Michioon State I‘
University

THESIS

mm

 

 

 

 

 

 

[l [Mlfli"'lfl"llHl'fllTll"WLlWI '

293 10632 147

 

This is to certify that the
dissertation entitled
Isozyme Analysis of the Blue—Engelmann Spruce
Comp lex in Southwes tern Colorado
presented by

Stephen Gerard Ernst ~

has been accepted towards fulfillment ‘ t

of the requirements for
Ph . D . degrée in Forest Genetics

mum

Major professor

Date Sept. 26, 1985

uen‘ lib-a: . ~ -- un'r L . , . 042771

 

MSU

 

LIBRARIES
”

 

 

RETURNING MATERIALS:
Place in book drop to
remove this checkout from
your record. FINES will
be charged if book is

 

returned after the date
stamped below.

 

W2) 8: g

b

71 K150
“W08“?! 1
3 2.‘ K 221%

W

587%

\
rum

 

'77/

 

3?’ K2}?

in"?

K

 

Mme;

 

 

 

ISOZYME ANALYSIS OF THE BLUE-ENGELMANN SPRUCE COMPLEX
IN SOUTHWESTERN COLORADO

BY

Stephen Gerard Ernst

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Forestry

1985

4&9

‘

:ba’7<p./

ABSTRACT

ISOZYME ANALYSIS OF THE BLUE-ENGELMANN SPRUCE COMPLEX
IN SOUTHWESTERN COLORADO

BY

Stephen Gerard Ernst

A partial diallel mating design incorporating 20
parents each of blue and Engelmann spruce in the Dolores
River drainage of southwestern Colorado was utilized to
generate intraspecific and interspecific full-sib progeny.
Percent germinated and percent ungerminated-but-full seed
on a total seed basis were used as measures of crossability
and abnormal seed production, respectively.

Production of viable intraspecific seed was primarily
under the control of nonadditive sources of genetic
variation in both blue and Engelmann spruce. Maternal
effects were also very important in the production of
viable intraspecific seed in blue spruce, but not in
Engelmann spruce. Interspecific crosses were successful
only with Engelmann spruce as the female parent; no viable
seed were produced from the reciprocal species cross.
Production of viable interspecific seed with Engelmann
spruce as the female parent was primarily under the control
of additive sources of genetic variation, but control by
nonadditive sources of variation was also evident.

The inheritance of isozymes representing 13 loci from

11 enzyme systems was determined in both blue and Engelmann

spruce. The isozymic composition of the two species was
quite different in the Dolores River drainage, and several
species—specific alleles and strong frequency differences
were observed at several loci. Positive identification of
control-pollinated interspecific hybrids was achieved using
isozyme analysis.

No natural F1 hybrids were observed in the field among
sampled mature trees based on isozyme analysis, nor among
open-pollinated embryos from mother trees located at a site
where both blue and Engelmann spruce are present and pollen
shed and female strobilus receptivity are coincident
between the two species in the spring. However, based on
the low crossability between the two species, if natural
hybridization does occur between blue and Engelmann spruce
it must be of very low frequency and beyond the resolution
of this study based on the number of enzyme loci and the
number of individuals sampled. Therefore, if natural
hybrids do exist and are fertile, evidence for

introgression would be rapidly diluted by backcrossing.

AC KN OW LE DGEM ENTS

I would like to thank Dr. JIW. Hanover (Chairman) for
the support, patience, trust and freedom he has extended to
me during my tenure at Michigan State University. I would
also like to thank the members of my guidance committee:
Dr. D.E. Keathley for his encouragement, advise and open
door; Dr. I.L. Mao for his generous help in the statistical
analysis; and Dr. J. Hancock for his help in presenting the
results of the isozyme analysis. Also, the help of many
Other members of the Department of Forestry, including
staff members, fellow graduate students and work-study
students, is much appreciated.

To my parents, I could never thank them enough for all
they have given and imparted to me, and I hope my work is a
favorable reflection on their efforts and love. Also, I
would like to thank my fiancee, Jane, for the encouragement

only a water ouzel can bring.

ii

TABLE OF CONTENTS
Page
LIST OF TABLES...................................... iv
LIST OF FIGURES..................................... vi
CHAPTER
I. Introduction................................... 1

II. Genetic Variation and Control of Intraspecific
Crossability in Blue and Engelmann Spruce

Abstract....................................... 10
Introduction................................... 11
Materials and Methods.......................... 14
Results........................................ 24
Discussion..................................... 28

III. Inheritance of Isozymes in Seed and Bud Tissues of
Blue and Engelmann Spruce

Abstract....................................... 34
Introduction................................... 35
Materials and Methods.......................... 36
Results and Discussion......................... 42

IV. Allozyme Variation of Blue and Engelmann Spruce in
Southwestern Colorado

Abstract....................................... 60
Introduction................................... 61
Materials and Methods.......................... 63
Results........................................ 67
Discussion..................................... 72

V. Assessment of Natural Hybridization and Introgression
Between Blue and Engelmann Spruce in Southwestern
Colorado

Abstract....................................... 80
Introduction................................... 81
Materials and Methods.......................... 84
Results........................................ 90
Discussion..................................... 97

VI. Recommendations for Future Study............... 107

LIST OF REFERENCES.OOOOOOOOOOOOOOOOOOOOCCOOOOOOOOOOO 110

iii

TABLE

LIST OF TABLES

Page

Chapter II

1.

Mean values of percent germinated and percent
ungerminated-but—full seed on a total seed

basis for open-pollinated and control-

pollinated collections..................... 26

Variance component and narrow-sense heritability
estimates for production of viable and abnormal
full-sib seed in blue and Engelmann spruce.. 26

Product-moment correlations between percent
germinated and percent ungerminated-but-full
values when parents served as females or

maleSOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO 29

Chapter III

1.

Electrophoresis buffers and respective power
requ1rementSOOOOOOOOOOOCOOOOOOOOOOIOOOOO0.. 41

2. Enzymeisystems assayed in respective buffer

SYStemSooooooooooooooooooeooooooooooooooooo 41
3. Single-locus segregation tests of genotypes
expressed in bud tissue of blue and Engelmann
spruce full-sib progeny.................... 44
4. Pooled segregation ratios observed in
megagametophyte tissue of seed collected from
heterozygous mother trees in the Dolores River
drainageOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO... 47
Chapter IV
1. Allelic frequencies and expected heterozygosities
among the respective loci and overall for blue
and Engelmann spruce open-pollinated collections
in the Dolores River drainage.............. 68
2. Allelic frequencies and expected heterozygosities

for the seven polymorphic loci observed among
three blue spruce subpopulations from the Dolores
River drainage...0.000IOOOOOOOOOOOOOOOOOOOO 70

iv

Page

3. Allelic frequencies and expected heterozygosities
for the eight polymorphic loci observed among the
three Engelmann spruce subpopulations from the
Dolores River drainage..................... 71

4. Nei's corrected genetic distance estimates and
their standard errors, and Rogers‘ distance
coefficient among the six blue and Engelmann
spruce subpopulations...................... 73

Chapter V

l. The eleven loci from nine enzyme systems analyzed
among the Engelmann x blue spruce hybrids... 89

2. Mean values of percent germination and percent
ungerminated-but-full seed on a total seed basis
for open-pollinated and control-pollinated
COIIxtionSOOCOOOOOOOOOOOOOOOOOOOOOOOOOOOO. 92

3. Variance component and narrow-sense heritability
estimates for production of viable and abnormal
full-sib seed in Engelmann x blue spruce
hYbridSOOOOOOOOOOOOOOOOOOOOOOOO0.00.0000... 92

4. Single-locus segregation tests of isozyme

genotypes expressed in bud tissue of Engelamnn
x blue spruce full-sib progeny............. 94

LSIT OF FIGURES

FIGURE

Chapter II
1. A portion of the partial diallel mating

design layout used to make the controlled
mllinationso.OOOOOOOOOOOOOOOOOOOOOOOOOOOOO

Chapter III
1. Zymograms of 11 enzyme systems in blue and

Engelmann spruce and their respective
genotypeSOOOOO0.0.0.00000000000000000000000

Chapter V
1. Unique isozyme phenotypes expressed in bud
tissue of Engelmann x blue spruce full-sib

progenYOOOOOOO0..C0.0000IOOOOOOOOOOOOCOOOOO

vi

Page

16

45

96

CHAPTER I

INTRODUCTION

A biological species has been defined as a group of
natural populations which can actually or potentially
interbreed but are reproductively isolated from other such
groups (Ayala 1982). Reproductive isolation is imperative
for the genetic divergence of two populations into distinct
species because gene frequency changes in one population
cannot then be incorporated into the other through
interbreeding. Two species can be reproductively isolated
either through geographical separation or as a result of
internal genetic mechanisms. When reproductively isolated,
each species represents an independent and discrete
evolutionary unit.

Mechanisms of reproductive isolation have been grouped
into two classifications based on the occurrence and
success of interspecific matings--prezygotic reproductive
isolating mechanisms which prevent the formation of
interspecific zygotes, and postzygotic mechanisms which
reduce the viability or fertility of interspecific hybrids
(Ayala 1982; Mayr 1963). Prezygotic isolating mechanisms
include: seasonal (allochronic) and habitat (allopatric)
-isolation such that potential mates do not meet; behavioral
(ethological) isolation where potential mates meet but do

not mate; mechanical isolation where transfer of gametes is

attempted but is not successful; and gametic isolation
where sperm or pollen is transferred but the egg is not
fertilized. Postzygotic isolating mechanisms include
hybrid inviability where hybrid zygotes die or fail to
reach sexual maturity; hybrid sterility where the hybrid
zygotes are viable but partially or completely sterile; and
hybrid breakdown where the F2 or backcross hybrid progeny
have reduced viability or fertility. Prezygotic mechanisms
are considered to be more efficient evolutionarily because
they reduce unneeded expenditure of limited resources in
the development of nonviable or infertile progeny (Grant
1981).

Reproductive isolation can occur as a by-product of
evolutionary divergence between populations or species, or
by direct selection for these mechanisms. If two
populations become separated geographically, they may
accumulate enough divergent characters independently to be
considered separate races or species. If the two
populations then come into sympatry (naturally or
artificially), they may or may not be capable or producing
viable hybrid progeny, depending on the degree and types of
genetic divergence. If the two groups became genetically
reproductively isolated during separation, these mechanisms
may have been the result of pleiotropic effects associated
with other genetic changes rather than direct selection for

them. Conversely, the two groups may be capable of

3

interbreeding, with geographic separation the only cause
for their independent genetic integrities. Many examples
of these two scenarios occur in nature (see Futuyma 1979).

While geographic isolation is regarded as the most
common mode of speciation (Mayr 1963; Lewontin 1974; Grant
1981), putative examples of direct selection for
reproductive isolation do exist in nature (Futuyma 1979;
Grant 1981). Grant (1981) has given five special
conditions for selection to enhance hybridization barriers
in plants. First, the races or species must be sympatric
in distribution. Second, hybridization must be
deleterious, where the hybrid progeny are less viable or
infertile. Third, hybridization results in a loss of
reproducitve potential to the two groups, and that loss is
disadvantageous. Fourth, absolute incompatibility may not
necessarily result, the extent of hybridization dependent
on how effectively the hybrids are eliminated by natural
selection. Lastly, only those isolating mechanisms which
are manifest in the parental generation will respond
effectively to selection for isolation.

The genus £1233 (spruce) presents some interesting
examples of reproductive isolation and speciation. Wright
(1955) observed that £1233 species which were the most
widely separated geographically and the most divergent
morhpologically generally crossed with the least success.
However, there do exist species combinations in £1333 which

occupy very similar geographic distributions and are

morphologically quite similar, but still possess very
strong hybridization barriers; emy, among the North
American spruces, white and black spruce (g; glauca
(Moench) V055. and g; mariana (Mill.) B.S.P.,
respectively), and blue and Engelmann spruce (g; pungens

Engelm. and g: engelmannii Parry ex Engelm., respectively).

 

These species combinations may represent instances of
selection for reproductive isolation. Whether reproductive
isolation in such species combinations resulted as a by-
product of geographic (allopatric) speciation and was
enhanced when the two species came into sympatry, or
occurred while in sympatry and caused their evolutionary
divergence into distinct species (sympatric speciation) can
only be speculated until further study. A

There is considerable overlap in the ranges of blue
and Engelmann spruce, with the range of blue spruce
contained almost entirely within that of Engelmann spruce.
They are both montane species native to the Rocky Mountain
region of North America, but occupy somewhat different
habitats. Blue spruce is generally found in the semiarid
riparian habitats of the montane zone at elevations of 2000
to 3000 meters. Associated species include trembling aspen

(Populus tremuloides Michx.), ponderosa pine (Pinus

 

 

ponderosa Dougl. ex Laws.) and Douglas-fir (Pseudotsuga

 

 

menziesii (Mirb.) Franco). Engelmann spruce generally

 

occurs above 3000 meters along the upper valleys and

hillsides in pure stands or mixed with subalpine fir (Abigs
lasiocarpa (HookJ thnflu). In the elevationally
intermediate zones both species are often present in close
proximity, with ample opportunity for cross-pollination.

Blue and Engelmann spruce are quite similar
morphologically, but can be distinguished using a
combination of traits (Daubenmire 1972; Jones and Bernard
1977; Schaefer and Hanover 1985a). The most diagnostic
characters are usually associated with female cone
morphology, bark texture, twig pubescence, shape of bud
scales and needle sharpness (Mitton and Andalora 1981).
Chemical traits have also been effective in distinguishing
the two species (Taylor et al. 1975; Schaefer and Hanover
1985b). However, there is enough overlap in morphological
and biochemical traits between between blue and Engelmann
spruce to suggest either introgression or coadaptation
between the two species (Daubenmire 1972; Taylor et al.
1975; Mitton and Andalora 1981). Schaefer and Hanover
(1985c) combined morphological and terpene data using
discriminant analysis and observed intermediacy between the
two species in southwestern Colorado.

Artificial hybrids between blue and Engelmann spruce
have been produced, but viable seed yields from the
interspecific crosses were very low. Fechner and Clark
(1969) obtained two percent viable hybrid seed for this
interspecific cross, but only with Engelmann spruce as the

female parent; the reciprocal cross did not produce any

viable seed. Kossuth and Fechner (1973) obtained less than
one percent viable hybrid seed, but only with blue spruce
as the female parent; the reciprocal cross was not
successful. However, in each of these studies only one or
two parents of each species were used to make the
controlled crosses. The selection of parents was shown to
influence seed yields in both intraspecific (King et al.
1970) and interspecific (Kudray and Hanover 1980) crosses
in spruce. Therefore, a large number of parents is
required to accurately quantify interspecific crossability
in this genus. It is reasonable to assume, based on
previous morphological, chemical and crossability studies
that the interspecific crossability between blue and
Engelmann spruce is very low.

