
build up of fungicide tolerance in late season
pathogens such as Botrytis cinerea .

Other Cone and Seed Fungi

Besides the previously mentioned fungi and the
Fusaria that Dr. James is covering in his paper,
many other fungi have been reported in and on
cones and seeds of local tree species (Bloomberg
1969; Harvey and Carpenter 1975; Rediske and
Shea 1965; Richardson 1979; Richardson 1981;
Shea 1960) . Many of these are potential
pathogens to cones, seeds or seedlings while

others such as Heterobasidion anno sum (Fr.)
Bref. (Richardson 1979; Richardson 1981), which
causes a butt rot of older trees, are obviously
incidental. The real gray area in our knowledge
concerns the role of the most common and
so-called saprophytic fungi such as Penicillium
and Aspergillus that have been found on cones
and especially seeds. Most research on these
fungi has been of a single, short-term nature
and the results and interpretations to date are
conflicting. Consequently few conclusions can
be drawn from the literature.

Another shortfall is that since none of the
studies have fulfilled Koch's Postulates it is
impossible to establish a cause and effect
relationship for any of these fungi. Although
cones and in turn seeds apparently acquire these
various fungi while still on the tree, there is
no evidence that fungus development occurs until
the cones have ripened and are collected.
Numerous fungi can then develop on the stored
cones prior to seed extraction, but mold
incidence and abundance on cones appears not to
subsequently affect seed quality (Bloomberg
1969). It is only after seed extraction and
particularly during germination tests that molds
appear (Bloomberg 1969) . At least in
Douglas-fir, the low percentage of healthy,
non-germinable seeds indicates that the fungi
and possibly bacteria are facultative parasites
of low quality seeds (Bloomberg 1969) . Based on
our limited knowledge this seems to be true for
seeds of other conifers too, e.g. reducing the
abundance of molds on Abies seeds failed to
increase germination (Edwards and Sutherland
1979). Examples of factors that can lower seed
quality and thereby increase susceptibility to
these facultative parasites include both
internal and seed coat damage acquired during
extraction from cones, insect damage, and
probably most importantly, lack of maturity
(Bloomberg 1969) . The low temperatures and low
moisture content of both seeds and the
environment inhibit mold development in
long-term storage (Holmes and Buszewicz 1958) .
However, in the local context nothing is known
about what happens afterward during seed
stratification or subsequently when seeds are
stored before sowing. Present evidence
indicates that incidence and abundance of
so-called saprophytic fungi such as Asperftillus
and Penicillium are indicators rather than the
cause of poor quality seeds.

REFERENCES

Bloomberg, W. J. Disease of Douglas-fir seeds
during cone storage. Forest Science. 15(2):
176-181; 1969.

Byler, J. W. ; Piatt, W. D. Cone infection by
Peridermium harknessii . Canadian Journal of
Botany. 50(6): 1429-1430; 1972.

Cummings, G. B. Annotated checklist and host
index of the rusts of Guatemala. Plant
Disease Reporter Sup. 142; 1943. 131 p.

264

Dwinell, L. D.; Barrows Broaddus , J. Pine wilt
and pitch canker of Virginia pines in seed
orchards. In: 17th Proceedings Southern Tree
Improvement Conference. 1983; 55-62.

Dwinell, L. D . ; Kuhlman, E. G . ; Blakeslee, G.
M. Pitch canker of southern pines. In:
Nelson, P. E . ; Toussoun, T. A.; Cook, R. J.
eds . Fusarium: diseases, biology and
taxonomy. University Park and London:
Pennsylvania State University Press; 1981:
188-194.

Edwards, D. G. W.; Sutherland, J. R. Hydrogen
peroxide treatment of Abies seeds . Canadian
Forestry Service, Bi-monthly Research Notes
35(1): 3-4; 1979.

Epners, Z. A new psychrophilic fungus causing
germination failure of conifer seeds.
Canadian Journal of Botany. 42(12):
1589-1604; 1964.

Funk, A. Sirococcus shoot-blight of western

hemlock in British Columbia and Alaska. Plant
Disease Reporter. 56(8): 645-647; 1972.

Funk, A. Parasitic microfungi of western trees.
Bulletin BC-X-222 . Victoria , BC: Canadian
Forestry Service, Pacific Forest Research
Centre, Victoria, B.C., 1981. 190 p.

Funk, A.; Woods, T. A. D.; Hopkinson, Sarah J.
Hormonema merioides n. sp . , on Douglas-fir
needles. Canadian Journal of Botany, [in
press )

Ginns, J. H. Caloscypha fulgens . Fungi
Canadensis No. 66, National Mycological
Herbarium, Biosystematics Research Institute,
Agriculture Canada, Ottawa, Ontario; 1975.

Gordon, A. G.; Salt, G. A.; Brown, R. M. Effect
of pre-sowing moist-chilling treatments on
seedbed emergence of Sitka spruce seed
infected by Geniculodendron pyrif orme Salt.
Forestry. 49(2): 143-151; 1976.

Harvey, G. M. ; Carpenter, Larry A. Fungi on
stored Douglas-fir cones--a problem? Tree
Planter's Notes. 26(4): 16-17, 22; 1975.

Harvey, R. D., Jr. Mortality from Caloscypha
fulgens and other fungi on spruce seed in
Oregon and Washington. Plant Disease. 64(2):
223-224; 1980.

Holmes, G. D; Buszewicz, G. The storage of seed
of temperate forest tree species. Part I
Forestry Abstracts. 19(3): 313-322; 1958.

McBeth, Jenifer Huang. Symptomology on spruce
trees and spore characteristics of a bud rust
pathogen. Phytopathology. 74(4): 456-461;
1984.

Nelson, David L.; Krebill, Richard G. Occurrence
and effect of Chrysomyxa pirolata cone rust
on Picea pungens in Utah. Great Basin
Naturalist. 42(2): 262-272; 1982.

Paden, J. W. ; Sutherland, Jack R. ; Woods, T. A. D.
Caloscypha fulgens (Ascomycetidae , Pezizales):
the perfect state of the conifer seed pathogen
Geniculodendron pyrif orme (Deuteromycotina,
Hyphomycetes) . Canadian Journal of Botany.
56(19): 2375-2379; 1978.

Rediske, J. H.; Shea, K. R. Loss of Douglas-fir
seed viability during cone storage. Forest
Science. 11(4): 463 472; 1965.

Richardson, M. J. An annotated list of

seed-borne diseases. 3d ed . Kew, Surrey,
England: Commonwealth Mycological Institute;
1979. 320 p.

Richardson, M. J. Supplement 1 to an annotated
list of seed-borne diseases. 3d ed . Zurich,
Switzerland: The International Seed Testing
Association: 1981. 78 p.

Riffle, J. W. ; Smith, R. S., Jr. Nursery

diseases of western conifers. Forest Insect
and Disease Leaflet 157, Washington, DC:
U.S. Department of Agriculture, Forest
Service, U.S. Government Printing Office;
1979. 11 p.

Salt, G. A. Conifer seedling pathology. In:
Report on forest research, 1970, Forestry
Commission, Her Majesty's Stationery Office,
London, England; 1970. 174-175.

Salt, G. A. Etiology and morphology of

Geniculodendron pyrif orme gen. et sp . nov . , a
pathogen of conifer seeds. Transactions of
the British Mycological Society. 63(2):
339-351; 1974.

Salt, G . A.; Brown, R. M. Conifer seedling
pathology. In: Report on forest research,
1969, Forestry Commission, Her Majesty's
Stationery Office, London, England; 1969. 147 p

Schaffer, Bruce; Hawksworth, F. G.; Jacobi, W.
R. Effects of commandra blister rust and
dwarf mistletoe on cone and seed production
of lodgepole pine. Plant Disease 67 (2):
215-217; 1983.

Shea, Keith R. Mold fungi on forest tree seed.
Forestry Research Note Number 31. Centralia,
WA: Weyerhaeuser Co.; 1960. 10 p.

Smith, Richard S., Jr. Sirococcus tip blight.
In: Peterson, Glenn W. ; Smith, Richard S.,
Jr. technical coordinators. Forest nursery
diseases in the United States. Agriculture
Handbook No. 4 70, Washington, DC: U.S.
Department of Agriculture, Forest Service;
1975. 125 p.

Srago, Michael D. Nursery disease

problems- -Sirococcus strobilinus . B134-B137.
Western Forest Nursery Council and
Intermountain Nurseryman's Association,
combined Nurseryman's Conference and Seed
Processing Workshop, Eureka, CA; 1978.

265

Sullivan, Thomas P.; Sutherland, Jack R.; Woods,
T. A. D. ; Sullivan, Drucilla S. Dissemination
of the conifer seed fungus Caloscypha fulgens
by small mammals . Canadian Journal of Forest
Research. 14(1): 134-137; 1984.

Summers, D.; Sutherland, Jack R. ; Woods, T. A.
D. Inland spruce cone rust (Chrysomyxa
pirolata Wint.) control: ferbam fungicide
application related to basidiospore
production rainfall and cone phenology.
Canadian Journal of Forest Research, [in
press] .

Sutherland, Jack R. The pathogenic fungus

Caloscypha f ulgens in stored conifer seeds in
British Columbia and relation of its
incidence to ground and squirrel-cache
collected cones. Canadian Journal of Forest
Research. 9(1): 129-132; 1979.

Sutherland, Jack R. Effects of Inland spruce

cone rust, Chrysomyxa pirolata Wint., or seed
yield, weight and germination. Canadian
Forestry Service Research Notes. 1(2): 8-9;
1981A.

Sutherland, Jack R. Time, temperature, and
moisture effects on incidence of seed
infected by Caloscypha fulgens in Sitka
spruce cones . Canadian Journal of Forest
Research. 11(3): 727-730; 1981B.

Sutherland, Jack R; Hopkinson, Sarah J.; Farris,
S. H. Inland spruce cone rust, Chrysomyxa
pirolata, in Pyrola asarif olia and cones of
Picea glauca, and morphology of the spore
stages. Canadian Journal of Botany. 62(11):
2441-2447; 1984.

Sutherland, Jack R. ; Lock, W. ; Farris, S. H.
Sirococcus blight: a seed-borne disease of
container-grown spruce in coastal British
Columbia forest nurseries. Canadian Journal
of Botany. 59: 559-562; 1981A.

Sutherland, Jack R. ; Lock, W. ; Woods, T. A. D. ;
Suttill, Terri. Sirococcus blight: not
seed-borne on serotinous lodgepole pine.
Canadian Forestry Service Research Notes.
2(3)L: 20-21; 1982.

Sutherland, Jack R. : Rink, Ute; McMullan, E. E. ;
Woods, T. A. D. Isozyme characteristics of
Caloscypha fulsens infested and pathogen-free
spruce seed samples and use of alkaline

phosphatase activity for qualitative and
quantitative disease incidence assays.
Canadian Journal of Forest Research. 11(2):
200-205; 1981B.

Sutherland, Jack R. ; Van Eerden, Evert. Diseases
and insects in British Columbia forest
nurseries. British Columbia Ministry of
Forests/Canadian Forestry Service Joint
Report No. 12; 1980. 55 p.

Sutherland, Jack R.; Woods, T. A. D. The fungus
Geniculodendron pyrif orme in stored Sitka
spruce seeds: effects of seed extraction and
cone collection methods on disease incidence.
Phytopathology. 68(5): 747-750; 1978.

Sutherland, Jack R. ; Woods, T. A. D. ; Lock, W. ;
Gaudet, Denis A. Evaluation of surface
sterilants for isolation of the fungus
Geniculodendron pyrif orme from Sitka spruce
seeds. Canadian Forestry Service Bi-Monthly
Research Notes. 34(4): 20-21; 1978.

Thomson, A. J.; Sutherland, Jack R.; Woods,

T. A. D., Moncrieff, S. M . Evaluation of seed
disease effects in container-sown Sitka
spruce. Forest Science. 29(1): 59-65; 1983.

Wall, R. E . ; Magasi, L. P. Environmental factors
affecting Sirococcus shoot blight of black
spruce. Canadian Journal of Forest Research.
6(4): 448-452; 1976.

Wicker, Ed F.; Laurent, Thomas H. ; Israelson,
Spencer. Sirococcus shoot blight damage to
western hemlock regeneration at Thomas Bay,
Alaska. Research Paper. INT-198. Ogden, UT:
U.S. Department of Agriculture, Forest
Service, Intermountain Forest and Range
Experiment Station; 1978. 11 p.

Wicklow-Howard, M. C: Sku j ins , J. Infection of
Engelmann-spruce seed by Geniculodendron
pyrif orme in western North America.
Mycologia. 72(2): 406-410; 1980.

Woods, T. A. D.; Farris, S. H. ; Sutherland,

Jack R. Penetration of Sitka spruce seeds by
the pathogenic fungus Caloscypha f ulgens .
Canadian Journal of Botany. 60(5): 544-548;
1982.

Ziller, Wolf G. The tree rusts of western

Canada. Publication No. 1329. Victoria, BC:
Canadian Forestry Service; 1974. 272 p.

