Upload
others
View
7
Download
0
Embed Size (px)
Citation preview
Teacher’s Notes 4.2 Does selective logging degrade the genetic quality of succeeding generations through dysgenic selection?
C
Forest Genetic resources traininG Guide
MODULE 4 Forest Management
Teacher’s Notes 4.2
Does selective logging degrade the genetic quality of succeeding generations through dysgenic selection?Jonathan Cornelius
Acknowledgements
The editors of this Forest Genetic Resources Training Guide wish to thank Jarkko Koskela and Barbara Vinceti for their contribution in identifying the need for the guide and for their continuous support during its preparation. We acknowledge the important advice from a reference group of scientists at Bioversity International - Elizabeth Goldberg, Jozef Turok and Laura Snook - who at various stages supported this project.
This training guide was tested during several training events around the world. We would like to acknowledge the valuable feedback received from many students and their teachers.
We would like to give special thanks to Ricardo Alía, National Institute of Agriculture and Food Research (INIA), Spain, for his review of the Case studies presented in this module.
The photos and maps in the PowerPoint presentation are the copyright of Jonathan Cornelius, Leonardo F. Freitas, Josiah Townsend and the American Geographical Society.
Finally, the production of the Forest Genetic Resources Training Guide would never have been possible without the financial support of the Austrian Development Cooperation through the project, ‘Developing training capacity and human resources for the management of forest biodiversity’, implemented by Bioversity International during 2004-2010. We would also like to thank the European Commission-funded “SEEDSOURCE” project for additional financial support.
All cover illustrations were drawn by Rosemary Wise and the layout was created by Patrizia Tazza. We are grateful for their beautiful work.
in collaboration with
Financed by
Citation:
Cornelius J. 2014. Does selective logging degrade the genetic quality of succeeding generations through dysgenic selection? A case study and teacher’s notes. In: Forest Genetic Resources Training Guide. Edited by Boshier D, Bozzano M, Loo J, Rudebjer P. Bioversity International, Rome, Italy.
http://forest-genetic-resources-training-guide.bioversityinternational.org/
ISBN 978-92-9043-994-3 ISSN 2223-0211
Bioversity International Via dei Tre Denari, 472/a 00057 Maccarese Rome, Italy © Bioversity International, 2014 Bioversity International is the operating name of the International Plant Genetic Resources Institute (IPGRI).
1
Module 4Forest Management
Teacher’s notes 4.2
Does selective logging degrade the genetic quality of succeeding generations through dysgenic selection?
Jonathan Cornelius, World Agroforestry Centre (ICRAF), Lima, Peru
Introduction
These Teacher’s notes aim to assist teachers in using Case Study 4.2 Dysgenic selection: can logging degrade the genetic quality of succeeding generations? in the classes. The notes:
• describe the key concepts covered in the case study, referring to forest genetics textbooks and other sources where explanations can be found (full references are given at the end of these notes)
• give tips on how to prepare and run the exercise and discuss the main learning points (genetic and other) that students should be able to derive from the case study
• outline the PowerPoint presentation, which introduces the case study to the students, and provides commentary the teacher can use (the presentation contains pictures of the species, sites where it occurs, relevant land-use issues in the area and figures/tables from the exercise)
• include an explanatory commentary on the case study document• summarize key elements to be covered in the tasks.
The case study document is intended to be a stand-alone document, assuming that students are undergraduates or graduates in forestry (or in related disciplines). However, prior reading on basic quantitative genetics (polygenic traits, response to selection) might be advantageous (see, for example, Jansson 2005).
Support materials are provided on the accompanying DVD and are available on the Forest Genetic Resources Training Guide webpage at www.bioversityinternational.org
These include:• The case study• The Teacher’s notes• The Teacher’s PowerPoint presentation.
Key concepts to cover/introduce in this case study
General conservation• Selective logging: see FAO et al. (2001, pp. 14–18, 25–231).Genetic concepts• Dysgenic selection: see FAO et al. (2001, pp. 20, 26–31); FAO et al. (2004a,
p. 104); FAO et al. (2004b, pp. 18–19); Finkeldey (2005, pp. 177, 181–198); Geburek and Turok (2005, pp. 443–444).
Module 4 Forest Management
2
• Heritability and genetic gain: see Jansson (2005); Walsh and Lynch (2008a; 2008b).
There is widespread perception of dysgenic1 selection as a threat to forest genetic resources, based on the belief that species or populations will degenerate if only the ‘worst individuals’ reproduce. However, this perception is largely the result of the simplicity and apparent reasonableness of the central belief rather than being based on experimental observations or study. Indeed, there are no proven examples of dysgenic selection having occurred, at least in the forestry sector, and the necessary conditions for it to occur in a forest management context appear to be rather stringent. This exercise aims to lead students towards this conclusion, while fulfilling two specific learning objectives:
• That the student should be able to identify situations in which dysgenic selection is likely to occur (or not occur)
• That the student should be able to explain to forest managers, certifying agencies and others why a given silvicultural regime might or might not lead to dysgenic selection.
As dysgenic selection is a form of phenotypic selection, the potential for dysgenic selection is explored using the framework of the ‘breeders’ equation.’ This equation encapsulates the theory of phenotypic selection on quantitative traits and facilitates a clear and logical presentation of the issues.
The case study includes background material on:• the breeders' equation• mahogany ecology and genetics• the mahogany-rich Marajoara site in Brazil.
How to run the exercise
The exercise can be run in a number of ways, depending on the time available, size of the class and level of knowledge of the students. The approach described below corresponds to a three-hour block session, e.g. during a training course or university practical class. In each case, the exercise is likely to work best if students work in groups of three to five and no more than six. Teachers should distribute the case study document at least 48 hours before the scheduled session. Careful reading of the document will take up to one hour and should be completed before the class. Teachers and any assistants should be fully familiar with both the case study and these Teacher’s notes.
• Introduction: Begin the session by using the PowerPoint presentation to introduce the theme and the exercise (30 minutes). The presentation is intended as an overview and ‘appetizer,’ rather than as an exhaustive briefing. The table in Annex 1 describes the function of each slide and includes a commentary and notes that presenters may find useful. Teachers should encourage students to ask questions during the presentation rather than simply reciting the text to them. The text is also included on the notes panel of each slide.
• Group work: The group work focuses on three tasks: Task 1. To discuss whether, in this logging operation, dysgenic selection would be likely to
1 Students may question whether “dysgenic” is a real word! The term dates from the early 20th century, and was initially used in the context of human societies and genetics. For example, the novelist Aldous Huxley referred to “dysgenics—the carrying on of the species by the worst members”. “Dysgenics” is the opposite of eugenics, i.e. selective breeding to improve the genetic quality of the human race, which was widely advocated in the late 19th and early 20th centuries, but has since been recognized as both unethical (at best) and based on dubious scientific assumptions. Neither term is derived directly from “gene”, which came into use later.
Teacher’s Notes 4.2 Does selective logging degrade the genetic quality of succeeding generations through dysgenic selection?
3
occur; Task 2. To formulate a general set of risk factors that would suggest susceptibility to dysgenic selection; Task 3. To discuss the consequences of dysgenic selection. Assign one of the three tasks to each group (but note that more advanced students may be able to cover Task 1 or 2 and Task 3). Students should discuss the case study amongst themselves, responding to the specific points and assigned tasks. The teacher should be present to answer questions (45 minutes discussion, 15 minutes to prepare presentations).