In the elevationally intermediate zones where both
blue and Engelmann spruce are present, there is enough
overlap between the phenology of pollen shed and female
strobilus receptivity of these two species for natural
cross-pollination to occur (Fechner and Clark 1969; Ernst,
unpublished data). Therefore allopatric or allochronic
prezygotic reproductive isolating mechanisms are not
isolating factors at present. Nor are ethological
mechanisms, as the spruces are anemophilous. Kossuth and
Fechner (1973) did document very clearly that gametic
isolation is manifest in this interspecific cross. Normal

development of the female gametophyte apparently requires

some type of stimulus from the intraspecific pollen.
However, the incompatibility mechanism between these two
species is more complex than this, as breakdown was
observed at several stages of ovule and embryo development,
and crossability barriers were not complete (Kossuth and
Fechner 1973). Hybrid inviability has also been observed
among blue-Engelmann hybrid embryos and seedlings (Fechner
and Clark 1969; Kossuth and Fechner 1973). Unfortunately,
the hybrids generated in the two studies mentioned were
destroyed in a greenhouse fire (Fechner, personal
communication), and no known mature hybrids exist to assess
hybrid fertility in the F1 and later generations.

Blue and Engelmann spruce are believed to be closely
related phylogenetically, blue spruce possibly the result
of a single speciation event from the progenitor Engelmann
spruce (Nienstaedt and Teich 1971; Daubenmire 1972; Taylor
et al. 1975). Blue spruce may have evolved from an ecotype
of Engelmann spruce which was adapted to the semiarid
lowlands and was partially (sympatrically or
parapatrically) or completely (allopatrically)
geographically isolated. Taylor et a1. (1975) suggested
the two species resulted from a geographical (allopatric)
speciation event, and incompatibility was a by-product of
evolutionary divergence. When the two species came into
sympatry, the hybridization barriers may not have been as
strong as they are today, but reproductive isolation was

advantageous due to habitat requirements of the two

incipient species and thus was enhanced to its present
state. Whether or not hybrid inviability or infertility
also had an effect cannot be determined until known hybrids
are observed through maturity and F2 crosses are attempted.
Hybrid inviability or infertility seems essential if strong
incompatibility barriers were not present when the two
incipient species came into sympatry.

Another possibility is that blue and Engelmann spruce
speciated while in sympatry. At least two different
scenarios can be envisioned for such an event. A quantum
speciation event may have occurred which resulted in a race
or species much better adapted to the more arid environment
now occupied by blue spruce. Again, reproductive isolation
may have been advantageous because of ecological divergence
between the two incipient species, or possibly the same
mutational event made the incipient species incompatibile
with Engelmann spruce but compatible with itself. Another
scenario is where a mutation occurred in a marginal
population of Engelmann spruce--located in a more arid
environment--which rendered it at least partially
incompatible with Engelmann spruce but compatible with
itself. The ensuing population then exploited the more
arid habitat and incompatibility was further enhanced due
to niche separation and hybrid inviability.

All speculation aside, it is necessary to first

quantify the genetic variation in blue and Engelmann spruce

across their respective ranges and determine how the
environment influences the expression of those traits
before any speciation theories can be adequately evaluated.
Also, the genetic variability and control of intraspecific
and interspecific crossability and its physiological basis
must be quantified using a large number of parents from
throughout the respective ranges of the two species.
Morphological, biochemical and physiological traits need to
be quantified in common garden experiments and the degree
of genotype x environment interaction assessed. Also, a
technique which will accurately identify interspecific
hybrids needs to be developed.

The objectives of this study were to (l) assess the
genetic variation and control of intraspecific and
interspecific crossability in blue and Engelmann spruce;
(2) determine the inheritance of isozymes in
megagametophyte, embryo and bud tissues in blue and
Engelmann spruce; (3) describe and quantify the isozyme
variability in blue and Engelmann spruce in the Dolores
River drainage in southwestern Colorado; (4) determine if
isozyme analysis can be used to identify blue-Engelmann
spruce hybrids; and (5) determine if natural hybrids
between blue and Engelmann spruce exist in the Dolores
River drainage and look for evidence of introgression.
This work is part of a long-range study investigating the
genetic basis of adaptation in the North American spruce

complex (Hanover 1975).

CHAPTER II

GENETIC VARIATION AND CONTROL OF INTRASPECIFIC CROSSABILITY
IN BLUE AND ENGELMANN SPRUCE

ABSTRACT

To assess the degree of genetic control for
intraspecific crossability in blue and Engelmann spruce,
two 20-parent partial diallel matings, one for each
species, were conducted in the field during the spring of
1983. As measures of crossability, percent germinated and
percent ungerminated-but-full seed on a total seed basis--
full and empty-~were determined for the individual crosses.
Best linear unbiased prediction (BLUP) was used to estimate
the individual fixed and random effects of the mixed model,
and restricted maximum likelihood (REML) methods were used
to estimate general combining ability (GCA), specific
combining ability (SCA), maternal effects and error
variances.

Engelmann spruce exhibited slightly higher
intraspecific crossability than was observed for blue
spruce. ‘Viable seed yields from selfed crosses were
approximately 50 percent less than control—pollinated
biparental seed yields. Nonadditive sources of variation
apparently exert major genetic influence in the production
of viable seed in both species. Maternal effects are very

important in the production of viable seed in blue spruce,

lO

11

but not so in Engelmann spruce. Therefore, to maximize
production of viable seed in seed orchards of both species,
parents should be selected on the basis of specific cross
combinations which produce proportionately high amounts of
viable seed. The practical effect of maternal influence in
blue spruce seed production should be studied further.

The use of best linear unbiased prediction techniques
should have wide utility in plant breeding because of their

ready application to unbalanced multifactor data.

INTRODUCTION

Phenotypic improvement of a trait requires an
understanding of the genetic crontrol governing the
character of interest and how the environment influences
its expression. Only then can appropriate selection and
breeding methods be applied to the population or
individuals to bring about the desired gains. In forest
trees, as in all commercially important plant species,
there are a wide variety of traits for which improvement is
desired. In conjunction with these efforts, the improved
plant material must be propagated either sexually—-via
seed--or asexually--via vegetative propagation--in large
numbers for commercial use. Therefore an equally important

trait is how well do the individuals selected on the basis

12

of yield or some other trait propagate and what is its
genetic control.

Many of the commercially important characters in
forest trees are under polygenic control (Wright 1976).
Therefore long-term gains are best realized through
breeding programs and mass selection techniques which
utililze the additive genetic variance associated with
these traits (Hallauer 1981). Because of this, most forest
trees in North America are propagated by seed rather than
vegetatively. Also, vegetative propagation is not yet
feasible on a commercial scale for many important forest
tree species, especially among conifers (Timmis and Ritchie
1984; Bonga 1981). In addition, because it takes so long
for many tree species to reach commercial maturity (from 20
to 200 years), a variety of genotypes is generally more
desirable than a monoculture (Tigerstedt 1974).

The objectives of this study were to (l) assess the
genetic variation and control of intraspecific crossability
in blue and Engelmann spruce, and (2) generate full-sib
progeny for future studies on the inheritance of a wide
variety of traits in both species. The information on
intraspecific crossability is desirable for two reasons.

It will serve as a reference for studying the interspecific
crossability between blue and Engelmann spruce (see Ernst

et al. 1985c). Also, it will provide information important
to seed orchard selection and management techniques for the

two species. Blue spruce is one of the most widely planted

13

horticultural conifers in the world (Hanover 1975), and
Engelmann spruce is a commercially valuable timber species
in the Rocky Mountain region of North America (Fowler and
Roche 1975).

The Dolores River drainage in southwestern Colorado
was selected as the study site because both blue and
Engelmann spruce occur in the drainage, often in close
proximity. Also, previous studies investigating the
genetic variability in morphological and terpenoid
characters of blue and Engelmann spruce have been conducted
in the drainage (Hanover 1975; Reed and Hanover 1983;
Schaefer and Hanover 1985a and b).

The partial diallel mating design is theoretically
efficient in regards to the amount of information per cross
and accuracy of the estimates of the genetic parameters
(Namkoong and Roberds 1974; Pederson 1972; Gordon 1980).
Empirical evidence has also shown the partial diallel to be
generally efficient (Chaudharey et al. 1977; Anand and
Murty 1969; Murty et al. 1967; Kearsey 1965).

Best linear unbiased prediction (BLUP), as developed
by Henderson (1973) and described by Mao (1982), can be
viewed in practice as an extension of generalized least-
squares (GLS)--or more specifically, best linear unbiased
estimation (BLUE)—-techniques which allow for simultaneous
estimation of fixed effects and prediction of random

effects. While BLUE is applicable only to fixed model

14

analysis, BLUP was developed for analysis of mixed models
which incorporate a mixture of random and fixed factors.
In BLUP, a priori variances and covariances--from previous
studies or estimates--associated with the different random
factors are incorporated into the mixed model equations
(MME) used to obtain the solutions. Using a maximum
likelihood or restricted maximum likelihood (REML)
technique such as described by Schaeffer (1976), new
variance and covariance estimates can be obtained from the
BLUP solutions, and the process is repeated until
convergence. BLUP is especially well suited to unbalanced
data situations, which are very prevalent in plant breeding

experiments.

MATERIALS AND METHODS

The Dolores River was divided elevationally into five
species-occupation zones, the middle three zones (2,3 and
4) comprising the sample area for this study. Zone 2
extends from 2400 to 2590 meters (m) in elevation and blue
spruce is the primary occupant relative to the occurrence
of Engelmann spruce. Zone 3 extends from 2590 to 2770 m,
an elevationally intermediate zone with respect to the
habitats of blue and Engelmann spruce, both species
occurring in zone 3 and often in close proximity. Zone 4

extends from 2770 to 2960 m and is occupied primarily by

15

Engelmann spruce relative to the presence of blue spruce.
The respective parents in zones 2 and 4 represent "pure"
species subpopulations and parents in zone 3 represent
putative introgressed subpopulations. All parents were
readily identifiable as to species. During the spring of
1983, ten blue spruce trees each in zones 2 and 3 and ten
Engelmann spruce trees each in zones 3 and 4-—for a total
of 40 trees, 20 of each species--were selected primarily on
the basis of fecundity, accessability and climbability to
be used as parents in the controlled pollinations.

The partial diallel design used in this study
incorporated three intraspecific crosses--including selfs--
and three interspecific crosses per parent (Figure 1).
During the spring of 1983, female strobili were isolated
before pollen shed. Also, pollen strobili were collected
just prior to anthesis, dried and the pollen extracted.
Standard pollination methods for conifers were utilized and
the trees were debagged following scale closure of the
female strobili. The female strobili on a few of the blue
spruce parents were beginning to close when pollinated, and
therefore possible receptivity differences based on scale
closure were recorded. Female strobili with scales
completely open and fully receptive were scored as l,
strobili with up to 50 percent of the scales beginning to
close were scored as 2, and strobili with more than 50

percent of the scales beginning to close scored as 3. The

16

Male
Female 1 2 3 4 5 6 7 8 9 10 11

 

l X X X

10 X X

11 X X X

 

Figure 1. A portion of the partial diallel mating design layout used
to make the controlled pollinations. An "X" denotes an
attempted full-sib cross.

17

ages of the respective parents were also determined using
increment cores taken from the hole at a height of one
meter.

In the fall of 1983, control and open—pollinated cones
were collected. The control-pollinated cones were kept
separate by isolation bag. The cones were dried and seed
extracted by hand, recording the total number of cones per
bag and the number of cones damaged by insects per
isolation bag. The extracted seed was blown to separate
empty and putatively full seed and both portions were
counted and kept in cold storage (4’C).

During the sumnmx of 1984, germination tests were
conducted using a maximum-—depending on availability--of 30
seeds per isolation bag. Blue and Engelmann spruce seed do
not have any special dormancy requirements (Heit 1961).

The seeds were placed in 6 cm diameter petri dishes, using
two filter paper discs (Whatman, No. 3) as the germination
medium. A solution of 2.59/1 Captan and Benlate was used
to keep the seeds moist and deter fungal growth during
germination. The seeds were germinated in partial light at
a day-night temperature cycle of 27 and 18° C,
respectively. The seeds were observed daily for 30 days,
and the number of newly germinated seeds recorded each day.
Upon germination, the seeds were removed and planted in 5 x
5 x 25 cm plant bands containing 3:1:1
(peat:vermiculite:perlite) soil mix. At the end of the 30—

day germination period, the number of ungerminated seeds

18

were counted and then dissected to determine the number of

full but ungerminated seed and empty seed. The percentages

of germinated and ungerminated—but-full seed were
determined from the germination test (subsample), and these
values were then extrapolated to a total seed basis—-the
total number of full and empty seed per isolation bag—-to
serve as the dependent variables in the analysis. Percent
germinated seed on a total seed basis was used as a measure
of intraspecific crossability because it measures the
number of viable seed produced for a given cross. Percent
ungerminated-but-full seed was measured primarily to detect
postzygotic abnormalities.