266

DISEASES OF CONIFER SEEDLINGS CAUSED BY SEED-BORNE J^USARIUM SPECIES^,

R. L. |james

ABSTRACT: The genus Fusarium includes many common
soil-borne fungi that may colonize conifer seed,
especially if cones are collected from the ground
or squirrel caches. These fungi most commonly
infect the seed coat, but can also colonize the
seed embryo and endosperm. Fusarium spp. cause
a wide variety of diseases, most of which affect
the roots of susceptible plants. Types of diseases
commonly affecting conifer seedlings in nurseries
include (1) seed decay, (2) pre-emergence damping-
off or germination failure, (3) postemergence
damping-off, (4) topdamping-of f or cotyledon blight,
and (5) root diseases or late damping-off. The
most common species of seed-borne fusaria include
F, oxysporum, F. solani , F. monilif orme, and F.
"roseum. " Diseases caused by Fusarium can be
reduced by seed treatments such as running water
rinses, surface sterilants, and fungicides.

INTRODUCTION

Conifer seeds are storehouses of food and energy
and many microorganisms have evolved mechanisms
for invading and utilizing them. Many different
fungi commonly infect conifer seeds. Infection
frequently damages seed and also provides a means
by which fungi may be transferred from one sub-
strate or geographic location to another (Harman
1983) .

Fusarium spp. are common soil-inhabiting plant
pathogens (Booth 1971; Gerlach and Nirenberg 1982),
which also frequently infect conifer seed
(Neergaard 1977). These fungi attack a wide range
of hosts and cause economically important diseases
of many commercial crops, including conifer seed-
lings (Bloomberg 1971; Tint 1945). Fusarium spp.
commonly occur within many types of soils. Popu-
lations frequently increase in cultivated soils
(Booth 1971); low levels of these fungi often
occur in undisturbed natural soils (Smith 1967).

Fusarium diseases of conifer seedlings have tradi-
tionally been most important in bareroot nurseries
(Bloomberg 1971). Pathogen populations are often
reduced in nursery soils by using fumigants such
as methyl bromide and chloropicrin (Miller and
Norris 1970). However, Fusarium- caused diseases
sometimes occur despite soil fumigation (Cooley
1982). Investigations of diseases incited by
Fusarium indicate that these fungi may be intro-

Paper presented at the Conifer Tree Seed in the
Inland Mountain West Symposium, Missoula, MT,
August 5-6, 1985.

R. L. James is Plant Pathologist, Cooperative For-
estry and Pest Management, USDA Forest Service,
Missoula, MT.

duced into both bareroot and container nurseries
on conifer seed, causing extensive losses (Cooley
1983b; Graham and Lindennan 1983; James 1983a).

Although Fusarium spp. can infect conifer seed
during flowering and cone formation, (Anderson
and others 1980; Mason and Van Arsdel 1978; Sharma
1978), probably most infection occurs when cones
or seed contact soil that harbors inoculum (James
1983c; Karrfalt 1983). Cones collected from
squirrel caches often contain large populations of
fungi including many pathogenic fusaria (James
1984c; James and Genz 1981; James and Genz 1982).
During the seed extraction process, infection by
fusaria may intensify (Salisbury 1955) , resulting
in both seedcoat and endosperm colonization (James
1984b; James 1984c). Diseases of seeds often in-
crease during prolonged seed and cone storage
(Bloomberg 1969; Harmon and others 1978; Harvey
and Carpenter 1975) . Seed colonization by patho-
gens can also increase during the extended seed
stratification periods that are common in conifer
nurseries (Bloomberg and Trelawny 1970).

TYPES OF DISEASES

Fusarium spp. cause several different kinds of
diseases, the most important of which affect roots
of susceptible plants (Booth 1971; Gerlach and
Nirenberg 1982) . Five types of diseases caused by
these fungi are generally recognized on conifer
seedlings. These include seed decay, pre-emergence
damping-off or germination failure, postemergence
damping-off, top damping-off or cotyledon blight,
and root disease or late damping-off (Bloomberg
1971; Matuo and Chiba 1966).

Seed decay occurs when fungi penetrate the seed-
coat, colonize it and break down internal seed
contents (Bloomberg 1969) . Seeds with damaged
seedcoats are especially vulnerable to rapid
fungal invasion (Gibson 1957; Neergaard 1977).
Decayed seed may or may not be detectable from
outward appearance (Bloomberg 1966) . However,
x-rays, which reveal hollow or partially deterior-
ated endosperms, can aid detection (Anderson and
others 1980) . Decayed seed may also be detected
during water or air separation operations because
of their reduced densities (James and Genz 1981;
James and Genz 1982; Neergaard 1977) . If decayed
seed are sown, decreased germination will result
and potentially pathogenic fungi are introduced
into seedbeds or containers (James 1984a; Landis
1976a) .

Pre-emergence damping-off occurs when the emerging
radicle of germinating seed is attacked by fungi
either carried on the seedcoat or present in soil
(Bloomberg 1971; Graham and Linderman 1983). If
the radicle is colonized by virulent fungi, decay
results and no germinant emerges (Rathbun-Gravatt

267

1931) . Most losses to pre-emergence damplng-off
are never detected and are often attributed to "bad
seed." Investigations of nonemergence of germlings
are usually necessary to determine if pathogenic
fungi are involved.

Postemergence damping-off refers to disease of
newly emerged germinants. Lesions often appear at
the ground line, causing infected germinants to
fall over (Bloomberg 1971; Landis 1976a). Decay of
the germinant follows and sporulation may occur on
decayed tissues. Seed-borne fusaria may incite
postemergence damping-off, resulting in reduced
seedling densities (Graham and Linderman 1983;
Matuo and Chiba 1966; Urosevic 1961).

Top damping-off caused by Fusarium spp. occurs as
colytedon blight (Mason and van Arsdel 1978), hypo-
cotyl rot (Brownell and Schneider 1983; Hamm, per-
sonal communication), or stem rot (Morgan 1983).
Cotyledon blight is especially common on pine
species that retain their seedcoats on the tips of
cotyledons for extended periods after germination
(Mason and van Arsdel 1978) . Seed-borne fusaria
move from attached seedcoats and colonize cotyle-
dons, causing decay and eventual mortality. Hypo-
cotyl and stem rots are caused by either natural
populations of soil-borne fusaria or pathogens
introduced on infected seed.

Root disease caused by Fusarium usually occurs on
seedlings that are several months old. Disease
results from decay of feeder roots (Pawuk and
Barnett 1975) ; affected seedlings become slow grow-
ing and chlorotic (Landis 1976b) and may develop
wilt symptoms and needle tip dieback (James 1983a;
James 1984c; James 1984d) . Seedling deterioration
may occur either gradually or rapidly (Merrill
and others 1981) . The disease may cause seedling
mortality or reduced seedling vigor, which adversely
affect outplanting survival (LaMadeleine 1979) .
Seedlings may become infected during or shortly
after establishment, but infecting fungi may remain
inactive for several months (Bloomberg 1966) . When
seedlings become stressed during crown closure,
periods of heat or moisture stress, or during hard-
ening off, the infecting fungi may become active
and induce disease (James 1984c; James 1984d) .
Another possibility is that soil-borne fusaria may
become more pathogenic when seedlings are stressed.
In any event, losses from root disease can continue
for several months in containerized stock (James
1983c; Landis 1976b) and throughout the first and
second growing seasons in bareroot stock (James
1983b; James 1983d) .

SPECIES OF FUSARIUM

The most common species of Fusarium isolated from
conifer seed is F. oxysporum Schlect. (Graham and
Linderman 1983; James 1984b; James 1983c; James
and Genz 1982) . This fungus is an important seed-
or soil-borne pathogen of many different plants
including conifer seedlings (Booth 1971; Cooley
1983a; Gerlach and Nirenberg 1982). It is capable
of causing vascular wilts (Booth 1971; Neergaard
1977) and cortical rots of seedling stems (Brownell
and Schneider 1983; Morgan 1983) and roots (James

1984b; James 1983d). Although _F. oxysporum exhi-
bits a wide host range (Booth 1971; Gerlach and
Nirenberg 1982) individual strains of the fungus,
called formae specialis (f. sp.), usually infect
only a few selective hosts (Gordon 1965; Snyder and
Hansen 1940). Only one f. sp. (designated pini) is
usually recognized for isolates of _F. oxysporum
that attack conifers (Gordon 1965) .

Isolates that cause diseases of conifers are gen-
erally not thought to infect other plant species
(Brownell and Schneider 1983) . However, responses
of different conifer species to infection by
several F_. oxysporum isolates have sometimes been
sufficiently variable to indicate that designation
of additional f . sp. (other than pini) which attack
conifers might be warranted (James and Gilligan
1984; Matuo and Chiba 1966) . Additional pathogen-
icity tests on a wide range of conifer hosts will
be needed to help clarify this issue. Pathogenic
isolates of F. oxysporum have been obtained from
conifer seed (Graham and Linderman 1983) . However,
nonpathogenic isolates have also been frequently
isolated. Therefore, occurrence of F_. oxysporum on
seed does not necessarily mean that disease will
result (James 1984a; James and Genz 1982) .

Another Fusarium species commonly isolated from
conifer seed is J£. solani (Mart.) Sacc. (James
1983a; James 1983c; James 1984a) . It is a common
root decay organism that is especially damaging on
certain agricultural crops (Booth 1971; Gerlach
and Nirenberg 1982; Neergaard 1977). The fungus is
occasionally associated with diseases of conifer
seedlings (Landis 1976b; Merrill and others 1981;
Tint 1945). However, the pathogenic potential of
seed-borne sources of this fungus is unclear for
conifer seedlings.

Other species of Fusarium frequently isolated from
conifer seed include F. monilif orme Sheldon and F_.
roseum (Lk. ) Sacc. (James 1983c; James 1983e; James
1984a; James and Genz 1982) . Fusarium monilif orme
causes root decay in several types of plants (Booth
1971; Gerlach and Nirenberg 1982), but is infre-
quently associated with conifer diseases (James
1984a; Rowan 1982) . Fusarium roseum is actually a
complex of organisms that produce distinctive pig-
ments in culture (Booth 1971) . Members of this
group are frequently isolated from conifer seed
(James 1983c; James 1983e; James and Genz 1981) and
less frequently from diseased seedlings (James 1983e
James 1984d; Morgan 1983) . Although some of these
fungi may be pathogenic (James and Gilligan 1984;
Morgan 1983), most are saprophytic (Booth 1971;
Gerlach and Nirenberg 1982) . Seed-borne isolates
of jF. roseum have generally not been evaluated for
their pathogenic potential.

DISEASE CONTROL

The extent of Fusarium contamination on seed varies
greatly among conifer species and seedlots (James
1984a; James and Genz 1982) . Differences among
seedlots may be related to cone collection, storage,
and seed extraction practices. Cones collected from
squirrel caches often have high levels of fungal
contamination. Also, cones and seed stored under

268

damp conditions for longer time periods are more
prone to damage by fungi.

Seed treatment before sowing may reduce disease
losses caused by seed-borne fusaria (Johnson and
Harvey 1975; Johnson and Linton 1942). Most growers
soak seed in wa-ter to condition them for sowing;
some use standing water and others a running water
rinse (James 1984a). If infected seed is soaked in
standing water, fungal propagules can spread,
causing widespread infection (James 1983e) . How-
ever, placing seed under a running water rinse can
reduce seedcoat contamination and does not spread
infection (James 1983e; James 1984a).

Surface sterilants, such as hydrogen peroxide and
sodium hypochlorite (commercial bleach), have fre-
quently been used to reduce fungal contaminations
and enhance germination of conifer seed (Advincula
and others 1983; James and Genz 1981). Hydrogen
peroxide usually reduces or eliminates fungal con-
taminants (Barnett 1976; James and Genz 1981). The
effect of hydrogen peroxide on conifer seed germin-
ation has been variable. For example, some investi-
gators (Edwards and Sutherland 1979; James 1983a)
report reduced seed germination; others (Ching and
Parker 1958; James and Genz 1981; Mason and van
Arsdel 1978) report improved germination. Detri-
mental effects of H202 generally increase with
chemical concentration and exposure period. Sodium
hypochlorite usually reduces fungal contamination
(James and Genz 1981) and sometimes enhances seed
germination (Advincula and others 1983).

Several fungicides have been used for seed treat-
ments to reduce damping-off caused by seed-borne
pathogens (Mittal and Sharma 1981; Strong 1952);
however, reports of fungicide toxicity to seed and
germinants have limited their use (Cooley 1983a;
James 1983e; Lock and others 1975). For example,
use of captan has resulted in reduced seed germina-
tion (Peterson 1970), and has caused seedling in-
jury following germination (Cayford and Waldron
1967; Lock and others 1975). Thiram, another common
seed-treatment fungicide, has reduced seed germina-
tion (Dick and others 1958; Shea 1959) and caused
deformed germinants (Hedderwick and Gadgil 1966).
Effectiveness of seed-treatment fungicides is ap-
parently related to dosage levels (Hamilton and
Jackson 1951) , activity spectrum against target or-
ganisms, development of resistant fungal strains,
and persistence on seed (Sutherland and van Eerden
1980) .

CONCLUSIONS

1. The genus Fusarium causes a wide variety of
diseases of conifer seedlings.

2. Several Fusarium spp. have been shown to be
carried both externally and internally by conifer
seed .

3. The best method to reduce Fusarium contamin-
ation of seed is unclear, although running water
rinses may be effective.

