Content uploaded by Katharina Liepe
Author content
All content in this area was uploaded by Katharina Liepe on Jul 06, 2021
Content may be subject to copyright.
Content uploaded by Heike Liesebach
Author content
All content in this area was uploaded by Heike Liesebach on May 12, 2021
Content may be subject to copyright.
84
DOI:10.2478/sg-2021-0007
edited by the Thünen Institute of Forest Genetics
Heike Liesebach1, Katharina Liepe1 and Cornelia Bäucker2
Abstract
New rst and 1.5 generation seed orchards are to be created in
Germany based on recently assembled breeding populations
of Acer pseudoplatanus, Larix sp., Picea abies, Pinus sylvestris,
Pseudotsuga menziesii, and Quercus sp. To justify the high
expenses in time and cost for orchard establishment and main-
tenance, planning should make use of consolidated know-
ledge and experience of both the national and international
scientic community. Here, we briey describe advances in
genetic gains achieved through tree breeding, and resume
population genetic aspects and design considerations to draw
conclusions for clonal composition and spatial design of the
new orchards.
We conclude that to avoid outbreeding depression separ-
ate orchards are required for each breeding zone. The zones
are species-specic and dened by ecological and climatic
aspects. A minimum of 60-80 clones per orchard is recom-
mended for native tree species with high proportions of natur-
al regeneration in forest practice. This would allow future selec-
tive thinning based on estimated breeding values from
progeny testing. It would also permit the transfer of seed
orchard progenies into a naturally regenerating forest stands
without the risk of a genetic bottleneck. Lower clone numbers
are appropriate for non-native species and hybrids. It is import-
ant to strictly avoid inbreeding depression, achieved by using
only one clone per progeny or population, from which the plus
trees were selected. Further, the spatial layout should promote
random mating by optimizing the neighbourhood of each
clone. With all of these considerations taken into account, we
expect superior quality traits and at least 10-15 % more volume
from the new seed orchards.
Towards new seed orchard designs in Germany
– A review
1 Thuenen Institute of Forest Genetics, Sieker Landstrasse 2, 22927 Grosshansdorf, Germany
2 Thuenen Institute of Forest Genetics, Eberswalder Chaussee 3A, 15377 Waldsieversdorf, Germany
* Corresponding author: Heike Liesebach, E-mail: heike.liesebach@thuenen.de
Keywords: Forest tree breeding, breeding population, plus trees,
relatedness, inbreeding depression, outbreeding depression, hete-
rosis, seed orchard design, clone number
Motivation
In forestry, natural regeneration is not always possible, or is not
sucient, so that planting with improved forest reproductive
material is required. For that purpose, seed orchards and
approved seed stands can produce seed for articial regenera-
tion and aorestation. In comparison to seed stands, seed
orchards are time- and cost-intensive due to the necessary pro-
cedures for their establishment and maintenance. The higher
expenses for seed orchards, therefore, have to be justied not
only by a good ecological stability and health of the tree popu-
lations originating from seed orchards’ seeds, but simultane-
ously by higher genetic gains, i.e., a superior value in timber
quality and/or growth performance. Usually, this will be achie-
ved by the selection of so-called plus trees, their vegetative
propagation by graftings or cuttings, and the planting of these
ramets to form synthetic populations, which allow mutual pol-
lination.
The German Forest Tree Breeding Strategy was a result of
in-depth discussions between experts from federal and state
forest research institutes including international contributors.
It was formulated and published in 2013 (Liesebach M et al.
2013). “The important goal of breeding is to provide reproduc-
tive material that is adaptable and powerful enough to meet
expected environmental changes in performing all forest func-
tions. … Experiences show that … signicantly increased mass
and value performance can be expected through breeding.”
Liesebach et al. Silvae Genetica (2021) 70, 84 - 98
85
(cited from the German Forest Tree Breeding Strategy, p. 1).
Thus, it proposes the establishment of healthy, vital and adapt-
able forest tree populations as a precondition for the simulta-
neously desired good growth performance resulting in high
yield. In addition to adaptation and adaptability, a good timber
quality (e.g., stem form) should promote a cascade utilization
of wood. In this regard, new seed orchards producing such
high-quality seed are the anticipated outcome to transfer the
results of forest tree breeding eorts into practice. Species-spe-
cic strategies for each of for the six focus tree species consider
dierent levels of breeding intensity. For Pseudotsuga menziesii
and Larix sp., more intense measures are planned, medium
intensities are provided for Acer pseudoplatanus, Picea abies
and Pinus sylvestris, and lowest intensity is intended for Quercus
sp.. Since 2014, the nationally funded joint research projects
“FitForClim” and “AdaptForClim” were conducted to realize rst
steps of the forest tree breeding strategy. They include the
selection of plus trees, preferably in eld trials, but also in forest
stands, and the establishment of breeding populations. Three
to four breeding zones for each of the six species were devel-
oped and delineated based on climatic, ecological and experi-
mental data from eld trials within Germany, e.g., for Pseudot-
suga menziesii (Liepe and Liesebach M 2017, Liesebach M et al.
2020), for Quercus sp. (Meißner et al. 2020), and for Pinus sylves-
tris (Liesebach M et al. 2020). The plus tree selection was carried
out using a multi-stage procedure that started with a compre-
hensive evaluation of eld trial data at the progeny level, fol-
lowed by an in-depth visual characterization to select individu-
als based on multiple traits (Liesebach M et al. 2020, Meißner et
al. 2020). Selection criteria were an above-average stem form
and growth performance, drought and late frost tolerance, and
an above-average resistance against the most widespread
typical fungal diseases (e. g., Lophodermium seditiosum and
Diploidia sp. in Scots pine, Lachnellula willkommii in larch, Pha-
eocryptopus gaeumannii in the coastal variety of Douglas r
and Verticillium sp. in sycamore maple). In total, about 500-
1000 plus trees per specieswere selected, grafted and planted
in breeding populations with at least 200 per breeding zone.
The established breeding populations are replicated at two
locations. Thus, the two projects prepared the preconditions
for subsequent seed orchards assembled with grafted clones
from each breeding zone.
These new seed orchards should be created in the light of
consolidated knowledge and experience of both the national
and international scientic community. Especially the aspect of
relatedness of seed orchard clones should be considered as
international studies have repeatedly shown disadvantageous
eects of crossbreeding between relatives on the growth per-
formance of seed orchards progenies. Besides superior quality
traits, a range of 10-15 % more volume, at least for the conifer
species, is the most likely we would expect as a result of new
seed orchards assembled from the recently established bree-
ding populations in Germany.
Here, after a short description of the baseline situation, we
resume population genetic aspects and the current relevant
literature with the objective to draw conclusions for the gene-
tic composition and design of new seed orchards in Germany.
Existing seed orchards at the national
and international scale
More than 300 seed orchards already exist in Germany, estab-
lished over many decades, mainly for the species Pinus sylvest-
ris (54), Larix sp. (49), Picea abies (44), Pseudotsuga menziesii
(23), Alnus glutinosa (23), Prunus avium (20), and Tilia cordata
(20). In total 255 seed orchards belong to the category “Quali-
ed” and 50 to the category “Tested” (BLE 2019). They are very
dierent in design and size, dependent on the locally available
material and the desired growing region. All of these orchards
have been established prior to the formulation of the long-
term, continuous and trans-regional breeding strategy in 2013.
Further, neither the breeding populations were consolidated
in a common place until recently nor were breeding values of
selected plus trees determined. In several cases, seed orchard
progenies were tested in eld trials according to the German
forest legislation to evaluate them in comparison with ocially
recommended standards and progenies from approved seed
stands. These tests showed that progenies of existing German
seed orchards partially reveal superior quality traits. The ave-
rage performance across seed orchard progenies, however, is
more or less comparable with the average across progenies
from approved seed stands and the well-growing standards.
The latter being populations, approved for their performance
by an expert advisory board and included in eld trials for com-
parative purposes (gGA 2019). Top performing progenies are
derived from seed orchards as well as from approved seed
stands (e.g., Hüller et al. 1995, Stephan and Liesebach M 1996,
Grotehusmann 1998, Rau 1998, Rau and Schulzke 2001,
Schneck 2001, Grotehusmann 2014a, b). Up until now, it has
not been possible to compare and evaluate dierent designs in
the existing seed orchards (number and origin of clones, selec-
tion criteria) with the currently available progeny testing data
in Germany (V Schneck, personal communication) because of
highly variable preconditions.
At the international level, several countries have develo-
ped long-term and continuous breeding strategies for a num-
ber of economically important tree species. Breeding histories
are reviewed, e.g., for the Scandinavian countries (Picea abies,
Pinus sylvestris, Betula pendula by Jansson et al. 2017); France
(Pinus pinaster by Merzeau et al. 2005); North America (e.g.,
Pinus taeda, Pinus elliottii by Li et al. 1999) or Australia and New
Zealand (Pinus radiata by Burdon et al. 2008, Wu et al. 2008).
Seed orchards are always the means to realize the so-called
genetic gain as the result of breeding eorts. In this regard,
genetic gain is quantied as the increase in average genetic
value between two consecutive generations, being measured
by the change in population means. Genetic gain varies from
trait to trait and is greater if the selection is very intensive (only
the very best individuals are selected) and the trait is under
strong genetic control, i.e., has a high heritability.
Noteworthy timber volume gains of rst-generation seed
orchards have been achieved, for example, for Pinus elliottii in
the US (+ 10 %, Vergara et al. 2004); Pinus radiata in New Zeal-
and (+ 15 %, Carson et al. 1999); Pinus sylvestris in Finland (+ 15
86
%, Ahtikoski et al. 2012) or Picea abies in dierent Scandinavian
countries (+ 9-15 %), but also for broadleaved species such as
Alnus glutinosa (+ 10 %) and Betula pendula (+ 10-15 %) in Fin-
land and Sweden (Haapanen et al. 2015).
Consecutive breeding cycles followed after the initial
selection of plus trees, building on the established breeding
populations and the estimation of breeding values by intensi-
ve progeny testing. It is generally expected that genetic gain
improves with each generation of tree improvement compa-
red to the unimproved material, while the magnitude of impro-
vement decreases from one generation to the next. Even
though growth traits have moderate to low heritabilities (h2 =
0.10-0.30), they were originally and are still the main focus in
many breeding programs (White et al. 2007). For the Swedish
Scots pine breeding program, for example, Rosvall et al. (2001)
summarize achieved gains of 10 % volume increase per unit
area for seed from rst-generation orchards, 10-25 % (the latter
for intense selections from a large number of eld-tested plus
trees) from the second generation and expect 20-25 % from
third generation orchards established at the beginning of the
Twenty-rst Century. For the next generation, currently under
establishment, they predict even up to 35 % gain.
A strict economic viewpoint on seed orchards is common
practice to increase yield and value of future forest stands to
meet the rising demand of forest products. Tree breeding does,
however, not only focus on growth traits. Quality traits such as
wood density, stem straightness and branchiness are frequent
breeding objectives, e.g., for European and hybrid larch (Klein-
schmit 1988), Scots pine (Kohlstock and Schneck 1992), Nor-
way spruce (Hansen and Roulund 1997) or Sitka spruce (Hann-
rup et al. 2004). In particular with climate change, adaptive
traits (e.g., phenology, drought stress resistance, frost resis-
tance) have become a new priority (Bastien et al. 2013, Cham-
bel et al. 2013, Pâques 2013b, Ivetić et al. 2016). Increasing
forest health by breeding for tolerance against pests and disea-
ses has recently also been considered as a way to prepare for
future challenges (Pâques 2013a, Haapanen et al. 2015). Taking
climate change into account, German (Liesebach M et al. 2013)
and Norwegian breeding programs (Edvardsen et al. 2010)
have stated the objective to improve growth without a loss in
wood quality and maintain genetic variation to sustain the
adaptive potential of future forests.