The model equation used to account for the identified
fixed and random factors was:

Yijklmn = C + LiLk + Fijm + Mk1 + (Fika) jl + Rm + baA

+ bCC + bdD + eijklmn
where:

Yijklmn = the nth germination record (percent germinated or
percent ungerminated—but-full seed on a total
seed basis) of the cross between the jth female,
found at location i and of receptivity class m
and of age A, with male 1 of location k, that
record (bag) having C cones and D (percent) of

those cones damaged by insects;

c a constant common to all parents;

Fijm = the random effect of female j which resides in

19

location i and of receptivity class m (j =
1,...,20);

Mk1 = the random effect of male 1 which resides in location
k (l = 1,...,20);

the random interaction effect common to all

(Fika)j1
records (bags) of the subclass corresponding to
the cross of female j (from location i and of
receptivity class m) and male 1 (from location
k);

LiLk = a subclass effect combining the fixed effects of the
location of female j, the location of male 1, and
the location interaction (i,k = 1,2 or 3 for
zones 2, 3 and 4, respectively);

Rm = the fixed effect of the receptivity class of female j
(m = 1,2 or 3, as described previously);

A = the age of female j in years;

ba = the regression coefficient corresponding to the age
of female j;

C = the number of cones in isolation bag n of cross
(Fika)jl7

bC = the regression coefficient corresponding to the number
of cones in record (bag) n of cross (Fika)jl;

D = the percentage of cones in isolation bag n of cross
(Fika)jl that were damaged by insects;

bd = the regression coefficient corresponding to the
percent of insect-damaged cones in record (bag) n

of cross (Fika) j17

20

eijklmn = the random residual (error) associated with

record Yijklmn-

The model equation is rewritten in matrix form to
facilitate the computation:
y = Xb + Zu + e

where:

y = an a x 1 vector of the germination records, where a =
91 for blue spruce and 96 for Engelmann spruce
intraspecific crosses;

b = an unknown b x 1 vector containing the different levels
of the fixed factors and the regression
coefficients as described previously, where b =
11 for blue spruce and 8 for Engelmann spruce
crosses;

a known a x b matrix, its elements denoting the levels

N
u

of the fixed factors and the values of the
covariates in the b vector which are associated
with a given y observation (record);

an unknown c x 1 vector containing the different

I:
II

levels of the random factors--female, male and
female x male effects--where c = 98 for blue
spruce and 100 for Engelmann spruce crosses;

a known a x c matrix of ones and zeros, its elements

N
ll

denoting the levels of the random factors in the
u vector which are associated with a given y

observation (record);

21

e = an unknown a x 1 vector of random residuals.
The two groups of intraspecific crosses were analyzed
separately.

Assumptions pertaining to the Operational model were:
(1) the parents were unrelated; (2) the parents of each
species were from a single, randomly mating population (and
therefore share common female, male and female x male
variances for each trait); and (3) the control pollinations
were made at random (no selection). Based on these

assumptions, the mathematical expectations of the variances

were:
Var y V 26 R
u = 62' (3 0
e R O R
where:

G = 1612, where I is a c x c identity matrix, and 612
= on, 6M2 or GFMZ' corresponding to variances
associated with the female, male and female x

male random effects, respectively;

162, where I is an a x a identity matrix and o’2

w
II

the error variance associated with the random
residuals;
and V = ZGZ' + R.

The fixed effects and covariates were considered to be
"nuisance" factors because they were included in the model
only to improve the accuracy of the variance estimates
associated with the random factors. Estimation of the

variances was the primary focus of this study.

22

To solve a set of mixed model equations (MME),
restrictions must be imposed in order to obtain solutions
for the fixed effects. For this model, the female, male
and female x male (random) effects have unique solutions,
but the fixed effects are not individually estimable
(unique). However, certain linear contrasts between the
levels of the fixed effects are best (minimum variance of
the estimator), linear, unbiased and estimable or unique
(BLUE).

The solutions for the random factors in mixed model
prediction (BLUP) are unique. However, nested predictands
are the sum of fixed and random effects in mixed model
prediction--e4b, location i + female j,and location k +
male l--and are not unique. Therefore, direct comparisons
of the female, male and female x male predictands are
possible only within a common location or receptivity
class. If the effects of the nested fixed factors are
accounted for (iJL, adjusted in linear combination),
comparisons of parents or crosses from different locations
or classes are estimable or unique.

The female, male and female x male predictands
correspond to general and specific combining ability
estimates of the respective parents and crosses in the
partial diallel mating design. Under the assumptions that
the population was sampled at random, it is randomly

mating, and there is no inbreeding, linkage or epistasis,

23

the general combining ability variance (dbz) corresponds to
one-fourth the total additive genetic variance (1/4 6A2)
for that trait, and the specific combining ability variance
(652) corresponds to one-fourth the total nonadditive
(dominance) variance (1/46b2) in the partial diallel mating
design (Kempthorne and Curnow 1961). Maternal effects were
contained within the female general combining ability
estimate, and therefore the maternal effect variance (GMEZ)
is estimable by subtracting the male general combining
ability variance from the female general combining ability
variance: 6ME2 = chz - dénz.

Variance component estimates were obtained using
iterative expectation maximization restricted maximum
likelihood (EM-REML) algorithms (Banks et al. 1985) under
the assumption of normality of female, male and female x
male effects. The EM-REML algorithms used were:

82 = (1")! - B'X'y - fi'z'yvn - 1

312 = (6-6 - 82tr(C))/pi — 1
where B and a were the solutions to the MME. The vector (ii
corresponds to the female, male or female x male portion of
6 used to estimate the respective variances. The C matrix
is the portion of the inverse matrix from the MME
pertaining to female, male or female x male effects. The
value n is the total number of records, and Pi is the
number of parents or crosses associated with the female,
male or female x male effects in each set of intraspecific

crosses. The variance estimates were incorporated into the

24

MME, and new solutions for b and u and variance estimates
were obtained. Iterations were carried out until the
estimated variances changed less than one percent from the

previous iteration.

RESULTS

Viable seeds were obtained from 46 blue spruce and 56
Engelmann spruce full-sib families, out of a total 60
possible full-sib families for each species. Mean values
of the percent germinated and percent ungerminated-but-full
seed from open—pollinated and control-pollinated selfed and
biparental collections are given in Table 1. Viable seed
yields for open-pollinated crosses exceeded those for
control-pollinated biparental crosses by a factor of two or
more. Larger yields for open-pollinated collections were
expected because only one pollen parent was used in each
pollination bag, whereas there is a diverse mixture of
pollen genotypes under open-pollinated conditions. Viable
seed yields for sel fed crosses in both species were 80 to
90 percent less than those for open-pollinated accessions,
and approximately 50 percent less than control—pollinated
biparental cross yields. Slightly higher intraspecific
crossability was observed for Engelmann spruce than for
blue spruce among both open and control-pollinated

collections (Table 1).

25

The percent ungerminated-but-full seed yields, when
compared to percent germinated values, were proportionately
consistent across all pollination types for both species
(Table 1). There was only a slight proportionate increase
in full-but-ungerminated seed among the selfed crosses
relative to open-pollinated and biparental crosses.

All variance component estimates converged rapidly
using restricted maximum likelihood techniques. Due to the
small magnitude of the ungerminated-but-full male general
combining ability variance in Engelmann spruce, iterations
for this variance component were terminated after a change
of less than five percent rather than one percent. The
variance component and narrow-sense heritability estimates
are given in Table 2.

Among the blue spruce crosses, general combining
ability (GCA) variance accounted for only one percent of
the total variance observed in percent germination (Table
2). Maternal effects and specific combining ability (SCA)
variances greatly exceeded the GCA variance for this trait,
accounting for 15 and 56 percent of the total variance,
respectively. For percent germination in Engelmann spruce,
GCA varaince was also much smaller than SCA variance, each
accounting for one and 47 percent of the total variance in
this trait, respectively (Table 2). The most striking
difference observed between the species in regards to the

production of viable seed was that the maternal effects

26

Table 1. Mean values of percent germinated (Z Germ) and percent
ungerminated-but-full (Z Ungf) seed on a total seed basis
for Open-pollinated (Open) and control-pollinated (selfed
and biparen tal) col lec tions.

 

 

 

 

Qpen Self Biparental
Species ZGerm ZUngf ZGerm ZUngf ZGerm ZUngf
Blue spruce Mean 43.1 4.9 4.8 1.0 11.9 2.2
Engel. spruce MEan 48.7 8.0 9.5 2.3 19.7 3.6

 

Table 2. Variance component and narrow-sense heritability estimates
for production of viable and abnormal full-Bib seed in blue
and Engelmann spruce. Numbers in parentheses represent
percentages of the respective variance components as
compared to the total observed variance for that trait.

 

Variance Component

 

 

Traitf_, Speciesb GCA SCA Maternal Error h2

ZGerm BS 1.890 30.992 118.482 57.796 0.04
(1) (15) (56) (28)

ES 3.987 135.299 -0.025 150.086 0.06
(1) (47) ----- ° (52)

ZUngf BS 2.642 1.028 -o.357 5.711 1.12
(28) (11) ----- ° (61)

ES 0.687 2.712 2.834 15.000 0.12
(3) (13) (13) (71)

 

a ZGerm = percent germinated, ZUngf = percent ungerminated-but-full
seed on a total seed basis.
BS = blue spruce, ES = Engelmann spruce.

c Net calculated due to small, negative variance components (assumed
to be zero).

27

variance was essentially zero in Engelmann spruce. These
results suggest maternal influences are very important in
the production of viable seed in blue spruce but are not in
Engelmann spruce. In both species, sources of nonadditive
(dominance) variance apparently exert a great deal more
control over viable seed yields than do additive sources of
genetic variation, and this is reflected in the very small
narrow-sense heritability estimates for this trait in both
blue and Engelmann spruce (Table 2).

GCA and SCA variances were both relatively large for
percent ungerminated-but-full seed yields in blue spruce,
accounting for 28 and 11 percent, respectively, of the
total variance observed in this trait (Table 2). However,
GCA variance was disproportionately large, resulting in a
narrow-sense heritability estimate exceeding unity.
Maternal effects variance was essentially zero for this
trait in blue spruce. In Engelmann spruce, GCA variance
was relatively small, accounting for only three percent of
the total variance. However, both SCA and maternal effects
variance were moderately large and approximately equivalent
in magnitude for percent ungerminated-but-full seed yields
in Engelmann spruce, each accounting for 13 percent of the
total observed variance (Table 2). In both species, the
number of full but nonviable seed may be under the control
of nonadditive sources of variation as well. Only for
Engelmann spruce was there any evidence of maternal

influences in the production of ungerminated-but-full seed.

28

Product-moment correlations between the two
germination traits, calculated using parental means or GCA
estimates, were somewhat contrasting for the two species
(Table 3). Blue spruce parents—-serving as either female
or male-~which produced higher viable seed yields also
produced higher ungerminated-but-full seed yields.
However, for Engelmann spruce there was no correlation
between the two germination traits, or possibly a slight
negative association when Engelmann spruce parents served
as males.

Product-moment correlations between male and female
mean responses for percent germination in both species were
calculated. Both blue and Engelmann spruce indicated a
slight degree of positive association between male and
female responses on a parental mean basis--r = 0.28 for

blue spruce and 0.23 for Engelmann spruce.

DISCUSSION

In both blue and Engelmann spruce, the production of
viable seed is primarily under the control of nonadditive
(dominance) sources of genetic variation. A large amount
of the total variance observed in both species for percent
germination on a total seed basis was attributable to SCA
variance, while very little was accounted for by GCA

variance. Therefore, very little gain will be achieved in

29

 

 

 

Table 3. Product-moment correlations between percent germinated and
percent ungerminated-but-full values when parents served as
females or males, derived using parental means and
general combining ability evaluations (BS = blue spruce,

ES = Engelmann spruce).
Females Males
BS ES BS ES
Parental means 0.46 -0.01 0.24 -O.15
GCA estimates 0.40 -0.30 0.44 -0.33

 

30

the production of viable intraspecific seed in blue and
Engelmann spruce by using half-sib selection techniques.
Rather, selection must be based on specific cross
combinations which show high crossability rates as measured
by germination tests.

Morgenstern (1974) observed in a diallel cross of

seven black spruce (Picea mariana (Mill.) B.S.P.) parents

 

that SCA variance accounted for six times the total
variance observed for percent germination than did GCA
variance. Morgenstern attributed this to the fact that
percent germination and percent survival, for which similar
results were obtained, are fitness or survival traits,
which are less likely to be controlled by additive sources
of variation (Falconer 1960). Morgenstern (1974) observed
in the same study that for traits related to growth and
size-~rate of germination, first and second—year height-—
additive sources of variation were of primary importance.

In a six parent diallel cross of flax (Linum usitatissimum

 

I"), percent germination was also controlled primarily by
genes showing dominance (Gupta and Basak 1983). Both
additive and nonadditive sources of variation were
important in low-temperature germination percentages in

cucumber (Cucumis sativus Ln) (Wehner 1984). Additive

 

sources of variation were more important than nonadditive

sources in seed yield of tall fescue (Festuca arundinacea

 

Schreb.) (Nguyen and Sleper 1983). Therefore, while

31

nonadditive sources of variation do not exert major control
in the production of viable seed among all plant species,
results similar to those observed in this study have been
reported for a variety of other plant species.

Maternal influences were very important in the
production of viable seed in blue spruce, but of no
apparent importance in Engelmann spruce. Large maternal
influences in seed germination have also been observed in
black spruce (Morgenstern 1974), Virginia pine (Pinus

virginiana EH33") (Bramlett et al. 1983), and Douglas-fir

 

(Pseudotsuga xnenziesii (Mirb.) Franco) (Greathouse 1966).

 

 

Falconer (1960) has interpreted maternal effects as common
environmental influences of a given female parent.
Bramlett et al. (1983) have further subdivided maternal
effects for members of the Pinaceae family into the effect
of the local environment of the female parent and the
effect of the seedcoat and megagametophyte tissue-—both
maternally derived--of the germinating seed. The
megagametophyte seed tissue in members of Pinaceae is
haploid and identical in genetic composition to the female
egg nuclei. Therefore the only paternal influence in the
developing seed is through the embryo. Whether the
observed difference between blue and Engelmann spruce in
maternal influence of seed set is due only to how each
species responds to the local environment or species-
specific responses to seed coat and megagametophyte effects

is uncertain. As Bramlett et al. (1983) suggested, the

32

only way to test the relative importance of the two effects
would be to replicate given crosses in clonally derived
seed orchards located on a variety of sites. The effect of
local environment only on the production of viable seed
could be assessed by replicating given crosses in separate
seedling seed orchards located on two or more sites, each
orchard made up of the same half-sib or full-sib families.
The magnitude of maternal effects should be investigated
further for seed orchards of blue spruce.

The variance of female parental means for percent
germination in blue spruce was 75 percent greater than that
observed among male responses. In Engelmann spruce the
variance of female means exceeded that observed among males
by 37 percent. The large maternal influences observed in
blue spruce may account for some of the increased variance
observed among females of this species.

Selfing reduced viable seed yields approximately 50
percent relative to biparental controlled crosses in both
blue and Engelmann spruce. This is consistent with other
reports on the effect of selfing in blue spruce (Cram
1984). Conifers are generally intolerant of inbreeding,
resulting in reduced levels of seed set and poor survival
and growth of selfed seedlings (Franklin 1970; Shaw and
Allard 1981). While selfed crosses in both species did
have lower yields of viable seed, they did not produce

proportionately more ungerminated-but-full seed relative to

33

the proportions observed among biparental controlled
crosses. Therefore reductions in seed set due to selfing
in blue and Engelmann spruce were probably the result of
prezygotic incompatibility mechanisms that prevented normal
fertilization, as has been reported for other conifers
(Fechner 1979). Some loss of seed may occur between
fertilization and early embryo development (Cecich 1979;
Kossuth and Fechner 1973; Allen and Owens 1972). Both
prezygotic incompatibility and postzygotic inviability have
been observed among members of the Pinaceae family (Fechner
1979).