4. Seedlots of susceptible species should be
bioassayed for presence of Fusarium after extrac-
tion to identify problem lots.

RESEARCH NEEDS

Effects of cone collection, storage, and seed
handling techniques on disease caused by seed-
borne fusaria need investigation. Several pertin-
ent questions need to be answered, for example,
should squirrel cache collections be permitted for
susceptible species, and should cones be stored
under specific conditions to reduce spread of
Fusarium? What are the best temperatures and
seed moisture levels for storage of Fusarium-
ihfested seedlots? Should infested seedlots be
stratified? Will stratification improve or reduce
germination?

Another important research need concerns taxonomy
of F. oxysporum strains that cause diseases of
conifer seedlings. Pathogenicity tests on a wide
range of conifer hosts are needed to determine
host specificity characteristics of fungal strains.

REFERENCES

Advincula, B. A.; Woo, J. Y.; Partridge, A. D. A
method for germination of western white pine
seeds. Moscow, ID: University of Idaho; 1983;
4 p. Unpublished report.

Anderson, R. L . ; Belcher, E.; Miller, T. Occur-
rence of internal seed fungi in slash pine seed
produced in seed orchards. Atlanta, GA: U. S.
Department of Agriculture, Forest Service,
Southeastern Area, State and Private Forestry;
1980; Rept. 81-1-4; 3 p.

Barnett, J. P. Sterilizing southern pine seeds
with hydrogen peroxide. Tree Planters' Notes.
27(3): 17-19; 1976.

Bloomberg, W. J. The occurrence of endophytic

fungi in Douglas-fir seedlings and seeds. Can.
J. Bot. 44: 413-420; 1966.

Bloomberg, W. J. Disease of Douglas-fir seeds
during cone storage. Forest Science. 15:
176-181; 1969.

Bloomberg, W. J. Diseases of Douglas-fir seedlings
caused by Fusarium oxysporum . Phytopathology.
61: 467-470; 1971.

Bloomberg, W. J.; Trelawny, J. Effect of thiram
on germination of Douglas-fir seed. Phyto-
pathology. 60: 1111-1116; 1970.

Booth, C. The genus Fusarium. Kew, Surrey,
England: Commonwealth Mycological Institute;
1971. 237 p.

Brownell, K. H.; Schneider, R. W. Fusarium hypo-
cotyl rot of sugar pine in California. Plant
Disease. 67: 105-107; 1983.

Cayford, J. H.; Waldron, R. M. Effects of captan
on germination of white spruce, jack and red
pine seed. For. Chron. 43: 381-384; 1967.

269

Ching, T. M. ; Parker, M. C. Hydrogen peroxide for
rapid viability tests of some coniferous tree
seeds. For. Sci. 4: 128-134; 1958.

Cooley, S. J. Evaluation of the fungicide Banrot
to prevent root rot of sugar pine seedlings.
Portland, OR: U. S. Department of Agriculture,
Forest Service, Pacific Northwest Region; 1982.
10 p.

Cooley, S. J. Fungicide trials on sugar pine at a
southern Oregon nursery. Tree Planters' Notes.
34(3): 15-18; 1983a.

Cooley, S. J. Seed and soil treatments to reduce
seed decay and Fusarium root rot of sugar pine.
Portland, OR: U. S. Department of Agriculture,
Forest Service, Pacific Northwest Region; 1983b;
8 p.

Dick, J.; Finnis, J. M. ; Hunt, L. 0.; Kverno, N. B.
Treatment of Douglas-fir seed to reduce loss to
rodents. J. Forestry. 56: 660-661; 1958.

Edwards, D. G. W. ; Sutherland, J. R. Hydrogen
peroxide treatment of Abies seeds. Can. For.
Ser. Bi-monthly Res. Notes 35: 3-4; 1979.

Gerlach, W. ; Nirenberg, H. The genus Fusarium -

a pictorial atlas. Berlin, Germany: Paul Parey;
1982. 406.

Gibson, I. A. S. Saprophtic fungi and destroyers
of germinating pine seeds. East African Agri.
J. 22: 203-206; 1957.

Gordon, W. L. Pathogenic strains of Fusarium
oxysporum. Can. J. Bot. 43(11): 1309-1318;
1965.

Graham, J. H. ; Linderman, R. G. Pathogenic seed-
borne Fusarium oxysporum from Douglas-fir.
Plant Disease. 67: 323-325. 1983.

Hamilton, J. R. ; Jackson, L. W. R. Treatment of
shortleaf pine and loblolly pine seed with
fungicidal dusts. Plant Dis. Reptr. 35: 274-
276; 1951.

Hamm, P. B. Personal communication. Corvallis,
OR: Oregon State University, Department of
Botany and Plant Pathology: 1985, February 13.

Harman, G. E. Mechanisms of seed infection and
pathogensis. Phytopathology. 73: 326-329;
1983.

Harman, G. E. ; Nedrow, B. ; Nash, G. Stimulation
of fungal spore germination by volatiles from
aged seeds. Can. J. Bot. 56: 2124-2127; 1978.

Harvey, G. M. ; Carpenter, L. R. Fungi on stored
Douglas-fir cones - a problem? Tree Planters'
Notes. 26(4): 16-17, 22: 1975.

Hedderwick, G. W. ; Gadgil, P. D. Effects of
fungicidal treatment on Pinus radiata seed.
New Zealand For. Res. Inst.; 1966; Ann. Rept. ,
p. 34-35.

James, R. L. Fusarium root disease of container-
ized seedlings at the Montana State Nursery,
Missoula. Missoula, MT: U. S. Department of
Agriculture, Forest Service, Northern Region;
1983a. 9 p.

James, R. L. Fusarium root disease of western
white pine seedlings at the Coeur d'Alene
Nursery, Missoula, MT: U. S. Department of
Agriculture, Forest Service, Northern Region;
1983b. 5 p.

James, R. L. Fungal contamination of ponderosa
pine cones and seed from the Coeur d'Alene
Nursery, Idaho. Missoula, MT: U.S. Department
of Agriculture, Forest Service, Northern Region;
1983c. 6 p.

James, R. L. Mortality of white fir seedlings at
the Fantasy Farms Nursery, Peck, Idaho.
Missoula, MT: U. S. Department of Agriculture,
Forest Service, Northern Region; 1983d. 7 p.

James, R. L. Occurrence of Fusarium on Douglas-fir
seed from the Montana State Nursery. Missoula,
MT: U. S. Department of Agriculture, Forest
Service, Northern Region; 1983e. 11 p.

James, R. L. Diseases associated with container-
ized seedling soil mixes. Missoula, MT: U. S.
Department of Agriculture, Forest Service,
Northern Region; 1984a. 7 p.

James, R. L. Fungi colonizing Douglas-fir seed at
the Champion Timber lands Nursery, Plains, Montana.
Missoula, MT: U. S. Department of Agriculture,
Forest Service, Northern Region; 1984b; Rept.
84-13. 3 p.

James, R. L. Needle tip dieback of containerized
Douglas-fir seedlings at the Coeur d'Alene
Nursery, Idaho. Missoula, MT: U. S. Department
of Agriculture, Forest Service, Northern Region;
1984c. 5 p.

James, R. L. Tip dieback of containerized Douglas-
fir seedlings at the Montana State Nursery,
Missoula. Missoula, MT: U. S. Department of
Agriculture, Forest Service, Northern Region;
1984d. 6 p.

James, R. L.; Genz, D. Ponderosa pine seed treat-
ments: effects on seed germination and disease
incidence. Missoula, MT: U. S. Department of
Agriculture, Forest Service, Northern Region;
1981: Rept. 81-16; 13 p.

James, R. L.; Genz D. Evaluation of fungal popu-
lations on ponderosa pine seed. Missoula, MT:
U. S. Department of Agriculture, Forest Ser-
vice, Northern Region; 1982; Rept. 82-22, 21 p.

James, R. L. ; Gilligan, C. J. Studies of Fusarium
associated with containerized conifer seedling
diseases: pathogenicity tests of isolates from
the Alpine Nursery, Kalispell, Montana.
Missoula, MT: U. S. Department of Agriculture,
Forest Service, Northern Region; 1984; Rept.
84-14. 29 p.

270

Johnson, D. W. ; Harvey, R. D. Seed protectant
fungicides for control of Douglas-fir and
ponderosa pine seedling root rots. Tree
Planters' Notes. 26(2): 3-5; 1975.

Johnson, L. P. V. ; Linton, G. M. Experiments on
chemical control of damping-off in Pinus
resinosa seedbeds. Can. J. Res. C. 20: 559-571;
1942.

Jones, L. Germination of repellent-treated south-
ern pine seed before and after storage. Res.
Note SE-15. Asheville, NC: U. S. Department
of Agriculture, Forest Service, Southeastern
Forest Experiment Station; 1963. 4 p.

Karrfalt, R. P. Fungus-damaged seeds can be re-
moved from slash pine seedlots. Tree Planters'
Notes. 34(2): 38-39; 1983.

LaMadeleine, L. A. Evaluation of 3-0 and 4-0 red
pine in Eveleth Nursery, Minnesota, 1979.
Broomall, PA: U. S. Department of Agriculture,
Forest Service, Northeastern Area, State and
Private Forestry; 1979; Rept. NA-TP-2. 4 p.

Landis, I. D. An analysis of seed and seedling
losses at Mt. Sopris Tree Nursery. Lakewood,
CO: U. S. Department of Agriculture, Forest
Service, Rocky Mountain Region; 1976a; Bio.
Eval. R2-76-18. 13 p.

Landis, T. D. Fusarium root disease of container-
ized tree seedlings - Colorado State Forest
Service Nursery. Lakewood, CO: U. S. Depart-
ment of Agriculture, Forest Service, Rocky
Mountain Region; 1976b; Bio. Eval. R2-76-16.
6 p.

Lock, W. ; Sutherland, J. R. ; Sluggett, L. J.
Fungicide treatment of seeds for damping-off
control in British Columbia lorest nurseries.
Tree Planters' Notes. 26(3): 16-18; 1975.

Mason, G. N.; Van Arsdel, E. P. Fungi associated
with Pinus taeda seed development. Plant Dis.
Reptr. 62: 864-867; 1978.

Matuo, T.; Chiba, 0. Species and formae specialis
of Fusaria causing damping-off and root rot of
coniferous seedlings in Japan. Ann. Phytopath.
Soc. Japan 32: 14-22; 1966.

Merrill, W. ; McCall, K; Zang, L. Fusarium root
rot of Douglas-fir and Fraser fir seedlings in
Pennsylvania. Plant Disease. 65: 913-914.
1981.

Miller, W. 0.; Norris, M. G. A new review of soil
fumigation practices for use in forest nurser-
ies. Down to Earth. 26(3): 9-12; 1970.

Mittal, R. K.; Sharma, M. R. Evaluation of fungi-
cides to control some common seed-borne fungi.
Indian Forester. 35: 589-591; 1981.

Morgan, P. Fusarium stem rot of Douglas-fir seed-
lings. Plant Disease. 67: 441-442; 1983.

Neergaard, P. Seed pathology. New York: John
Wiley & Sons; 1977. 1187 p.

Pawuk, W. H. ; Barnett, J. P. Root rot and damping-
off of container-grown southern pine seedlings.
In: Tinus, R. W., Stein, W. I., Balmer, W. E.,
eds. North American containerized forest tree
seedling symposium: proceedings; 1974. Denver,
CO; Great Plains Agr. Council Pub. No. 68;
1975: 173-176.

Peterson, G. W. Seed-protectant chemicals affect
germination of ponderosa pine seed. Tree
Planters' Notes. 21(4): 25-29; 1970.

Rathbun-Gravatt, A. Germination loss of coniferous
seeds due to parasites. J. Agric. Res. 42: 71-
92; 1931.

Rowan, S. J. Tip dieback in southern pine nurser-
ies. Plant Disease. 66: 258-259; 1982.

Salisbury, P. J. Moulds of stored Douglas-fir seed
in British Columbia. Victoria, B.C.: Canadian
Department of Agriculture Forest Biology
Division; 1955; Interim Rept. 10 p.

Sharma, A. D. Fungi associated with conifer

(Pinus roxiburgii) seeds. Indian J. Mycol. and
Plant Pathol. 10(1): 106-107; 1978.

Shea, K. R. Phy totoxicity of thiram to Douglas-fir
seed. Weyerhaeuser Timber Co., For. Res. Note
21; 1959. 5 p.

Smith, R. S., Jr. Decline of Fusarium oxysporum
in the roots of Pinus lamber tiana seedlings
transplanted into forest soils. Phytopathology.
57: 1265; 1967.

Snyder, W. C; Hansen, H. N. The species concept
in Fusarium. Am. J. Bot. 27(2): 64-67; 1940.

Strong, F. C. Damping-off in the forest tree

nursery and its control. Mich. Agr. Exp. Sta.
Quart. Bull. 34: 280-296; 1952.

Sutherland, J. R.; Van Eerden, E. Diseases and
insect pests in British Columbia forest nurser-
ies. British Columbia Ministry of Forests,
Canadian Forestry Service; 1980; Joint Report
No. 12. 55 p.