From the retrospective point of view, MacLachlan et al.
(2017) evaluated whether selective breeding (with the objec-
tive of an increased growth) did compromise the adaptive
potential of Pinus contorta in Western Canada. Comparing
seedlings from seed orchards and natural populations in a
common garden they quantied the eect of selection on
adaptive traits and whether these aect adaptive pheno-
type-climate associations. According to their conclusions,
selection, breeding and progeny testing have produced taller
seedlings that were not adaptively compromised relative to
their natural seedling counterparts (MacLachlan et al. 2017).
This gives selectively bred reproductive material a positive tes-
timonial in light of climate change.
Population genetic background
Comparison of natural vs. articial regeneration
Remarkable dierences between natural and articial regene-
ration exist in forest tree populations (reviewed, e. g., by Eriks-
son 1998, Hattemer and Ziehe 2018). Good professional practi-
ce can take care of the use of appropriate forest reproductive
material and for the application of adequate timing and plan-
ting techniques. The genetic composition of the young forest
generation, however, is always changed after articial regene-
ration compared to natural processes, whereby some aspects
must be considered.
In the natural regeneration of tree populations, a large
amount of seed may be produced over long periods of time
and several reproduction events cover annually uctuating
conditions. In contrast, articial regeneration is usually based
on a single reproduction event under singular conditions. In
addition, the commercial harvest in approved seed stands pre-
fers a seed collection from mothers with above average seed
yields and, therefore, enhances naturally occurring dierences
in reproductive success.
To date, only a few studies exist on the amount of seed
produced by natural forests. Sarvas (1962) analyzed 13 popula-
tions of Pinus sylvestris in Finland with seed traps for a maxi-
mum of 13 years and determined a mean value of 625,000
seeds per ha*year (total amount including lled and empty
seed). Hofgaard (1993) studied a 400-year-old Picea abies natu-
ral stand in the northern boreal zone of Sweden over a nine-
year period. The mean seed rain amounted to 767,000 per
ha*year, but only very little seedling recruitment was observed
due to large predation and unfavorable climatic conditions.
Further, Saksa (2004) reported results of ve unevenly-aged,
spruce-dominated boreal forest stands in Finland. Depending
on the fructication, he found a mean seed production ran-
ging from 11 kg/ha*year up to 28 kg/ha in mast year, which
corresponds to about 4.5 million seeds.
Commercial seed harvests in Germany collect up to 150
kg/ha for Picea abies and 80 kg/ha for Pinus sylvestris (BLE 2017)
in mast years. That means 24 million and 13.6 million seeds per
ha, respectively, to give an idea of the extent of reproduction
excess (the larger seed harvests in Germany in comparison to
Finland may be caused by more favorable growing conditions).
Under natural conditions, such an extremely high number
of seed and seedlings experiences high random losses, but
also a very strong viability selection against (partial) inbree-
ding. Thus, young generations will be drastically decimated by
natural processes. In contrast, the nursery cultivation aims to
achieve low random and selection losses of a given amount of
seed. This can be realized due to more or less optimal growing
conditions avoiding strong competition with irrigation, weed
and pest control and the application of fertilizer. Nevertheless,
there is a weak human- induced selection towards growth and
vitality, because less vigorous seedlings are eliminated. In the
next step of articial regeneration, usually a few thousand
seedlings per hectare are transferred into the forest, which is a
relatively low number of seedlings compared to natural rege-
neration.
87
In the case of articial regeneration, the eorts of foresters
are focused on the phenotypic selection for desired traits of
adult trees at the population or at an individual level (strong
selection). Then, the harvest of forest reproductive material is
restricted to approved seed stands or seed orchards to use
their seeds for the next forest generation. But, once this seed is
available, further selection is largely avoided. At least in the
nursery stage and in the early stand development, forest
practice diminishes selection pressure by competition due to
favorable plant spacing and even supports the desired species
by manual treatments.
Therefore, particular attention is required for the produc-
tion of forest seed to compensate or even overcompensate low
rates of natural selection in the early phase of the ospring
generation in the case of articial regeneration.
Eect of mating on growth and tness of forest
trees
Most trees are outbreeding species that have developed die-
rent mechanisms against selng: dioecism (e. g., Taxus, Popu-
lus, Fraxinus in parts), gametophytic self-incompatibility (in
Rosaceae), the avoidance of contemporary owering of male
and female parts in hermaphrodite owers (protandry, proto-
gyny, e.g., in Juglans, Acer), the separation of female and male
owers in dierent parts of the crown (Picea), or frequent abor-
tion of embryos after selng resulting in empty seeds (in coni-
fers). In contrast to annual plants, due to their longevity trees
have many chances to produce ospring during their life.
Tree populations with continuous natural regeneration
often show typical structures with correlated genetic and spa-
tial distances (e.g., Gonzáles-Martínez et al. 2002 for Pinus pin-
aster, Cavers et al. 2005 for Symphonia globulifera, Eusemann
and Liesebach H 2021 for Quercus petraea). These family struc-
tures result from restricted pollen ow and seed dispersal.
More frequent pollination events between neighboring trees
promote partial inbreeding, which is permanently present in
natural populations. However, remarkable inbreeding depres-
sion leads to a strong selection against inbred ospring in early
stages. As an example, in Scots pine, Danusevičius et al. (2016)
found no more substantial changes in xation index from the
age of 20 years, indicating that the selection process against
inbreeding has nearly nished.
Figure 1 shows a schematic and hypothetical curve sum-
marizing the whole spectrum of possible constellations of
mating within a species. On the x-axis, the possible degree of
relatedness between mating parents ranges from selng (pos-
sible in monoecious species) over mating among relatives of
several degrees to mating of unrelated members within a
population under natural conditions. Mating between indivi-
duals from separated and even long distant populations from
the same species is also possible, which may lead to outbree-
ding depression in case of sexual reproduction or articial cros-
sing of local with transferred material from divergent environ-
mental conditions. The y-axis quanties the magnitude of
growth and tness-related traits.
1. Inbreeding depression
Without using the term “inbreeding depression,” Charles Dar-
win wrote in his book “The Eects of Cross and Self-Fertilization
in the Vegetable Kingdom” (Darwin 1877): “The rst and most
important of the conclusions which may be drawn from the
observations given in this volume, is that cross-fertilization is
generally benecial, and self-fertilization injurious. This is
shown by the dierence in height, weight, constitutional vigor,
and fertility of the ospring from crossed and self-fertilized o-
wers, and in the number of seeds produced by the parent-
plants.” Nowadays, inbreeding depression is dened as redu-
ced tness (survival and fecundity) and/or reduced growth of
progenies derived from selng or mating of relatives compa-
red to the parent(s). It is explained by an accumulation of dele-
terious, partially recessive alleles in mainly outcrossing species
such as trees (Namkoong and Bishir 1987, Williams and Savolai-
nen 1996).
According to a review of Pyhäjärvi et al. (2020) high inbree-
ding depression is very pronounced in Scots pine and many
other conifers with large distribution areas (e.g., Picea abies,
Pseudotsuga menziesii). They assessed the level of inbreeding
depression to be lower in most angiosperms. In a study with
Eucalyptus grandis, however, Hedrick et al. (2016) also obser ved
a high number of deleterious alleles and a strong selection
against homozygosity in a selfed progeny using SNP genoty-
ping of about 10,000 heterozygous genes.
A large body of papers exists describing the negative
eect of inbreeding in numerous tree species after controlled
pollination experiments. Among them are some studies with
specically designed crossing schemes in order to quantify the
eect of several levels of inbreeding. Mostly they include self-
pollination (inbreeding coecient F=0.5), crosses between
full-sibs or between parents and ospring (F=0.25), crosses
between half-sibs (F=0.125) and outcrosses (F=0). Such experi-
ments require a breeding history longer than one generation
to have access to parent trees with known pedigrees.
The following experimental examples prepared several
levels of inbreeding. However, for comparative purposes the
amount of depression is reported here only for an inbreeding
coecient of 0.25. One of the rst experiments with controlled
crosses was started in 1978 with Pinus pinaster in France. At age
11, the inbreeding depression amounted to 5 % for height
growth, 20 % for volume and 37 % for fecundity measured by
number of cones (Durel et al. 1996). In the mid-1980s, Ford et
al. (2014) conducted a similar experiment with Pinus taeda in
the United States. At age 9, reductions for height growth of
6-13 % and for volume of 11-26 % depending on test sites were
found. Stem straightness or fusiform rust disease incidence,
however, were not aected by inbreeding. In Canada, Woods
and Heaman (1989) carried out crossings for Pseudotsuga men-
ziesii in 1987 and reported rst results on the negative eect of
inbreeding on the production of lled seeds. At the nursery
stage, signicant inbreeding eects were found for germinati-
on, seedling survival and growth (Woods et al. 2002). At age 26,
the inbreeding depression of these progenies was nally esti-
mated for height, diameter and volume growth accounting
percentages of 12, 22 and 43 % respectively, while there was
88
nearly no inuence on wood density (Stoehr et al. 2015).
Recently, in 2013 and 2014, a new series of crosses was conduc-
ted in Sweden with Pinus sylvestris creating 9 dierent inbree-
ding levels. As a preliminary result, 48 % of the variance in the
percentage of lled seeds was explained by inbreeding level
(Mullin et al. 2019).
Inbreeding depression has also been reported for the o-
spring of seed orchards. As an example, Pupin et al. (2019) stu-
died eects of inbreeding depression and found serious decre-
ases of 18 % for height, 26 % for diameter and 20 % for survival
in Eucalyptus urophylla at an inbreeding level of 0.25. Doerksen
et al. (2014) estimated the reduction of Picea glauca height
growth to approximate 6 % for every 0.1 increase in inbreeding
coecient.
Studies regarding natural populations often compare
parameters of dierent ontogenetic stages. A few research
papers report higher mean heterozygosities and outcrossing
rates in adults compared to lower levels of heterozygosity and
outcrossing rates in seeds or seedlings, e.g., in Pinus leucoder-
mis (Morgante et al. 1993), in the neotropical tree Platypodium
elegans (Huord and Hamrick 2003) or in Picea jezoensis (but
not in Abies sachalinensis) by Okada et al. (2015). The authors
discussed their results in relation to viability selection against
inbred ospring in natural populations.
2. Outbreeding depression
“An outbreeding depression occurs when fecundity and/or via-
bility decline following an intraspecic hybridization. Coadap-
tation provides an explanation for outbreeding depression
based upon the tness interactions between genes.” (cited
from Templeton 1986). Outbreeding depression in the narrow
sense is dened as a reduction in tness of progenies below
the average of the parents (Edmands 2007). However, there are
dierent points of views, and sometimes outbreeding depres-
sion is considered as a signicant decline in hybrid tness rela-
tive to either parent. Two main mechanisms of outbreeding
depression were referred to by Templeton (1986): The “… dis-
ruption of benecial interactions between loci (intrinsic coad-
aptation or “coadapted gene complexes”), and disruption of
benecial interactions between genes and the environment
(extrinsic or local adaptation)” (cited from Edmands and Tim-
merman 2003). The latter may be very typical for stand-form-
ing tree species with large distribution areas growing under
variable environmental conditions. Here, the adaptive and
evolutionary potential of long-distance pollen ow has to be
mentioned, since single individuals originating from such
events may survive and reach the reproductive stage. In prac-
tice, it seems to be very dicult to develop experimental
designs for tree species to separate the two dierent
Figure 1
Scheme of the whole spectrum of possible constellaons of within-species mangs in relaon to growth
and tness traits.