For some as yet unknown reason, there was a slight
tendency for blue spruce parents which produced more viable
seed on average to also produce greater numbers of
nonviable full seed. There was no such trend--or possibly
a slight reversal--observed among Engelmann spruce parents.
This species difference is apparently not due to the
maternal effects observed to influence seed set in blue
spruce, as the trend was consistent in both species when
the parent served as either male or female. Also, it will
not account for the slightly higher intraspecific
crossability observed for Engelmann spruce, as the
proportion of ungerminated-but-full seed relative to viable
seed was the same in both species for the respective

pollination types.

CHAPTER III

INHERITANCE OF ISOZYMES IN SEED AND BUD TISSUES OF BLUE AND
ENGELMANN SPRUCE

ABSTRACT

Thirteen loci from 11 enzyme systems wereridentified
among full—sib and half-sib progeny of blue and Engelmann
spruce. Eleven of the loci were expressed in bud, embryo
and megagametophyte tissue; the reamaining two loci were
expressed only in embryo and megagametophyte tissue. There
were no mobility differences observed between loci
expressed in seed and bud tissues.

The mode of inheritance for ten of the loci was
confirmed based on progeny genotypic distributions. For
the two loci not expressed in bud tissue, acid
phosphatase(2) (ACP(2)) and diaphorase(2) (DIA(2)),
inheritance was inferred from pooled segregation ratios of
megagametophytes from open-pollinated half-sib seed from
heterozygous females. The inheritance of glutamate
oxaloacetate transaminase(3) (GOT(3)) was also inferred
from segregation ratios and diploid embryo phenotypes of
open-pollinated progeny due to a lack of variability at
this locus among the 40 parents in the mating design. Two
loci, aldolase (ALD) and malate dehydrogenaSe (MDH(2)),

were monomorphic among the 20 parents of both species.

34

35

INTRODUCTION

Isoenzyme analysis is a useful tool in assessing the
population genetic structure of a species and its
evolutionary relatedness to other species. Allelic
frequencies can be determined for a large number of loci
which have various-~and generally unknown--degrees of
selective neutrality: 64L, 71 loci sampled in a study of
humans in Europe (Harris and Hopkinson 1972); 42 loci

sampled in a study of lodgepole pine (Pinus contorta

 

Dougla) in western North America (Wheeler and Guries 1982).
Gene flow, drift and mutations among subpopulations of a
species or between species can be quantified (exp, Millar
1983; Shaw and Allard 1981; Muller 1977; Mitton et al.
1977; Adams and Coutinho 1977). However, the mode of
isozyme inheritance must be known before one can accurately
estimate and interpret measures of variation, gene flow,
fitness, mutation or genetic relatedness.

The question of whether blue and Engelmann spruce

(Picea pungens Engelm. and g; engelmannii Parry ex Engelm.,

 

respectively) hybridize naturally has been studied
extensively but no clear conclusions have been drawn.
Morphological and biochemical data have proven relatively
effective in distinguishing the two species and trees of
intermediate phenotypes have been identified (Daubenmire

1972; Taylor et al. 1975; Mitton and Andalora 1981;

36

Schaefer and Hanover 1985a and b). Isozymes may also prove
useful in distinguishing hybrids between blue and Engelmann
spruce and subsequent introgression because they represent
more closely the variability at the DNA level, but first
the inheritance patterns of the isozymes must be
determined. The objective of this study was to determine
the inheritance patterns of isozymes in megagametophyte,
embryo and bud tissues in blue and Engelmann spruce.

The Dolores River drainage in southwestern Colorado
was selected as the study site because both blue and
Engelmann spruce occur in the drainage, often in close
proximity. Also, previous studies investigating the
genetic variabililty in morphological and terpenoid
characters of blue and Engelmann spruce have been conducted
in the drainage (Hanover 1975; Reed and Hanover 1983;

Schaefer and Hanover 1985a and b).

MATERIALS AND METHODS

The Dolores River was subjectively divided into five
elevational species-occupation zones. The parents used to
generate the full-sib progeny in this study were located in
the middle three zones (2,3 and 4). Zone 2 represents an
essentially pure blue spruce zone relative to the
occurrence of Engelmann spruce and extends from 2400 to

2590 meters (m) in elevation. Zone 3 is an elevationally

37

intermediate zone relative to the habitats of blue and
Engelmann spruce, extending from 2590 to 2770 m, with both
species present in this zone and often in close proximity.
Zone 4 is a predominately Engelmann spruce zone and extends
from 2770 to 2960 min elevation. Twenty trees of each
species—-ten blue spruce trees each in zones 2 and 3, and
ten Engelmann spruce trees each in zones 3 and 4--were
selected to serve as parents for the controlled
pollinations. Selection was based primarily on fecundity,
accessability and climbability. All parents were readily
identifiable as to species.

A partial diallel mating design was utilized to carry
out the pollinations in the field during the spring of
1983. Both intra- and interspecific crosses were made.
Only the intraspecific progeny will be evaluated in this
paper, and the results from the interspecific crosses will
be reported elsewhere ( Ernst et al. 1985c). For a more
detailed description of the partial diallel mating design
used in this study, see Ernst et al. (1985). Three
intraspecific crosses were made per female, including one
self and two biparental crosses. The biparental crosses
were reciprocated on the respective females.

In the fall of 1983, control-pollinated and Open-
pollinated cones were collected from each parent. The
cones were dried and seed extracted by hand. In the fall

of 1984, dormant vegetative buds were collected from each

38

parent tree and stored at -20°C until used in the
electrophoreth: analysis.

The extracted seed from the controlled crosses were
blown to remove empty seeds. Germination tests were
conducted with the blown seed during the summer of 1984
(see Ernst et al. (1985) for results of the germination
tests and measures of crossability for the intraspecific
crosses), and then each seedling was planted in a 5 x 5 x
25 cm plant band containing 3:1:1
(peat:vermiculite:perlite) soil mix. The seedlings were
grown under accelerated-optimal-growth conditions (Hanover
et al. 1976) from August, 1984, until January, 1985, when
the seedlings were allowed to go dormant. Dormant
vegetative buds were collected from the progeny seedlings
in March, 1985, and stored at -20°C until used in the
electrophoretic analysis.

Parental genotypes were determined by simultaneous
comparison of isozymes in bud, embryo and gametophyte
tissues. After partial germination--radic1e protruding
more than 3 mm—-and removal of the seed coat, the
megagametophyte and embryo from ten open-pollinated seeds
from each female were separated and each placed into a 0.5
ml polystyrene sample vial. Apical domes from five dormant
vegetative buds from the same female tree were placed into
separate sample vials. The sample vials were then put in
foam storage blocks, wrapped in plastic wrap and frozen at

-20‘C until the day they were used.

39

Because of the relatively small number of seed
available from each cross (family), only bud tissue from
the seedlings was used to characterize full-sib progeny
genotypes. They were determined using one dormant
vegetative bud from each full-sib seedling, with a maximum
of ten seedlings per family analyzed. Each bud was placed
in a 0.5 ml sample vial and stored at -20°C until used.

To determine the inheritance of loci not expressed in
bud tissue (i.e., expressed only in megagametophyte and/or
embryo tissue), isozymes were assayed using
megagametophytes from seed of open-pollinated half-sib
collections of blue and Engelmann spruce. These
collections were made in the Dolores River drainage,
primarily for use in another study (Ernst et al. 1985b).
Eight megagametophytes were prepared as previously
described and used in electrophoresis.

Table 1 gives the composition of the electrophoresis
buffers, power requirements, and references for each.
Table 2 lists the enzymes analyzed in each buffer system
and references for the staining recipes. Connaught starch,
lot 400-1 (Connaught Laboratories, Willowdale, Ontario,
Canada), was utilized throughout the study to prepare the
12.5 percent w/v starch gels. The thickness of the gel
varied with the buffer system used: 20 mm gels for buffer
system I and 16 mm gels for buffer system II. The
electrophoresis equipment and techniques were similar to

those described elsewhere (O'Malley et al. 1980).

40

In preparation for electrophoresis, the sample vials
containing bud, megagametophyte and embryo tissues were
removed from the freezer and placed in a plexiglass block
embedded in ice. Two or more drops of extraction buffer
(Wendel and Parks 1982) were placed in each vial and the
tissues homogenized using a motor-driven teflon grinding
tip. The homogenate was absorbed onto 2 x 20 mm filter
paper wicks (Whatman, No. 3) and inserted into the vertical
slice at the gel origin.

Electrophoresis was conducted in a forced—air
refrigerator at 4‘C, maintaining the amperage settings as
given in Table l. Wicks were removed after 15 minutes, and
then electrophoresis was continued for 5.0 and 4.5 hours,
respectively, for buffer systems I and II. These run
durations ensured a front migration of 8 cm, as determined
by bromophenol blue dye markers. The gels were then sliced
horizontally, immersed in staining solution and incubated
at 37‘C in the dark until resolved. The stain recipes used
were similar to those referenced in Table 2. All enzyme
systems assayed in this study migrated anodally--no
isozymes appeared in stained cathodal gel slices.

For enzyme systems controlled by more than one locus,
the fastest migrating zone was designated as locus l, the
next fastest 2, etc.. Multiple allozymes within each locus
were numbered in the same manner; the most anodal was
labeled as allele 1, etc.. Mobilities of the different

isozymes were quantified relative to the buffer front (Rf).

41

Table l. Electrophoresis buffers and respective power requirements.

 

Electrophoresis Buffer System

 

 

No. Electrode Gel Power
Ia 0.125 M Tris 0.05 M DL-histidine-HCl 50 ms
pH 7.0 with 1.0 M 1.40 mM EDTA
citric acid ph 7.0 with 1.0 M Tris
Dilute 5:1 (4 water:
1 stock)
IIb 0.029 M lithium hydroxide 102 electrode buffer 75 ms
0.19 M boric acid 0.05 M Tris
pH 8.3 with 10 N NaOH 0.0076 M citric acid

pH 8.3 with 10 N NaOH
Dilute 10:1 (9 water:
1 stock)

 

a Cheliak and Pitel 1984.
b Scandal ios 1969.

Table 2. Enzyme systems assayed in respective buffer systems.

 

 

 

Buffer System Enzyme E.C. No. Referencea
I ACO 4.2.1.3 l
ACP 3.1.3.2 l
ALD 4.1.2 13 2
CG? 1.1.1.49 2
IDH 1.1.1.42 1
MDH 1.1.1.37 l
6PG 1.1.1.44 2
SKDH 2
II DIA 1.6.4.3 1
GDH 1.4.1.3 I
GOT 2.6.1.1 2
PGI 5.3.1.9 2
PGM 2.7.5.1 2

 

a 1 = O°Malley et a1. (1980); 2 = Conkle et al. (1982).

42

The modes of inheritance of isozymes expressed only in
bud tissue were evaluated by comparing the observed
distribution of progeny genotypes to the expected
distribution--based on parental genotype designations-—
using the log-linear G-statistic (Sokal and Rohlf 1969).
For polymorphic loci which were expressed only in
megagametophyte-~and sometimes embryo-~tissueq inheritance
was determined by comparing pooled segregation ratios of
megagametophytes from half-sib open-pollinated families of
heterozygous mother trees using the same log-linear G-
statistic. The probability of misclassifying a
heterozygote at a given locus is (1/2)n'1, where n equals
the number of megagametophytes analyzed per family. In
this study, n = 8, and therefore the probability of

misclassifying a heterozygote was (1/2)7 = 0.0078.

RESULTS AND DISCUSSION

From a total 60 possible intraspecific full-sib
families for each species group, viable progeny were
obtained for 49 blue spruce and 56 Engelmann spruce
families. Of the 13 enzyme systems assayed (Table 2), 11
produced sufficiently clear band patterns using seed or bud
tissues to make putative locus and allele assignments; only

glucose—6-phosphate dehydrogenase (G6P) and shikimic

43

dehydrogenase (SKDH) did not. GGP was weakly staining in
all blue spruce tissues assayed, while in Engelmann spruce
only megagametophyte tissue produced clear band patterns.
The band patterns of G6P in blue and Engelamnn spruce

resembled those observed in ponderosa pine (Pinus ponderosa

 

Dougl. ex Laws.) (O'Malley et a1. 1979). SKDH produced
clear bands in all tissues, but no genetic interpretation
could be discerned. Another buffer system for this enzyme
may help achieve a genetically explainable separation of
electromorphs (e.g., see Conkle et a1. 1982).