Tint, H. Studies in the Fusarium damping-off of
conifers. I. The comparative virulence of
certain Fusaria. Phytopathology. 35: 421-439;
1945.

Urosevic, B. The influence of saprophytic and
semiparasitic fungi on the germination of
nursery spruce and Scots pine seeds. Proc.
Int. Seed Test. Assoc. 26: 537-556; 1961.

271

A.

7)

MONOCLONAL ANTIBODIES RECOGNIZING THE SEED-BORNE FUNGAL PATHOGEN

. Sirococcus strobilinus

<^ ■/

Leslie Ann) Mitchell

ABSTRACT: Hybridoma cell lines producing
monoclonal antibodies (McAbs) recognizing
antigens of the seed-borne conifer pathogen,
Sirococcus strobilinus have been prepared by
cell fusion techniques. Specificity of these
McAbs has been ascertained in enzyme- linked
immunosorbent and surface immunofluorescence
assays using a panel of nonrelated commensal
fungi. These McAbs will serve as reliable
diagnostic probes in assaying seeds for S.
strobilinus .

INTRODUCTION

Sirococcus shoot blight caused by the fungus ,
Sirococcus strobilinus Preuss (syn. Ascochyta
piniperda Lindau) is an important seed-borne
disease in coastal British Columbia container
nurseries where it affects Sitka and white
spruce (Sutherland and others 1981) . The
pathogenesis of Sirococcus blight and its
confinement to specific spruce seedlots are
indicative that the disease is seed-borne and
recently this was confirmed by Sutherland and
others (1981) .

Current methods for detecting seed-borne
pathogens such as S. strobilinus involve plating
surface sterilized seeds onto nutrient media and
identifying fungal outgrowth on the basis of
general morphology and the production of
distinctive spores. These techniques are
time-consuming and insufficiently sensitive for
detecting low levels of pathogens. The success
of Sirococcus detection by these methods is low
(0.5 to 3 percent) as often rapidly growing
saprobes mask the slower-growing Sirococcus ■
Also, as Sirococcus often fails to sporulate in
culture, definitive identification based on
spore morphology is not always possible.
However, the consequences of failure to diagnose
Sirococcus infestation in a seedlot are often
severe as nursery conditions are highly
favorable to the disease. Thus, a more reliable
test for this pathogen is needed. The
specificity, speed and relative economy of an
immunoassay would fulfill these requirements.

Paper presented at the Conifer Tree Seed in the
Inland Mountain West Symposium, Missoula, MT ,
August 5-6, 1985.

Leslie Ann Mitchell is NSERC Visiting Fellow at
the Pacific Forestry Centre, Canadian Forestry
Service, Victoria, B.C.

The primary requisite of such an immunologic
test is specific antibody to serve as a probe
for locating Sirococcus strobilinus components
(antigens) in or on seeds. Techniques devised
by Kohler and Milstein (Kohler and Milstein
1975; Milstein 1980) permit the isolation of
cells which produce antibodies of unique
specificities (called monoclonal antibodies).
In this procedure, antibody-forming cells taken
from the spleen of a mouse that has been
immunized with antigen are fused to cancerous
myeloma cells to form new cells (called
hybridomas) which have the ability to live
indefinitely in culture and to produce and to
secrete into the culture fluids monoclonal
antibodies (McAbs) directed to the antigen.

This paper describes the derivation, by cell
fusion techniques, of hybridoma cell lines which
produce McAbs recognizing antigens of S.
strobilinus . The McAbs recognize antigens from
several S. strobilinus isolates obtained from
diverse host species and tissues but do not
react with other nonrelated seed-associated
fungi when tested in enzyme-linked immunosorbent
(ELISA) or by indirect surface immuno-
fluorescence (IFA) assays and thus will be
useful diagnostic probes for Sirococcus in seeds
and other tissues.

MATERIALS AND METHODS
Fungal Isolation and Culture

Axenic cultures of Sirococcus strobilinus and 10
other saprophytic fungi isolated from
surface-sterilized (Sutherland and others 1978)
seeds obtained from an Engelmann spruce (Picea
engelmanni Parry) seedlot were prepared as
previously described (Sutherland and others
1981) . Fungi used for immunizing mice or for
the preparation of soluble antigens were grown
in 1.25% malt extract broth (MB) containing
trace amounts of B, Cu, Fe, Mn, Mo and Zn
(Vezina and others 1965).

Preparation of Fungal Antigens

Antigens for immunizing mice and for testing
McAbs were prepared as follows. Mycelium from
axenic cultures of S. strobilinus and commensal
fungi, grown in MB, was separated from the
culture medium, washed with 100 mL of distilled
water by vacuum filtration and used directly or
frozen at -20°C.

272

Total antiftens .--Mycelium (5-10 g) was
homogenized immediately in a volume of 3-4 mL of
phosphate buffered saline (PBS, pH 7.4
containing 10 mM Na and K phosphates, 3 mM KC1
and 140 mM NaCl) in a glass homogenizer, then
dried for 24 hr at 40°C in glass Petri dishes.
The dried mycelium was scraped from the Petri
dish, ground into a fine powder and stored at 4*C.

Soluble antigens . --Five to 10 g of frozen
mycelium was suspended in 20 mL of extraction
buffer (50 mM NH4HC03/2% w/v polyvinylpyrrolidone)
and homogenized for 2 min at maximum speed in a
polytron homogenizer (Type PT 20-0D, Kinematica,
Lucerne, Switzerland). The disrupted mycelium was
held at 0°C for 1 hr, rehomogenized , then
centrifuged for 20 min at 600 x g. The
supernatant was saved and the pellet re-extracted
twice with 15 mL of extraction buffer.
Supernatants from all extractions were pooled,
cleared by centrifugation at 12000 x g and stored
-20°C. Protein concentrations were determined as
described by Lowry and others (1951).

Secreted antigens . --Antigens secreted into
culture fluids were prepared by mixing two
volumes of 95% ethanol with one volume of
filtered culture fluids. Precipitation was
allowed to occur for 24-48 hours at 4°C then the
mixture was centrifuged for 30 min at 10000 x g
and the pellet redissolved in 10-20 mL of
distilled water and stored at -20°C.

Animals

BALB/c mice were obtained from Charles River
(Canada) Inc., St. Constant, Que, or bred from
parental stock purchased from the same source.
Both male and female mice, aged 6-12 weeks were
used .

Immunization of Animals

Mice were immunized by intraperitoneal (i.p.)
injection of 5 mg of air-dried mycelium
emulsified with Freund's complete adjuvant in a
volume of 0.2 mL/animal and challenged five times
at 4-6 week intervals by i.p. injection of 1 mg
of mycelium in Freund's incomplete adjuvant in
the same volume. Three days before cell fusion
mice were given 0.1 mL of S. strobilinus soluble
antigens intravenously.

Sera from mice hyper immunized with S. strobilinus
dried mycelium were pooled and used as positive
reference sera in enzyme- linked immunosorbent and
immunofluorescence assays as described below.
Sera pooled from another group of animals
immunized in the same way with air-dried mycelium
from Trichoderma viride (a common saprophyte on
seeds) served as a negative reference serum.

Enzyme-linked Immunosorbent Assays (ELISAs)

Enzyme-linked immunosorbent assays were performed
as described by Voller and others (1976) with the

following modifications. Wells of 96 well,
flat-bottomed polystyrene microtiter plates (no.
3590 Serocluster, Costar, Cambridge, MA) were
coated with 50 pL/well of fungal soluble
antigens adjusted to a protein concentration of
lOpg/mL in distilled water. The plates were
dried overnight at 37°C, then well sites not
coated with antigen were blocked by adding 200
pL of 1% bovine serum albumin (Fraction V) in
PBS (pH 7.4) to each well and incubating for 2 hr
at room temperature. All other aspects of the
assay were conducted as described by Voller and
others (1976) using 50 pL volumes/well of
hybridoma culture fluids or dilutions of ascites
fluids in the first antibody layer and the same
volumes of optimal dilutions of enzyme- labelled
second antibody. Enzyme-labelled second
antibodies employed were anti-mouse IgF(ab')2-
alkaline phosphatase conjugate (Helix Biotech,
Vancouver, B.C.) or anti-mouse IgG (H + L)-
peroxidase conjugate (HyClone Labs., Logan, UT)
which were used to detect antibodies of all
classes. Alkaline phosphatase- labelled
chain-specific second antibodies, anti-mouse IgM
(y-chain) and anti-mouse IgG (y-chain) were
used to detect antibodies of the IgM and IgG
classes, respectively. All second antibodies
were affinity-purified and no binding to
Sirococcus or other fungal antigens was observed
in the absence of specific antibody. In assays
where alkaline phosphatase- labelled second
antibodies were employed 4-nitrophenyl phosphate
(Boehringer-Mannheim, Dorval , Que . ) , 1 mg/mL in
diethanolamine buffer pH 9.8 (0.97% v/v
diethanolamine containing 0.02% NaN3 and 0.1 mg/mL
MgCl2.6 H20) was used as the substrate. The
enzymatic reaction was allowed to proceed for 45
min in the dark at room temperature then stopped
by adding 50 yL/well of 3 M NaOH and A405 was
determined for each well with a MicroELISA
Minireader (Series MR 590, Dynatech, Alexandria,
VA) . In assays where peroxidase conjugate was
used the substrate solution consisted of
o-phenylenediamine (Fisher, Vancouver, B.C.),
0.4 mg/mL and 3% H202, 4 yL/mL (v/v) in
citrate buffer (pH 5.0). The microplates were
incubated in darkness for 15 min at room
temperature. The enzymatic reaction was stopped
by adding 50 yL of 2.5 M H2S04 to all
wells and A490 was determined for each well.

Indirect Immunofluorescence Assays (IFA)

Unit layers of mycelium were grown on acid
washed sterile 18 mm2 glass coverslips
(mounted on autoclaved filter paper disks
moistened with MB in covered Petri dishes) by
inoculating them with 1 drop of medium from
sporulated cultures. When hyphae were visible
the coverslips were removed individually to 60
mm glass Petri dishes and washed with 5 mL PBS
containing 10% fetal calf serum (FCS) for 15 min
at room temperature. The PBS/FCS was suctioned
off and replaced with 100 yL of undiluted
culture fluids and incubated in a moist chamber
at room temperature for 1 hr. The coverslips
were washed with 10 mL PBS/FCS and suctioned

273

dry. One hundred pL of fluorescein (FITC) -
conjugated second antibody (anti-mouse IgG or
anti-mouse IgM chain-specific sera, Kirkegaard
and Perry, Gaithersburg, MD diluted to 1/20 in
PBS/FCS and centrifuged for 5 min at 12000 x g)
were added to each covers lip and incubated for 1
hr at room temperature. The coverslips were
washed as before and inverted onto glass slides
with 1 drop of mounting medium (90% glycerol in
PBS, with 1 mg/mL o-phenylenediamine) . The
slides were examined with a Zeiss Photoscope II
equipped with a 100 illuminator (fitted with an
HBO 50 W high pressure Hg source) ,
epif luorescence condenser III RS, 390-440 nm
excitation and 4 75 nm barrier filters, lOx and
40x Neofluar objectives and lOx Kpl-w eyepieces.

Derivation of Hybridomas

Media and supplements . --RPMI -1640 medium,
glutamine (200 mM) , gentamycin (50 mg/mL),
4-(2-hydroxyethyl)-l-piperazine ethanesulf onic
acid (HEPES, 1 M) were obtained as sterile
solutions from Gibco Canada Ltd. , Burlington,
Ont. Sodium pyruvate, 2-mercaptoethanol ,
hypoxan thine, aminopterin, thymidine,
dimethylsulf oxide (DMSO) and 2, 6, 10, 14-
tetramethylpentadecane (Pristane) were obtained
from Sigma, St. Louis, MO. Polyethylene glycol
(Serva 1550) was from Terochem Labs., Edmonton,
Alta. Fetal calf serum (FCS) , preselected for
its ability to support optimal growth of the
myeloma cells used in fusion (see below) was
supplied by Animal Health Labs., Islington,
Ont. FCS was heat-activated (56°C for 30 min)
prior to use as a medium supplement.

Myeloma cells ,--SP2/0-Agl4 , a nonsynthesizing
myeloma cell line of BALB/c mouse origin
(Shulman and others 1978) was obtained from Dr.
T.W. Pearson (Department of Biochemistry and
Microbiology, University of Victoria, B.C.).
Prior to their use in cell fusion, the cells
were grown for several passages in log phase in
growth medium consisting of RPMI-1640
supplemented with 10% FCS, 0.5 mg/mL gentamycin,
2 mM glutamine, 1 mM Na puruvate and 0.1 mM
2-mercaptoethanol .