89
mechanisms resulting in outbreeding depression from each
other. There are not only possible intrinsic genetic eects and
the inuence of the maladapted parent to the ospring perfor-
mance at a given growing site, but also epigenetic eects that
may modify experimental results. Although numerous experi-
mental results of outbreeding depression are given for herba-
ceous plants (e.g., Edmands 2007 or Oakley et al. 2015), the
phenomenon has rarely been studied in tree or shrub species.
Various crossings of Pinus sylvestris clones from divergent
climatic regions were conducted in Germany in the 1970s and
1980s, however in an irregular non-reciprocal design. They
resulted in reduced growth performance of progenies compa-
red to combinations of locally adjacent parents (V Schneck
unpublished). In a full diallel crossing scheme between six dis-
tant populations of Lotus scoparius, fruit set and seedling emer-
gence were inversely related to the genetic distance between
populations measured with isozyme markers (Montalvo and
Ellstrand 2001). In controlled pollination experiments with
neighboring and distant pollen donors, Stacy (2001) found an
optimal distance for maximum fruit set of 1-2 km for Syzygium
rubicundum and 2-10 km for Shorea cordifolia. Smaller and lar-
ger distances between the origin of parents led to a reduced
reproductive tness assigned to biparental inbreeding and
outbreeding depression, respectively. A similar experiment
was done by Forrest et al. (2011) with the shrub Grevillea
mucronulata. They also found the best seed set and seedling
performance for intermediate distance pollination compared
to distant and open pollination treatments. Goto et al. (2011)
carried out controlled reciprocal crossings with local (low-ele-
vation) and non-local (high-elevation) parents of Abies sachali-
nensis and grew them at a low-elevation test site. They obser-
ved a reduced height and diameter growth of 25-year-old
ospring families derived from one or two non-local parents
and considered this as outbreeding depression. In this examp-
le, however, the distance relates to remarkable dierences in
ecological conditions due to elevation rather than to a pure
geographic distance.
3. Heterosis eect
Heterosis (also called as hybrid vigor or outbreeding enhance-
ment) is dened as superior growth or tness of ospring com-
pared to the mid-parent values. This term was mentioned the
rst time by Shull in 1914 in the context of maize breeding
(Shull 1952), and, originally, it was restricted to enhanced
growth after crosses between inbred lines. Meanwhile, the
term heterosis is used in a much broader sense. Considering
tree species, this term is sometimes also used for the positive
eects due to hybridization among closely related species, e.
g., for Larix (Pâques 2009), Liriodendron (Yao et al. 2016) or
Populus (Rood et al. 2017, Zanewich et al. 2018). However, we
regard heterosis as a within-species eect and look for the
maximum of the curve in the scheme of Figure 1. Therefore, we
refer to heterosis as an eect of so-called inter-provenance
hybridization. Recessive or nearly recessive deleterious muta-
tions that have become xed within populations because of
genetic drift may be responsible for heterosis after crosses bet-
ween natural populations (Oakley et al. 2015).
There are some experimental results from controlled cros-
sings, which investigate ospring families for their potential
heterosis. An early attempt was made to exploit the heterosis
eect for forest tree breeding in Sweden in the 1950s. Cont-
rolled pollinations were carried out with several selected clo-
nes of Picea abies derived from Sweden (used as females) and
Germany (as pollen donors). Remarkable dierences between
families were found for stem volume at age 25. A clear superio-
rity of the inter-provenance families and indication for hetero-
sis was observed at one out of three Swedish test sites (Eriks-
son and Ilstedt 1986). The authors explained the missing
heterosis at the two other sites in relation to dierences in
growth rhythm and harsh conditions.
In France, a comprehensive diallel mating scheme was car-
ried out for Pinus pinaster with 10 provenances from three geo-
graphic regions in 1978. The mean value of heterosis for height
growth amounted to 8.4 % at age 13, however, no relationship
between heterosis and stem crookedness or insect resistance
was detected (Harfouche et al. 2000, Harfouche and Kremer
2000). Another full diallel crossing design with natural Eucalyp-
tus globulosa trees in Tasmania was conducted in 1997-1998 to
generate families from within-region and long-distance out-
crossings (Lopez et al. 2003). At age 13, Costa e Silva et al.
(2014) observed signicant heterosis eects for family means
of growth of long-distance crossings for one of the two test
sites.
A second theoretical approach without known pedigrees
is based on the reconstruction of relatedness by the applica-
tion of genetic markers. Doerksen et al. (2014) used an array of
nearly 6000 SNPs to determine genetic relationships in a eld
experiment including intra-provenance and inter-provenance
crosses of Picea glauca. After removing partial inbreeding
eects by detection of weak relatedness in the intra-prove-
nance crosses, a heterosis eect of about 6% for height growth
at age 15 had been found by comparing intra-provenance and
inter-provenance progenies.
Practical aspects for establishing seed
orchards
Number of clones
Many publications in forestry deal with the question of the
adequate clone number for seed orchards, although the topic
itself has never been fully discussed in the literature (Lindgren
and Prescher 2005). This is surprising, given the fact that the
number of parents in a seed orchard is one of the most funda-
mental questions for seed orchard genetics as mentioned by
Funda and El-Kassaby (2012). A universal answer to this questi-
on, however, may not exist, because seeds from seed orchards
are used for dierent purposes, under dierent local condi-
tions and for native as well as for introduced tree species. On
the one hand, seeds produced in seed orchards do only serve
90
into advanced generation seed orchards after the testing of
clones, but also managing seed orchard progenies as perma-
nent forest with natural regeneration.
Moreover, a large number of clones ensures genetic
diversity, which is essential for the long-term survival of
forests under a changing environment (Ivetić et al. 2016). In
this respect, the general tendency that a lower number of clo-
nes in a seed orchard causes a reduced genetic diversity of
the resultant seed crops has been emphasized in some stu-
dies. For example, Sønstebø et al. (2018) investigated the
genetic diversity of seed from two Norway spruce seed
orchards with dierent number of parents (25 and 60) and
compared them with seed from approved seed stands and
natural forests. They found a decrease in allelic richness (-6 %)
and remarkably lower eective population sizes (-71 %) for
the seed originating from the seed orchards compared to the
other studied samples.
Another important aspect for the quality and genetic
diversity of seed from seed orchards of wind-pollinated tree
species is pollen ow from nearby forests, as shown by a large
body of literature (e.g., Adams and Burczyk 2000, Fries et al.
2008, Fernandes et al. 2008, Torimaru et al. 2009, Korecký and
El-Kassaby 2016, Sønstebø et al. 2018). These marker-based
studies estimated a pollen contamination of up to about 50 %
in case of common tree species within their natural range.
Besides the detrimental eect of losses in genetic gain
through pollen contamination (e. g., Di-Giovanni and Kevan
1991), pollen ow from outside has also been regarded to be
of great importance for the genetic diversity of seed produ-
ced by seed orchards with a lower number of clones because
gene ow always increases eective population size (e.g.,
Sønstebø et al. 2018). According to Ingvarsson and Dahlberg
(2019), a higher genetic variation determines tness in natu-
ral populations and the persistence under changing environ-
mental conditions. The current tendency is that the speed of
climatic shifts will surpass forest rotations of 40 to 60 years.
Therefore, we suggest including the aspect of high genetic
diversity into future orchards. To achieve this, we propose to
establish the new seed orchards with clone numbers between
60 and 80 for species native to Germany. These numbers
might be a compromise between feasibility from a practical
point of view and sucient genetic diversity as estimated by
Wojacki et al. (2019). For the non-native species Douglas r,
the number of 40 clones per seed orchard may be sucient.
Compared to the native tree species, this reduced number of
clones must be seen in the context of planned seed imports
from the native range to supplement the genetic diversity, as
outlined in the German breeding strategy. Recently, a mark-
er-based study in older Douglas r seed orchards and their
ospring was carried out in Germany. The results show a bet-
ter transfer of genetic diversity to the ospring compared
with Douglas r seed stands having comparable numbers of
genotypes in the adult population (Pakull et al. unpublished).
Contrary to this suggestion, seed orchards to produce
species hybrids should be assembled dierently since they
represent a special case. Here, the objective is to combine
selected clones of European and Japanese larch in order to
for commercial timber production, where improvement of
genetic gain and consequently shorter rotation periods are the
main foci. Such a strict economic viewpoint of seed orchards is
common practice in countries with a long history in forest tree
breeding. On the other hand, however, seed from seed
orchards might also function as starting point for new forest
tree populations, which should be able to regenerate naturally
without a severe genetic bottleneck and with a low risk of a
reduced adaptability to changing environmental conditions.
Therefore, the number of seed orchard clones highly depends
on the purpose of the resultant seed.
Regarding concrete recommendations for clone numbers
for advanced breeding cycles, very dierent statements and
details can be found in the literature. For example, Libby (1982)
reported mixtures of 2-3 clones as the worst strategy and pro-
posed numbers of 7-25 clones as robust strategy for seed
orchards. Lindgren and El-Kassaby (1989) analyzed the ques-
tion of optimal clone number from the viewpoint of maximiz-
ing genetic gain though genetic thinning. As initial scenario,
the authors used a hypothetical seed orchard established with
50 progeny-tested clones. According to theoretical models of
Bishir and Roberds (1997, 1999), moderate numbers of clones
between 20 and 40 seem to be sucient under general condi-
tions and provide at least equivalent protection against cata-
strophic loss as does a large number of clones. Johnson and
Lipow (2002), moreover, considered that 90 % of isozyme vari-
ation found in natural populations can be achieved with seed
orchards assembled of 20 or more clones under the assump-
tion that clones are used in their native breeding zone. For Bri-
tish Columbia (Canada), it has been suggested that orchard
seed lots for public land reforestation have been suggested to
exceed an eective population size of 10 clones to realize a
minimum level of diversity (Stoehr et al. 2004). Referring to
conifers in Sweden, Lindgren and Prescher (2005) recommend
20 tested clones as a rule of thumb. This number of 20 is a com-
monly accepted minimum margin of unrelated clones being
combined in an orchard that is supposed to serve for approxi-
mately 20 years, before being replaced by an advanced orchard
(e.g. Danusevicius and Lindgren 2002, Lindgren and Prescher
2005).
In contrast to advanced generation seed orchards where
the breeding values of clones are known, recommendations
on the number of clones for rst-generation seed orchards
have seldom been published. For early seed orchards of Pinus
sylvestris in northern Sweden, Andersson et al. (2003) compa-
red growth performance and survival of progenies from 11
orchards each composed of 20-59 clones at the time of proge-
ny testing. They mentioned that some of these orchards were
later extended leading to clone numbers ranging from 20 to
112 in the year 2003. Recently, a number greater than hundred
has also been reported by Yang et al. (2020) designing a rst-
generation seed orchard of P.sylvestris var. mongolica with 108
clones in total. Sønstebø et al. (2018) mentioned 60-200 untes-
ted and unrelated plus trees in Norwegian Norway spruce seed
orchards belonging to the rst generation. Such high numbers
of clones denitely create more options and greater exibility
for later changes and decisions. This may include the transfer
91
breed seed with a high percentage of hybrids, which are supe-
rior to both parental species (hybrid vigor or heterosis eect in
its broadest sense). For this purpose, seed are generally only
harvested from one parental species, while the other is only
the pollen donor and represented by a larger number of clo-
nes. Crucial for successful hybridization is a synchronous ow-
ering time of the parental genotypes. Extreme, but very suc-
cessful in terms of hybrid percentage and performance of the
ospring, is the design proposed by Langner combining one
single clone of Larix decidua with more than 200 clones of Larix
kaempferi as pollen donors (Langner and Schneck 1998). In this
case, inbred seed due to selng of the mother clone are either
empty or highly mortal in the seed bed.