Aconitase

 

One zone of activity was observed on gels stained for
ACO, with variation in this zone occurring in both blue and
Engelmann spruces Bands were best resolved in
megagametophyte tissue, but also resolved well in embryo
and bud tissue. The same two electromorphs were observed
in both species, and they are inherited as expected for a
one locus system (Table 3). Heterozygous diploid tissue
expressed a three-banded phenotype (Figure 1), indicating
the functional enzyme may be a dimer. The functional form
of ACO in humans is monomeric (Slaughter et al. 1975;
Zouros 1976), but its subunit structure has not been
determined in conifers (El-Kassaby et al. 1982; Guries and
Ledig 1978).
Acid phosphatase

Two zones of activity were observed on gels stained

for ACP. Only the slower of the two zones, ACP(2),

44

.Auv NyuN.o pew Awe qvco.o
Amp Nvoo.n use Awe ~v¢w.n

.ooaumu cause?» I a moo—Can 33 I mm A

u~o>u~ oucoomuacmwo _o.o
"~o>o~ oocuodudcwwo no.o

noun sovoouu mo noouwoe use a~o>o~ cocoomumcwmo oau cu ucwuconuouuoo ao:_w> U o
.caozm ac: and mucuouosu vac oauu coup u>n3~w cocoa-a ozowxnoao; coulomb noououu scum mcmu~auou acowOLm

 

 

 

 

adv¢~.o Ao.~mvan Ao.smvmm egg N— «\N x «\—
A~v~c.q Am.o_vm Ana.mv- Amu.mv- an c N\_ x N\_ mm
Amve~.N Am.odv- Am~.ovc Ans.ov- on a n\N x n\u
A~V_c.o Am.oxvow Am.o~voh an ad n\u x «\N mm Aevznr
Advnw.o Am.m~vo~ an.auvmm on o «\N u ~\~
A~vm9.~ Ac.¢nVOn Ac.s~vo~ Ac.-vm— no «u N\_ x «\—
Advuu.n Ao.~N~vom_ Ac.NN~vmo~ ecu mu ~\~ x _\— mm
A~v~c.o Am.omV~n an.onvon mu m «\N x «\u
A~v_o.o Ao.~_vm~ Ao.ovn Ao.ove aw m N\~ u N\~
A~vq~.~ Am.onvmn an.onvnc ~o~ N_ ~\~ I _\~ mm Anvzaz
A_voo.o Ao.c¢vo¢ Ac.o¢vo¢ co m «\N x N\~
Amvmo.n Ao.omen Ac.c~vmu Ao.m~vc~ as dd N\— x ~\~
A~v3_.~ Am.csvwo Am.¢nv~m oc~ n— ~\~ x ~\~ mm A~Vzom
A_v~o.o Ao.m~V—fi Ao.n~Vm~ on m «\m x n\n
A_vmo.o Am.¢~ve~ Am.c~vm_ ow n nxn x n\~ mm Anymom
A_vm¢.n A0.0mvon Ao.cnvg~ oo o «\N x «\—
A~v_n.c Am.-vo_ Ann.wvo An~.mvs mm m ~\~ x «\—
Aavoo.. Am.oo_veu_ Am.eo~v¢m ndu NN N\~ x _\~ mm zoo
Amy—m.o Ao.m_vm_ Am.mva Am.mv~_ an s Nx~ x «\—
A_V__.~ Ao.movmm Ao.movno— oau cu ~\~ x ~\~ mm
Auvom.o Amu.mvo~ Am~.wvn mm o n\_ x n\~
A_Vma.o Ao.emvmm mo~ Nfi n\— x ~\~ mm AAVUmo
A_voo.u Ao.~¢voc Ao.s¢vcn «a o— «\N x «\—
vamo.m Am.mvo Ann.~V— Am~.~vc - N N\~ x «\—
Amvmo._ Ao.mvm Ac.mvn o~ ~ ~\~ x _\~ mm Auvzau
~.v~c.L Ao.~ovcm Ao.~ovmo «Nd nu nxw w N\~
ANV._.o Ao.m~vo~ Ao.¢~vm~ Ao.c~vc~ on m ~\~ x ~\~
A_vwm.~ Am.o~vm~ Am.o~v¢m on o ~\— x _\~ mm
A_v~o.o Am.omen Am.onvoc mu m ~\~ x ~\~
A~v~¢.~ Am.muvm~ Amu.u~vo~ Ans.-vo_ ~m m N\_ x ~\—
Adv—o.o Am.snvnm Am.hmem nu c— N\~ x ~\~ mm ou<
oAuvvo ¢\m «\N ~\~ M\n N\~ ~\~ .1Nmmmoum ooenouo amouu quomoomm Ansoodv
.oz .02 ww onus osxucm
Auouoomxoveo>uomnonnnomhuocow Ncowoum
.acowOua

nau1~—=u museum ccuegowcm one on~n mo oncomu can ca vocauuaxo nonhuocuw we wanna acmuowouwoa esoo~10~wcmm

.m manna

45

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

.- a a a a a n a a n a n a n .33on

S S S S S S S S S S S S S S 3.105

S S S S S S S 3 S S S S S S .3102

c «({J 11) co

m .. u m nu an .n n .
.... .u .o m .miw "n dam .m. .. u... “mu .. n u.
.3on oomu Audits 33 new: a" I u I .0 am nan-am .H .m u”.
moESNoHHm so :owumsmwe czocx mo .

mosxnoafim Has: “communes flmoxoov umguem

moses xcmam .oossmm :cmEHowcm 0 mm

moospmm mafia 0 mm .mhoawposo: .
.a. mm was .msoEHoOHouo: acomopaou .

.3. mm ooxsms mocmm .monxaocom .
o>wuoodmou agony can possum 1
:cmsfiomcm ocm oan :w mEoumxm no.9

o m Hwoex . mesmH
25.8 S m as N H .a 5:92 5.; .3192
use a nun a unnnn an can: a unnumfluoam
a a 8 a. a. .3 8 a .a a .. a a n a .a.. n a 8.: 2.8 a
S S S S .3 S S 5. S S S S S S S S S S S S S S S. S S S. S S S 3 S S S S S mafia—LOO
“Ho
MM» on no I- 00000.1
0 n W I 0 0 I. a u 1
0 - - I. U . - u I- -
I . I
ImAv
|.un.l . em
10;.
3.105%... .529. 5.0.. 8:60 zoo REE Eons 5:9 9 3:3 oo<

 

 

 

 

 

 

46

produced resolvable bands, and then only in megagametophyte
and embryo tissue, but not in bud tissue. Therefore the
inheritance of this enzyme system was inferred from
segregation ratios of half-sib open-pollinated families
from heterozygous mother trees instead of bud tissue from
the seedling progeny.

In blue spruce, two allozymes were observed among the
twenty parents in the mating study, and an additional null
allele was found among the open-pollinated collections.

The banding patterns observed in embryo tissue for both
homozygotes and heterozygotes, except for individuals with
null alleles, are shown in Figure 1. A single large,
diffuse band of intermediate migration was observed in
heterozygous embryo tissue for ACP(2) rather than the
typical three-banded dimer pattern, possibly because there
was so little migration distance between the two allozymes
and staining for this enzyme system was relatively diffuse.
The intermediate band of the heterozygous phenotype concurs
with the dimeric structure of this enzyme as proposed for
Norway spruce (g; abies (L.) Karst.) by Lundkvist (1975).
The pooled segregation ratio of open-pollinated half-sib
progeny from heterozygous females infers that ACP(2) is
inherited as a one locus system (Table 4). Only two
individuals (mother trees) from the open-pollinated
collection possessed the null allele, and both of these
parents were heterozygotes as determined by segregation of

the haploid megagametophytes.

47

Table 4. Pooled segregation ratios observed in megagametophyte tissue
of seed collected from heterozygous mother trees in the
Dolores River drainage for two enzyme systems, and G-
statistic from goodness-of-fit test.

 

 

 

No. of Observed
Locus Speciesa Genotype Families Segregation C(l df)
ACP(2) BS 1/3 15 64:67 0.07
1/4 2 8:8 0.00
ES 1/2 3 12: 12 0.00
2/4 4 18:14- 0.50
DIA(2) BS 1/2 21 88:92 0.09

 

a BS 8 blue spruce; ES 8 Engelmann spruce.

48

Among the 20 Engelmann spruce parents in the mating
study, only one allozyme was observed, and it was
intermediate in mobility to the two non—null allozymes
observed in blue spruce (Figure 1). Among the open-
pollinated collections, parents were found which possessed
both the faster allozyme observed in blue spruce and a null
allozyme. The pooled segregation ratio for these
heterozygotes also infers ACP(2) is inherited as a one
locus system (Table 4). It must be noted that mobility
differences among the different ACP(2) allozymes in blue
and Engelmann spruce are relatively small, and further
study under a broader range of electrophoretic conditions
may disclose even more alleles than observed in this study.
Aldolase

There were four zones of activity for all tissue types
on gels stained for ALD, but only one zone produced clear
bands. This zone was monomorphic among all parents of both
species in the mating study (Figure 1; Table 3). Using the
same buffer system among five white spruce Qh_glauca
(Moench) Voss) parents, Cheliak and Pitel (1984) observed a
monomorphic zone at a similar migration distance. Also,
Wendel and Parks (1982) observed three monomorphic zones

for aldolase in Camelia japonica L.. The three non-

 

resolvable zones in blue and Engelmann spruce all appear to
be polymorphic, but unfortunately the staining in these

zones was too diffuse to score reliably.

49

Isocitrate dehydrogenase

 

Three zones of activity were observed on gels stained
for IDH. The fastest zone, IDH(1), did not produce clear
bands in any tissue and therefore was not scored. The two
slower zones, IDH(2) and IDH(3), were well resolved in
megagametophyte, embryo and bud tissue. Two allozymes were
observed at IDH(2) in Engelmann spruce (Figure 1), and they
are inherited as a single locus (Table 3). The three-
banded heterozygous phenotype suggests the functional form
of IDH(2) is a dimer. These results are consistent with
those presented for IDH in other conifer species (Cheliak
and Pitel 1984; Neale et al. 1984; Neale and Adams 1981;
O'Malley et al. 1979; Guries and Ledig 1978). Only one
allozyme was observed among the 20 blue spruce parents, all
being homozygous for the slower allozyme found in Engelmann
spruce (Figure 1). Electromorphs in the slowest of the
three zones, IDH(3), form heterodimers--as identified in
haploid megagametophyte tissue--with electromorphs in the
intermediate zone, IDH(2), but there were not enough
individuals variable at IDH(3) to deduce a mode of
inheritance.
6-phosphogluconate dehydrogenase

Two zones of activity were resolved for all tissue
types on gels stained for 6PG. The slower zone appears to
be controlled by two loci which form heterodimers much in

the same manner as IDH(2) and IDH(3). However, there were

50

not enough individuals variable in this slower zone to
determine the mode of inheritance.

Two bands were observed at the faster zone, 6PG(1),
among the 20 blue spruce parents. The heterozygotes
expressed a three-banded phenotype (Figure 1), indicating
the functional form of 6PG(1) is a dimer. The distribution
of progeny genotypes infers 6PG(1) is controlled as a
single locus system (Table 3). These results are
consistent with those presented for 6PG in other conifers
(Cheliak and Pitel 1984; Neale et al. 1984).

Two bands were also observed at 6PG(1) among the 20
Engelmann spruce parents, but only the faster allozyme
appears to be in common with that found in blue spruce.
The slower variant in Engelmann spruce migrates somewhat
faster than the slower allozyme in blue spruce (Figure 1).
Based on heterozygote intermediacy, the functional 6PG(1)
enzyme in Engelmann spruce is also a dimer, and the progeny
distributions indicate a one locus system (Table 3). As
observed for ACP(2), heterozygotes expressed a large,
diffuse band rather than the typical three-banded dimer
pattern, possibly because there was so little migration
distance between the two allozymes and the relatively
diffuse staining for this enzyme system.

Diaphorase

 

Several zones of activity were observed on gels
stained for DIA, but only one zone, DIA(2), produced clear

band patterns, and then only in megagametophyte and embryo

51

tissues. Therefore, the mode of inheritance of DIA(2) was
determined from pooled segregation ratios of open-
pollinated half-sib families from heterozygous mother
trees.

Two variants were observed at DIA(2) among the 20 blue
spruce parents, while Engelmann spruce was monomorphic for
the slower allozyme (Figure 1). The pooled segregation
ratio of half—sib families from heterozygous mother trees
(Table 4) infers a single locus mode of inheritance for
DIA(2). Heterozygotes were identified based on this
segregation of the allozymes in haploid megagametophyte
tissue. The heterozygote, as observed in embryo tissue,
produces an intermediate phenotype relative to the two
homozygotes, suggesting a multimeric structure for the
functional DIA(2) enzyme. Reports of DIA in Douglas-fir

(Pseudotsuga menziesii (Mirb.) Franco) (Neale et al. 1984;

 

El-Kassaby et al. 1982) and Camelia japonica L. (Wendel and
Parks 1982) have shown DIA to be a monomeric enzyme.
Therefore the multimeric structure suggested by the band
patterns observed in this study for DIA(2) may be
incorrect. However, the intermediate band pattern is
readily apparent, although diffuse, in diploid embryo
tissue of blue and Engelmann spruce, so further study is
warranted. Variability observed for DIA among full-sib
progeny of white spruce was not heritable (Cheliak and

Pitel 1984).

52

Glutamate’dehydrogenase

One zone of activity was observed on gels stained for
GDH, with good resolution for megagametophyte, embryo and
bud tissues. Two variants were observed among the 20
Engelmann spruce parents, while only one allozyme--
corresponding to the slower variant in Engelmann spruce--
was observed in blue spruce (Figure 1). Distributions of
the Engelmann spruce progeny infer GDH is inherited as a
one locus system (Table 3). The heterozygous phenotype is
intermediate in mobility and somewhat more»diffuse relative
to the two homozygotes, indicating GDH is functionally
multimeric. Heterozygotes were also identifiable based on
segregation of the allozymes in haploid megagametophyte
tissue. Similar results have been reported for GDH in
other conifers (Cheliak and Pitel 1984; Neale et a1. 1984;
Adams and Joly 1980; Mitton et al. 1979) and in maize
(Pryor 1974).
Glutamate oxaloacetate transaminase

Three zones of activity were observed on gels stained
for GOT. The two faster zones, GOT(1) and GOT(2), were
observed across all tissue types, while the slowest zone,
GOT(3), was best resolved in megagametophyte tissue, but
also with good definition in embryo and bud tissue. There
was insufficient variability in GOT(1) and GOT(2) among the
20 parental trees of either species to determine the mode

of inheritance of these putative loci.

53

Triple-banded allozymes were observed at GOT(3) in
blue and Engelmann spruce haploid megagametophyte tissue,
the blue spruce phenotype migrating slower--approximately
0.1 Rf unit-—relative to that of Engelmann spruce. Double-
banded allozymes were observed at GOT(3) in embryo and bud
tissue for both species (Figure 1). The double and triple-
banded allozymes are apparently the product of a single
allele, as they are inherited as a single unit. Possibly
the multiple bands represent post-translational
modification products of a single allozyme (Finnerty and
Johnson 1979; Newton 1979). Double and triple-banded
allozymes have been observed for GOT in eastern white pine

(Pinus strobus Lu) (Eckert et a1. 1981), balsam fir (Abies

 

balsamea (Linn.) Mill.) (Neale and Adams 1981), loblolly
pine (_P_. taeda L.) (Adams and Joly 1980), pitch pine (_P_._
rigida Pull”) (Guries and Ledig 1978), and Scotch pine (P;

sylvestris L.) (Rudin and Ekberg 1978).