• Presentations and teacher feedback: Each group presents the outcomes of its work (10 minutes per presentation, with around 5 minutes after each presentation for questions and comments from the rest of the class and the teacher (based on five groups; more time can be allocated if there are fewer groups).
There are a number of alternative approaches to running the exercise. In the context of a training course, groups could be asked to reconvene to make their presentations the following day. In a university context, all three tasks could be assigned to each group as a take-home assessment task for completion over all or part of a one-semester course. In this case, one 50-minute lecture period could be devoted to the introduction and preliminary discussion.
Notes and commentary by case study headings
IntroductionThis section introduces the concept of dysgenic selection, noting that the conditions under which it might occur are not as widespread as might be expected. It then states the tasks and learning objectives.
Exploring dysgenic selection: response to selection and the ‘breeders’ equation’The breeders’ equation is used to predict response to selection. For this reason, it is important to explain clearly what we mean by the ‘response to selection.’ The approach taken here is to begin with realized response, as this is easily understood. We use an example from tree improvement because (a) tree improvement techniques may be familiar to some students and (b) there are no available examples of realized response to dysgenic selection. The idea that, just as tree improvement produces a positive response to selection, dysgenic selection produces a negative response has already been presented in the introduction and should be readily appreciated here.
Some students may struggle with the concept of response to selection as a genetic response because it is measured on phenotypic values. One way to deal with this is to explain that, at a given site, genetic and environmental deviations from the phenotypic mean would sum to zero and that therefore genetic and phenotypic means are identical. However, that concept would probably only be grasped readily by students who had already understood the initial point. A more effective approach would be to ask “what other factors could have caused the response?” and then, if necessary, point out why the approach to measuring the realized response in fact rules these out.
The response to selectionThe general approach mentioned at the start of this section describes improvement from one generation to the next. Some students may notice that in the example given in Box 1 both seedlots come from the same generation. It may be necessary to point out that it is the improved progeny that would actually be carried forward to form a new generation. The “SSO [seedling seed orchard] routine” seedlot represents what would happen if the parental population were used to form a new generation without any selection being applied.
Module 4 Forest Management
4
Predicting response: the breeders’ equationThis section introduces the breeders’ equation, providing a simple form of the equation before going on to discuss its application in subsequent subsections.
Selection differentialThe selection differential (S) is commonly expressed in units of the phenotypic standard deviation, i.e. as the selection intensity. This approach has two advantages: (a) it permits direct comparison of the strength of selection in different traits, or in the same trait when populations have different standard deviations, and (b) under truncation selection (i.e. when all individuals to the right or left of a given point in a frequency distribution are selected), values can be derived from the proportion selected, assuming that values of the trait in question are normally distributed.
In spite of these advantages, we use the unstandardized selection differential here, i.e. expressed in terms of the measurement units rather than the standard deviation units of the selection intensity. We have chosen to do this because using selection intensity would add complexity (truncation selection, normal probability function, an extra term in the breeders’ equation). This is because selection intensity rests on a dubious assumption (normal distribution), and because it would give an erroneous impression that dysgenic selection is necessarily truncation selection. In any case, the selection differential is a more readily understandable measure of phenotypic superiority.
Students often confuse the selection differential with the response to selection. The key here is to stress that S is a phenotypic, within-generation difference, whereas response to selection (R) is a genetic and between-generation difference.
HeritabilityStudents frequently have problems with heritability concepts. The familiarity and the common use of the word ‘heritable’ cause problems; students may wrongly assume that, because they know the meaning of ‘heritable,’ they also know the meaning of ‘heritability.’ This misunderstanding should be addressed by ensuring that the students read the case study document carefully. In their interactions with individual groups, teachers may find it useful to introduce a parent–progeny regression concept of heritability, and/or the concept of heritability as the coefficient of determination in a regression of genotype and phenotype. However, both approaches assume additional knowledge and the teacher will need to assess whether the students possess such knowledge.
The breeders’ equation applied to dysgenic selectionThe example used is hypothetical, partly to facilitate learning and partly to stimulate thinking about the factors that might mitigate apparent dysgenic selection in real situations.
Case study species and case study site and management regimesThese sections provide a reasonably complete account of the case study species, site and management regimes. The information provided is sufficient for the students to appreciate the context and to respond to the tasks. Not all the details included are needed to do the tasks; students will have to distinguish directly relevant information from less relevant data and background details to complete the tasks successfully.
Note that the scenario presented is based on logging at Marajoara, rather than being an exact case study. As is pointed out in the case study document, a ‘business-as-usual’ scenario would probably lead to zero regeneration. In that case, dysgenic selection would be unimportant.
Teacher’s Notes 4.2 Does selective logging degrade the genetic quality of succeeding generations through dysgenic selection?
5
Notes on the tasks
Task 1. Discuss whether, in this logging operation, dysgenic selection would be likely to occurThe best way to approach this question is by working backwards from hypothesized postlogging regeneration. You can guide the students by posing a number of questions, as follows:
Where will regeneration be found?Postlogging regeneration is likely to be made up almost exclusively of seedlings in the cleared areas around remnant trees, as elsewhere most seedlings will be shaded out.
Which trees will be the seed parents of the regeneration?Clearly, remnant trees will be the seed parents of all, or practically all, of the seedlings.
Which trees will be the pollen parents of the regeneration?As mahogany trees are usually outcrossing (see ‘Ecology’ in the case study), remnant trees are unlikely to be both pollen parents and seed parents of seedlings in their immediate vicinity. Based on the information given under the heading ‘Regeneration and management’ in the case study, and assuming that the logging that occurred between 1992 and 1994 was completed in one season, there are six flowering events that could give rise to seed during the ‘regeneration window’ afforded by opening the canopy around the remnant mother trees: the first would occur in the dry season before felling, the second following felling, and then annually in the years corresponding to the four subsequent annual weedings. Logged trees could contribute to the first of these, but not to the others. Populations around Marajoara had already been heavily logged, so it is unlikely that a significant proportion of pollen parents in any of these six events came from trees in the original populations.
Given the expected parentage of the regeneration, what is the selection differential?Some groups may choose to disregard the contribution of logged trees, as this would occur in just one of six flowering events. They could then proceed to calculate selection differential based only on the mean diameter at breast height (dbh) of remnant trees:
Selection differential (S) = mean dbh of remnant trees –mean dbh before logging
S = 41 cm – 67 cm = –26 cm.
Some groups may take other approaches. For example, the annual variation in fruit production mentioned in the case study may imply annual variation in flower production. So, if the dry season before felling (that is, the year in which logged trees would have been able to contribute as pollen parents) were a heavy-flowering year, then the influence of logged trees would be greater than if it were a year of little flowering. Furthermore, if the role of logged trees as pollen producers is to be taken into account, then groups might attempt to allow for differential contribution by different size classes. One approach would be to weight size classes crudely using the near four-fold greater fruit production of trees in the >60 cm dbh class (assuming proportionality of flower and fruit production).