Cell Fusions . --Media, supplements and solutions
used in cell fusion were prepared as described
by Pearson and others (1980) . All procedures
were carried out under aseptic conditions.
BALB/c mice that had been hyperimmunized with
dried mycelium and rechallenged by intravenous
injection of soluble antigens from S.
strobilinus were used as spleen cell donors.
Spleens were dispersed into single cell
suspensions in fusion medium consisting of
RPMI-1640 medium supplemented with 20% FCS, 0.5
mg/mL gentamycin, 2 mM glutamine, 1 mM Na
pyruvate, 0 . 1 mM 2-mercaptoethanol and 10 mM
HEPES, washed twice by centrif ugation at 500 x g
for 10 min at 4°C and adjusted to 1 x 107 or
2 x 10 7 cells/mL in the same medium. Log
phase SP2/0 cells were washed once by
centrif ugation and adjusted to 1 x 106 or 2
x 10* cells/mL in fusion medium. Immune

spleen cells and myeloma cells were mixed at a
ratio of 10:1 and fused in the presence of
polyethylene glycol as described by Pearson and
others (1980). The fused cells were immediately
dispensed in 50 pL aliquots into wells of 96
well microculture plates (no. 25860, Corning,
Palo Alto, CA) , supplemented with 160 pL/well
of fusion medium containing 6.25 x 10*
normal BALB/c mouse thymocytes as feeder cells
and cultured for 18 hr at 37 °C in 10% C02 in
air. At the end of this period, the medium was
removed and replaced with 150 pL/well of
selective medium (HAT medium, consisting of
fusion medium with 13 pg/mL hypoxan thine, 0.2
Pg/mL aminopterin and 3.9 pg/mL thymidine).
The microplates were returned to the incubator
and fresh HAT medium added to the wells on days
7 and 14 after cell fusion. During this period
the microplate wells were checked frequently for
cell growth. When colonies were visible,
culture fluid from each well was tested in ELISA
for specific antibody. The contents of wells
producing Sirococcus- specif ic antibody were
removed to 1 mL of HT medium (fusion medium
supplemented with 13 pg/mL hypoxanthine and
3.9 pg/mL thymidine) containing 1 x 106
thymocytes/mL in 24 well culture plates (no.
3524, Costar) . After cell expansion, antibody
specificity was checked by testing the medium in
each well in ELISA using soluble antigens from
either Sirococcus or Trichoderma . Hybridoma
cells from wells that were positive for
Sirococcus but negative for Trichoderma were
cloned by limiting dilution as described below.

Limiting Dilution Cloning of Hybridomas

Hybridomas were cloned twice by diluting cells
to 80, 20 and 10 cells/mL in cloning medium
consisting of HT medium (first cloning) or
fusion medium (second cloning) with 1 x 106
thymocytes/mL. For each dilution, 0.1 mL/well
was pipetted into 12 wells of microculture
plates to give cell densities of 8, 2, and 1
cell/well, respectively. After incubation at
37°C in 10% C02 in air for 7-14 days, wells
were screened for specific antibody production
and selected clones were expanded and
cryopreserved or injected into mice for ascites
production as described below.

Freezing and Thawing of Hybridomas

Hybridoma cells obtained from log phase culture
were adjusted to 2 x 106 to 1 x 107
cells/mL in 90% FCS/10% glycerol and cryo-
preserved in 1 ml aliquots in liquid nitrogen.
To re-establish hybridomas" in culture, the cells
were thawed rapidly in a 37 °C water bath and
immediately centrifuged through 10 mL of warmed
fusion medium. Cells were resuspended in 3-5 mL
of the same medium in 25 cm2 culture flasks
which were gassed with 5% C02 in air, sealed
and incubated as described above.

274

Production of Ascites

For production of ascites tumors, 1-2 x 107
hybridoma cells in a volume of 0.2 mL were
injected i.p. into BALB/c mice primed (by i.p.
injection) 10 days previously with 0.5 mL of
Pristane. Ten to 14 days later the animals were
killed and the ascites fluids removed
aseptically from the peritoneal cavity. Cells
were pelleted by centrifugation at 500 x g for
10 min at 4*C and cryopreserved as described
above. The supernatant was filtered through
cotton wool then stored in 0.2 mL aliquots at
-20'C.

RESULTS

Derivation of Hybridoma Cell Lines Producing
Monoclonal Antibodies Directed to Sirococcus
strobilinus .

Two independent cell fusions (SDMI and SDM2)
were conducted using different spleen cell
(SPLC) and myeloma cell (SP2/0) cell densities
as described in Methods. Macroscopic colonies
were visible in microculture plate wells within
7 days and culture fluid from each well was
tested for specific antibody by ELISA 11 days
after fusion. In fusion SDMI in which SPLC and
myeloma cells were mixed at a ratio of 1 x
107:1 x 10* , 31 wells out of 480
contained antibody-producing hybidomas. The
second fusion (SDM2) conducted at higher
SPLC.myeloma cell densities (2 x 107:2 x 10')
resulted in 7/480 wells which were positive for
antibody production.

Cells producing specific antibody were expanded
in 24 well culture plates and five days later,
culture medium from each well was tested for
specific antibody in ELISA using soluble
antigens from S. strobi 1 inus or Trichodenna
viride or extraction buffer to coat ELISA plate
wells. In 11 of the wells tested, antibodies in
the culture fluids bound only to Sirococcus
antigens, while in nine of the wells, antibodies
which bound to both Sirococcus and Trichodenna
antigens as well as extraction buffer-coated
wells, were present. Those cells producing
antibodies which recognized only Sirococcus
soluble antigens were cloned twice by limiting
dilution. After each cloning, hybridoma cells
were tested for specific antibody production by
ELISA. As a result, 30 stable cloned hybridoma
cell lines producing Sirococcus- spec if ic
monoclonal antibodies (McAbs) were established.
These cell lines are clones originating from
five different (parental) wells in the initial
fusion plates (indicated by the numbers 475,
441, 434, 171 and 392 in table 1). For each
line, cells were cryopreserved in liquid
nitrogen and McAb isolated from culture or
ascites fluids. Twelve of the hybridoma lines
(of the series 441 and 171) produce McAbs of the
IgM class, while 18 (series 434, 475 and 392)
hybridoma lines produce IgG McAbs as determined
in ELISA using chain-specific second antibodies.

Analysis of Monoclonal Antibody Specificity

Antigen recoRnition in other Sirococcus
strobilinus isolates . --As the primary goal of
producing S. strobilinus-specif ic McAbs was to
develop a diagnostic reagent for detecting this

Table 1. — Reactivity of Sirococcus -directed monoclonal antibodies with soluble
antigens from other Sirococcus strobilinus isolates

A49C nm/Hybridoma1

Sirococcus Host2 SDM SDM SDM SDM SDM

isolate 1/475. 6a. 9 1/441.4.4 1/434.15.21 1.171.27.13 2/392.2.4

2456

ES

2.00

2.00

2.00

2.00

2.00

8652

SS

1.33

0.39

0.35

2.00

1.25

2231

SS

2.00

0. 74

2.00

1.67

0.27

8615

WS

1.41

1.96

0.49

2.00

1.92

4321

IS

1.55

2.00

0.39

2.00

1. 79

RR

LP

1.75

1.85

0.73

2.00

0. 79

2311

LP

2.00

1.11

0. 71

2.00

0.42

4248

LP

1.46

1.19

1.44

2.00

1.15

2247

WH

0.46

0.00

0.00

1.89

0.08

1 McAbs produced by SDM hybridoma cell lines were tested in ELISA using soluble

antigens (at 10 yg/mL) prepared from the isolate used in their derivation (2456)
and eight other S. strobilinus isolates, to coat ELISA plate wells. A positive
A4,0 nm value indicates reactivity of the McAb with its antigen in the soluble
antigen mixture.

2 ES = Engelmann spruce, SS - Sitka spruce, WS = white spruce, IS = Interior spruce,

LP = Lodgepole pine, WH = western hemlock.

275

pathogen in seeds, it was important to ensure
that the McAbs were not isolate-specif ic .
Therefore, culture fluids obtained from each
hybridoma line were tested in ELISA using
soluble antigens prepared from nine different
isolates of S. strobilinus from several host
species and tissues. The results of assays done
on representative clones of each parental cell
line are shown in table 1. The results obtained
with McAbs from other clones within each
parental line were similar. Thus, the McAbs
produced by the hybridomas collectively
recognized antigen(s) present in all the
Sirococcus isolates tested. However,
differential patterns of reactivity were
observed with different Sirococcus isolates and
McAbs originating from different parental lines.

Specificity of monoclonal antibodies for
Sirococcus strobilinus . --As Sirococcus is
frequently associated with nonpathogenic fungi
in or on seeds, it was necessary to ensure that
the McAbs did not recognize antigens common to
Sirococcus and other saprobes . Therefore, the
McAbs were tested in ELISAs using soluble
antigens prepared from Sirococcus or soluble
antigens from seven different genera of
commensal fungi all isolated from the same
infested seedlot (2456) to coat microtiter plate
wells. The results (table 2) showed that all
McAbs were specific for S. strobilinus antigens
and did not crossreact with antigens from

commonly found saprophytic fungi. Although all
the McAbs produced by the hybridoma lines were
tested for specificity, only data from
representative clones of each parental line are
shown .

Monoclonal antibodies from representative
hybridoma lines were also tested in indirect
surface immunofluorescence assays (IFA) against
mycelium from Sirococcus and the same commensal
isolates. Although some of the McAbs recognized
antigens exposed on the surface of Sirococcus
mycelium (see below), none of the McAbs bound to
mycelium of other fungal genera.

Morphological Distribution of Antigens
Recognized by Sirococcus-Directed Monoclonal
Antibodies .

Recognition of secreted antigens . --Monoclonals
from representative hybridoma lines were tested
in ELISAs in which ethanol-precipitated antigens
from fungal culture medium (secreted antigens)
were used to coat microplate wells. Several
clones representing different parental hybridoma
lines were tested. The data in table 3 show
that two of the hybridoma lines (SDM 1/441.8.2
and SDM 1/171.27.13) produced McAbs which
recognize antigen(s) that are also secreted into
the culture medium by Sirococcus . No binding to

Table 2. --Reactivity of Sirococcus-directed monoclonal antibodies with
soluble antigens from Sirococcus and other fungi

A49D nm/Hybridoma1

Fungus SDM SDM SDM SDM SDM

Genus2 1/475. 6a. 5 1/441.4.4 1/434.15.21 1.171.27.22 2/392.2.4

Sirococcus

1

91

1.06

1

23

0

71

0.87

Trichoderma

0

05

0.14

0

04

0

10

0.03

Altemaria

0

04

0.05

0

03

0

04

0.03

Paecilomyces

0

05

0.05

0

05

0

11

0.04

Sclerophoma

0

03

0.03

0

02

0

03

0.02

Penici Ilium

0

04

0.04

0

04

0

07

0.03

Rhizopus

0

02

0.03

0

01

0

01

0.01

Mucor

0

02

0.02

0

02

0

02

0.01

Extraction Buffer

0

02

ND

0

01

0

02

0.04

McAbs antibodies produced by SDM hybridomas were tested in ELISA using
soluble antigens from S. strobilinus or from seven different fungal genera
to coat microplate wells.

Soluble antigens prepared from S. strobilinus and seven other fungal
isolates from seedlot 2456 were coated onto ELISA microplate wells at a
concentration of 10 yg protein/mL. Extraction buffer (see Methods) coated
at 1/10 dilution served as a control.

276

secreted antigens of Penicil 1 ium was observed
with any of the McAbs tested.

Recognition of surface antigens .- Monoclonals
from representative hybridoma lines were tested
in I FA for their ability to recognize antigens
exposed on the surface of mycelium grown on MB.
Pale yellow autof luorescence in mycelial
preparations caused a background which
interfered with interpretation of low levels of
FITC (green) fluorescence due to specific

Table 3. Reactivity of Sirococcus -directed
monoclonal antibodies
with fungal secreted antigens

SDM
Hybridoma

A49G nm/Antigen1
Sirococcus Penici Ilium

1/434.15.21
1/171.27 .13
1/441.8.20
1/475 .6a. 10
2/392.10.20

0.08
2.00
0.68
0.02
0.10

0.05
0.07
0.05
0.01
0.04

1 Fungal secreted antigens were prepared by
ethanol- precipitation of spent culture medium
( see Methods ) .

antigen- antibody interactions. Therefore, the
McAbs were tested several times in separate
assays. Although the majority of the McAbs did
not bind to the surface of Sirococcus , some
McAbs produced two distinct patterns of
fluorescence. Some McAbs produced bright
patches of fluorescence at what resembled thin
areas of the cell wall often near the hyphal tip

or at broken ends of the hyphac ("patchy
surface" fluorescence). Other McAbs produced
bright punctuate fluorescence at structures
resembling holes in the cell wall and on
fibrillar material surrounding the hyphae
("patchy surface + fibrils"). These
observations are summarized in table 4.