 

A single open-pollinated half-sib blue spruce family
was heterozygous (segregating at a 5:3 ratio).
Heterozygous embryos from this female produced a phenotype
clearly indicating that GOT(3) is functionally dimeric,
consistent with results reported for GOT in other species
(Neale et al. 1984; El-Kassaby et a1. 1982; Wendel and
Parks 1982; OHMalley et al. 1979; Guries and Ledig 1978).
Only the slower migrating bands of the double-banded
allozymes stained well enough in the heterozygote to

produce a clear dimer pattern (Figure 1). The staining of

54

the faster bands in the heterozygote was too diffuse to
discern any band patterns.
Phosphoglucose isomerase

Gels stained for PGI exhibited two zones of activity,
but the more anodal zone, PGI(1), did not produce
sufficiently clear bands to score. The more cathodal zone,
PGI(2), resolved well in all tissues, and exhibited three
phenotypes among the 20 blue spruce parents but only one
phenotype among the 20 Engelmann spruce parents. Three and
four-banded allozymes were observed in megagametophyte
tissue, while one, two and three—banded phenotypes were
observed in embryo and bud tissue from homozygous
individuals (Figure 1). As suggested for GOT(3), the
multiple-banded allozymes for PGI(2) may be the result of
post-translational modifications of these allozymes
(Finnerty and Johnson 1979; Newton 1979). The multiple-
banded nature of these allozymes were manifest in the
dimers as well (thure 1%. The multibanded allozymes
observed in this study for PGI(2) are consistent with those
observed in Douoglas-fir (Neale et a1. 1984).

The progeny distributions for PGI(2) in blue spruce
infer a single locus mode of inheritance (Table 3). The
heterozygous phenotypes (Figure 1) are consistent with
reports for other species that PGI(2) is functionally
dimeric (Cheliak and Pitel 1984; Neale et al. 1984; Adams

and Joly 1980; Mitton et al. 1979; Guries and Ledig 1978).

55

Phosphoglucomutase

 

Two zones of activity were observed on gels stained
for PGM, but only the fastest zone, PGM(1), was
consistently resolvable among all three tissue types. The
20 Engelmann spruce parents expressed two alleles, and
progeny distributions infer a one locus mode of inheritance
for PGM(1) (Table 3). The heterozygous phenotype exhibits
both bands found in the respective homozygous phenotypes
(Figure 1), indicating that PGM(1) is functionally
monomeric. These results are consistent with those
reported for other conifers (Cheliak and Pitel 1984; Neale
et al. 1984;14itton et a1. 1979).

Only one allozyme, the slower variant observed in
Engelmann spruce, was observed among the 20 blue spruce
parents. One blue spruce parent exhibited segregation of
an electromorph somewhat intermediate to the two allozymes
observed in Engelmann spruce, but was not expressed in
embryo or bud tissue; the diploid phenotype resembled that
of the other 19 blue spruce parents. Therefore this
individual was scored as a homozygote corresponding to the
slower allozyme.

Malate dehydrogenase

 

Four zones of activity were observed on gels stained
for MDH, all zones equally resolvable in megagametophyte,
embryo and bud tissues. The most anodally migrating zone,
MDH(1), is double-banded, and only one of the Engelmann

spruce parents segregated at this putative locus. It was

56

not possible to accurately determine the mode of
inheritance of MDH(1) based on the progeny of only one
individual and therefore will require further study. The
second most anodal zone, MDH(Z), was monomorphic among both
sets of parents, and is represented by a single-banded
phenotype (Figure 1). MDH(Z) in blue and Engelmann spruce
is similar to the MDH(1) locus as described for several
other conifer speices by El-Kassaby (1981).

The third most anodal zone, MDH(3), was variable among
both sets of parents. Progeny distributions of
intraspecific crosses for both species show MDH(3) to be
inherited as a single locus system (Table 3). The
heterozygous phenotype exhibits three bands, including a
band intermediate in migration to the two homozygotes
(Figure 1), indicating MDH(3) is functionally dimeric in
blue and Engelmann spruce. The banding patterns of MDH are
relatively complicated because MDH(3) and MDH(4) form
heterodimers (Figure 1), which is consistent with
observations in other conifer species (El-Kassaby et a1.
1982; El-Kassaby 1981; CPMalley et a1. 1979; Guries and
Ledig 1978).

The most cathodal zone, MDH(4), is the most variable
and complicated of the four. Two allozymes were observed
among the 20 Engelmann spruce parents, the most anodal of
the two, MDH(4)—l, being null (Figure l). The functional
MDH(4) enzyme is also a dimer and forms heterodimers with

the allozymes of the MDH(3) locus. However, in the

57

the allozymes of the MDH(3) locus. However, in the
heterozygote MDH(4)-l/2, MDH(4)-2 appears to have a higher
affinity for allozymes of the MDH(3) locus than for
allozymes of the MDH(4) locus (Figure 1). Also, the higher
affinity of MDH(4)-2 precludes MDH(4)-l from forming a
heterodimer with MDH(3) allozymes. Therefore, in MDH(4)-
1/2 heterozygotes, MDH(4)—2 heterodimers apparently form at
the expense of homodimers at this locus and MDH(4)—1
heterodimers.

The 20 blue spruce parents also expressed two alleles
for MDH(4) (Figure 1), including the more cathodal allozyme
found in Engelmann spruce, MDH(4)-2, and an even slower
migrating allozyme, MDH(4)-3. Intralocus and interlocus
interaction of MDH(4) heterozygotes results in a wide array
of band patterns (Figure 1). Heterozygotes at this locus
may express (i) only the heterodimer(s), (ii) the
heterodimer(s) and intralocus dimer, or (iii)
heterodimer(s), dimer and homozygous bands. As yet we have
no explanation for the inconsistency in expression of the
allozymes at this locus in the heterozygous condition for
both blue and Engelmann spruce, but progeny distributions
of both species for MDH(4) show it to be inherited as a
single locus system (Table 3), indicating our
interpretation of the phenotypes is probably correct.

There were no mobility differences observed among all
tissue types for loci expressed in megagametophyte, embryo

and bud tissues. The same observation was made in white

58

spruce (Cheliak and Pitel 1984), but mobility differences
were observed for loci when expressed in embryo and needle
tissue of Douglas-fir (Neale et al. 1984). The same
extraction buffer was used for all tissues in the present
study to minimize mobility differences due to preparative
procedures.

A total of five electrophoresis buffers were screened
before selecting the final buffers to be used in this
study. The two selected buffers resulted in far superior
resolution for a maximum number of enzyme systems. The
remaining three buffers which were not reported in this
study include a morpholine citrate buffer (Clayton and
Tretiak 1972), another variation of the tris
citrate/lithium borate buffer (Ridgway et a1. 1970), and a
tris citrate buffer (Nichols and Ruddle 1973).

A total of 26 different enzyme systems were screened
initially in blue and Engelmann spruce, of which only the
13 listed produced consistent resolution. The remaining 13
enzymes which did not produce consistent resolution are
alcohol dehydrogenase (ADH), fluorescent esterase (FLE),
fructose—1,6-diphosphatase (FDP), fumarase (FUM),
superoxide dismutase (SOD), glycerate-Z-dehydrogenase
(GZD), malic enzyme (ME), menadione reductase (MNR--the
same as diaphorase, DIA), glutathione reductase (GLR--
equivalent to the faster electromorphs of diaphorase),

mannose-6-phosphate isomerase (MPI), sorbitol dehydrogenase

59

(SDH), uridine diphosphoglucose pyrophosphorylase (UDP),
and glyceraldehyde-phosphate dehydrogenase (GPD). These
are listed merely as background information for others
interested in assaying these enzymes in blue and Engelmann

spruce in hopes that resolution can be improved.

CHAPTER IV

ALLOZYME VARIATION OF BLUE AND ENGELMANN SPRUCE IN
SOUTHWESTERN COLORADO

ABSTRACT

Open-pollinated single-tree cone collections of blue
and Engelmann spruce were made in the Dolores River
drainage in southwestern Colorado during the fall of 1983
from three elevational subpopulations of each species,
including a zone where both species were present. Thirteen
isozyme loci were assayed using megagametophytes from the
half-sib seed collections. Fifty-four percent and 62
percent polymorphic loci (0.99 criterion) were observed for
blue and Engelmann spruce, respectively. An average of 1.6
alleles per locus was observed in beth species. Based on
observed allele frequencies, average expected
heterozygosities were 0.193 and 0.203 for blue and
Engelmann spruce, respectively. Observed genotypic
distributions at all loci conformed to Hardy-Weinberg
expectations, indicating both populations are panmictic.
Seventeen species-specific alleles were observed between blue
and Engelmann spruce, and strong frequency differences were
also observed between the two species at seven of the 13

loci.

60

61

Average genetic distance estimates indicated very
little intraspecific genetic differentiation in the Dolores
River drainage, while high degrees of divergence were
observed between blue and Engelmann spruce. The greatest
degree of genetic divergence between blue and Engelmann
spruce was observed among the respective species
subpopulations in the zone of overlap where both species
were present. This suggests the two species do not
hybridize naturally, or at least only very rarely (beyond
the limits of the sample sizes in this study), because if
they did this intermediate zone of overlap should indicate
a convergence in their respective genetic compositions.

The average genetic distance estimate between blue and
Engelmann spruce (0.46) is comparable to sibling species or
morphologically distinct species, indicating both
prezygotic and postzygotic reproductive isolating
mechanisms may be functioning to maintain the observed

species integrity.

INTRODUCTION

Allozyme variation is readily observable and
quantifiable at a large number of loci thanks to the rapid
development of electrophoretic and associated biometrical
techniques. Measures of isozyme variability and its

partitioning--tng., F-statistics (Wright 1951, 1965; Nei

62

1977) and genetic differentiation (Nei 1972, 1978; Rogers
1972)--are readily applicable to isozyme data and allow the
observed variability within a species or between species to
be partitioned according to various levels of population
structure. These measures may also indicate how long two
or more species have been separated phylogenetically
(Sarich 1977) and the stage of speciation they exhibit
(Ayala 1975, 1982).

The possibility of natural hybridization between blue
and Engelmann spruce (£1233 pungens Engelm. and g;

engelmannii Parry ex Engelm., respectively) is interesting

 

in regards to speciation and reproductive isolation in
conifers. The two species are thought to be closely
related phylogenetically (Daubenmire 1972; Nienstaedt and
Teich 1972). While they are somewhat similar
morphologically, a combination of traits will usually
distinguish the two species (Schaefer and Hanover 1985a;
Jones and Bernard 1977). There is some degree of
elevational overlap in their respective habitats with ample
opportunity for cross-pollination, and the results of many
studies have suggested natural hybrids between blue and
Engelmann spruce may exist, although no natural hybrids
have been identified based on morphological and biochemical
traits (eéh: Habeck and weaver 1969; Daubenmire 1972;
Taylor et al. 1975; Mitton and Andalora 1981; Schaefer and

Hanover 1985a and b). A few artificial hybrids have been

63

produced and indicated the crossability between the two
species is very low (Fechner and Clark 1969; Kossuth and
Fechner 1973). Isozymes represent more closely the
variability at the DNA level relative to morphological and
biochemical traits and therefore may be more useful in
distinguishing the two species and their hybrids. The
objective of this study was to describe and quantify the
observed variability among 13 loci both within and between
blue and Engelmann spruce in a drainage in southwestern
Colorado as part of a broader study to determine if natural

interspecific hybrids do exist.

MATERIALS AND METHODS

During the fall of 1983, open-pollinated cones were
collected from individual blue and Engelmann spruce trees
in the Dolores River drainage in southwestern Colorado.
The Dolores River drainage encompasses elevationally the
habitats of both blue and Engelmann spruce and there exist
many sites in the drainage where both species occur in
close proximity. The drainage is easily accessible, and
genetic studies investigating morphological and terpenoid
variability of the two species have been conducted
previously in the study area (Hanover 1975; Reed and

Hanover 1983; Schaefer and Hanover 1985a and b).

64

Before collections were made in 1983, the Dolores
River and some of its tributaries were subjectively divided
into five elevational species-occupation zones: Zone 1 -
the zone of lowest elevation, extending from 2225 to 2400
meters (m), and in which blue spruce occurs but Engelmann
spruce does not; Zone 2 - extending from 2400 to 2590 m in
elevation, and where blue spruce is almost exclusively
predominant relative to the occurrence of Engelmann spruce,
but scattered individuals of Engelmann spruce are present
on the adjacent hillsides of north aspect; Zone 3 - an
elevationally intermediate zone relative to the habitats of
blue and Engelmann spruce, extending from 2590 to 2770 m,
and where both species are present and often in close
proximity; Zone 4 - extending from 2770 to 2960 m in
elevation, and where Engelmann spruce is almost exclusively
predominant but with a few blue spruce individuals present,
primarily on hillsides of south aspect; Zone 5 - the zone
of highest elevation, extending from 2960 to 3140 m and in
which Engelmann spruce is present but blue spruce is not.
The breakdown of single-tree collections made in each zone

during the fall of 1983 is as follows:

 

 

Species Zone 1 Zone 2 Zone 3 Zone 4 Zone 5 Total
Blue spruce 16 22 18 56
Engelmann spruce 23 31 22 76

The cones were kept separate by mother tree and taken back

to the nursery to be dried and the seed extracted and

65

blown. Collections were made in all five zones along the

Dolores River and at a site in zone 3 of Scotch Creek, a
tributary of the Dolores River. Both blue and Engelmann
spruce occur at the Scotch Creek site, often side-by-side,
and pollen shed and female strobilus receptivity are
coincident among both species in the spring.

Isozymic genotypes of the mother trees were determined
using haploid megagametophyte tissue from germinated seed,
eight megagametophytes per parent. This sample size gives
a probability of (1/2)7 = 0.0078 of misclassifying a
heterozygote at a given locus. The inheritance of the
respective loci are described elsewhere (Ernst et al.
1985a). The individual megagametophytes were placed in
0.5ml sample vials, and these were put in foam storage
blocks, wrapped in plastic wrap, and frozen at -20‘C until
used.

Electrophoretic conditions were as described elsewhere
(Ernst et al. 1985a). A total of 13 loci from 11 enzyme
systems were analyzed: Aconitase (ACO), acid phosphatase
(ACP(2)), isocitrate dehydrogenase (IDH(2)), 6-
phosphogluconate dehydrogenase (6PG(1)), diaphorase
(DIA(2)), glutamate dehydrogenase (GDH), glutamate
oxaloacetate transaminase (GOT(3)), phosphoglucose
isomerase (PGI(2)), phophoglucomutase (PGM(1)), and malate
dehydrogenase (MDH(Z), MDH(3), and MDH(4)).

Prior to electrophoresis, the sample vials were

removed from the freezer and placed in a plexiglass block

66

embedded in ice. Two drops of extraction buffer (Wendel
and Parks 1982) were placed in each vial and the
megagametophyte homogenized using a motor-driven teflon
grinding tip. Filter paper wicks (Whatman, No. 3) were
used to absorb the homogenate and these were inserted at
the vertically sliced gel origin.