The prelogging mean dbh of trees ≤60 cm dbh and >60 cm dbh was 40.6 cm and 83.6 cm respectively. There were 804 and 1609 stems respectively in these two broad size classes (values can be estimated from Figure 1 in the case study document). If trees in the >60 cm dbh group produce, on average, three times as much pollen as those in the ≤60 cm dbh group and they are approximately
Module 4 Forest Management
6
twice as numerous, then in the calculation of weighted mean dbh of pollen parents contributing to postlogging regeneration in that year, the former would be accorded a weight six times as great as that of the latter:
[40.6 + (83.6)6]/7 = 77.46 cm]
Consequently, the selection differential for the same year would be:
S = mean dbh of pollen and seed parents of next generation – unlogged population mean dbh
S = (77.46 + 40.6)/2 – 67 cm = 59.03 – 67 = –7.97 cm
The overall selection differential applying to the future regeneration would be the mean over the six flowering events. One possible estimate of selection differential (S = –20.0 cm) is shown in Table 1, in which it is assumed that two of the years (the first, to which logged trees contribute, and year 5) are heavy-flowering years in which flowering and fruiting are four times the long-term mean. Here, no within-year, between-size class weighting has been applied for the second and subsequent years, because after logging there is only one size class in the higher fertility group >60 cm dbh). Very ambitious groups might decide to tackle this issue by assuming some continuous relationship between size class and reproductive contribution, such that there are differences in reproductive contribution between the ≤60 cm dbh classes. Arguably, unlogged population mean dbh should also be calculated by weighting the contribution of different size classes; this would give higher values of the selection differential.
Table 1. Illustration of calculation of average selection differential over six years
Year 11 Year 2 Year 3 Year 4 Year 5 Year 6
Weighting factor for flowering2
4 1 1 1 4 1
Mean dbh of parents of regeneration3
59.03 41 41 41 41 41
Selection differential (cm)
−7.97 −26 −26 −26 −26 −26
Mean selection differential = [(-7.97)(4) − 26 − 26 − 26 − (26)(4) − 26]/12 = −20.00 cm
1 Year 1 corresponds to the flowering season before felling.2 Assuming that trees produce on average four times as many fruits in heavy-flowering years as in normal
years.3 Mean in Year 1 reflects contribution of subsequently logged trees, which have a higher mean and are
assumed to have higher pollen production (see text).
What are the likely heritability values?Following the calculation of selection differential, groups will need to make assumptions about the value of the heritability for growth traits, based on the comments in the ‘Heritability’ and ‘Genetics’ sections of the case study. Values less than or equal to 0.1 would appear most reasonable, as hinted at in the case study. Based on this, they should conclude that dysgenic selection is unlikely to be significant, principally because of low heritability values.
Teachers might also point out the implications of variation in fertility even between trees of similar dbh (even when, as is the case here, only trees that produce ≥25 capsules are considered). The implication is that in each size class relatively few trees actually make a significant reproductive contribution. This
Teacher’s Notes 4.2 Does selective logging degrade the genetic quality of succeeding generations through dysgenic selection?
7
decreases the reliability of heritability as a guide to the relationship between phenotype and genotype. Consequently, the dysgenic response could be either lower (including zero) or greater than predicted.
How might the logging reforms introduced in 2003 affect your conclusions?As far as dbh is concerned, the 2003 reforms should further reduce the likelihood of dysgenic effects. Assuming the new regime is adequately overseen, no trees ≤60 cm dbh would be logged and larger numbers of trees ≥60 cm dbh would be retained (i.e. around 205, based on five trees per 100 ha over the total area at Marajoara of 4100 ha). However, the lack of restrictions on the form of seed trees is likely to mean that these are the worst formed of the legally harvestable trees. These 205 seed trees about 17% of the remnant trees; the others (the 1000 trees that are <60 cm) are expected to be of average form (because they were left on size criteria), implying that around 17% (202/1205) of the remaining trees are likely to be poorly formed trees of relatively high average fertility (because they are relatively large). It is possible, therefore, that without further fine-tuning, the new regime might lead to an increased probability of dysgenic selection. However, further study would be needed before concluding that there is a major risk of this occurring.
Task 2. Formulate a general set of risk factors that would suggest susceptibility to dysgenic selectionA model answer is provided below (Figure 1). Some groups may arrive at something similar without any guidance beyond the template included in the case study document.
PROCESS OR CHARACTERISTICS RISK FACTORS EXPLANATION
HARVESTING REGIME
1.Remnant trees are strongly inferior in one or more phenotypic characteristic: logging not highly selective
The less selective the logging, the more likely loggers are to take everything but the worst. If remnant trees are not inferior, then there
can be no dysgenic effect
SURROUNDING POPULATIONS
REPRODUCTIVE ECOLOGY
GENETICS
2. Logging is large-scale, with many remnant trees; or multiple small-scale operations with similar logging
criteria
3.Logging is carried out before flowering or before seed dispersal
4.No nearby populations of the same species, or only logged populations
6. There is little pollen flow (inter- or intrapopulation)
7.There is no seed bank or there is little advance regeneration
9.Inferior remnant trees are highly fertile
11.Selected trait(s) moderately or strongly genetically controlled
10.Selected trait(s) has high genetic variation
12.Trait non-quantitative with simple Mendelian inheritance, little environmental effect on phenotype
With small numbers of selections, mean phenotype is a poor guide to mean genotype. However, small numbers of remnants implies
unpredictability rather than absence of effects
No opportunity for logged trees to contribute to subsequent generations from “beyond the grave” (either through pollen or seed)
8.Logged trees do not coppice
No mitigating gene flow from unlogged populations
Implying that their contribution to future generations will not be outweighed by smaller numbers of non-inferior but more fertile trees
All implying that future generations will be derived solely or very largely from remnant trees
In a highly variable population, the worst trees are likely to be more inferior than in a less variable population
In many cases, genetic control of commercial traits is weak. In such cases, dysgenic effects between successive generations will be
negligible
Could produce strong dysgenic effects between successive generations. Possible example: destructive harvesting of aguaje palm
5.Logged species is a facultative or obligate selfer
Figure 1. Risk factors for dysgenic selection, classified according to four classes of process or characteristics.
Module 4 Forest Management
8
Task 3. Discuss the consequences of dysgenic selection and how it can be mitigatedIs dysgenic selection a conservation genetics issue?The question is meant to be provocative. Nevertheless, in the case of quantitative traits, and given both that selection on any one locus is weak and that this is reduced still further by low heritabilities, changes in allele frequencies over a few generations are likely to be negligible. Although a large number of such negligible changes can have a perceptible effect on the mean value of the trait selected for, there is little chance of a given allele being lost due to selection, particularly at realistic levels of heritability and over the number of generations applicable in the case of long-lived organisms such as timber trees.
To say that dysgenic selection is not a conservation genetics issue would overstate the case, but at least in timber trees it appears likely to be of limited short-term importance. Its importance has probably been overstated in the past.
If dysgenic selection occurs, how can it be mitigated or reversed?Groups should be thinking in terms of both management (given that most commercial traits are heavily influenced by environment as well as genetics) and genetic measures. Mitigation could be through silvicultural techniques such as thinning, pruning and fertilization that can be used to improve product quality and quantity. Positive selection could reverse dysgenic selection by being built into forest management regimes or by being implemented through supplemental planting with improved or ‘average’ stock. However, even such simple measures may be difficult to achieve where institutions (either executing or enforcing) are weak.
Further information
Cornelius JP, Navarro CM, Wightman KW, Ward SE. 2005. Is mahogany dysgenically selected? Environmental Conservation 32:129–139. (This is a useful reference for instructors, but should not be given to students until after completion of the exercise).
FAO, DFSC, IPGRI. 2001. Forest Genetic Resources Conservation and Management, vol. 2: In managed natural forests and protected areas (in situ). International Plant Genetic Resources Institute, Rome, Italy.
FAO, FLD, IPGRI. 2004a. Forest Genetic Resources Conservation and Management, vol. 1: Overview, concepts and some systematic approaches. International Plant Genetic Resources Institute, Rome, Italy.