Relatedness
The avoidance of genetic relatedness or co-ancestry among
individuals is an important aspect in forest tree breeding and
seed production in clonal seed orchards. Since the 1960s, the
topic is under discussion mainly in North America (e. g., Pinus
taeda, Pinus elliottii), in Scandinavian countries (Picea abies,
Pinus sylvestris) and New Zealand (Pinus radiata). Opposite
eects would arise between an enrichment of desired growth-
enhancing alleles and inbreeding depression in continued
breeding cycles. Theoretical backgrounds were considered,
e.g., by Namkoong (1966) and Burrows (1970).
Since the 1970s, several eorts were made always with the
objective to avoid co-ancestry among individuals in the pro-
duction seed orchards (Zobel et al. 1972, Talbert 1979, McKe-
and and Beineke 1980). In this context, ideas have been deve-
loped to keep breeding populations unrelated, like nested
polycross designs, disconnected diallels, breeding groups, or
sublining. “Selections within a breeding group are unrelated to
selections in any other breeding group in order to allow com-
plete avoidance of inbreeding in seed from wind-pollinated
clonal seed orchards established with 1 clone from each group”
as stated by White et al. (1993) for the example of Pinus elliottii.
With the increasing establishment of advanced generati-
on seed orchards, tree breeders aimed to nd the optimal
balance between maximization of genetic gain and mainte-
nance of genetic diversity through control of relatedness. In
the 1990s, some models from animal breeding (Wray and God-
dard 1994) were adapted to forestry, and Lindgren and Mullin
(1997) were among the rst, who implemented an advanced
generation tree breeding strategy considering a compromise
between genetic gain and an acceptable level of relatedness.
They proposed to take a weighting factor into account consi-
dering breeding values and penalties for co-ancestry of the
considered clones. An example is given by Olsson et al. (2001),
who reported the successful application in Pinus taeda bree-
ding. They compensate a certain amount of relatedness with a
larger number of clones in order to obtain an additional gene-
tic gain. Danusevičius and Lindgren (2002), further, investiga-
ted genetic gain by using results of experiments on Norway
spruce and calculated various parameters like diversity loss per
breeding cycle or costs per plant in $ for dierent scenarios of
three breeding strategies (phenotype, clone, progeny
strategy). Danusevičius and Lindgren (2008), moreover, com-
pared several selection models for candidate clones from unre-
lated half-sib families with known breeding values and with a
predicted level of inbreeding depression. In this comprehensi-
ve simulation study, they found a maximized genetic gain for a
given gene diversity in case of an optimal proportion deploy-
ment strategy, which is more ecient at higher levels of rela-
tedness in the candidate population. Besides, Lindgren et al.
(2009), focusing on dierent selection strategies such as trun-
cation selection and linear deployment, reported a clear supe-
riority of the latter approach concerning genetic gain. Interes-
tingly, the linear deployment strategy, which uses clones in
dierent proportions according to their breeding value (Lindg-
ren and Matheson 1986), outperformed the classical trunca-
tion approach under the conditions of both no relatives and
allowing related clones. Linear deployment had also been
emphasized by Danusevičius and Lindgren (2008) to be an
applicable method for seed orchard designs in advanced bree-
ding cycles. Further, several computer-based methods simula-
ting selection over generations were developed in order to
support an optimal clone selection in tree breeding. The most
recent is the program “OPSEL” introduced by Mullin (2014,
2017). The focus of “OPSEL” is to maximize the genetic value
through the selection of certain cohorts, while a constraint on
acceptable genetic diversity has to be maintained. The pro-
gram can manage balanced and unbalanced numbers of
ramets per clone (Yamashita et al. 2018).
The search for optimal strategies and compromises consi-
dering genetic gain as well as genetic diversity is still under dis-
cussion. In contrast to citations above, Silva et al. (2018) recom-
mend the restriction to one tree per family in Eucalyptus
urophylla breeding, accepting a slight decrease (0.9-1.5 %) in
genetic gain to insure genetic variation for next generations.
Regarding the new seed orchards in Germany, the related-
ness of clones selected from a progeny of a eld test or within
a population is unknown, but a potential relatedness of clones
may exist. The existence of distinct family structures, especially
half-siblings, in a standard provenance eld trial was shown in
a case study with European beech (Liesebach H et al. 2015).
Since the vast majority of plus trees was selected in prove-
nance eld trials, we strongly recommend avoiding arrange-
ments where several clones from one progeny or one popula-
tion are planted together in an orchard. Thus, the new seed
orchards should be assembled using only one clone per prog-
eny or population, if the desired number of clones can be
achieved in this way. In cases with low available numbers of
progenies or populations, from which the plus trees were
selected, several seed orchards with non-overlapping clonal
compositions could be established to avoid crossbreeding of
potentially related clones. To establish new forest stands, the
resulting seed lots should then be combined to achieve a larg-
er genetic variation. Furthermore, we suggest composing the
orchards with clones in equal proportions because information
on breeding values, which might oer options for linear
deployment, does not exist yet.
92
Spatial design of seed orchards
Seed orchard designs aim to promote many dierent cross-
pollinations by using a repeated and randomized layout of clo-
nes. Since the beginning of forest tree breeding, a wide variety
of spatial designs of seed orchards have been developed, and
the availability of computational tools (Tab.1), as well as the
development of molecular methods opened up new perspec-
tives regarding the layouts. In practice, therefore, clones had
been planted according to numerous dierent seed orchard
layouts (SOL), which makes direct comparisons between seed
orchards very complex. Moreover, layouts are seldom publis-
hed (but see D’Amico et al. 2019) and only some studies report
specic details about the arrangement and spacing of clones
and their ramets, respectively (e.g., Torimaru et al. 2013).
Further, there is still an ongoing discussion on the topic of the
best layout for seed orchards; even though, an all-in-one solu-
tion may not exist. As Lstibůrek and El-Kassaby (2008) concre-
tely mentioned “neither design will be optimal in every situati-
on due to year-to-year variability in reproductive output”. Also,
circumstances under which seed orchards are established vary
from case to case, so that the decision for a layout has to be
based on the consideration of dierent aspects: biology of the
tree species, for example monoecious/dioecious or wind-/
insect-pollinated (e.g., Van Buijtenen 1975), environmental
conditions like size and shape of the orchard, wind direction by
which the pollen cloud might be transferred, as well as number
of selected clones, number of ramets per clone (balanced/
unbalanced), and the presence of related clones as sib-mating
or parent-ospring mating in advanced generations.
Historically, tree breeders focused primarily on the rando-
mization of clones during the initial phase of a seed orchard’s
establishment. Giertych (1975) gives an excellent overview
about these early strategies by comparing 14 dierent approa-
ches, which include, for example, the Randomized Complete
Block design as realized in a Prosopis alba seed orchard in
Argentina (D’Amico et al. 2019) or the Systematic design as
used for the arrangement of ramets in a seed orchard of P.men-
ziesii in western Oregon (Slavov et al. 2005). Systematic and
simple designs may be also useful if the clones need to be
identied for use in controlled crossings. Further, some of the-
se early designs already explicitly considered both the pollen
ow by wind and the aspect of panmixia in the seed orchard
like the Directional Cyclic Balanced Incomplete-Block (Free-
man 1967) or the Permutated Neighborhood Design (PND) by
La Bastide (1967). In particular, the computerized approach of
Permutated Neighborhood has to be emphasized here
because it provided the basis for later developments of soft-
ware codes. One of those later programs was introduced by
Bell and Fletcher (1978) with the name Computer Organized
Orchard Layouts (COOL) and used for the establishment of
some prominent seed orchards existing in British Columbia
(Canada). A well-studied example is the second-generation
seed orchard of P.menziesii established in 1990 (Lai et al. 2010,
Kess and El-Kassaby 2015, Korecký and El-Kassaby 2016, Song
et al. 2018), another one is a rst-generation orchard of Larix
occidentalis from 1989 (Funda et al. 2008). In the following
years, the COOL approach was further developed under the
name of Permutated Neighborhood Seed Orchard Design by
Chakravarty and Bagchi (1993, 1994). But, neither the original
COOL software code nor its enhanced version were designed
to avoid inbreeding or to incorporate relatedness among clo-
nes. These features, however, were already considered by Van-
clay (1991), who developed the computer program Seed
Orchard Designer (SOD). This software was explicitly intended
for second generation seed orchards with a high proportion of
related clones.
Due to the accumulating evidence that any kind of rela-
tedness of clones compromises the long-term genetic gain of
seed orchards (see chapter “Inbreeding depression”), the mini-
mization of inbreeding eects became more and more the
focus in forest tree breeding during the rst decade of the new
century. In 2010, the Minimum-inbreeding seed orchard
design (MI), which calculates spatial distributions among indi-
vidual trees as a function of the extent of their genetic related-
ness was published by Lstibůrek and El-Kassaby (2010). Still
today, the MI-design is superior to other approaches for alloca-
ting ramets of related clones or ramets of the same clone with
the greatest possible distance along the orchard’s grid (Cha-
loupková et al. 2016). The software can be used for various
seed orchard scenarios such as balanced/unbalanced numbers
of ramets per clone, including situations where clones are
strongly underrepresented; dierent deployment strategies
like linear or proportional deployment, or a mixture of genetic
relatedness between clones, for example presence of both
half-sib clones and parent-ospring mating. The MI-code, fur-
thermore, is also suitable for large and complex advanced
generations through its subsequent improvement to an exten-
ded global algorithm (MI-EGA) by Lstibůrek et al. (2015). Paral-
lel to the development of the MI-design, another software pro-
gram, the Randomized, Replicated, Staggered Clonal-Row
design (R2SCR), has also been introduced (Lstibůrek and El-Kas-
saby 2010, El-Kassaby et al. 2014). The R2SCR-design overco-
mes diculties arising from severe correlated mating in the
classical clonal-row design (El-Kassaby 2003, El-Kassaby et al.
2007) because the staggering and randomization of rows pro-
motes outcrossing among clones. The big advantage of an
arrangement of clones in rows is the facilitation of individual
clone management during harvest, pollination and pest con-
trol. The software, moreover, is exible and can handle geneti-
cally related clones via exclusion zones, empty positions along
the orchards’ grid, as well as non-plantable spots. However, the
R2SCR-design is primarily intended for seed orchards of coni-
fers with severe inbreeding depression after selng and poly-
embryony (El-Kassaby et al. 2014). Recently, another algorithm
with the name Improved Adaptive Parallel Genetic Algorithm
(IAPGA) was demonstrated by Yang et al. (2020) using microsa-
tellite data of P. sylvestris var. mongolica. The IAPGA approach,
which was originally developed by Wang et al. (2018), uses the
genetic distance among clones to minimize the eect of
inbreeding depression through improved spatial clonal
deployment. Hence, the aim of IAPGA is maximum spatial dis-
tance between clones of closer genetic relationship to main-
tain genetic diversity in the next generation.