Multiple locus enzyme systems were scored with the
fastest—-most anodally--migrating zone labeled as locus 1,
next fastest as locus 2, etc. Multiple allozymes at a
locus were numbered in the same manner, with the fastest
allozyme»labeled as allele 1, etc.. Mobilities of the
different allozymes presented here have been quantified--
Rf-values--and are presented elsewhere (Ernst et al. 1985a)
as diagramatic zymograms of the loci as expressed in
diploid tissue.

Allele frequencies were determined for each of the six
subpopulations, three for each species, and for the species
groups in total. Observed genotypic ratios were compared
to those expected under Hardy-Weinberg equilibrium
conditions using the log—linear G statistic (Sokal and
Rohlf 1969). Expected heterozygosities for each locus
corrected for small sample size, average expected
heterozygosities and their standard errors, genetic
distance estimates for small sample sizes and their
standard errors (Nei 1972, 1978; Nei and Roychoudhury

1974), and Rogers' coefficient of distance (Rogers 1972)

67

were computed using the program written by Dowling and

Moore (1984).

RESULTS

Of the 13 loci surveyed among the two species in the
Dolores River drainage, seven were found to be polymorphic
in blue spruce and eight in Engelmann spruce (Table 1).
Only two loci, ALD and MDH(2), were monomorphic across both
species. The remainder of the monomorphic loci within each
species were either fixed for opposing alleles--GOT(3)--or
fixed in one species and polymorphic in the other--IDH(2),
DIA(2), GDH, PGI(2), PGM(1)--based on a presence frequency
of 0.01 or greater (0.99 criterion). Rare alleles were
observed which would render some of these loci polymorphic
at less restrictive criteria (Table 1), but the two species
are still quite different in allelic composition at these
and other loci. For example, strong frequency differences
occur between the two species at ACP(2), where allele 1
predominates in blue sprucebut is somewhat rare in
Engelmann spruce, while allele 2 predominates in Engelmann
spruce but was not observed in blue spruce (Table 1). The
average number of alleles per locus and percent polymorphic
loci at 0.99 and absolute criterion are listed in Table 1.

Blue spruce possessed a greater number of rare

alleles--frequency less than 0.01-~than Engelmann spruce

68

 

 

 

 

 

 

 

 

 

 

 

Table 1. Allelic frequencies and expected heterozygosities (corrected
for small sample size) among the respective loci and overall
for blue and Engelmann spruce Open-pollinated collections in
the Dolores River drainage in southwestern Colorado “L99
criterion).

No. Allele Std.
Locus Speciesa Families 1 2 3 4 H error
ACO BS 56 0.652 0.348 0.458
ES 76 0.461 0.539 0.499
ACP(2) BS 56 0.77 ----- 0.205 0.018b 0.357
ES 76 0.033 0.941 ---- 0.026 0.114
ALD BS 56 1.000 0.000
ES 76 1.000 0.000
IDH(2) BS 54 ----- 1.000 *C--- 0.000
ES 76 0.165 0.835 ---- 0.277
6PG(1) BS 55 0.786 ----- 0.214 0.340
ES 75 0.900 0.100 ----- 0.181
DIA(2) BS 55 0.282 0.718 0.408
ES 76 ----- 1.000 0.000
GDH BS 55 ---- 1.000 **--- 0.000
ES 76 0.730 0.270 ---- 0.397
GOT(3) BS 53 *b--- 1.000 *b--- 0.000
ES 76 l 000 0.000
PGI(2) BS 55 ----- 0.036 0.964 *b--- 0.071
ES 74 l 000 *b--- 0.000
PGM(1) BS 56 ----- 1.000 0.000
ES 76 0.592 0.408 0.486
MDH(Z) BS 56 1.000 0.000
ES 76 1.000 0s000
MDH(3) BS 56 0.446 0.554 ----- 0.499
ES 74 0.689 0.311 *b--- 0.431
MDH(4) BS 54 *b--- 0.750 0.250 0.378
ES 73 0.144 0.856 ----- 0.248
Average BS 55.2 0.193 0.058
ES 75.4 0.203 0.055
Average no. alleleskper locus Percent polymorphic loci
Criterion Criterion
0.99 Absolute 0.99 Absolute
BS 1.62 2.08 BS 54% 77%
ES 1.69 1.85 ES 621 69%

 

a BS 8 blue spruce; E3 = Engelmann spruce.
Null allele.

c * designates an allele present at a frequency less than 0.01 «L99
criterion).

69

among the individuals surveyed in the Dolores River
drainage (Table 1). The slight disparity in sample sizes
between the two species groups may account for some of this
difference, although the larger sample size for Engelmann
spruce should result in a greater number of rare alleles
sampled relative to blue spruce.

Expected heterozygosities ranged from 0.000 to 0.499
among the loci of the two species groups, with Engelmann
spruce only slightly more variable than blue spruce at the
0.99 criterion (Table 1). Observed genotypic frequencies
at the polymorphic loci of the two species did not deviate
significantly from those expected, indicating the two
populations conform to Hardy-Weinberg equilibrium. When
broken down into subpopulations (by zone, Tables 2 and 3),
observed genotypic frequencies did deviate significantly
from those expected for the ACO locus in the zone 2
subpopulation of blue spruce and the MDH(3) locus in the
zone 5 subpopulation of Engelmann spruce. However, these
were the only significant departures from Hardy-Weinberg
equilibrium among the loci of the respective
subpopulations.

Some elevational trends were evident in allelic
frequencies among the subpopulations (Tables 2 and 3). In
blue spruce, a fairly marked elevational change occurs for
DIA(2) (Table 2). More subtle elevational changes occur

for ACO, 6PG(1) amd MDH(3). Expected heterozygosities of

70

 

 

 

 

Table 2. Allelic frequencies and expected heterozygosities (corrected
for small sample size) for the seven polymorphic loci
observed among three blue spruce subpopulations from the
Dolores River drainage in southwestern Colorado.

No. Alleleb
Locus Zonea Families 1 2 3 4 H error
A00 1 16 0.750 0.250 0.387
2 22 0.591 0.409 0.495d
3 18 0.639 0.361 0.475
ACP(2) l 16 0.781 ---- 0.188 0.031 0.365
2 22 0.682 ---- 0.273 0.023 0.471
3 18 0.861 ---- 0.139 ---- 0.246
6PG(1) 1 16 0.750 ---- 0.250 0.387
2 22 0.795 ---- 0.205 0.334
3 18 0.833 ---- 0.167 0.286
DIA(2) 1 16 0.313 0.687 0.444
2 22 0.364 0.636 0.474
3 18 0 750 0.250 0.386
PGI(2) 1 16 ----- 0.031 0.969 0.062
2 21 ----- 0.048 0.952 *°--- 0.094
3 18 ----- 0.028 0.972 0.056
MDH(3) 1 16 0.375 0.625 0.484
2 22 0.455 0.545 0.499
3 18 0.500 0.500 0.500
MDH(4) 1 16 *h--- 0.800 0.200 0.330
2 22 ----- 0.727 0.273 .0.406
3 l7 ----- 0.735 0.265 0.401
Average8 1 15.9 0.189 0.057
2 21.9 0.214 0.064
3 17.9 0.182 0.057
a See text.
b

C

d

e

Alleles shown based on 0.99 criterion for total species sample.

* designates an allele present at a frequency less than 01n.in the
total species sample.
Only for ACO (zone 2) did genotypic ratios deviate significantly

(0.05) from those expected.
Includes monomorphic loci.

71

 

 

 

 

 

Table 3. Allelic frequencies and expected heterozygosities (corrected
for small sample size) for the eight polymorphic loci
observed among three Engelmann spruce subpopulations from
the Dolores River drainage in southwestern Colorado.

No. Alleleb
Locus Zonea Families 1 2 3 4 H error
A00 3 23 0.543 0.457 0.499
4 31 0.468 0.532 0.499
5 22 0.364 0.636 0.474
ACP(2) 3 23 ----- 1.000 ---- ---- 0.000
4 31 ----- 0.968 ---- 0.032 0.063
5 22 0 114 0.814 ---- 0.045 0.284
IDH(2) 3 23 0.109 0.891 0.199
4 31 0.210 0.790 0.337
5 22 0.159 0.841 0.274
6PG(1) 3 22 0.932 0.068 0.130
4 31 0.871 0.129 0.228
5 22 0.909 0.091 0.169
GDH 3 23 0.696 0.304 0.433
4 31 0.710 0.290 0.419
5 22 0.795 0.205 0.337
PGM(1) 3 23 0.630 0.370 0.477
4 31 0.661 0.339 0.456
5 22 0.455 0.545 0.499
MDH(3) 3 22 0.682 0.318 0.444
4 30 0.800 0.200 *°--- 0.325
5 22 0.545 0.455 0.499d
MDH(4) 3 21 0.143 0.857 0.251
4 31 0.129 0.871 0.228
5 21 0.167 0.833 0.285
Averagee 3 22.5 0.188 0.058
4 30.9 0.197 0.054
5 21.9 0.218 0.057
a
b See text.

C

d

e

Alleles shown based on 0.99 criterion for total species sample.

* designates an allele present at a frequency less than 0.01 in the
total species sample.
Only for MDH(3) (zone 5) did genotypic ratios deviate significantly

(0.05) from those expected.
Includes monomorphic loci.

72

the Engelmann spruce subpopulations tended to be more
variable with increasing elevation (Table 3). For
Engelmann spruce, consistent elevational shifts in allelic
frequency were observed at ACO and GDH (Table 3).

Nei‘s (1978) and Rogers‘ (1972) measures of genetic
distance are presented in Table 4. While the relative
values within each are essentially equivalent between the
two techniques, the Nei estimates are consistently greater
than the Rogers estimates. This was expected based on the
results of other empirical studies (Futuyma 1979). The
large values obtained for comparisons among blue and
Engelmann spruce subpopulations and between the two species
groups (Table 4) indicate the strong genetic divergence of
these two species which are thought to be closely related
and possibly hybridize naturally. Distances between
subpopulations within each species are very small, with the
estimates exceeded by their respective standard errors

(Table 4).

DISCUSSION

The observed percentages of polymorphic loci (0.99
criterion) of 54 percent and 62 percent, respectively, for
blue and Engelmann spruce correspond well to a mean value
of 67 percent reported in a survey of 20 conifer species

(Hamrick et al. 1981). In the present study, 13 loci were

73

 

 

Table 4. Nei’s (1978) corrected genetic distance estimates (below
diagonal) and their standard errors (in parentheses), and
Rogers’ (1972) distance coefficients (above diagonal) among
the six blue and Engelmann spruce suprpulations.

B31 B82 BS3 E33 E84 E85

BS 13 0.0039 0.0687 0.4274 0.4513 0.4183

BS 2 0.0000 0.0541 0.4112 0.4358 0.4020
(0.0216)

BS 3 0.0155 0.0109 0.4432 0.4685 0.4342
(0.0221) (0.0189)

ES 3 0.4671 0.4603 0.5308 0.0343 0.0638
(0.2024) (0.2013) (0.2191)

ES 4 0.4993 0.4866 0.5587 0.0000 0.0691
(0.2054) (0.2034) (0.2210) (0.0098)

ES 5 0.4444 0.4338 0.5057 0.0052 0.0089

(0.1941) (0.1937) (0.2095) (0.0117) (0.0122)

 

a BS 8 blue spruce; ES 8 Engelmann spruce; numerals are zone
designations.

74

assayed among blue and Engelmann spruce individuals
comprising essentially single populations of each species.
The individual studies which made up the survey (Hamrick et
al. 1981) assayed an average of 20 loci per species and
sampled anywhere from one to 34 populations. In general,
higher amounts of isozyme variability have been observed in
conifers than in dicots or monocots, possibly because of
the life history characteristics of conifers (Hamrick et
al. 1981; Shaw and Allard 1981). Conifers rely primarily
on outcrossing and are relatively intolerant of selfing,
rely upon wind for pollination and seed dispersal, exhibit
high fecundity, are generally widespread in distribution
and have long generation intervals. The same survey
(Hamrick et al. 1981) reported an average of 2.2 alleles
per locus among the 20 conifer species, slightly greater
than the 1.6 alleles per locus observed in this study for
both blue and Engelmann spruce.

The average expected heterozygosities--the average
number of heterozygous loci per individual--of 0.193 for
blue spruce and 0.203 for Engelmann spruce were higher than
some estimates reported for other conifers; 0.116 for

lodgepole pine (Pinus contorta Dougl.) (Wheeler and Guries

 

1982), 0.157 for Douglas—fir (Pseudotsuga menziesii (Mier
Franco) (Yeh and O'Malley 1980), 0.146 for pitch pine (P;-
rigida PMJJu) (Guries and Ledig 1981), 0.123 for ponderosa

pine (& ponderosa Dougl. ex Laws.) (O'Malley et al. 1979)

 

75

and 0.147 for Sitka spruce (Picea sitchensis (Bong.) Carr.)

 

(Yeh and El-Kassaby 1979). The slightly higher estimates
for blue and Eng elmann spruce may be due to the fewer
number of loci sampled in this study (13) relative to those
cited above (20 to 42 loci) (Leigh Brown and Langley 1979),
the number of progeny analyzed per tree and the number of
trees sampled per population (Morris and Spieth 1978), or
simply because blue and Engelmann spruce are more variable.
As mentioned previously, the percentage of polymorphic loci
observed in this study was consistent with other conifer
species, and the number of alleles per locus was less than
generally reported elsewhere. Therefore, the loci sampled
in this study must be nearer to allelic equilibrium--
frequencies approaching 0.5 in a two allele system, with
associated higher expected heterozygosity--than those
assayed in other species; iJL, fewer "common" alleles.
Blue spruce may exhibit even more "latent" variability than
Engelmann spruce based on the large number of rare alleles
observed in this species.

The high degree of conformity to Hardy-Weinberg
equilibrium expectations among the sampled loci suggests
the two species and the sampled subpopulations are randomly
mating with sufficient amounts of gene flow to minimize the
effects of genetic drift or selection. This appears to be
true in general for conifers (Wheeler and Guries 1982),
possibly because of their life history characteristics

(Hamrick et al. 1981), as mentioned previously. Only

76

slight trends were observed between subpopulations within
each species at the individual loci (Tables 2 and 3), but
Engelmann spruce subpopulations did exhibit greater
heterozygosities with increasing elevation, indicating the
higher elevation populations of this species may be more
diverse.