FAO, FLD, IPGRI. 2004b. Forest Genetic Resources Conservation and Management, vol. 3: In plantations and genebanks (ex situ). International Plant Genetic Resources Institute, Rome, Italy.
Finkeldey R. 2005. An Introduction to Tropical Forest Genetics. Georg-August-University, Göttingen, Germany.
Geburek T, Turok J, editors. 2005. Conservation and Management of Forest Genetic Resources in Europe. Arbora Publishers, Zvolen, Slovakia.
Jansson G. 2005. Quantitative genetics. In: Geburek T, Turok J, editors. Conservation and Management of Forest Genetic Resources in Europe, Arbora Publishers, Zvolen, Slovakia. pp. 213–235.
Walsh B, Lynch M. 2008a. 7. The population genetics of selection. In: Walsh, B. and Lynch, M. Evolution and Selection of Quantitative Traits:
Teacher’s Notes 4.2 Does selective logging degrade the genetic quality of succeeding generations through dysgenic selection?
9
I. Foundations.. Online prepublication draft. Version: 12 December 2009. Available from: http://nitro.biosci.arizona.edu/zbook/NewVolume_2/newvol2.html#2A Date accessed: 21 November 2013.
Walsh B, Lynch M. 2008b. 12. Short-term changes in the mean: 1. The breeders’ equation. In: Walsh, B. and Lynch, M. Evolution and Selection of Quantitative Traits: I. Foundations. Online prepublication draft. Version: 17 November 2008. Available from: http://nitro.biosci.arizona.edu/zbook/NewVolume_2/newvol2.html#2A Date accessed: 21 November 2013.
Module 4 Forest Management
10
Annex
1.
Sum
mary
of
the a
udio
visu
al
pre
senta
tion
Tab
le A
1. S
umm
ary
of t
he a
udio
visu
al p
rese
ntat
ion.
Slid
e no
./T
itle
Func
tio
nS
ugg
este
d s
po
ken
text
No
tes
1/D
ysge
nic
sele
ctio
n:
can
logg
ing
be
a fo
rm
of ‘n
egat
ive
gene
tic
imp
rove
men
t’?
Intr
oduc
es t
he t
hem
e.Th
ere
are
vario
us w
ays
in w
hich
fore
st g
enet
ic r
esou
rces
can
be
deg
rad
ed o
r d
estr
oyed
. Tod
ay w
e w
ill lo
ok a
t on
e of
the
se: d
ysge
nic
sele
ctio
n. It
is c
lose
ly a
ssoc
iate
d w
ith h
arve
stin
g an
d r
egen
erat
ion
pra
ctic
es. W
e’ll
defi
ne d
ysge
nic
sele
ctio
n in
just
a fe
w m
omen
ts.
Firs
t ho
wev
er, l
et’s
hav
e a
look
at
som
e ha
rves
ting
pra
ctic
es, b
oth
curr
ent
and
his
toric
.
A d
efini
tion
at t
his
stag
e w
ould
be
‘ove
r-w
eigh
ty’ a
nd is
unn
eces
sary
. But
do
say
that
w
e w
ill d
efine
it, a
s th
is r
aise
s ex
pec
tatio
ns
and
inte
rest
(and
mak
es it
cle
ar t
hat
defi
nitio
n is
imp
orta
nt).
The
cont
ent
that
follo
ws
shou
ld
get
stud
ents
thi
nkin
g in
the
rig
ht d
irect
ion
(if t
hey
are
not
alre
ady
awar
e of
the
sub
ject
) an
d w
ill le
ad u
p t
o a
form
al d
efini
tion.
If
gene
tic d
rift
has
alre
ady
bee
n co
nsid
ered
el
sew
here
, the
n it
mig
ht b
e he
lpfu
l to
men
tion
at t
he o
utse
t th
at d
ysge
nic
sele
ctio
n (D
S) i
s a
diff
eren
t fo
rm o
f gen
etic
det
erio
ratio
n.
2/La
Mos
qui
tia,
Hon
dur
asE
xam
ple
of l
oggi
ng
sugg
estiv
e (s
uper
ficia
lly,
per
hap
s) o
f DS
.
Late
r in
thi
s ca
se s
tud
y w
e w
ill b
e lo
okin
g in
som
e d
epth
at
a q
uite
rec
ent
case
of m
ahog
any
logg
ing
in t
he M
araj
oara
For
est
in t
he B
razi
lian
Am
azon
. Now
, tho
ugh,
let’s
go
bac
k in
tim
e an
d a
fe
w t
hous
and
kilo
met
res
to t
he n
orth
, to
the
Car
ibb
ean
coas
t of
H
ond
uras
in C
entr
al A
mer
ica,
par
t of
an
area
kno
wn
as L
a M
osq
uitia
or
, in
Eng
lish,
the
‘Mos
qui
to C
oast
.’ In
land
of t
he la
goon
s an
d…
Forw
ard
to
slid
e th
ree
on r
each
ing
the
wor
d
“lag
oons
” in
tex
t.
Figu
re fr
om V
on H
agen
WV.
194
0. T
he
Mos
qui
to C
oast
of H
ond
uras
and
its
inha
bita
nts.
Geo
grap
hica
l Rev
iew
30:
238–
259.
3/’M
ahog
any
bus
h’E
xam
ple
of l
oggi
ng
sugg
estiv
e (s
uper
ficia
lly,
per
hap
s) o
f DS
.
… a
nd t
he P
inus
car
ibae
a sa
vann
ah o
f the
coa
st, c
an (o
r co
uld
) b
e fo
und
mah
ogan
y-ric
h ar
eas
of fo
rest
, whi
ch, b
y th
e 19
40s,
and
p
rob
ably
long
bef
ore,
had
att
ract
ed t
he a
tten
tion
of fo
reig
n lo
ggin
g co
mp
anie
s.
Pho
tos
from
Von
Hag
en W
V. 1
940
The
Mos
qui
to C
oast
of H
ond
uras
and
its
inha
bita
nts.
Geo
grap
hica
l Rev
iew
30:
238–
259.
S
ee a
lso
Slid
e 32
(alte
rnat
ive)
.
Teacher’s Notes 4.2 Does selective logging degrade the genetic quality of succeeding generations through dysgenic selection?
11
Slid
e no
./T
itle
Func
tio
nS
ugg
este
d s
po
ken
text
No
tes
4/N
o tit
leE
xam
ple
of l
oggi
ng
sugg
estiv
e (s
uper
ficia
lly,
per
hap
s) o
f DS
.
Von
Hag
en (1
940)
, writ
ing
in t
he G
eogr
aphi
cal R
evie
w, d
escr
ibed
a
logg
ing
oper
atio
n as
a r
esul
t of
whi
ch, i
n hi
s w
ord
s “f
or t
he fi
rst
time
sinc
e 19
00 t
he w
ood
s ar
e co
min
g to
life
aga
in.”
The
com
pan
y op
erat
ing
the
conc
essi
on w
ould
fell
all c
omm
erci
ally
val
uab
le
mah
ogan
y tr
ees
(i.e.
tre
es m
ore
than
ab
out
30 y
ears
old
). Th
ese
wer
e flo
ated
to
the
mou
th o
f the
Car
atas
ca L
agoo
n an
d fr
om t
here
wer
e sh
ipp
ed t
o Fl
orid
a. H
owev
er, V
on H
agen
rep
orte
d t
hat
rege
nera
tion
of t
he fo
rest
was
, non
ethe
less
, ass
ured
.
The
rich
(uni
nten
ded
) iro
ny is
wor
th
high
light
ing!