DISCUSSION

This paper has described the derivation by cell
fusion techniques of monoclonal antibodies
(McAbs) which recognize antigens of the fungus,
S. strobi linus . This approach was necessary as
preliminary investigations (unpublished) with
polyclonal antisera raised in rabbits and mice
by immunizing them with Sirococcus extracts
revealed a considerable amount of cross-
reactivity between Sirococcus and other
nonrelated seed- associated fungi, presumably due
to shared antigenic determinants. McAbs, which
recognize a single site on an antigen molecule
(Milstein 1980) , may be selected for their
ability to recognize unique antigenic sites in
related organisms and therefore circumvent the
problems of crossreact ivity often observed with
antisera. The Sirococcus- directed McAbs
described herein did not recognize internal or
surface antigens of nine different nonrelated
fungi often associated with Sirococcus in
seeds. Therefore, these McAbs are probably
directed to unique antigens of S. strobi 1 inus
and will serve as useful diagnostic reagents for
detecting this conifer pathogen in seeds or
other tissues. These McAbs, as a group, also
recognized antigens of eight other S.
strobilinus isolates obtained from diverse host
species and tissues. Therefore, the McAbs,
although derived

Table

4.- Surface

immunofluorescence patterns

observed

with Sirococcus-

directed

monoc lonal

antibodies

SDM Monoclonal

Fluorescence Distribution1

Antibody

Negative

Patchy Surface

Patchy

Surface & Fibrils

1/441

4.2

• •

• •

1/441

8.6

1/441

8.18

•

1/171

21.16

• •

•

1/171

21.22

•

1/434

15.9

• •

•

1/434

28.5

1/475

6a. 2

1/475

6a. 4

•

2/392

5.31

• •

2/392

2.4

• •

•

2/392

10.20

•

McAbs from representative SDM hybridoma cell lines were tested for
surface binding to unit layers of Sirococcus strobilinus mycelium. Bound
McAbs were detected with a mixture of FITC-conjugated goat anti- mouse IgM
and goat anti-mouse IgG second antibodies at 1/20 dilutions. Each dot
represents the results of one assay conducted with the monoclonal antibody.

277

using a single isolate, were not isolate-
specif ic and can be used generally to detect
this fungus. The observed differential patterns
of reactivity of the monoclonals with other
Sirococcus isolates suggest that they recognize
more than one antigen and/or there is
differential expression of these antigens in
Sirococcus obtained from varied sources.
Although all McAbs tested recognized antigens in
isolates from spruce and pine hosts, only five
hybridoma clones produced McAbs which recognized
antigen(s) of Sirococcus isolated from western
hemlock .

A preliminary characterization of the
morphologic distribution of the antigens
recognized by these McAbs indicated that some
antigens may be exposed on the surface of the
fungus and/or may be secreted into culture
fluids. As mice were immunized with whole
mycelium and therefore were exposed to the
whole antigenic repertoire of the fungus,
antibodies directed to both internal and surface
antigens would have been generated. However, as
antibodies produced by the hybridomas were
selected in ELISA using only soluble antigens, a
predominance of antibodies directed to internal
components would be expected. Indeed, this was
the case, as the majority of the McAbs did not
react with the mycelial surface in IFA.
However, several of the monoclonals exhibited a
patchy distribution of surface fluorescence in
these assays. Whether this represents
reactivity of the McAbs with cell wall
antigen(s) (that were also present in the
soluble antigen preparation used in ELISA) or
accessibility of the antibodies to an internal
component through a lesion or thinning of the
cell wall, is speculative. The variability in
IFA fluorescence patterns may also reflect the
age and physiologic state of the mycelial
material used in the assays. Two monoclonals
(SDM 1.172.27.13 and SDM 1/441.8.20) reacted in
ELISA with Sirococcus secreted antigens.
Another clone (SDM 1/441.8.18) produced antibody
which reacted in IFA with both the mycelial
surface and associated fibrillar material.
Perhaps the antigens recognized by these McAbs
may be enzymes or secondary metabolites that are
transported to the exterior of the fungus.

The specificity of the Sirococcus-directed McAbs
described here will allow their use as
diagnostic probes for this pathogen in seeds and
other plant tissues. Sensitive immunologic
assays employing these McAbs are currently being
developed with the objective of establishing a
simple, rapid and accurate method for screening
seedlots destined for use in forest nurseries or
for export.

REFERENCES

Kohler, G.; Milstein, C. Continuous cultures of
fused cells secreting antibody of predefined
specificity. Nature. 256 (5517): 495-497; 1975.

Lowry, 0. H.; Rosebrough, N. J.; Farr, A. L.;
Randall, R. J. Protein measurement with the
Folin phenol reagent. J. Biol. Chem. 193;
265-275; 1951.

Milstein, Cesar. Monoclonal Antibodies.
Scientific American; 1980, Oct.: 66-74.

Pearson, Terry W. ; Pinder, Margaret; Roelants,
Georges E. ; Kar, Santosh K. ; Lundin, Lena B. ;
Mayo r-Whi they , Kathleen S.; Hewett, Rosemary S.
Methods for derivation and detection of
anti-parasite monoclonal antibodies. J.
Immunol. Meth. 34: 141-151; 1980.

Shulman, Marc; Wilde, CD.; Kohler, Georges.
A better cell line for making hybridomas
secreting specific antibodies. Nature. 276:
269-270; 1978.

Sutherland, Jack R. ; Woods, T. A. D. ; Lock, W. ;
Gaudet, D. A. Evaluation of surface
sterilants for isolation of the fungus,
Geniculodendon pyiforme. from Sitka spruce
seed. Can. For. Serv. Bi-monthly Res. Notes
34: 20-21; 1978.

Sutherland, Jack R. ; Lock, W. ; Farris, S. H.
Sirococcus blight: a seed-borne disease of
container-grown spruce seedlings in Coastal
British Columbia forest nurseries. Can. J.
Bot. 59(5): 559-562; 1981.

Vezina, Claude; Singh, Kartar; Sehgal, S. N.
Sporulation of filamentous fungi in submerged
culture. Mycologia. 57: 722-726; 1965.

Voller, A. D. ; Bidwell, D.; Bartlett, A.
Microplate enzyme immunoassays for the
diagnosis of viral infections. In: Rose, N. ;
Freedman, H. (eds) , Manual of Clinical
Immunology, American Society for Microbiology;
1976: 200-206.

ACKNOWLEDGMENTS

The author thanks T.A.D. Woods and W. Lock for
providing axenic cultures of Sirococcus
strobilinus and commensal fungi and M. Griffin
and G. Jensen for help with immunofluorescence
assays .

278

POSTER PAPER ABSTRACTS

IMPACT OF INSECTS ON SEED PRODUCTION IN A
DOUGLAS-FIR SEED ORCHARD

Scott A. Dombrosky and Timothy D. Schowalter

ABSTRACT: The impact of various factors on seed
production in a Douglas-fir (Pseudotsuga
menziesii) seed orchard in western Oregon was
examined by monitoring the fate of seeds in 30
cones, stratified into three crown levels, on
each of 10 trees during the 1984 growing season.
Cones were examined monthly between April and
September for mortality or evidence of insect
damage. In September a sample of mature cones
was collected and completely dissected. Each
seed was examined for extractability , insect
damage, or unexplained abortion. These data were
used to measure the relative impacts of various
cone and seed mortality agents, and to develop an
inventory monitoring system for management of
Douglas-fir seed production. The results of this
study indicate that, in addition to the recog-
nized importance of insects tc seed loss at cone
maturity, insects also cause seed production
losses through early conelet abortion in
Douglas-fir.

Scott A. Dombrosky is Graduate Research Assist-
ant, Department of Entomology, Oregon State
University, Corvallis, OR.

Timothy D. Schowalter is Assistant Professor,
Department of Entomology, Oregon State
University, Corvallis, OR.

SELECTIVELY HARVESTING CONES WITH A FANDRICH
AERIAL RAKE

Helmut Fandrich

ABSTRACT: Selecting the trees to be harvested
from the air allows the forester to specify har-
vesting areas, elevations, type of trees and even
specific trees. Aerially picking dominant trees
at selected intervals with a Fandrich rake gives
the forester the opportunity not only to freely
specify the best trees from a broad genetic base,
but also to rapidly collect large volumes during
heavy crop years when the yields are higher and
the costs are lower. The rate of picking can be
estimated if one knows the quantity of cones near
the top of the trees since the pilot will rake
approximately one tree per minute.

Helmut Fandrich is a professional engineer and
acts as a Machine Design Consultant, Clearbrook,

BC, Canada.

ASSESSING INDIVIDUAL SPECIES' IMPACT ON SEED
YIELD AND COMMUNITY INTERACTIONS AMONG CONE AND
SEED INSECTS

Nancy Rappaport

ABSTRACT: Methods are described for assessing
individual species' impact on seed yield and
community interactions among cone and seed
insects. The procedure involves sequentially
bagging cones to exclude all possible combina-
tions of species. These exclusions may be
viewed as a series of simulated pesticide appli-
cations. Thus, in our study, we excluded each
of the three major pests of Douglas-fir cones
singly and in combination with the other pest
species. This was possible because there was a
degree of nonoverlap in the oviposition phenolo-
gies of the principal pest species. Cones were
left bagged during the oviposition periods of
each species, then were exposed to allow ovi-
position by the other species. This permits
assessment of impact by individual species and
also yields information about potential associ-
ative and competitive interactions among the
species. Such interactions may have important
consequences for our pest management strategies,
particularly if our treatments remove only
certain of the pest species.

Nancy Rappaport is Biological Technician, Pacific
Southwest Forest and Range Experiment Station,
Berkeley, CA.

USE OF SYSTEMIC INSECTICIDES TO PROTECT INDI-
VIDUAL TREES FROM WESTERN SPRUCE BUDWORM

Richard C. Reardon

ABSTRACT: Implantation and injection methods are
described for protection of individual trees
from western spruce budworm. Implantation is
with Medicaps which contain a dry chemical;
injection is with Mauget Systemic Injector Units
which contain a liquid formulation. Impact of
consecutive yearly applications in reducing
western spruce budworm populations and in
response of trees to wounding is described. The
advantages and disadvantages of each method are
compared .

Richard C. Reardon is Entomologist, Forest Pest
Management, Northeastern Area, USDA Forest Service,
Morgantown, WV.

279

YIELD AND VIABILITY OF NORTHERN LODGEPOLE PINE
SEED

Allen Richmond, John Alden, Thomas Malone, and
Robert Van Veldhuizen

ABSTRACT: Lodgepole pine (Pinus contorta Dougl.
ex Loud.) is the most widely planted conifer in
boreal forests. Demand for seed from northern
populations has accelerated cutting of trees for
cone harvest. Populations from central Yukon are
discontinuous and may not produce enough seed to
meet demands unless utilization of cone crops is
improved. At the present time, only cones from
the 2- to 5-year age class are usually recom-
mended for commercial collection.

Serotinous cones were found to contain viable
seed to 55 years of age. Seed age had no effect
on seed vigor as measured by rate of real (viable
seed) germination. Seed recovery was not affect-
ed by age of cones. Seed viability declined with
age, but still maintained relatively high germi-
nation rates to 20 years of age. The seed not
collected in the 6- to 20-year age classes repre-
sents a wasted seed resource. Portions of this
resource should be utilized as its viability and
vigor are only slightly lower than seed from 2-
to 5-year-old cones.

Allen Richmond is Research Associate, School of
Agriculture and Land Resources Management,
University of Alaska, Fairbanks, AK.

John Alden is Research Forester, Pacific North-
west Research Station, Fairbanks, AK.

Thomas Malone is Research Aid, School of Agri-
culture and Land Resources Management, University
of Alaska, Fairbanks, AK.

Robert Van Veldhuizen is Research Aid, School of
Agriculture and Land Resources Management,
University of Alaska, Fairbanks, AK.

DEFOLIATION AND DOUGLAS-FIR CONE AND SEED LOSSES
CAUSED BY BUDWORM IN THE NORTHERN U.S. ROCKY
MOUNTAINS

Raymond C. Shearer

ABSTRACT: In September 1979, counts of ovulate
buds on 25 randomly selected branchlets on each
of 22 Douglas-fir (Pseudotsuga menziesii [Mirb.]
Franco) trees at 20 locations in the Northern
U.S. Rockies predicted a good cone crop for 1980
throughout the study area. However, potential
cone crop was nearly eliminated in areas of
moderate to high defoliation by western spruce
budworm (Choristoneura oocidentalis Freeman),
especially in Montana east of the Continental
Divide and in nearby east-central Idaho. Few
cones matured in stands with a 1980 defolia-
tion-rating index greater than 40 percent.
Stands with low to moderate defoliation-rating

indexes (from 30 to 40 percent) had from 30 to
90 percent of the potential cones killed by
budworm larvae. The remaining stands had low
defoliation rating indexes (from 2 to 21 per-
cent) and low cone mortality caused by budworm
larvae (from 2 to 25 percent of the cone poten-
tial). Most of the mortality occurred during
the early stages of cone development (92 percent
were killed by budworm larvae in the bud or
erect conelet stage) . These losses accounted
for about 80 percent of all seed mortality
caused by budworm.

The remaining 20 percent of budworm-caused seed
mortality occurred in mature cones. Budworm
larvae reduced the number of viable seed in
mature cones by 1 to 32 percent, depending on
location. Seed loss in live cones increased in
areas of greater defoliation. Other insects
causing seed losses in mature cones were a cone
moth (probably Barbara colfaxiana [Kearfott]), a
cone worm (probably Dioryotria sp.), a scale
midge (probably Contarinia washingtonensis
Johnson) , a seed chalcid (probably Megastigmus
spermotrophus Wachtl), and a seedbug (probably
Leptoglossus occidentalis Heidemann) .