93
Soware Reference and code availability Soware features
COOL
Computer
Organized
Orchard Layout
Bell and Fletcher (1978),
code on request from authors (in Fortran)
The code is based on the Permutated Neighborhood Design (PND) of La Basde
(1967). It is applicable for both regular and irregular types of SOL. Planng places
are rectangularly arranged and dierent degrees of separaon between the clones
(design types) are possible. Up to 100 clones with dierent numbers of ramets can
be handled, but no consideraon of clone relatedness.
Permutated
Neighbourhood
Seed Orchard
Design
Chakravarty and Bagchi (1993, 1994),
code on request from authors
The code is based on the PND (La Basde 1967). It can be used for rectangular,
triangular or hexagonal plots. Up to 1000 clones with an equal number of ramets
can be handled. Ramets per clone are separated by an inner and outer isolaon
ring. No consideraon of clone relatedness.
SOD
Seed Orchard
Designer
Vanclay (1991),
code on request from author (in Fortran 77)
The code was designed for 2nd generaon SOL. It creates PNDs and can handle
high proporon of related clones, which are separated by maximum distance. The
program is exible and informaon on owering or biology (dioecy) can also be
incorporated.
MI
Minimum-
inbreeding seed
orchard design
Lsbůrek and El-Kassaby (2010),
calculaons possible in cooperaon with Lsbůrek
The algorithm is suitable for SOL of rst and advanced generaons with a rectan-
gular arrangement of planng places. Various degrees of relatedness and dierent
numbers of ramets per clone are possible. Inbreeding and mang among relaves
are avoided by maximizing the distance between ramets of a clone and related
clones, respecvely.
R2SCR
Randomized,
Replicated,
Staggered Clonal-
Row Design
El-Kassaby et al. (2014),
calculaons possible in cooperaon with El-Kassaby
The algorithm is suitable for all generaon types of seed orchards with a rectangu-
lar arrangement of planng places. The design facilitates a selecve seed harvest
by clone because several ramets of one clone are systemacally planted in a row (=
clonal-row). The impact of selng and correlated mangs is minimized by stagge-
ring and maximal separaon of clonal-rows of the same clone and related clones,
respecvely.
MI-EGA
Minimum-
inbreeding seed
orchard design-
extended global
algorithm
Lsbůrek et al. (2015),
code included in supplementary material (in R)
The code is based on the MI-approach and has been explicitly developed for large
and complex advanced generaon seed orchards, seed orchards designs, e.g. SOL
with 900 planng places. The procedure includes: 1) subdividing the orchard’s
grid into independent blocks and using the original MI-algorithm for each block
independently, 2) rotaon and merging of blocks using a random operator. Various
numbers of ramets per clone and dierent degrees of relatedness are possible.
ONA
Opmum
Neighbourhood
Algorithm
Chaloupková et al. (2016),
free download at:
hps://katedry.czu.cz/en/kgd/soware (in R)
The algorithm is suitable for all generaon types of seed orchards with a rectangu-
lar arrangement of planng places. The design enhances panmixia (random mang)
by opmizaon of the clone’s’ neighborhood. This is achieved through a reduced
variance in the number of direct neighbors of a clone. It is the most appropriate
design, when a clone has to be surrounded by as many dierent neighbors as
possible.
IAPGA
Improved
Adapve Parallel
Genec
Algorithm
Wang et al. (2018), Yang et al. (2020),
formula and soware tools included in publicaons
The approach is suitable for all generaon types of seed orchards. The method is
genec distance-dependent and uses molecular data (SSR) to achieve an opmal
clone deployment in a seed orchard by maximizing spaal distance between clones
of closer genec relaonship. Applicable for rectangular layouts, where each clone
is surrounded by eight neighbors.
Further, an alternative approach to ensure high genetic
diversity of seed orchards’ progenies has been suggested by
Chaloupková et al. (2016) under the name Optimum Neighbor-
hood Algorithm (ONA). Compared to previous software tools,
which concentrated on maximum distance of ramets of the
same or related clones, the focus of ONA is the promotion of
panmixia. This is achieved by optimization of the neighbor-
hood of each clone and their ramets, respectively. This means
that each ramet is surrounded by a more or less equal number
of diering clones; hence, the variance of direct neighbor
counts is minimized and achieves zero when all surrounding
counts are equal. The pollen cloud, therefore, can be expected
to reach the greatest level of optimal mixing, which promotes
panmixia. In this regard, the ONA is signicantly superior to
other software tools, such as MI or R2SCR (Chaloupková et al.
2016). The algorithm, furthermore, can handle variable clonal
sizes, irregular plots, relatedness, assortative mating, and it is
applicable for the upgrading of existing seed orchards, when
new parents are introduced to replace failed clones or those of
lower quality (Chaloupková et al. 2019). Moreover, the code for
ONA is freely available (Tab.1), and an additional ONA version
has been developed for MATLAB (M. Lstibůrek, personal com-
munication).
Tab. 1
Overview of the most relevant soware programs for the design of seed orchard layouts (SOL).
94
potential inbreeding depression in the seed orchard pro-
genies. In cases of multiple selection of plus trees from one
progeny in a eld test or from one population, a potential
relatedness of selected clones has to be assumed. For most
species considered here, there is no need for a combination
of clones, which might be potential relatives, regarding the
high number of selected plus trees for the respective bree-
ding zones and the number of planned seed orchards. Alter-
natively, admixtures of seed lots from smaller seed orchards
with non-overlapping clonal composition could be produ-
ced.
4. The clones should be used in equal proportions of
ramets because breeding values giving information about
the best clones to incorporate other approaches, e.g., linear
deployment, are not available at the moment. Hence, the
major focus of the new round of seed orchards will be ran-
dom mating (i.e., panmixia). We recommend the software
ONA developed by Chaloupková et al. (2016), which is cur-
rently the best method to promote panmixia in seed orchards
to create optimal spatial layouts for planting the seed
orchards.
5. Our future work in relation to seed orchards should be
focused on the estimation of breeding values of single clo-
nes by testing their ospring families. These breeding values
will help to plan genetic thinnings of the new seed orchards,
if the spacing of plants is not too wide, and to design the
genetic composition of an advanced generation of orchards
in the forthcoming. Further, comprehensive eld tests of pro-
genies should be conducted to achieve the category “Tested”
for superior seed orchards. Simultaneously, these tests should
compare the ospring from the new orchards with already
existing ones to draw inferences concerning their design
from practical experience in addition to the stated theoretical
considerations. Last, but not least, the expected new experi-
ences will lead to an updated breeding strategy.
Acknowledgments
This review was produced in the framework of two national
joint research projects “FitForClim” and “AdaptForClim” fun-
ded by the German Federal Ministry of Food and Agriculture
and the Federal Ministry of the Environment, Nature Conser-
vation and Nuclear Safety (Waldklimafonds, Grant numbers
22WB400704 and 22WB415204, administrated by the Agen-
cy for Renewable Resources and the Federal Oce for Agri-
culture and Food). For helpful discussions we thank our colle-
agues Volker Schneck and Mirko Liesebach and all project
partners. We do much appreciate the valuable review of Dar-
ius Danusevičius (Alexandras Stulginskis University, Lithua-
nia) and Doris Krabel (Technical University Dresden, Germa-
ny), which further improved the manuscript.
Conclusions for new seed orchards in
Germany
Currently, new rst-generation clonal seed orchards are
planned in Germany. For tree species, where plus trees of the
recently installed breeding populations were selected
predominantly in eld trials, the planned seed orchards could
be considered as the 1.5 generation. This is reasonable, since
plus trees were selected for their superiority under comparab-
le growing conditions, although their breeding values are so
far unknown. The following criteria should be considered for
the establishment of new seed orchards:
1. All seed orchards should consist of clone collections suitable
for one of the three or four species specic breeding zones to
avoid outbreeding depression. Germany is located partially
in the Atlantic climate and partially in the transition to the con-
tinental climate; as a result, species-specic reactions to die-
rent environmental conditions are not uniform. Therefore, the-
se zones were delineated considering climatic and ecological
data and, if possible, veried by retrospective analysis of exis-
ting transplant experiments.
2. Generally, a number of 40 clones per seed orchard is recom-
mended in the current German regularities for forest reproduc-
tive material. However, the number of clones to be used in the
new seed orchards should consider species-specic aspects.
We recommend a number of 60-80 clones for the native tree
species Pinus sylvestris, Picea abies, Larix decidua, Acer pseudo-
platanus and Quercus sp.. Such numbers of clones are feasible
and will keep options open, e.g., for future selective thinning or
to transfer seed orchard progenies into a naturally regenera-
ting forest stand without the risk of a genetic bottleneck. At
present, we cannot expect a better growth performance of
seed orchard progenies by reducing clone numbers with a
stronger selection because selection criteria (i.e., breeding
values) are not yet available. Further, it is currently not possible
to create a conclusive performance ranking among the selec-
ted plus trees as they originated from stands and trials, which
include various trial series with dierent source collections and
sites of divergent growing conditions. For the non-native spe-
cies Douglas r, the number of 40 clones per seed orchard may
be sucient. In case of hybrid larch, the objective is to produce
seed with a high percentage of high-performance hybrids bet-
ween selected European and Japanese clones. This should be
realized by harvesting seeds from one individual, but replica-
ted, clone of the maternal species, while the other species
functions as the pollen donor and may be represented by a lar-
ger number of clones. A natural regeneration of the planted
hybrid progeny is out of question.
3. With regard to the assembly of each single orchard, a very
important point is to use only one clone per progeny or per
population existing in the clonal archives. This procedure will
ensure the promotion of the heterosis eect in the sense of
inter-provenance hybridizations and a strict avoidance of
95
D’Amico I, Vilardi JC, Saidman BO, Ewens M, Bessega C (2019) Pollen contamina-
tion and mating patterns in a Prosopis alba clonal orchard: impact on seed
orchards establishment. iForest - Biogeosciences and Forestry 12(3):330-
337. https://dx.doi.org/10.3832/ifor2936-012
Danusevičius D, Lindgren D (2002) Eciency of selection based on phenotype,
clone and progeny testing in long-term breeding. Silvae Genetica 5(1):19–
26
Danusevičius D, Lindgren D (2008) Strategies for optimal deployment of related
clones into seed orchards. Silvae Genetica 57(1-6):119.
https://doi.org/10.1515/sg-2008-0018
Danusevicius D, Kerpauskaite V, Kavaliauskas D, Fussi B, Konnert M, Baliuckas V
(2016) The eect of tending and commercial thinning on the genetic diver-
sity of Scots pine stands. European Journal of Forest Research 135(6):1159-
1174. https://dx.doi.org/10.1007/s10342-016-1002-7
Darwin C (1877) The eects of cross and self fertilisation in the vegetable king-
dom. London: D. Appleton, 486 p
Di-Giovanni F, Kevan PG (1991) Factors aecting pollen dynamics and its impor-
tance to pollen contamination: a review. Canadian Journal of Forest Re-
search 21(8):1155-1170. https://dx.doi.org/10.1139/x91-163
Doerksen TK, Bousquet J, Beaulieu J (2014) Inbreeding depression in intra-prov-
enance crosses driven by founder relatedness in white spruce. Tree Genet-
ics & Genomes 10(1):203-212.