P
hoto
: Leo
nard
o F
Frei
tas,
Par
á, B
razi
l. Fl
ona
de
Jam
anxi
m P
ara.
htt
p:/
/ww
w.fl
ickr
.com
/p
hoto
s/le
offr
eita
s/14
6942
7761
/siz
es/m
/
5/R
egen
erat
ion
assu
red
?Th
e ai
m h
ere
is t
hat
stud
ents
sho
uld
th
emse
lves
won
der
w
heth
er s
usce
ptib
ility
to
term
ite a
ttac
k co
uld
hav
e so
me
gene
tic b
asis
.
[Rea
d o
r p
arap
hras
e th
e q
uote
from
Von
Hag
en o
n th
e sl
ide]
, the
n:
...so
, the
fore
st w
ould
be
rege
nera
ted
exc
lusi
vely
by
thos
e m
ahog
any
tree
s th
at h
ad b
een
infe
sted
by
term
ites.
Le
t’s n
ow g
o so
uth
to t
he A
maz
on, b
ut t
o th
e w
est
rath
er t
han
the
east
.
Que
stio
ns o
r co
mm
ents
mig
ht b
e in
vite
d a
t th
is s
tage
(par
ticul
arly
if a
n ob
viou
s ‘p
regn
ant
pau
se’ i
s m
ade
afte
r “t
erm
ites”
), or
mig
ht
be
rais
ed a
nyw
ay. S
usce
ptib
ility
of t
rees
to
atta
ck b
y te
rmite
s co
uld
per
hap
s b
e re
late
d
to g
ener
al v
igou
r, w
hich
in t
urn
wou
ld b
e ex
pec
ted
to
be
influ
ence
d b
y ge
noty
pe.
6/Iq
uito
s, P
eruv
ian
Am
azon
Lead
ing
into
sub
top
ic o
n ag
uaje
pal
m.
Iqui
tos
is a
dyn
amic
, bus
tling
city
in t
he P
eruv
ian
Am
azon
, in
spite
of
hav
ing
no r
oad
con
nect
ions
to
the
outs
ide
wor
ld. M
any
kind
s of
fo
rest
pro
duc
ts c
an b
e fo
und
in t
he m
arke
ts o
f Iq
uito
s…
Forw
ard
to
slid
e 7
on “
mar
kets
of I
qui
tos.
..”
7/M
edic
inal
sp
ecie
s,
Iqui
tos
Inte
rest
/var
iety
, lin
king
to
the
mai
n su
bto
pic
....
incl
udin
g m
edic
inal
sp
ecie
s an
d…
Forw
ard
to
slid
e 8
on t
he w
ord
“an
d”
[Thi
s sl
ide
shou
ld la
st fo
r ab
out
one
seco
nd.]
8/Fr
uit
spec
ies
Link
ing.
…of
cou
rse,
frui
ts, p
erha
ps
the
mos
t im
por
tant
of w
hich
is…
Forw
ard
to
slid
e 9
on “
of w
hich
is…
” [T
his
slid
e sh
ould
last
for
abou
t on
e se
cond
.]
9/Th
e ag
uaje
pal
m
(Mau
ritia
flex
uosa
)E
xam
ple
of h
arve
stin
g su
gges
tive
(sup
erfic
ially
, p
erha
ps)
of D
S. M
aurit
ia
flexu
osa
was
sel
ecte
d t
o em
pha
size
tha
t co
ncer
ns
are
not
rest
ricte
d t
o ge
netic
res
ourc
es o
f tim
ber
sp
ecie
s.
... t
he a
guaj
e p
alm
, Mau
ritia
flex
uosa
.
Module 4 Forest Management
12
Slid
e no
./T
itle
Func
tio
nS
ugg
este
d s
po
ken
text
No
tes
10/A
maj
or lo
cal
ind
ustr
yLi
nkin
g th
roug
h to
the
ha
rves
ting
met
hod
.In
Iqui
tos
a si
gnifi
cant
ind
ustr
y is
bui
lt ar
ound
the
agu
aje
pal
m, g
ivin
g em
plo
ymen
t to
ab
out
5000
peo
ple
in d
iffer
ent
stag
es o
f the
sup
ply
ch
ain,
incl
udin
g ha
rves
ters
, mid
dle
men
, str
eet-
corn
er v
end
ors,
and
m
anuf
actu
re a
nd s
ale
of p
roce
ssed
pro
duc
ts li
ke t
his
ice
crea
m.
Aro
und
20
tonn
es o
f agu
aje
frui
ts a
re c
onsu
med
eac
h d
ay in
the
city
. O
ur in
tere
st, t
houg
h, is
in t
he fi
rst
stag
e of
the
sup
ply
cha
in.
11/H
arve
stin
g ag
uaje
Set
ting
up t
he s
ugge
sted
d
ysge
nic
scen
ario
.Th
ose
20 t
onne
s ar
e eq
uiva
lent
to
abou
t 40
0 tr
ees,
whi
ch a
re fe
lled
to
har
vest
the
frui
t. A
roun
d 1
500
ha o
f agu
aje—
pur
e or
alm
ost
pur
e st
and
s—ar
e fe
lled
eac
h ye
ar. A
s th
e m
arke
t p
refe
rs la
rge
frui
t, t
rees
w
ith la
rge
frui
t ar
e p
refe
rent
ially
felle
d. F
elle
d t
rees
do
not
cop
pic
e,
and
all
thei
r se
ed is
col
lect
ed.
12/M
aurit
ia fl
exuo
sa
typ
esTh
e su
gges
ted
DS
sc
enar
io.
Fort
unat
ely,
in o
ne s
ense
, the
sp
ecie
s is
dio
ecio
us, s
o al
l the
mal
e tr
ees
are
left
, tog
ethe
r w
ith r
emna
nt fe
mal
es—
esca
pee
s (th
ose
that
did
not
pro
duc
e fr
uit
that
yea
r or
tha
t w
ere
‘mis
sed
’ by
the
harv
este
rs) o
r th
ose
with
sm
all f
ruit.
The
suc
ceed
ing
gene
ratio
n w
ill
be
pro
duc
ed fr
om t
hese
esc
apee
s an
d in
ferio
r tr
ees.
Ano
ther
thr
eat
is lo
w e
ffect
ive
pop
ulat
ion
size
res
ultin
g fr
om u
neq
ual s
ex r
atio
s (a
s on
ly
fem
ales
are
felle
d).
The
two
fact
ors
may
als
o ac
t sy
nerg
istic
ally
.
13/N
o tit
leS
umm
ing
up a
nd le
adin
g in
to t
he d
efini
tion.
So,
bot
h th
ese
harv
estin
g re
gim
es h
ave
in c
omm
on t
he p
ossi
bili
ty
that
onl
y th
e ‘w
orst
’ mem
ber
s of
the
pop
ulat
ion
will
rep
rod
uce.
..w
hich
brin
gs u
s to
our
defi
nitio
n of
dys
geni
c se
lect
ion.
"Bot
h" r
efer
s to
mah
ogan
y in
Hon
dur
as a
nd
agua
je p
alm
in P
eru.
14/D
ysge
nic
sele
ctio
nTh
e d
efini
tion.
Dys
geni
c se
lect
ion
is s
elec
tion
that
lead
s to
an
und
esira
ble
d
irect
iona
l cha
nge
in g
enet
ic q
ualit
y ov
er o
ne o
r m
ore
gene
ratio
ns.