Forest managers should not depend on Douglas-fir
cones to provide seed for artificial or natural
regeneration when the stand-defoliation index is
greater than 30 percent, even in years of poten-
tially heavy cone production. Budworm larvae
also decrease seed production in stands with
defoliation under 30 percent. To protect
Douglas-fir cone crops, insecticide should be
applied when the cone and pollen buds open.

Raymond C. Shearer is Principal Silviculturist ,
Intermountain Research Station, Missoula, MT.

REDUCING DOUGLAS-FIR SEED AND CONE DAMAGE
Lawrence E. Stipe

During 1979 and 1980, single, double, and triple
ground applications of acephate and carbaryl
were used to improve Douglas-fir {Pseudotsuga
menziesii [Mirb.] Franco) seed and cone produc-
tion in areas of moderate to heavy western
spruce budworm {Choristoneura occidentalis
Freeman) populations. Ten single-tree replica-
tions were used for all treatments. Cone-bear-
ing Douglas-fir trees 9 to 18 m (30 to 60 ft)
tall were selected by examining cone buds, then
randomly assigned one of the three treatments.
Mixing and application rates were according to
label instructions: acephate — 400 g (300 g Al) in
500 L water (2/3 lb. [1/2 lb. Al] in 100 gal.
water); and carbaryl — 2.5 L (1.2 kg Al) in 500 L
water (2 qt. [1 lb. Al ] in 100 gal. water).

280

Each tree was sprayed beyond the point of runoff
with a hydraulic pumper and handgun operated at
28 kg/cm* (400 lb/irfi). The single applica-
tion, timed to coincide with peak second instar
spring dispersal, was on May 21, 1979, when cone
buds had expanded to 2.54 ci" (1 in) long and
were still erect. Postspray densities of fourth
instars for acephate, carbaryl, and control were
3.6, 1.3, and 21.2 larvae per 100 shoots, respec-
tively; defoliations after pupation were 10.8,
2.2, and 73.0 percent, respectively. Of the
potential 68 seeds per cone, the single applica-
tion provided seed production per cone of 0.5
seed for acephate, 1.0 seed for carbaryl, and no
seeds for control. These less-than-expected
results were primarily because budworm feeding
had damaged over 30 percent of the cones before
treatment and because subsequent cone damage by
budworm and other seed-and-cone species had
caused additional damage before cone harvest on
August 23. To reduce prespray infestation losses
and provide longer protection, double and triple
applications were tested the next year. The
first applications were on May 8 and 9, 1980,
coinciding with the beginning of larval disper-
sal, which peaked on May 24. This earlier timing
reduced prespray cone damage to less than 14 per-
cent. The second applications were on May 27 and
28. The last of the triple application was on

June 16. Postspray, mature budworm larvae per
100 shoots for acephate, carbaryl, and control
were 0.9, 0.6, and 16.3 for double application
and 0.3, 0, and 16.3 for triple application;
defoliation was 4.8, 0.9, and 77.5 percent for
single application, 1.1, 0.6, and 77.5 percent
for double application, and 1.3, 0.6, and 77.5
percent for triple application. Mature cones
were collected on August 18. Total seed yield
for acephate, carbaryl, and control was 34.2,
41.4, and 22.5. Total seed yield for all spray
treatments was significantly different from the
control, but no significant differences were
found between acephate and carbaryl or between
double or triple applications. Seed condition
(sound, hollow, and insect-damaged) and germina-
tion differences were small, variable, and not
related to treatment. Where seed yields are
important and budworm populations are moderate to
heavy, a double application of either acephate or
carbaryl is recommended. Time the first applica-
tion when budworm larvae first begin spring dis-
persal, and follow with a second in about 14
days .

Lawrence E. Stipe is Entomologist, USDA Forest
Service, Cooperative Forestry and Pest Manage-
ment, Northern Region, Missoula, MT.

281

SYMPOSIUM SUMMARY: CONIFER TREE SEED IN THE INLAND MOUNTAIN WEST

Stanley L. Krugman

In spite of the recent advances in tree propaga-
tion, such as tissue culture and mass rooting,
seeds will remain the major source of planting
material for many years to come. Each year we
are dealing with more tree species from more
areas. Each year we are working with a larger
number and variety of seed lots as we attempt to
better match seed source to site. As the genet-
ics and tree improvement programs expand there is
a continuous need for additional information in-
cluding time of flowering, and improved methods
of collecting, processing, storing, and germinat-
ing seeds.

Basically such information is needed to improve
efficiency and to reduce field planting costs.
Speakers throughout this meeting have addressed
these issues from a number of different points of
view. We have been fortunate to have researchers
as well as regeneration specialists share their
experiences with us. Predicting and stimulating
seed production remains a major issue in western
forestry. A further understanding of flowering
biology and development is essential if a
rational program is to be initiated. The review
of cone and seed biology and development by
John N. Owens demonstrated that there are a
series of critical stages in cone and seed devel-
opment. We need to have an understanding of them
for our individual species under our local envi-
ronmental conditions if we are to obtain maximum
yield or at least reduce certain types of losses.
Closely related is the use of hormones to stimu-
late flowering in some of our western species.
Stephen D. Ross reviewed the current experience
with hormones, their value, and limitations.
Although the system is not perfect, for selected
species and purposes we now have a valuable tool.
Reports by Glenn L. Jacobsen, Allen S. Rowley,
and Raymond C. Shearer pointed out that many con-
ditions influence cone production and seed yield
and our understanding of environmental influences
is still very primitive. Interesting papers by
Katherine A. Yakimchuk, R. J. Hoff, Carole L.
Leadem, and A. K. Helium provide new data on seed
germination and methods for increasing germina-
tion. These papers and several others, however,
demonstrated we are not consistent in defining

germination. As noted by Katherine A. Yakimchuk,
we need to consider both radicle elongation and
seedling development. This is an important ele-
ment since older seeds and immature seeds, as
well as damaged seeds, when germinated may have
radicle elongation but seedling development is
abnormal and the resulting seedlings will perish
in the nursery bed.

Biotechnology that offers unique opportunities
in tree improvement may soon play a major role
in reducing seedborne diseases. The research by
Leslie Ann Mitchell suggests that the further
development of appropriate monoclonal antibodies
techniques will be a valuable tool in recogniz-
ing seedborne fungal pathogens. More tradi-
tional and nontraditional approaches to disease
and insect protection were discussed by
Michael I. Haverty, Patrick J. Shea, William F.
Johns, and Richard C. Reardon. Although the
methods are promising, most such techniques can
only be used in high-value areas such as seed
orchards or superior seed stands. As we in-
crease our seed crops annually through manipu-
lations we can expect an increase in disease and
insect problems; improved protection techniques
are badly needed.

A major value of this meeting, in addition to
the sharing of current information and experi-
ences, is that it brought the researchers and
practitioners together. For the nurserymen and
nurserywomen I would suggest that some of the
problems raised by your papers and questions
have already been answered by earlier research.
It would appear to me that we still have an
information gap. For the scientists in the
audience I would like to once again state that
your job does not end with the publication of a
scientific paper but must include at least an
attempt to translate the science into useful
and, if possible, practical methods for the
field user. To encourage such activities we
need to in the future, as we did here, bring
the field user groups and the scientists to-
gether in common meetings. I believe we all
found the meeting of use and I look forward to
the published proceedings.

Stanley L. Krugman is Director, Timber Management
Research, Forest Service, U.S. Department of
Agriculture, Washington, DC.

282

ATTENDANCE ROSTER

Aitken, Karen
USDA Forest Service
2480 Carson Rd .
Placerville CA 95667

Bodmer, Steven E.
Department of State Lands
2705 Spurgin Road
Missoula MT 59801

Alcorn, Cindy
Plum Creek Timber Co.
P 0 Box 188
Pablo MT 59855

Boes, Teresa
1415 N. 46th, #8
Lincoln NE 68503

Alden, John N.
USDA Forest Service
308 Tanana Drive
Fairbanks AK 99701

Bolander, Bruce
P O Box 165A
Bennet NE 68317

Andersen, Vic
Plum Creek Timber Co.
P 0 Box 1957
Kalispell MT 59901

Bollenbacher , Barry L.
Swan Lake Ranger District
P 0 Box 370
Bigfork MT 59911

Anderson, Howard G.

Alberta Energy and Natural Resources

1176 Switzer Dr.

Hinton Alberta TOE IBO

Bower, Ralph
MacMillan Bloedel, Ltd,
65 Front Street
Nanaimo B.C. V9 R 5H9

Arno, Steve

Intermountain Forestry Sciences Lab
University of Montana
Missoula MT 59807

Bramlett, David
Southeastern For. Exp. Stn,
Rt. 1, Box 182A
Dry Branch GA 31020

Attalla, Nabil S.
Bureau of Land Management
1980 Russell Rd
Merlin OR 97532

Brewer, J. Wayne
Dept. of Entomology
Leon Johnson Hall, MSU
Bozeman MT 59717

Au, Peter N.

Pine Ridge Forest Nursery

P 0 Box 750

Smoky Lake Alberta TOA 3CO

Brown, Kevin
Dept of Forest Science
University of Alberta
Edmonton Alberta T6E 2C9

Barth, Richard S.
Wallace Ranger District
USDA Forest Service
Silverton ID 83867

Bush, Robin

Institute of Forest Genetics
2480 Carson Rd.
Placerville CA 95667

Bewley, Derek

Biology Dept--2500 University Dr NW
University of Calgary
Calgary Alberta T2N 1N4

Campbell, Dale

Glacier View Ranger District

P 0 Box W

Columbia Falls MT 59912

Blair, Roger
Potlatch Corporation
P 0 Box 1016
Lewiston ID 83501

Castle, Linda

Institute of Forest Genetics
2480 Carson Rd
Placerville CA 95667

283

Chatterton, Cleve
Coeur d'Alene Nursery
3600 Nursery Rd .
Coeur d'Alene ID 83814

Donnelly, Steve
Payette National Forest
P 0 Box 1026
McCall ID 83638

Coffen, Dale
USDA Forest Service
1221 S. Main
Moscow ID 83843

Du, Wei

Department of Forestry
University of Idaho
Moscow ID 83843

Coles, Barry

Glacier View Ranger District
P 0 Box W

Columbia Falls MT 59912

Dumroese, R. Kasten
8058 North Main
Moscow ID 83843

Comfort, Jeff

Champion International Corporation
P 0 Box 8

Milltown MT 59851

Edwards, George W.
Pacific Forestry Centre
506 W Burnside Rd.
Victoria B.C.

Connor, Jerry

Colville Tribal Forestry-Nursery
P O Box 328
Nespelem WA 99155

Edwards, John

Proctor & Gamble Cellulose

P 0 Bag 1020

Grande Prairie Alberta T8V 3A9

Corda, Al

Kootenai National Forest
RR3, Box 700
Libby MT 59923

Elliott, Dennis E.
USDA Forest Service
P 0 Box 308
Kamiah ID 83536

Corse, Tom
P 0 Box 235
Ronan MT 59864

Erickson, James R.
Colville Tribal Forestry
P O Box 328
Nespelem WA 99155

Daniels, Jess D.
Daniels & Associates
611 E Street
Centralia WA 98531

Fandrich, Helmut
Fandrich, Inc.
P O Box 8000
Surnas WA 98295

Dewey, Jerald
6750 Driftwood Lane
Missoula MT 59801

Fay, Suzanne

USDA FS Coeur d'Alene Nursery

3600 Nursery Rd.

Coeur d'Alene ID 83814

Dexter, Barry N.

Champion International Corp.

P 0 Box V-X

Libby MT 59923

Fins, Lauren

Department of Forest Resources
University of Idaho
Moscow ID 83843

Dolezal, Sharon
P 0 Box 493
Plains MT 59859

Flannigan, Jim
Yaak Ranger District
Route 1

Troy MT 59935

Dombrosky, Scott A.
348A NW 15th
Corvallis OR 97330

Frank, Gerald C
2328 East D Street
Moscow ID 83843

284

Friedman, Sharon
Ochoco National Forest
P O Box 490
Prineville OR 97754
(503) 447-6247

Gansel, Charles

USFS Dorena Tree Improvement Center
P 0 Box 38

Cottage Grove OR 97424

Garner, C. F.