https://dx.doi.org/10.1007/s11295-013-0676-y
Durel CE, Bertin P, Kremer A (1996) Relationship between inbreeding depression
and inbreeding coecient in maritime pine (Pinus pinaster). Theoretical and
Applied Genetics 92(3):347-356. https://dx.doi.org/10.1007/BF00223678
Edmands S (2007) Between a rock and a hard place: evaluating the relative risks
of inbreeding and outbreeding for conservation and management. Molecu-
lar Ecology 16(3):463-475. https://doi.org/10.1111/j.1365-294x.2006.03148.x
Edmands S, Timmerman CC (2003) Modeling factors aecting the severity of
outbreeding depression. Conservation Biology 17(3):883-892.
https://dx.doi.org/10.1046/j.1523-1739.2003.02026.x
Edvardsen O, Steenrem A, Johnskås O, Johnsen Ø, Myking T, Kvaalen H (2010)
Strategi for Skogplanteforedling 2010-2040 (høringsutkast). Hamar, Nor-
way, 27 p
El-Kassaby YA (2003) Clonal-row vs. random seed orchard designs: mating pat-
tern and seed yield of western hemlock (Tsuga heterophylla (Raf.) Sarg.). For-
est Genetics 10:121-127
El-Kassaby YA, Fayed M, Klápště J, Lstibůrek M (2014) Randomized, replicated,
staggered clonal-row (R2SCR) seed orchard design. Tree Genetics & Ge-
nomes 10(3):555-563. https://dx.doi.org/10.1007/s11295-014-0703-7
El-Kassaby YA, Stoehr MU, Reid D, Walsh CG, Lee TE (2007) Clonal-row versus ran-
dom seed orchard designs: interior spruce mating system evaluation. Cana-
dian Journal of Forest Research 37(3):690-696.
https://dx.doi.org/10.1139/X06-248
Eriksson G (1998) Evolutionary forces inuencing variation among populations
of Pinus sylvestris. Silva Fennica 32:173-184.
https://doi.org/10.14214/sf.694
Eriksson G, Ilstedt B (1986) Stem volume of intraand interprovenance families
of Picea abies (L.) Karst. Scandinavian Journal of Forest Research 1(1-4):141-
152. https://doi.org/10.1080/02827588609382407
Eusemann P, Liesebach H (2021) Small-scale genetic structure and mating pat-
terns in an extensive sessile oak forest (Quercus petraea (MATT.) LIEBL.).
Ecology and Evolution, in press
https://dx.doi.org/10.002/ece3.7613
Fernandes L, Rocheta M, Cordeiro J, Pereira S, Gerber S, Oliveira MM, Ribeiro MM
(2008) Genetic variation, mating patterns and gene ow in a Pinus pinaster
Aiton clonal seed orchard. Annals of Forest Science 65(7):706-706. https://
dx.doi.org/10.1051/forest:2008049
Ford GA, McKeand SE, Jett JB, Isik F (2014) Eects of Inbreeding on Growth and
Quality Traits in Loblolly Pine. Forest Science 61(3):579-585.
https://dx.doi.org/10.5849/forsci.13-185
Forrest CN, Ottewell KM, Whelan RJ, Ayre DJ (2011) Tests for inbreeding and out-
breeding depression and estimation of population dierentiation in the
bird-pollinated shrub Grevillea mucronulata. Annals of Botany 108(1):185-
195. https://dx.doi.org/10.1093/aob/mcr100
References
Adams WT, Burczyk J (2000) Magnitude and implications of gene ow in gene
conservation reserves. In: Young A, D Boshier and T Boyle (eds). Forest con-
servation genetics: Principles and practice. Victoria, Australia: CSIRO Pub-
lishing, Collingwood, pp 215-244
https://doi.org/10.1079/9780851995045.0215
Ahtikoski A, Ojansuu R, Haapanen M, Hynynen J, Kärkkäinen K (2012) Financial
performance of using genetically improved regeneration material of Scots
pine (Pinus sylvestris L.) in Finland. New Forests 43(3):335-348.
https://dx.doi.org/10.1007/s11056-011-9284-6
Andersson B, Elfving B, Ericsson T, Persson T, Gregorsson B (2003) Performance
of Improved Pinus sylvestris in Northern Sweden. Scandinavian Journal of
Forest Research 18(3):199-206.
https://dx.doi.org/10.1080/02827581.2003.9728290
Bastien J-C, Sanchez L, Michaud D (2013) Douglas-Fir (Pseudotsuga menziesii
(Mirb.) Franco). In: Pâques LE (ed) Forest Tree Breeding in Europe: Current
State-of-the-Art and Perspectives. Dordrecht © Springer Science+Business
Media, pp 325-369. https://doi.org/10.1007/978-94-007-6146-9_7
Bell GD, Fletcher AM (1978) Computer organised orchard layouts (COOL) based
on the permutated neighbourhood design concept. Silvae Genetica
27(6):223-225
Bishir J, Roberds JH (1997) Limit theorems and a general framework for risk anal-
ysis in clonal forestry. Mathematical Biosciences 142(1):1-11.
https://doi.org/10.1016/s0025-5564(96)00184-8
Bishir J, Roberds JH (1999) On numbers of clones needed for managing risks in
clonal forestry. Forest Genetics 6(3):149-155
BLE (2017) Forstliches Vermehrungsgut - Informationen für die Praxis. 9. Aua-
ge. Bonn, Germany: BLE (Bundesanstalt für Landwirtschaft und Ernährung),
72 p, ISBN 978-3-8308-1291-3
BLE (2019) Übersicht über zugelassenes Ausgangsmaterial für forstliches Ver-
mehrungsgut in der Bundesrepublik Deutschland (Stand: 01.07.2019) [onli-
ne]. Available from https://www.ble.de/DE/Themen/Wald-Holz/Forstli-
ches-Vermehrungsgut/forstliches-vermehrungsgut_node.html [cited
August 8, 2020]
Burdon RD, Carson MJ, Shelbourne CJA (2008) Achievements in forest tree ge-
netic improvement in Australia and New Zealand 10: Pinus radiata in New
Zealand. Australian Forestry 71(4):263-279.
https://dx.doi.org/10.1080/00049158.2008.10675045
Burrows P (1970) Coancestry control in forest tree breeding plans. Proceedings
of the Second Meeting of the Working Group on Quantitative Genetics, Sec-
tion 22 IUFRO, August 18-19, 1969, Raleigh, North Carolina: 27-36
Carson SD, Garcia O, Hayes JD (1999) Realized Gain and Prediction of Yield with
Genetically Improved Pinus radiata in New Zealand. Forest Science
45(2):186-200. https://dx.doi.org/10.1093/forestscience/45.2.186
Cavers S, Degen B, Caron H, Lemes MR, Margis R, Salgueiro F, Lowe AJ (2005) Op-
timal sampling strategy for estimation of spatial genetic structure in tree
populations. Heredity 95(281):289. https://doi.org/10.1038/sj.hdy.6800709
Chakravarty GN, Bagchi SK (1993) A computer program for permutated neigh-
bourhood seed orchard design. Silvae Genetica 42:1-5
Chakravarty GN, Bagchi SK (1994) Short Note: Enhancement of the computer
program of the permutated neighbourhood seed orchard design. Silvae
Genetica 43(2):177-178
Chaloupková K, Stejskal J, El-Kassaby YA, Frampton J, Lstibůrek M (2019) Current
advances in seed orchard layouts: Two Case Studies in Conifers. Forests
10(2):1-6. https://dx.doi.org/10.3390/f10020093
Chaloupková K, Stejskal J, El-Kassaby YA, Lstibůrek M (2016) Optimum neighbor-
hood seed orchard design. Tree Genetics & Genomes 12(6):105.
https://dx.doi.org/10.1007/s11295-016-1067-y
Chambel MR, Climent J, Pichot C, Ducci F (2013) Mediterranean Pines (Pinus
halepensis Mill. and brutia Ten.). In: Pâques LE (ed) Forest Tree Breeding in
Europe: Current State-of-the-Art and Perspectives. Dordrecht © Springer
Science+Business Media, pp 229-265.
https://doi.org/10.1007/978-94-007-6146-9_5
Costa e Silva J, Potts BM, Lopez GA (2014) Heterosis may result in selection fa-
vouring the products of long-distance pollen dispersal in Eucalyptus. PLoS
ONE 9(4):e93811-e93811. https://dx.doi.org/10.1371/journal.pone.0093811
96
Huord KM, Hamrick JL (2003) Viability selection at three early life stages of the
tropical tree, Platypodium elegans (Fabaceae, Papilonideae). Evolution
57(3):518-526. https://doi.org/10.1111/j.0014-3820.2003.tb01543.x
Hüller W, Svolba J, Kleinschmit J (1995) Entwicklung von Kiefernplantagenab-
saaten in Niedersachsen. Forst und Holz 50:142-144
Ingvarsson PK, Dahlberg H (2019) The eects of clonal forestry on genetic diver-
sity in wild and domesticated stands of forest trees. Scandinavian Journal of
Forest Research 34(5):370-379.
https://dx.doi.org/10.1080/02827581.2018.1469665
Ivetić V, Devetaković J, Nonić M, Stanković D, Šijačić-Nikolić M (2016) Genetic di-
versity and forest reproductive material - from seed source selection to
planting. iForest - Biogeosciences and Forestry 9(5):801-812.
https://dx.doi.org/10.3832ifor1577-009
Jansson G, Hansen JK, Haapanen M, Kvaalen H, Steenrem A (2017) The genetic
and economic gains from forest tree breeding programmes in Scandinavia
and Finland. Scandinavian Journal of Forest Research 32(4):273-286.
https://dx.doi.org/10.1080/02827581.2016.1242770
Johnson R, Lipow S (2002) Compatibility of breeding for increased wood pro-
duction and longterm sustainability: the genetic variation of seed orchard
seed and associated risks. Proceedings of the Wood compatibility initiative
workshop Number 18, 169-179
Kess T, El-Kassaby YA (2015) Estimates of pollen contamination and selng in a
coastal Douglas-r seed orchard. Scandinavian Journal of Forest Research
30(4):266-275. https://dx.doi.org/10.1080/02827581.2015.1012112
Kleinschmit J (1988) Konsequenzen aus den Lärchen-Herkunftsversuchen für
die Lärchen-Züchtung. Forst und Holz 43(11):259-262
Kohlstock N, Schneck H (1992) Scots pine breeding (Pinus sylvestris L.) at
Waldsieversdorf and its impact on pine management in the Northeastern
German Lowland. Silvae Genetica 41(3):174-180
Korecký J, El-Kassaby YA (2016) Pollination dynamics variation in a Douglas-r
seed orchard as revealed by microsatellite analysis. Silva Fennica 50:1-12
La Bastide JGA (1967) A computer program for the layouts of seed orchards. Eu-
phytica 16(3):321-323. https://dx.doi.org/10.1007/BF00028937
Lai BSK, Funda T, Liewlaksaneeyanawin C, Klápště J, Van Niejenhuis A, Cook C,
Stoehr MU, Woods JH, El-Kassaby YA (2010) Pollination dynamics in a Doug-
las-r seed orchard as revealed by pedigree reconstruction. Annals of Forest
Science 67(8):808. https://doi.org/10.1051/forest/2010044
Langner W, Schneck V (1998) Ein Beitrag zur Züchtung von Hybridlärchen (Larix
x eurolepis HENRY). Frankfurt am Main: Sauerländer’s Verlag, 159 p
Li B, McKeand S, Weir R (1999) Tree improvement and sustainable forestry–im-
pact of two cycles of loblolly pine breeding in the USA. Forest Genetics
6(4):229-234
Libby WJ (1982) What is a safe number of clones per plantation? Proceedings of
the Genetics of host-parasite interactions in forestry, 14-21 September
1980, Wageningen, The Netherlands: 342-360
Liepe KJ, Liesebach M (2017) Verwendungszonen für Vermehrungsgut von Dou-
glasie auf Basis von Klimadaten und Herkunftsversuchen. Proceedings of
the Hochwertiges Forstvermehrungsgut im Klimawandel - Symposium des
Verbundprojektes FitForClim vom 14. bis 15. Juni 2016 in Chorin, Chorin,
Germany: 39-54
Liesebach H, Eusemann P, Liesebach M (2015) Verwandtschaftsbeziehungen in-
nerhalb von Prüfgliedern in Herkunftsversuchen – Beispiel Buche (Fagus syl-
vatica L.). Forstarchiv 86(6):174-182.