By
‘gen
etic
qua
lity’
we
usua
lly m
ean
per
form
ance
for
a gi
ven
com
mer
cial
ly im
por
tant
tra
it (s
uch
as g
row
th r
ate
or s
urvi
val),
and
b
y ‘d
irect
iona
l cha
nge’
we
mea
n a
pos
itive
or
nega
tive
chan
ge in
m
ean
valu
es. D
irect
iona
lity
dis
tingu
ishe
s d
ysge
nic
sele
ctio
n fr
om
gene
tic d
rift,
whi
ch c
an a
lso
lead
to
und
esira
ble
cha
nges
in g
enet
ic
qua
lity,
but
in a
ran
dom
way
and
in a
noth
er s
ense
(i.e
. ran
dom
loss
of
alle
les)
.
The
inst
ruct
or s
houl
d c
onsi
der
ask
ing
for
defi
nitio
ns o
f bot
h ‘g
enet
ic q
ualit
y’ a
nd
‘dire
ctio
nal,’
and
ask
ing
how
dire
ctio
nal d
iffer
s fr
om r
and
om d
rift
(or,
for
that
mat
ter,
‘gen
etic
co
ntam
inat
ion’
thr
ough
gen
e flo
w).
15/C
laim
s of
dys
geni
c se
lect
ion
and
var
iatio
n p
atte
rns’
Poi
ntin
g ou
t th
at t
he r
isks
ha
ve b
een
sugg
este
d in
th
e lit
erat
ure:
the
the
me
is
‘on
the
agen
da.
’
It ha
s b
een
sugg
este
d a
num
ber
of t
imes
tha
t d
ysge
nic
sele
ctio
n m
ight
be
of im
por
tanc
e in
fore
st t
rees
. For
exa
mp
le, P
alm
ber
g su
gges
ted
tha
t it
coul
d e
xpla
in w
hy p
rove
nanc
e va
riatio
n in
tw
o M
edite
rran
ean
pin
e sp
ecie
s fa
iled
to
show
rel
atio
nshi
ps
with
latit
ude
or lo
ngitu
de.
Pal
mb
erg
C. 1
975.
Geo
grap
hic
varia
tion
and
ear
ly g
row
th in
sou
th-e
aste
rn s
emi-
arid
A
ustr
alia
of P
inus
hal
apen
sis
Mill
. and
the
P.
bru
tia T
en. s
pec
ies
com
ple
x. S
ilvae
Gen
etic
a 24
: 150
–160
.
Teacher’s Notes 4.2 Does selective logging degrade the genetic quality of succeeding generations through dysgenic selection?
13
Slid
e no
./T
itle
Func
tio
nS
ugg
este
d s
po
ken
text
No
tes
16/C
laim
s of
dys
geni
c se
lect
ion
and
se
lect
ive
logg
ing
Poi
ntin
g ou
t th
at t
he r
isks
ha
ve b
een
sugg
este
d in
th
e lit
erat
ure:
the
the
me
is
‘on
the
agen
da.
’
Sim
ilarly
, it
has
bee
n cl
aim
ed t
hat
the
poo
r fo
rm o
f the
Car
ibb
ean
mah
ogan
y, S
wie
teni
a m
ahag
oni,
is t
he r
esul
t of
the
sam
e fa
ctor
….
17/C
laim
s of
dys
geni
c se
lect
ion
and
se
lect
ive
logg
ing
Poi
ntin
g ou
t th
at t
he r
isks
ha
ve b
een
sugg
este
d in
th
e lit
erat
ure:
the
the
me
is
‘on
the
agen
da.
’
...w
hile
, tur
ning
from
pos
sib
le p
asts
to
pos
sib
le fu
ture
s, L
emes
et
al.
sugg
este
d t
hat
this
mig
ht s
till b
e ha
pp
enin
g in
the
cas
e of
Sw
iete
nia
mac
rop
hylla
. W
e ha
ve o
f cou
rse
bee
n hi
ntin
g at
suc
h p
roce
sses
with
the
tw
o ex
amp
les
of L
a M
osq
uitia
and
Iqui
tos.
How
ever
, to
clar
ify t
hese
id
eas
a lit
tle w
e ne
ed t
o lo
ok c
aref
ully
at
the
term
‘sel
ectiv
e lo
ggin
g.’
Lem
es M
R, G
ribel
R, P
roct
or J
, Gra
ttap
aglia
D
. 200
3. P
opul
atio
n ge
netic
str
uctu
re o
f m
ahog
any
(Sw
iete
nia
mac
rop
hylla
Kin
g,
Mel
iace
ae) a
cros
s th
e B
razi
lian
Am
azon
, bas
ed
on v
aria
tion
at m
icro
sate
llite
loci
: im
plic
atio
ns
for
cons
erva
tion.
Mol
ecul
ar E
colo
gy 1
2:
2875
–288
3.
18/W
hat
is s
elec
tive
logg
ing?
Aim
s to
cla
rify
the
mea
ning
, par
ticul
arly
to
avoi
d t
he id
ea t
hat,
with
in
spec
ies,
it m
eans
‘hig
h gr
adin
g.’
‘Sel
ectiv
e lo
ggin
g’ r
efer
s, o
f cou
rse,
to
the
felli
ng a
nd e
xtra
ctio
n of
on
ly c
erta
in t
ypes
of t
rees
. How
ever
, for
our
pur
pos
es, i
t is
use
ful t
o d
istin
guis
h tw
o sl
ight
ly d
iffer
ent
typ
es o
f sel
ectio
n as
pra
ctis
ed b
y lo
gger
s an
d o
ther
har
vest
ers
of fo
rest
pro
duc
ts.
Firs
t, w
e ha
ve t
he s
elec
tive
logg
ing
of p
artic
ular
sp
ecie
s. T
ypic
ally
, lo
gger
s on
ly e
xtra
ct t
hose
sp
ecie
s th
at h
ave
com
mer
cial
val
ue, a
nd
this
is t
he u
sual
mea
ning
of ‘
sele
ctiv
e lo
ggin
g’ w
hen
used
in fo
rest
ry.
Pho
to:
Leon
ard
o F.
Fre
itas.
Fel
iz N
atal
, Mat
o G
ross
o, B
razi
l. ht
tp:/
/ww
w.fl
ickr
.com
/p
hoto
s/le
offr
eita
s/79
0041
598/
size
s/m
/in/s
et-
7215
7600
6059
3314
4/
19/W
ithin
-sp
ecie
s se
lect
ive
logg
ing
Aim
s to
cla
rify
the
mea
ning
, par
ticul
arly
to
avoi
d t
he id
ea t
hat,
with
in
spec
ies,
it m
eans
‘hig
h gr
adin
g.’
Sec
ond
, we
have
sel
ectio
n w
ithin
sp
ecie
s. In
thi
s ca
se, l
ogge
rs
don
’t us
ually
sel
ect
‘onl
y th
e b
est’
(as
the
quo
te in
slid
e 17
see
ms
to s
ugge
st).
Rat
her,
they
fell
ever
ythi
ng t
hat
it is
wor
thw
hile
to
fell,
an
d t
here
fore
leav
e on
ly t
he w
orst
. In
fact
, if l
ogge
rs t
ook
only
the
ve
ry b
est,
the
n d
ysge
nic
sele
ctio
n w
ould
not
be
a se
rious
wor
ry. I
t w
ould
be
like
tryi
ng t
o ca
rry
out
gene
tic im
pro
vem
ent
by
lett
ing
all
ind
ivid
uals
rep
rod
uce
exce
pt
the
very
wor
st. T
his
wou
ld n
ot w
ork,
b
ecau
se g
enet
ic im
pro
vem
ent
by
rem
oval
of j
ust
the
very
wor
st
wou
ld h
ave
a ne
glig
ible
effe
ct o
n th
e fr
eque
ncy
in t
he p
opul
atio
n of
th
e b
est
alle
les.