Mobay Chemical Corporation

P 0 Box 4913

Kansas City MO 64120

Gibbs, Carter B
4363 Taylor Avenue
Ogden UT 84403

Glen, David L.
P 0 Box 123

Moyie Springs ID 83845

Gottfried, Gerald J.
Rocky Mtn For & Rge Exp Stn — FSL
Northern Arizona University
Flagstaff AZ 86001

Graham, Russell T.
1221 South Main
Moscow ID 83843

Graham, Russell L.
Boise Cascade Corporation
90 South 21st Street
Elgin OR 97827
(503) 437-2611, ext. 270

Haas, Fred

USDA Forest Service

Red River Ranger Station

Elk City ID 83525

Hallman, Richard
3125 Eldora Lane
Missoula MT 59801

Halvorson, Curtis

US Fish & Wildlife Svc-Ayleswor th Hall
Colorado State University
Fort Collins CO 80525

Hamilton, Ron
USDA Forest Service
324 — 25th Street
Ogden UT 84401

Hardin, Ed

Director, Seed Testing Lab
Farm Crops Annex, O.S.U.
Corvallis OR 97331

Harper, David
Nez Perce National Forest
Route 2, Box 475
Grangeville ID 83530

Haverty, Michael

Pac SW For & Rge Exp Stn

P O Box 245

Berkeley CA 94549

Hayes, Steven

Champion International Corporation
P 0 Box 8

Milltown MT 59851

Heidmann, LeRoy J.
Rocky Mtn For & Rge Exp Stn
Northern Arizona University
Flagstaff AZ 86001

Helium, A. K.
Forest Science/U of Alberta
817 General Services Bldg.
Edmonton Alberta T6G 2H1

Hicks, Donald T.
Bureau of Land Management
3040 Biddle Rd .
Medford OR 97504

Howe, George E.
USDA Forest Service
P O Box 7669
Missoula MT 59807

Hoyle, Janet Sarah
Institute of Forest Genetics
2480 Carson Rd .
Placerville CA 95667

Hutchins, Harry
P O Box 686
Ganado AZ 86505

Jacobsen, Glenn
Payette National Forest
P 0 Box 1026
McCall ID 83638

Jeffers, Richard M
USDA Forest Service
P 0 Box 25127
Lakewood CO 80225

285

Jenkins, Michael
Department of Forestry-UMC 52
Utah State University
Logan UT 84322

Johns, William F
2780 Schuler Lane
Dillon MT 59725

Johnson, David R.
Institute of Forest Genetics
2480 Carson Rd.
Placerville CA 95667

Joy, John

Deerlodge National Forest
P 0 Box 400
Butte MT 59703

Joyce, Dennis

5041 Old Pullman Rd.

Moscow ID 83843

Kilpatrick, Kran

Rt. 1, Box 527

Trout Lake WA 98650

Koester, Harvey J.

USDI Bureau of Land Management

3040 Biddle Rd.

Medford OR 97501

Konishi, Jenji
B.C. Ministry of Forests
1120 Sluggett Rd.
Brentwood BC VOS 1A0

Krebill, Richard
Intermountain Research Station
P 0 Box 8089
Missoula MT 59807

Krugman, Stanley

Director, USDA Forest Service

P 0 Box 2417

Washington D.C. 20013

Labrie, Lorrie

Champion International Corporation
P 0 Box 8
Milltown MT 59851

Lanner, Ronald M.
Utah State University
Dept. of Forest Resources
Logan UT 84322

Lawyer, David A.
Lawyer Nursery, Inc.
950 Hwy 200 West
Plains MT 59859

Leadem, C. L.
B. C. Ministry of Forests
4300 North Road
Victoria B.C. V87 5J3

Lee, Robert M.
34935 Northernwood
Brownsville OR 97327

Lerandeau, Henry
Nez Perce National Forest
Rt. 2, Box 475
Grangeville ID 83530

Liane, Tony

Champion International Corporation
P O Box 8

Milltown MT 59851

Liengsiri, Chaiyasit
Dept. of Forest Science
University of Alberta
Edmonton Alberta T6G 2G5

Luibrand, Jon

Champion International Corporation
P. O. Box 8
Milltown MT 59851

Lukes, Andy

Champion International Corporation
P. 0. Box 8
Milltown MT 59851

Maahs, Wayne

3480 Timber Edge Drive

Missoula MT 59802

Magnuson, William B
424 Woodworth Avenue
Missoula MT 59801

McCaughey, Ward W.
Forestry Sciences Lab
Montana State University
Bozeman MT 59717

McGuire, Gerald
Rexford Ranger District
P. O. Box 666
Eureka MT 59917

286

McLuskie, Cliff

P. 0. Box 491

Thompson Falls MT 59873

Meagher, Michael D.
Canadian Forestry Service
506 W. Burnside Rd .
Victoria B.C. V8Z 1M5

Menkens, Randy

Tally Lake Ranger District

1335 Hwy 93 W

Whitefish MT 59937

Miller, Dan
P 0 Box 1016
Lewiston ID 83501

Miller, Gordon E.
Pacific Forestry Centre
506 West Burnside Road
Victoria B.C.

Miller, Richard G.
486 Bay Green Ct.
Arnold MD 21012

Mitchell, Leslie Ann
Pacific Forestry Centre
506 West Burnside Road
Victoria B.C. V8Z 1M5

Myers, Joseph

Coeur d'Alene Nursery/USDA

3600 Nursery Rd.

Coeur d'Alene ID 83814

Northcutt, Lee
4235 Sundown Rd.
Missoula MT 59801

Owens, John N.

Biology Dept-Univ of Victoria

P O Box 1700

Victoria B.C. V8W 2Y2

Pakulak, James A.
Pakulak Seed Nursery Co.
4293 W. Hansen
Ludington MI 49431

Pakulak, Nancy
Pakulak Seed Nursery Co.
4293 W. Hansen
Ludington MI 49431

Parent, Dennis R.
Inland Empire Paper Co.
N. 3320 Argonne Rd.
Spokane WA 99212

Parker, Mel

Champion International Nursery
P 0 Box 631
Libby MT 59923

Payne, Mark G

Champion International Corporation
P O Box V-X
Libby MT 59923

Pietroburgo, Stephen
BIA-Colville Agency
P O Box 111
Nespelem WA 99155

Pitcher, Alcine
A & D Seeds
Moose Creek Rd .
Polebridge MT 59928

Poriz, Jaines A.
Fording Coal Ltd.
P O Box 100

Elkford B.C. VOB 1H0

Quigley, Steve

Champion International Corporation
Star Route
Marion MT 59925

Rappaport, Nancy
PSW For & Rge Exp Stn
1960 Addison St
Berkeley CA 94704

Reardon, Richard
180 Canfield St
Morgantown WV 26505

Reedy, Terry
P 0 Box 440
Plains MT 59859

Reedy, Verna

Champion International Corporation
P O Box 939
Plains MT 59859

Reno, Jim
Weyerhaeuser Co.
P 0 Box 420
Centralia WA 98531

287

Roberts, Warren
USDA Forest Service
P 0 Box 147
Kalispell MT 59901

Rogers, Robert
Plum Creek Timber Co.
P 0 Box 1499
Newport WA 99156

Ross, Stephen D.
Research Laboratory
B. C. Ministry of Forests
Victoria B.C.

Rowley, Allen S.
USDA Forest Service
Rt. 2, Box 475
Grangeville ID 83530

Rozell, Eric
Plum Creek Timber Co.
P 0 Box 1957
Kalispell MT 59901

Rust, Marc
College of FWR
University of Idaho
Moscow ID 83843

Sandquist, Roger
319 SW Pine Street
P 0 Box 3623
Portland OR 97208

Schaefer, Richard M.
Potlatch Corporation
P 0 Box 1016
Lewiston ID 83501

Schmidt, Jack

USDA Intermountain Research Station
P O Box 8089
Missoula MT 59807

Schmidt, Wyman C.
Forestry Sciences Laboratory
Montana State University
Bozernan MT 59717

Schoeppach, Bill

Idaho Panhandle National Forest

HC, Box 1

Avery ID 83802

Schooley, Hugh 0.

Patawawa National Forestry Institute

Dept. of Environment

Chalk River Ontario KOJ 1JO

Schowalter, T. D.
Entomology Department
Oregon State University
Corvallis OR 97331

Seeley, Chuck

Champion International Corporation
P 0 Box 8

Milltown MT 59851

Shea, Patrick J.
P 0 Box 245
Berkeley CA 94701

Shearer, Raymond C.
500 West Crestline Drive
Missoula MT 59803

Sherman, Frank

Champion International Corporation
P 0 Box 8

Milltown MT 59851

Spencer, Oscar

Bureau of Indian Affairs

P O Box 111

Nespelem WA 99155

Stanger, Mark
Colville Tribal Forestry
P O Box 328
Nespelem WA 99155

Stanger, Lavell O. (Pete)
USDA Forest Service — S & PF, R 0
P 0 Box 3623
Portland OR 97208

Stipe, Lawrence
USDA Forest Service
P O Box 7669
Missoula MT 59807

Stutts, Roger A.
Institute of Forest Genetics
2480 Carson Rd .
Placerville CA 95667

Sutherland, J.
Canadian Forestry Service
506 W. Burnside
Victoria B.C.

Szumera, Lech
Plum Creek Timber Co.
700 South Avenue West
Missoula MT 59801

288

Ta.iaka, Yasuomi
Weyerhaeuser Corporation
534 N. Tower
Centralia WA 98531

White, Jeanne Pedro
USDA Forest Service
P 0 Box 120
Pierce ID 83546

Theoe, Donald R.

WA Department of Natural Resources
6425 Meridian Rd. SE
Olyrnpia WA 98503

White, Monte
USDA Forest Service
P 0 Box 120
Pierce ID 83546

Theroux, Leon

USDA Intermountain Research Stn.
P 0 Box 8089
Missoula MT 59807

Wilson, George

Glacier View Ranger District

P O Box W

Columbia Falls MT 59912

VanDenburg, James H.
Flathead National Forest
P 0 Box 147
Kalispell MT 59901

Winter, Al

Plus Tree Improvement

P 0 Box 5067

Coeur d'Alene ID 83814

Wagner, Michael R.
School of Forestry, N. A. U
P 0 Box 4098
Flagstaff AZ 86011

Wolfe, Warren D.
Creative Sales, Inc,
P 0 Box 501
Fremont NE 68025

Ward, Roger A.
Nez Perce National Forest
Rt. 2, Box 475
Grangeville ID 83530

Woods, John

Champion International Corporation
P O Box 8

Milltown MT 59851

Weaver, T.
Botany Department
Montana State University
Bozeman MT 59717

Yakimchuk, Katherine A.
Pine Ridge Forest Nursery
P O Box 750

Smoky Lake Alberta TOA 3C0

Weber, John C.

Dept of Forest Science — 0. S. U.
3015 Jefferson Way
Corvallis OR 97330

Ying, Cheng C.

Research Branch, BC Ministry/Forests
1450 Government Street
Victoria B.C. V8W 3E7

Wells, Pat
USDA Forest Service
1221 S. Main
Moscow ID 83843

Zack, Art

Idaho Panhandle National
1201 Ironwood Drive
Coeur d'Alene ID 83814

Forest

Wenny, David
College of Forestry
University of Idaho
Moscow ID 83843

i'U.S. GOVERNMENT PRINTING OFFICE 1986-678-039/20&41

289

Shearer, Raymond C, compiler. Proceedings — conifer tree
seed in the Inland Mountain West symposium; 1985 August
5-6; Missoula, MT. General Technical Report INT-203.
Ogden, UT: U.S. Department of Agriculture, Forest Service,
Intermountain Research Station; 1986. 289 p.

Includes six reviews and 34 papers presented in four
technical sessions: cone and seed biology; cone prediction,
collection, and processing; seed orchard and seed production
area management; and effects of vertebrates, insects, and
diseases on seed production. Current information is
presented on conifer tree cones and seeds native to the
Inland Mountain West (east slopes of the coastal ranges to
the plains of the United States and Canada) .

KEYWORDS: conifers, cone and seed biology, seed orchards,
seed production area, cone and seed damage

PESTICIDE PRECAUTIONARY STATEMENT

This publication reports research involving pesticides.
It does not contain recommendations for their use, nor
does it imply that the uses discussed here have been
registered. All uses of pesticides must be registered by
appropriate State and/or Federal agencies before they
can be recommended.

CAUTION: Pesticides can be injurious to humans,
domestic animals, desirable plants, and fish or other
wildlife— if they are not handled or applied properly.
Use all pesticides selectively and carefully. Follow
recommended practices for the disposal of surplus
pesticides and pesticide containers.

The use of trade, firm, or corporation names in this
paper is for information and convenience of the read-
er. Such use does not constitute an official endorse-
ment or approval by the U.S. Department of
Agriculture of any product or service to the exclusion
of others that may be suitable.

INTERMOUNTAIN RESEARCH STATION

The Intermountain Research Station provides scientific knowl-
edge and technology to improve management, protection, and use
of the forests and rangelands of the Intermountain West. Research
is designed to meet the needs of National Forest managers,
Federal and State agencies, industry, academic institutions, public
and private organizations, and individuals. Results of research are
made available through publications, symposia, workshops, training
sessions, and personal contacts.

The Intermountain Research Station territory includes Montana,
Idaho, Utah, Nevada, and western Wyoming. Eighty-five percent of
the lands in the Station area, about 231 million acres, are classified
as forest or rangeland. They include grasslands, deserts, shrub-
lands, alpine areas, and forests. They provide fiber for forest in-
dustries, minerals and fossil fuels for energy and industrial develop-
ment, water for domestic and industrial consumption, forage for
livestock and wildlife, and recreation opportunities for millions of
visitors.

Several Station units conduct research in additional western
States, or have missions that are national or international in scope.
Station laboratories are located in:

Boise, Idaho

Bozeman, Montana (in cooperation with Montana State
University)

Logan, Utah (in cooperation with Utah State University)

Missoula, Montana (in cooperation with the University of
Montana)

Moscow, Idaho (in cooperation with the University of Idaho)
Ogden, Utah

Provo, Utah (in cooperation with Brigham Young University)

Reno, Nevada (in cooperation with the University of Nevada)