https://dx.doi.org/10.4432/0300-4112-86-174
Liesebach M, Degen B, Grotehusmann H, Janßen A, Konnert M, Rau H-M, Schir-
mer R, Schneck D, Schneck V, Steiner W, Wolf H (2013) Strategie zur mittel-
und langfristigen Versorgung mit hochwertigem forstlichem Vermehrungs-
gut durch Züchtung in Deutschland. ISSN 2196-2324. Braunschweig,
Germany: Thünen Institute, 69 p, ISBN 978-3-86576-107-1
Liesebach M, Liesebach H, Schneck V, Bäucker C, Eusemann P, Heimpold C, Liepe
KJ, Pakull B, Rieckmann C, Schröder J, Wojacki J (2020) Bereitstellung von
leistungsfähigem und hochwertigem Forstvermehrungsgut für den klima-
und standortgerechten Wald der Zukunft (FitForClim) –Teilprojekt 4.
Großhansdorf, Germany, 66 p
Lindgren D, Danusevicius D, Rosvall O (2009) Unequal deployment of clones to
seed orchards by considering genetic gain, relatedness and gene diversity.
Forestry: An International Journal of Forest Research 82(1):17-28.
https://dx.doi.org/10.1093/forestry/cpn033
Freeman GH (1967) The use of cyclic balanced incomplete block designs for di-
rectional seed orchards. Biometrics:761-778.
https://doi.org/10.2307/2528427
Fries A, Lindgren D, Andersson B (2008) The Swedish Scots pine seed orchard
Västerhus: a study of linear deployment. Proceedings of the Seed Orchards
Conference, 26-28 September, 2007, Umeǻ, Sweden: 70-78
Funda T, Chen CC, Liewlaksaneeyanawin C, Kenawy AMA, El-Kassaby YA (2008)
Pedigree and mating system analyses in a western larch (Larix occidentalis
Nutt.) experimental population. Annals of Forest Science 65(7):705. https://
dx.doi.org/10.1051/forest:2008055
Funda T, El-Kassaby YA (2012) Seed orchard genetics. CAB Reviews 7(013):1-23.
https://doi.org/10.1079/pavsnnr20127013
Giertych M (1975) Seed orchard designs. In: Faulkner R (ed) Seed orchards. A
joint production by specialist members of the International Union of Forest
Research Organization’s Working Party on seed orchards (S2.03.3). London,
UK: Forestry Commission, pp 25–37, ISBN 0-11-710146-X
gGA [gemeinsamer Gutachterausschuss] (2019) Forstvermehrungsgutrecht:
Empfehlungen des gemeinsamen Gutachterausschusses (gGA) der Länder
für dessen Umsetzung. 86 p.
https://www.ble.de/DE/Themen/Wald-Holz/Forstliches-Vermehrungsgut/
Gonzáles-Martínez SC, Gerber S, Cervera M-T, Martínez-Zapater J, Gil L, Alía R
(2002) Seed gene ow and ne-scale structure in a Mediterranean pine (Pi-
nus pinaster Ait.) using nuclear microsatellite markers. TAG Theoretical and
Applied Genetics V104(8):1290-1297.
https://dx.doi.org/10.1007/s00122-002-0894-4
Goto S, Iijima H, Ogawa H, Ohya K (2011) Outbreeding depression caused by in-
traspecic hybridization between local and nonlocal genotypes in Abies
sachalinensis. Restoration Ecology 19(2):243-250.
https://dx.doi.org/10.1111/j.1526-100X.2009.00568.x
Grotehusmann H (1998) Geprüftes Vermehrungsgut aus Kiefern- und Er-
len-Samenplantagen. AFZ-Der Wald 53:240-242
Grotehusmann H (2014a) Prüfung von 25 jährigen Absaaten aus Kiefern-Samen-
plantagen. Landbauforschung - Applied Agricultural and Forestry Research
64(2):107-118. https://dx.doi.org/10.3220/LBF_2014_107-118
Grotehusmann H (2014b) Prüfung von Fichten-Samenplantagen. AFZ-Der Wald
69(5):6-9
Haapanen M, Jansson G, Nielsen UB, Steenrem A, Stener L-G (2015) The status
of tree breeding and its potential for improving biomass production: a re-
view of breeding activities and genetic gains in Scandinavia and Finland.
Uppsala, Sweden: SkogForsk, 56 p
Hannrup B, Cahalan C, Chantre G, Grabner M, Karlsson B, Bayon IL, Jones GL,
Müller U, Pereira H, Rodrigues JC, Rosner S, Rozenberg P, Wilhelmsson L,
Wimmer R (2004) Genetic parameters of growth and wood quality traits in
Picea abies. Scandinavian Journal of Forest Research 19(1):14-29.
https://dx.doi.org/10.1080/02827580310019536
Hansen J, Roulund H (1997) Genetic parameters for spiral grain, stem Form, Pilo-
dyn and growth in 13 years old clones of Sitka spruce (Picea sitchensis
(BONG.) CARR.). Silvae Genetica 46(2):107-112
Harfouche A, Bahrman N, Baradat P, Guyon JP, Petit RJ, Kremer A (2000) Prove-
nance hybridization in a diallel mating scheme of maritime pine (Pinus pin-
aster). II. Heterosis. Canadian Journal of Forest Research 30(1):10-16.
https://dx.doi.org/10.1139/x99-179
Harfouche A, Kremer A (2000) Provenance hybridization in a diallel mating
scheme of maritime pine (Pinus pinaster). I. Means and variance compo-
nents. Canadian Journal of Forest Research 30(1):1-9.
https://dx.doi.org/10.1139/x99-178
Hattemer HH, Ziehe M (2018) Erhaltung forstgenetischer Ressourcen - Grundla-
gen und Beispiele. Göttingen University Press, ISBN 978-3-86395-362-1.
https://doi.org/10.17875/gup2018-1094
Hedrick PW, Hellsten U, Grattapaglia D (2016) Examining the cause of high in-
breeding depression: analysis of whole-genome sequence data in 28 selfed
progeny of Eucalyptus grandis. New Phytologist 209(2):600-611.
https://dx.doi.org/10.1111/nph.13639
Hofgaard A (1993) Seed rain quantity and quality, 1984–1992, in a high altitude
old-growth spruce forest, northern Sweden. New Phytologist 125(3):635-
640. https://dx.doi.org/10.1111/j.1469-8137.1993.tb03913.x
97
Lindgren D, El-Kassaby YA (1989) Genetic consequences of combining selective
cone harvesting and genetic thinning in clonal seed orchards. Silvae Genet-
ica 38(2):65-70
Lindgren D, Matheson AC (1986) An algorithm for increasing the genetic quality
of seed from seed orchards by using the better clones in higher propor-
tions. Silvae Genetica 35(5-6):173-177
Lindgren D, Mullin TJ (1997) Balancing gain and relatedness in selection. Silvae
Genetica 46(2):124-128
Lindgren D, Prescher F (2005) Optimal clone number for seed orchards with
tested clones. Silvae Genetica 54(1-6):80.
https://dx.doi.org/10.1515/sg-2005-0013
Lopez GA, Potts BM, Vaillancourt RE, Apiolaza LA (2003) Maternal and carryover
eects on early growth of Eucalyptus globulus. Canadian Journal of Forest
Research 33(11):2108-2115. https://dx.doi.org/10.1139/x03-132
Lstibůrek M, El-Kassaby YA (2008) Advanced-generation seed orchard designs.
Proceedings of the Seed Orchard Conference, 26-28 September 2007,
Umeå, Sweden: 155-160
Lstibůrek M, El-Kassaby YA (2010) Minimum-inbreeding seed orchard design.
Forest Science 56(6):603-608.
https://dx.doi.org/10.1093/forestscience/56.6.603
Lstibůrek M, Stejskal J, Misevicius A, Korecký J, El-Kassaby YA (2015) Expansion
of the minimum-inbreeding seed orchard design to operational scale. Tree
Genetics & Genomes 11(1):1-8.
https://dx.doi.org/10.1007/s11295-015-0842-5
MacLachlan IR, Wang T, Hamann A, Smets P, Aitken SN (2017) Selective breeding
of lodgepole pine increases growth and maintains climatic adaptation. For-
est Ecology and Management 391:404-416.
https://dx.doi.org/https://doi.org/10.1016/j.foreco.2017.02.008
McKeand E, Beineke F (1980) Sublining for half-sib breeding populations of for-
est trees. Silvae Genetica 29(1):14-17
Meißner M, Volmer K, Hardtke A, Stiehm C, Steiner W (2020) Bereitstellung von
leistungsfähigem und hochwertigem Forstvermehrungsgut für den klima-
und standortgerechten Wald der Zukunft (FitForClim) – Teilprojekt 1 –.