Pho
to:
Leon
ard
o F.
Fre
itas.
Sen
ador
Jos
e P
orfir
io,
Par
á, B
razi
l. ht
tp:/
/ww
w.fl
ickr
.com
/pho
tos/
leof
frei
tas/
4476
1604
8/si
zes/
m/in
/set
-72
1576
0060
5933
144/
Module 4 Forest Management
14
Slid
e no
./T
itle
Func
tio
nS
ugg
este
d s
po
ken
text
No
tes
20/D
ysge
nic
sele
ctio
n: t
he m
irror
im
age
of g
enet
ic
imp
rove
men
t
Link
ing
bac
k to
DS
as
the
opp
osite
of g
enet
ic
imp
rove
men
t.
In g
enet
ic im
pro
vem
ent,
bre
eder
s of
ten
sele
ct t
he v
ery
bes
t an
d
allo
w o
nly
thos
e in
div
idua
ls t
o re
pro
duc
e.
In lo
ggin
g, lo
gger
s of
ten
leav
e on
ly t
he v
ery
wor
st, a
nd t
hese
‘s
elec
ted
ind
ivid
uals
’ (w
hich
of c
ours
e ar
e th
ose
left
by
logg
ers,
ra
ther
tha
n th
ose
sele
cted
by
them
) are
tho
se t
hat
rep
rod
uce
and
gi
ve r
ise
to t
he n
ext
gene
ratio
n.
So
dys
geni
c se
lect
ion
is li
ke t
he m
irror
imag
e of
gen
etic
im
pro
vem
ent.
..
21/P
ositi
ve
phe
noty
pic
sel
ectio
n ai
min
g at
gen
etic
im
pro
vem
ent
Con
solid
atin
g th
e ‘m
irror
im
age
idea
’ and
lead
ing
into
exp
licit
case
s of
ne
gativ
e an
d p
ositi
ve
sele
ctio
n.
…an
d w
e ca
n ill
ustr
ate
that
with
the
se d
ata
from
a V
ochy
sia
guat
emal
ensi
s p
lant
atio
n in
Sar
apiq
uí, C
osta
Ric
a… [b
riefly
exp
lain
w
hat
the
grap
h sh
ows]
In t
his
case
, we’
re il
lust
ratin
g p
ositi
ve
phe
noty
pic
sel
ectio
n. T
rees
ab
ove
7 cm
db
h ha
ve b
een
sele
cted
, an
d e
vent
ually
onl
y th
ese
will
be
allo
wed
to
rep
rod
uce.
22/N
egat
ive
phe
noty
pic
sel
ectio
n,
pos
sib
ly le
adin
g to
d
ysge
nic
sele
ctio
n
Neg
ativ
e se
lect
ion.
In t
his
case
, we
are
imag
inin
g th
e se
lect
ion
of o
nly
thos
e tr
ees
with
db
h <
2 c
m. I
f onl
y th
ese
35 o
r so
sm
all t
rees
wer
e al
low
ed t
o re
pro
duc
e, t
hen
ther
e w
ould
be
a ris
k of
dys
geni
c se
lect
ion.
23/T
here
are
no
doc
umen
ted
ex
amp
les
of d
ysge
nic
sele
ctio
n in
fore
st
tree
s
Lead
ing
in t
o th
e ne
ed
for
pre
dic
tion
and
the
b
reed
ers’
eq
uatio
n.
In fa
ct, t
here
are
no
doc
umen
ted
exa
mp
les
of d
ysge
nic
sele
ctio
n ac
tual
ly h
avin
g ha
pp
ened
in fo
rest
tre
es. T
his
shou
ld n
ot b
e se
en a
s an
ind
icat
ion
that
it is
not
imp
orta
nt—
rath
er, i
t re
flect
s th
e fa
ct t
hat
no o
ne h
as s
erio
usly
look
ed in
to it
. For
exa
mp
le, n
o on
e ha
s ev
er
com
par
ed t
he g
row
th r
ate
of p
roge
ny o
f log
ged
and
rem
nant
tre
es.
As
a re
sult,
we
need
to
app
roac
h th
is s
ubje
ct fr
om a
the
oret
ical
vi
ewp
oint
. For
tuna
tely
, as
dys
geni
c se
lect
ion
is li
ke g
enet
ic
imp
rove
men
t in
rev
erse
, the
hig
hly
succ
essf
ul t
heor
y th
at u
nder
lies
gene
tic im
pro
vem
ent
serv
es a
s ou
r fr
amew
ork.
Muc
h of
thi
s fr
amew
ork
is s
umm
ariz
ed in
one
rel
atio
nshi
p, k
now
n as
…
24/T
he b
reed
ers’
eq
uatio
n...
the
bre
eder
s’ e
qua
tion,
whi
ch a
llow
s us
to
pre
dic
t th
e ef
fect
of
sele
ctio
n, w
heth
er p
ositi
ve (i
mp
rove
men
t) or
neg
ativ
e (d
ysge
nic)
. Le
t’s lo
ok b
riefly
at
the
thre
e co
mp
onen
ts.
Teacher’s Notes 4.2 Does selective logging degrade the genetic quality of succeeding generations through dysgenic selection?
15
Slid
e no
./T
itle
Func
tio
nS
ugg
este
d s
po
ken
text
No
tes
25/T
he b
reed
ers’
eq
uatio
n:
Res
po
nse
to
sele
ctio
n (R
) =
sele
ctio
n d
iffer
entia
l (S
) x h
erita
bili
ty (h
2 )
Exp
lain
s th
e re
spon
se t
o se
lect
ion—
wha
t it
is t
hat
we
are
tryi
ng t
o p
red
ict.
The
effe
ct o
f sel
ectio
n is
qua
ntifi
ed b
y th
e re
spon
se t
o se
lect
ion,
w
hich
mea
sure
s th
e ch
ange
in p
erfo
rman
ce b
etw
een
one
gene
ratio
n (th
e on
e in
whi
ch s
elec
tion
is c
arrie
d o
ut) a
nd t
he n
ext
(the
imp
rove
d
[or
dys
geni
c] g
ener
atio
n). I
t is
mea
sure
d in
wha
teve
r un
its a
re
app
rop
riate
to
the
trai
t b
eing
sel
ecte
d (e
.g. d
bh—
cm, h
eigh
t—m
, st
em q
ualit
y-sc
ore,
den
sity
—kg
/m3)
, or
can
be
exp
ress
ed a
s a
per
cent
age.
For
exa
mp
le, i
f we
sele
ct p
lus-
tree
s in
pla
ntat
ions
, and
tr
ees
grow
n fr
om t
heir
seed
gro
w 1
0% m
ore
qui
ckly
tha
n av
erag
e se
ed fr
om t
he s
ame
pla
ntat
ions
, the
n th
e re
spon
se t
o se
lect
ion
is
10%
. B
ut b
efor
e w
e ca
rry
out
sele
ctio
n, w
e w
ill n
ot k
now
its
futu
re im
pac
t,
whi
ch is
why
we
need
to
pre
dic
t it.
We
do
so u
sing
tw
o p
aram
eter
s.
26/T
he b
reed
ers’
eq
uatio
n:
Res
pon
se t
o se
lect
ion
(R) =
sel
ecti
on
diff
eren
tial
(S) x
he
ritab
ility
h2 )
Exp
lain
s th
e se
lect
ion
diff
eren
tial.
Firs
t, le
t’s lo
ok a
t th
e se
lect
ion
diff
eren
tial,
whi
ch is
the
diff
eren
ce
bet
wee
n th
e se
lect
ed in
div
idua
ls a
nd t
he o
vera
ll m
ean.
For
exa
mp
le,
in t
he c
ase
just
men
tione
d, i
f the
plu
s-tr
ees
had
a m
ean
db
h of
40
cm a
nd t
he p
opul
atio
n m
ean
is 2
0 cm
, the
n th
e se
lect
ion
diff
eren
tial
is 2
0 cm
. In
itsel
f, th
e se
lect
ion
diff
eren
tial t
ells
us
noth
ing
abou
t w
heth
er t
he s
uper
iorit
y it
exp
ress
es is
due
to
gene
tics
or s
ome
othe
r ca
use—
it is
just
a w
ithin
-gen
erat
ion
mea
sure
of d
iffer
ence
.
27/T
he b
reed
ers’
eq
uatio
n:
Res
pon
se t
o se
lect
ion
(R) =
sel
ectio
n d
iffer
entia
l (S
) x
heri
tab
ility
(h2 )
Intr
oduc
es h
erita
bili
ty.
Sec
ond
, we
look
at
the
herit
abili
ty, w
hich
mea
sure
s th
e d
egre
e to
w
hich
the
sup
erio
rity
mea
sure
d in
the
sel
ectio
n d
iffer
entia
l is
pas
sed
on
to
the
pro
geny
. The
val
ue o
f her
itab
ility
var
ies
from
zer
o to
on
e—a
typ
ical
val
ue fo
r d
bh
mig
ht b
e ar
ound
0.2
. Put
ting
thes
e tw
o to
geth
er, w
e ca
n p
red
ict
the
resp
onse
to
sele
ctio
n.
28/T
he b
reed
ers’
eq
uatio
n:
Res
pon
se t
o se
lect
ion
(R) =
sel
ectio
n d
iffer
entia
l (S
) x
herit
abili
ty (h
2 )
Exa
mp
le o
f pre
dic
tion
of
resp
onse
to
sele
ctio
n.Th
e he
ritab
ility
tel
ls u
s th
at 2
0% o
f the
phe
noty
pic
sup
erio
rity
exp
ress
ed in
the
sel
ectio
n d
iffer
entia
l is
due
to
herit
able
gen
etic
ch
arac
teris
tics
(alle
les
that
the
plu
s-tr
ees
have
). W
e th
eref
ore
pre
dic
t a
bet
wee
n-ge
nera
tion
gene
tic r
esp
onse
of 2
0% o
f the
sel
ectio
n d
iffer
entia
l, i.e
. 4 c
m.
Module 4 Forest Management
16
Slid
e no
./T
itle
Func
tio
nS
ugg
este
d s
po
ken
text
No
tes
29/M
ahog
any
logg
ing
in B
razi
lFi
nish
ing
with
the
ch
alle
nge
pos
ed b
y th
e ex
erci
se, w
hich
invo
lves
ap
ply
ing
the
bre
eder
s’
equa
tion.
In o
ur e
xerc
ise,
we
will
use
thi
s fr
amew
ork
to e
xplo
re b
oth
the
natu
re
of d
ysge
nic
sele
ctio
n an
d t
he p
rob
abili
ty o
f it
occu
rrin
g in
a s
pec
ific
exam
ple
bas
ed o
n lo
ggin
g in
the
Mar
ajoa
ra F
ores
t, P
ará
Sta
te,
Bra
zil.
Mar
ajoa
ra is
a 4
100
ha in
dus
try-
man
aged
fore
st a
rea,
and
is
pro
bab
ly t
he m
ost
stud
ied
mah
ogan
y-ric
h fo
rest
any
whe
re.
30/M
araj
oara
bef
ore
logg
ing
Her
e is
the
dis
trib
utio
n of
db
h at
Mar
ajoa
ra b
efor
e lo
ggin
g...
The
grap
h ac
tual
ly s
how
s va
lues
for
the
sam
ple
d a
rea
at M
araj
oara
, ext
rap
olat
ed o
ver
the
entir
e 41
00 h
a, o
n th
e as
sum
ptio
n of
eq
ual
stoc
king
thr
ough
out.
31/M
araj
oara
bef
ore
logg
ing…
Mar
ajoa
ra
afte
r lo
ggin
g
Lead
s in
to t
he r
est
of t
he
sess
ion
(cas
e st
udy)
....
and
her
e is
the
db
h d
istr
ibut
ion
afte
r lo
ggin
g. W
ill t
he “
cont
inuo
us
exp
loita
tion
of la
rge,
sup
erio
r in
div
idua
ls”
lead
to
dys
geni
c se
lect
ion,
as
sug
gest
ed b
y so
me
rese
arch
ers?
One
of y
our
task
s is
to
answ
er
this
que
stio
n!
The
two
grap
hs a
re q
uite
str
ikin
g an
d s
houl
d
arou
se in
tere
st.
Dra
w a
tten
tion
to t
he o
bvi
ous
dem
ogra
phi
c im
pac
t of
the
logg
ing,
if it
is n
ot c
lear
whe
ther
th
is h
as a
lread
y b
een
notic
ed.
32/’
Mah
ogan
y b
ush’
Alte
rnat
ive
to s
lide
3.P
hoto
s:
Left
: Jos
iah
Tow
nsen
d. M
osq
uitia
Mar
sh,
Gra
cias
a D
ios
Dep
artm
ent,
Hon
dur
as.
http
://w
ww
.flic
kr.c
om/p
hoto
s/jo
siah
_to
wns
end
/275
8880
722/
size
s/m
/ R
ight
: Jos
iah
Tow
nsen
d. R
io T
apal
was
, W
amp
usirp
i, R
us R
us, G
raci
as a
Dio
s D
epar
tmen
t, H
ond
uras
. htt
p:/
/ww
w.fl
ickr
.com
/p
hoto
s/jo
siah
_tow
nsen
d/2
7589
4050
0/si
zes/
m/
Module 4 Forest Management
B
Forest Genetic Resources Training GuideMODULE 1 Species conservation strategies
1.1 Leucaena salvadorensis: genetic variation and conservation
1.2 Talbotiella gentii: genetic variation and conservation1.3 Shorea lumutensis: genetic variation and conservation
MODULE 2 Trees outside of forests
2.1 Conservation of tree species diversity in cocoa agroforests in Nigeria
2.2 Devising options for conservation of two tree species outside of forests
MODULE 3 Tree seed supply chains
3.1 Genetic bottlenecks in the restoration of Araucaria nemorosa
3.2 Tree planting on farms in East Africa: how to ensure genetic diversity?
MODULE 4 Forest management
4.1 Impacts of selective logging on the genetic diversity of two Amazonian timber species
4.2 Does selective logging degrade the genetic quality of succeeding generations through dysgenic selection?
4.3 Conserving Prunus africana: spatial analysis of genetic diversity for non-timber forest product management
MODULE 5 How local is local? – the scale of adaptation
5.1 Selecting planting material for forest restoration in the Pacific north-west of the USA
5.2 Local adaptation and forest restoration in Western Australia
Other modules to be published among the following: Plantation forestry, Tree domestication, Forest restoration, Genetic modification