Hann. Münden, Germany, 66 p
Merzeau D, Alazard P, Canteloup D, Crémière L, Daubet A, Lesgourgues Y, Pas-
tuszka P, Ran A (2005) Genetic breeding of the maritime pine in Aquitaine:
an exemplary success story. Pierroton, France: Groupe Pin Maritime du Fu-
tur, 31 p
Montalvo AM, Ellstrand NC (2001) Nonlocal transplantation and outbreeding
depression in the subshrub Lotus scoparius (Fabaceae). American Journal of
Botany 88(2):258-269. https://dx.doi.org/10.2307/2657017
Morgante M, Vendramin GG, Rossi P, Olivieri AM (1993) Selection against inbreds
in early life-cycle phases in Pinus leucodermis Ant. Heredity 70(6):622-627.
https://dx.doi.org/10.1038/hdy.1993.89
Mullin TJ (2014) OPSEL 1.0: a computer program for optimal selection in forest
tree breeding. Arbetsrapport från Skogforsk TechnicalReport Nr 841-2014,
Report No: ISSN 1404-305X, 19 p
Mullin TJ (2017) OPSEL 2.0: a computer program for optimal selection in tree
breeding. Report No: ISSN 1404-305X, 24 p
Mullin TJ, Persson T, Abrahamsson S, Andersson Gull B (2019) Eects of inbreed-
ing depression on seed production in Scots pine (Pinus sylvestris). Canadian
Journal of Forest Research 49(7):854-860.
https://dx.doi.org/10.1139/cjfr-2019-0049
Namkoong G (1966) Inbreeding eects on estimation of genetic additive vari-
ance. Forest Science 12(1):8-13.
https://dx.doi.org/10.1093/forestscience/12.1.8
Namkoong G, Bishir J (1987) The frequency of lethal alleles in forest tree popula-
tions. Evolution 41(5):1123-1126.
https://doi.org/10.1111/j.1558-5646.1987.tb05882.x
Oakley CG, Ågren J, Schemske DW (2015) Heterosis and outbreeding depression
in crosses between natural populations of Arabidopsis thaliana. Heredity
115(1):73-82. https://dx.doi.org/10.1038/hdy.2015.18
Okada M, Kitamura K, Lian C, Goto S (2015) The eects of multilocus heterozy-
gosity on the longevity of seedlings established on fallen logs in Picea
jezoensis and Abies sachalinensis. Open Journal of Forestry Vol.05No.04:9.
https://dx.doi.org/10.4236/ojf.2015.54036
Olsson T, Lindgren D, Li B (2001) Balancing genetic gain and relatedness in seed
orchards. Silvae Genetica 50(5-6):222-226
Pakull B, Eusemann P, Wojacki J, Ahnert D, Liesebach H: Genetic diversity of
seeds from four German Douglas r (Pseudotsuga menziesii) seed orchards.
unpublished
Pâques LE (2013a) Forest tree breeding in Europe - Current State-of-the-Art and
Perspectives. Springer, 527 p. https://doi.org/10.1007/978-94-007-6146-9
Pâques LE (2009) Growth rhythm parameters as components of hybrid vigour in
young seedlings of hybrid larch (Larix decidua x L. kaempferi). 58(1-6):42.
https://dx.doi.org/https://doi.org/10.1515/sg-2009-0006
Pâques LE (2013b) Introduction. In: Pâques EL (ed) Forest Tree Breeding in Eu-
rope: Current State-of-the-Art and Perspectives. Dordrecht © Springer Sci-
ence+Business Media, pp 1-9.
https://doi.org/10.1007/978-94-007-6146-9_1
Pupin S, Sebbenn AM, Cambuim J, da Silva AM, Zaruma DUG, Silva PHM, Rosse
LN, Souza ICG, Marino CL, Moraes MLT (2019) Eects of pollen contamina-
tion and non-random mating on inbreeding and outbreeding depression in
a seedling seed orchard of Eucalyptus urophylla. Forest Ecology and Man-
agement 437:272-281.
https://dx.doi.org/https://doi.org/10.1016/j.foreco.2019.01.050
Pyhäjärvi T, Kujala ST, Savolainen O (2020) 275years of forestry meets genomics
in Pinus sylvestris. Evolutionary Applications 13(1):11-30.
https://dx.doi.org/10.1111/eva.12809
Rau HM (1998) Vermehrungsgut von Samenplantagen im Vergleich zu handels-
üblichem Material. AFZ-Der Wald 53(5):236-239
Rau HM, Schulzke R (2001) Beitrag forstlicher Samenplantagen bei der Bereit-
stellung herkunftsgesicherten Vermehrungsgutes. Proceedings of the FO-
RUM Genetik-Wald-Forstwirtschaft: Herkunftssicherung und Zertizierung
von forstlichem Vermehrungsgut, June 11-13, 2001, Freiburg, Germany: 66-
71
Rood SB, Goater LA, McCarey D, Montgomery JS, Hopkinson C, Pearce DW
(2017) Growth of riparian cottonwoods: heterosis in some intersectional
Populus hybrids and clonal expansion of females. Trees 31(3):1069-1081.
https://dx.doi.org/10.1007/s00468-017-1531-9
Rosvall O, Jansson G, Andersson B, Ericsson T, Karlsson B, Sonesson J, Stener L-G
(2001) Genetiska vinster i nuvarande och framtida fröplantager och klon-
blandningar. Uppsala (Sweden): SkogForsk, Technical Report, Report Num-
ber: Skogforsk-RED-1-2001, Report No: ISSN 1103-4580; TRN: SE0107372, 41
p
Saksa T (2004) Regeneration process from seed crop to saplings-A case study in
uneven-aged norway spruce-dominated stands in southern Finland. Silva
Fennica 38(4):371-381. https://doi.org/10.14214/sf.405
Sarvas R (1962) Investigations on the owering and seed crop of Pinus silvestris.
Helsinki, Finland, 198 p, ISBN 0026-1610
Schneck V (2001) Bestände und Samenplantagen von Gemeiner Kiefer. AFZ-Der
Wald 56(5):232-233
Shull GH (1952) Beginnings of the heterosis concept. In: Gowen JW (ed) Hetero-
sis. Iowa State College Press pp 14-48
Silva P, Brune A, Pupin S, Moraes M, Sebbenn A, de Paula R (2018) Maintenance
of genetic diversity in Eucalyptus urophylla S. T. Blake populations with re-
striction of the number of trees per family. Silvae Genetica 67(1):34-40.
https://dx.doi.org/10.2478/sg-2018-0005
Slavov GT, Howe GT, Adams WT (2005) Pollen contamination and mating pat-
terns in a Douglas-r seed orchard as measured by simple sequence repeat
markers. Canadian Journal of Forest Research 35(7):1592-1603.
https://doi.org/10.1139/x05-082
Song J, Ratclie B, Kess T, Lai BS, Korecký J, El-Kassaby YA (2018) Temporal quan-
tication of mating system parameters in a coastal Douglas-r seed orchard
under manipulated pollination environment. Scientic Reports 8(1):11593.
https://dx.doi.org/10.1038/s41598-018-30041-4
Sønstebø JH, Tollefsrud MM, Myking T, Steenrem A, Nilsen AE, Edvardsen ØM,
Johnskås OR, El-Kassaby YA (2018) Genetic diversity of Norway spruce (Picea
abies (L.) Karst.) seed orchard crops: Eects of number of parents, seed year,
and pollen contamination. Forest Ecology and Management 411:132-141.
https://dx.doi.org/https://doi.org/10.1016/j.foreco.2018.01.009
Stacy EA (2001) Cross-fertility in two tropical tree species: evidence of inbreed-
ing depression within populations and genetic divergence among popula-
tions. American Journal of Botany 88(6):1041-1051.
https://dx.doi.org/10.2307/2657086
98
Stephan BR, Liesebach M (1996) Results of the IUFRO 1982 Scots pine (Pinus syl-
vestris L.) provenance experiment in southwestern Germany. Silvae Geneti-
ca 45(5-6):342-349
Stoehr MU, Ott P, Woods JH (2015) Inbreeding in mid-rotation coastal Doug-
las-r: implications for breeding. Annals of Forest Science 72(2):195-204.
https://dx.doi.org/10.1007/s13595-014-0414-0
Stoehr MU, Webber J, Woods JH (2004) Protocol for rating seed orchard seedlots
in British Columbia: quantifying genetic gain and diversity. Forestry: An In-
ternational Journal of Forest Research 77(4):297-303.
https://dx.doi.org/10.1093/forestry/77.4.297
Talbert J (1979) An advanced-generation breeding plan for the NC State Univer-
sity-Industry pine tree improvement cooperative. Silvae Genetica
28(2/3):72-75
Templeton AR (1986) Coadaptation and outbreeding depression. In: Soule ME
(ed) Conservation Biology-The Science of Scarcity and Diversity. Sinauer As-
sociates, pp 105-116, ISBN ISSN: 0878937943
Torimaru T, Wang X-R, Fries A, Andersson B, Lindgren D (2009) Evaluation of pol-
len contamination in an advanced Scots pine seed orchard. Silvae Genetica
58(1-6):262-269. https://doi.org/10.1515/sg-2009-0033
Torimaru T, Wennström U, Andersson B, Almqvist C, Wang X-R (2013) Reduction
of pollen contamination in Scots pine seed orchard crop by tent isolation.
Scandinavian Journal of Forest Research 28(8):715-723.
https://dx.doi.org/10.1080/02827581.2013.838298
Van Buijtenen JP (1975) The planning and strategy of seed orchard programmes,
including economics. In: Faulkner R (ed) Seed orchards. London, UK: Forest-
ry Commission, pp 9-24
Vanclay JK (1991) Seed orchard designs by computer. Silvae Genetica 40:89-91
Vergara R, White TL, Huber DA, Shiver BD, Rockwood DL (2004) Estimated real-
ized gains for rst-generation slash pine (Pinus elliottii var. elliottii) tree im-
provement in the southeastern United States. Canadian Journal of Forest
Research 34(12):2587-2600. https://dx.doi.org/10.1139/x04-136
Wang Q, Qi J, Cui X, Li W (2018) Application of improved genetic algorithm in
clonal deployment for seed orchard (in Chinese). Sci. Silv. Sin. 54(4):30-37
White TL, Hodge G, Powell G (1993) An advanced-generation tree improvement
plan for slash pine in the southeastern United States. Silvae Genetica
42(6):359-371
Williams CG, Savolainen O (1996) Inbreeding depression in conifers: Implica-
tions for breeding strategy. Forest Science 42(1):102-117.
https://dx.doi.org/10.1093/forestscience/42.1.102
Wojacki J, Eusemann P, Ahnert D, Pakull B, Liesebach H (2019) Genetic diversity
in seeds produced in articial Douglas-r (Pseudotsuga menziesii) stands of
dierent size. Forest Ecology and Management 438:18-24.
https://dx.doi.org/10.1016/j.foreco.2019.02.012
Woods JH, Heaman JC (1989) Eect of dierent inbreeding levels on lled seed
production in Douglas-r. Canadian Journal of Forest Research 19(1):54-59.
https://dx.doi.org/10.1139/x89-007
Woods JH, Wang T, Aitken SN (2002) Eects of inbreeding on coastal Douglas-r:
nursery performance. Silvae Genetica 51(4):163-170
Wray N, Goddard M (1994) Increasing long-term response to selection. Genetics
Selection Evolution 26(5):431. https://doi.org/10.1186/1297-9686-26-5-431
Wu H, Ivkovic M, Gapare W, Matheson A, Baltunis B, Powell M, McRae T (2008)
Breeding for wood quality and prot in Pinus radiata: a review of genetic
parameter estimates and implications for breeding and deployment. New
Zealand Journal of Forestry Science 38(1):56-87
Yamashita M, Mullin TJ, Safarina S (2018) An ecient second-order cone pro-
gramming approach for optimal selection in tree breeding. Optimization
Letters 12(7):1683-1697. https://dx.doi.org/10.1007/s11590-018-1229-y
Yang B, Sun H, Qi J, Niu S, El-Kassaby YA, Li W (2020) Improved genetic dis-
tance-based spatial deployment can eectively minimize inbreeding in
seed orchard. Forest Ecosystems 7(1):10.
https://dx.doi.org/10.1186/s40663-020-0220-0
Yao J, Li H, Ye J, Shi L (2016) Relationship between parental genetic distance and
ospring’s heterosis for early growth traits in Liriodendron: implication for
parent pair selection in cross breeding. New Forests 47(1):163-177.
https://dx.doi.org/10.1007/s11056-015-9508-2
Zanewich KP, Pearce DW, Rood SB (2018) Heterosis in poplar involves phenotyp-
ic stability: cottonwood hybrids outperform their parental species at sub-
optimal temperatures. Tree Physiology 38(6):789-800.
https://dx.doi.org/10.1093/treephys/tpy019
Zobel BJ, Weir RJ, Jett JB (1972) Breeding Methods to Produce Progeny for Ad-
vanced-generation Selection and to Evaluate Parent Trees. Canadian Jour-
nal of Forest Research 2(3):339-345. https://dx.doi.org/10.1139/x72-052