Blue and Engelmann spruce in the Dolores River
drainage possess some marked differences in allelic
frequencies and even some species-specific alleles. GOT(3)
was essentially fixed for opposing alleles between the two
species, although one blue spruce individual was
heterozygous for both alleles. The alleles coding for the
slower allozymes at IDH(2), GDH, and PGM(1) were
essentially fixed in blue spruce, while these loci were
variable in Engelmann spruce, and at GDH and PGM(1) the
allele observed in blue spruce was not the more common
allele found in Engelmann spruce (Table 1). In Engelmann
spruce, DIA(2) was fixed for the allele responsible for the
slower allozyme observed in the variable blue spruce
population. PGI(2) also presents an essentially
oppositely-fixed situation, where blue spruce possess three
alleles-—PGI(2)-2,3,4--and Engelmann spruce is essentially
fixed for PGI(2)-1, with one Engelmann spruce individual
found to be heterozygous for alleles 1 an 2 (Table 1). In
both instances where blue and Engelmann spruce possess

essentially oppositely-fixed alleles, the two observed

77

heterozygous individuals --one individual heterozygous for
GOT(3)-1/2 and the other heterozygous for PGI(2)—l/2--did
not possess at any other loci alleles characteristic of the
other species. Whether or not these two individuals were
backcrossed hybrids or pure species could not be determined
based on the number of loci and number of individuals
sampled in this study.

Other loci contained species-specific alleles, with
varying degrees of diagnostic value. For example,
Engelmann spruce possesses a very common and unique allele
at ACPWZ) (Table 1). Also, other species-specific alleles
with lower frequencies were observed at ACP(2), 6PG(1),
GDH, GOT(3), MDH(3), and MDH(4) (Table 1). A small survey
of 24 blue spruce individuals from outside the Dolores
River drainage--six from South Park, Colorado, eight from
the White River National Forest in Colorado, and ten
individuals from the Lincoln National Forest in New
Mexico-~did not reveal any alleles not observed in the
Dolores River collections (unpublished data).

The genetic distance statistic developed by Nei (1972,
1978) estimates the accumulated number of detectable gene
substitutions per locus among the sampled populations. The
intraspecific subpopulation comparisons indicated very
little genetic differentiation within the two species in
the Dolores River drainage. However, high degrees of
divergence were observed for the interspecific

subpopulation comparisons and between the two species as a

78

whole in the sample area. The average number of allelic
substitutions per locus between the two species as measured
in this study was 0.46, or 46 complete allelic
substitutions for every 100 gene loci. The zone 3
subpopulation of blue spruce--the zone in which both blue
and Engelmann spruce are common--exhibited the greatest
degree of genetic divergence from Engelmann spruce
subpopulations, while the zone 4 Engelmann spruce
subpopulation exhibited a slightly greater degree of
divergence from the blue spruce subpopulations than did the
zone 3 subpopulation. These trends are of interest,
because if blue and Engelmann spruce do hybridize
naturally, their zone of overlap should show the least
amount of divergence because of shared genes. These
statistics indicate there may be selection for just the
opposite, where blue and Engelmann spruce species
identities are even stronger in the zone of overlap than in
the peripheral zones.

The average genetic distance estimate of 0.46 between
blue and Engelmann spruce, computed using allelic
frequencies calculated at the 0.99 criterion, is comparable
to average genetic distance estimates observed among
sibling species--species which are morphologically similar
but quite distinct genetically and are reproductively
isolated--and morphologically distinct species across a

wide variety of organisms (Ayala 1975, 1982). The results

79

of this study indicate that blue and Engelmann spruce
should be strongly reproductively isolated, possessing both
prezygotic and postzygotic reproductive isolating
mechanisms (Ayala 1982), and will maintain their species
identities even in the presence of the other species. This
supports the poor crossability, prezygotic incompatibility
and hybrid inviability reported earlier for interspecific
crosses between blue and Engelmann spruce (Fechner and

Clark 1969; Kossuth and Fechner 1973).

CHAPTER V

ASSESSMENT OF NATURAL HYBRIDIZATION AND INTROGRESSION
BETWEEN BLUE AND ENGELMANN SPRUCE IN SOUTHWESTERN COLORADO

ABSTRACT

In a partial diallel mating design among 20 blue and
20 Engelmann spruce parents, the interspecific cross was
successful only with Engelmann spruce as the female parent.
No viable seed were obtained from the reciprocal cross
among the 60 full-sib families attempted. Under the
conditions of artificial pollination and a controlled
germination environment, very low interspecific
crossability was observed, with an average of 0.3 percent
germinated seed on a total seed basis across all 20
Engelmann spruce females. Many abnormalities were observed
among the hybrid germinants, suggesting hybrid inviability
also contributes to the low crossability between these two
species.

Isozyme analysis can be used as evidence for
interspecific hybridization between blue and Engelmann
spruce because of the unique genotypic compositions of the
hybrids relative to the two species. No natural F1 hybrids
between blue and Engelmann spruce were observed in this
study based on isozyme analysis of mature individuals or
their seedling progeny. Backcrossed hybrids may exist, but

determination of such was beyond the resolution of this

80

81

study based on the number of loci and number of individuals
sampled. Analyses included samples of open-pollinated seed
from blue and Engelmann spruce females located in an area
where both species are present in close proximity--often
side-by-side--and flowering phenology is coincident between
the two species. The probability of finding a mature
natural hybrid must be very small due to intraspecific
pollen competition, incompatibility, hybrid inviability and
the environmental conditions imposed upon the rare viable
interspecific seed during germination, seedling

establishment and growth in the field.

INTRODUCTION

Reproductive isolation is the primary criterion for
the definition of biological species, each species
representing an independent and discrete evolutionary
entity (Ayala 1982). Reproductive isolation may develop as
a by-product of evolutionary divergence when two incipient
species are separated geographically, or possibly while
they are in sympatry through mutations which prevent cross-
compatibility but maintain self-compatibility.
Intraspecific crossability differences also exist in many
organisms due to a wide variety of causes.

Blue and Engelmann spruce (Picea pungens Engelm. and

 

P; engelmannii Parry ex Engelm., respectively) represent an

82

interesting species combination in regards to reproductive
isolation and speciation. Studies of morphological and
chemical variability among blue and Engelmann spruce have
suggested the two species are phylogenetically closely
related, blue spruce possibly the result of a single
speciation event from the older Engelmann spruce
(Daubenmire 1972; Nienstaedt and Teich 1971; Taylor et al.
1975). The two species are morphologically quite similar,
although a combination of traits will generally distinguish
the two species (Schaefer and Hanover 1985a; Jones and
Bernard 1977). They are both montane species and occupy
overlapping habitats in western North America. Blue spruce
is primarily a riparian species, found along streams and
adjacent hillsides at elevations of 2000 to 3000 meters.
Engelmann spruce is generally found above this elevation,
occupying upper valleys, hillsides and plateaus in pure

stands or mixed with subalpine fir (Abies lasiocarpa

 

(Hook.) Nutt.). In elevationally intermediate zones, the
two species can often be found in close proximity with
ample opportunity for cross-pollination. In these
intermediate zones, there is enough overlap between blue
and Engelmann spruce in phenology of pollen shed and female
strobilus receptivity for natural cross-pollination to
occur (Fechner and Clark 1969; Ernst, unpublished data).
Some individuals of intermediate phenotype between the two

species have been identified based on morphological and

83

biochemical traits (Daubenmire 1972; Taylor et al. 1975;
Schaefer and Hanover 1985a and b). A few artificial
hybrids between blue and Engelmann spruce have been
produced, and interspecific crossability was very low
(Fechner and Clark 1969; Kossuth and Fechner 1973), but
only one or two parents of each species were used in the
hybridizations.

The objectives of this study were to (1) produce known
hybrids between blue and Engelmann spruce, (2) quantify the
crossability between the two species using a large number
of parents, (3) determine if isozyme analysis can be used
to identify blue-Engelmann hybrids, and (4) determine if
hybrids between blue and Engelmann spruce exist in nature.

The Dolores River drainage in southwestern Colorado
was chosen as the study site for two primary reasons.
First, there are many sites within the drainage where both
blue and Engelmann spruce are present, and at these sites
pollen flow and female strobilus receptivity are coincident
between the two species. Also, studies investigating the
genetic variability in morphological and terpenoid
characters of blue and Engelmann spruce have previously
been conducted in this drainage (Hanover 1975; Reed and
Hanover 1983; Schaefer and Hanover 1985a and b).

Blue and Engelmann spruce differ markedly in their
isozymic compositions (Ernst et al. 1985b). Several
species-specific alleles were observed, and strong

frequency differences were found between the two species at

84

seven of the 13 loci analyzed. Therefore Fl hybrids
between blue and Engelmann spruce, if they can be found in
nature or artificially produced, should exhibit unique
isozyme genotypes relative to the two species. However,
backcrossed hybrids may not be identifiable based on an
analysis of 13 enzymatic loci unless very large sample

sizes are obtained.

MATERIALS AND METHODS

The Dolores River and five of its tributaries were
divided elevationally into five species-occupation zones.
Zone 1, the zone of lowest elevation and extending from
2225 to 2400 meters (m), was a "pure" blue spruce zone
relative to the occurrence of Engelmann spruce. Zone 2,
extending form 2400 to 2590 m, was almost exclusively blue
spruce in composition with a few scattered Engelmann spruce
individuals. Zone 3, extending from 2590 to 2770 m, was an
elevationally intermediate zone relative to the habitats of
blue and Engelmann spruce, with both species present and
often in close proximity. Zone 4, extending from 2770 to
2960 m in elevation, was occupied primarily by Engelmann
spruce with a few scattered blue spruce individuals
present. Zone 5, the zone of highest elevation and

extending from 2960 to 3140 m, was a "pure"Enge1mann

85

spruce zone. The parents used to make the interspecific
matings were located in zones 2, 3 and 4--ten blue spruce
individuals from each of zones 2 and 3, and ten Engelmann
spruce individuals from each of zones 3 and 4, for a total
of 40 parents, 20 of each species. The 40 parents were
selected primarily on the basis of fecundity and
climbability. All parents were readily identifiable as to
species and no putative hybrids were found in any of the
zones along the Dolores River.

The partial diallel mating design used in this study
was comprised of three intraspecific matings--including
selfs--and three interspecific matings per parent.

The results of the intraspecific matings and details

of the pollination and cone collection procedures are
described elsewhere (Ernst et al. 1985). Each biparental
interspecific cross was replicated three times on a female
parent. The pollinations were carried out during the
Spring of 1983 using fresh pollen. In the fall of 1983,
the control-pollinated seed was collected and kept separate
by isolation bag. During this time single-tree open-
pollinated cone collections were also made from each of the
40 parents in the mating design and also from nine blue
spruce and 11 Engelmann spruce individuals in zone 3 of
Scotch Creek, a tributary of the Dolores River. The Scotch
Creek site serves as a putative hybrid swarm area, as both
blue and Engelmann spruce are present, often side-by-side,

and pollen shed and female strobilus receptivity occur

86

simultaneously among both species in the spring. Most
individuals at the Scotch Creek site were readily
identifiable as to species. In the fall of 1984, dormant
vegetative buds were collected from each of the 40 parents
used in the mating design and stored at —20’C until used in
the electrophoretic analysis.

The Open and control-pollinated cones were dried, the
number of cones per accession recorded, and the seed
extracted by hand and blown to separate empty and
putatively full seed. For the control-pollinated
accessions, the number of cones per bag damaged by insects
was also recorded, and both empty and putatively full seed
were counted separately; The seed was kept in cold storage
(4'C) until used.

Germination tests were conducted during the summer of
1984 using a maximum of 30 seed per isolation bag,
depending on availability of seed per bag. The number of
newly germinated seed was recorded daily, and the
germinants were then planted in individual plant bands in
the greenhouse. Germination was considered complete after
30 days, and the number of ungerminated seed were recorded
and then each was dissected to determine the number of full
but ungerminated seed versus empty seed. The percent
germinated and percent ungerminated-but-full seed were
determined from the germination test and then extrapolated

to a total seed basis--full and empty--to serve as the

87

dependent variables in the analysis. Details of the
germination procedures and results of the intraspecific
progeny are given elsewhere (Ernst et al. 1985). Percent
germination was used as the measure of interspecific
crossability because it estimates the number of viable seed
or progeny produced for a given cross. Percent
ungerminated-but-full seed was measured primarily to detect
postzygotic abnormalities.

The model equation used to estimate the fixed and
random effects for the germination data and its associated
assumptions are given elsewhere (Ernst et al. 1985). Using
best linear unbiased prediction (BLUP) techniques (Mao
1982), parental general combining ability (GCA) estimates
and individual—cross specific combining ability (SCA)
estimates were determined. From these, restricted maximun1
likelihood (REML) techniques (Schaeffer 1976) were used to
estimate the GCA, SCA and error variances. Under the
assumptions that the blue and Engelmann spruce populations
were sampled at random, each is randomly mating, and there
is no inbreeding, epistasis or linkage, the GCA variance
(625) corresponds to one-fourth the additive variance
(1/462A), and the SCA variance (623) corresponds to one-
fourth the nonadditive (dominance) variance (1/4629) for
the trait in the partial diallel mating design (Kempthorne
and Curnow 1961).

The seedlings from the germination test were grown

under accelerated—optimal-growth conditions (Hanover et al.

88

1976) in the greenhouse from August, 1984, until January,
1985, when the seedlings were allowed to go dormant.
Dormant vegetative buds were collected from each of the
seedling progeny in March, 1985, and stored at -20‘C until
used in electrophoresis.

Nine enzyme systems were assayed in the
electrophoretic analysis (Table l). Genorypes of the
parents in the controlled pollinations were determined by
simultaneous comparison of isozymes in bud, embryo and
megagametophyte tissues. Progeny genotypes were
characterized using dormant vegetative bud tissue. The
preparatory techniques and electrophoretic conditions
utilized in this study are reported elsewhere (Ernst et al.
1985a), including the inheritance of the 11 loci from the
nine enzyme systems analyzed in this study. For multiple
locus enzyme systems, the fastest migrating--most anodal--
zone was designated as locus 1, the next fastest 2, etc.
Multiple allozymes within each locus were numbered in the
same manner, with the fastest allozyme labeled as allele 1,
etc. Mobilities of the different allozymes were quantified
relative to the buffer front (Rf). Where possible,
segregation tests of observed progeny genotypes were made
using the log-linear G-statistic (Sokal and Rohlf 1969).

The single-tree open-pollinated collections from zone
3 of Scotch Creek were analyzed to determine if any blue-

Engelmann hybrid progeny could be identified

89

Table 1. The eleven loci from nine enzyme systems analyzed among the
Engelmann x blue Spruce hybrids. The numbers in parentheses
represent locus designations.

 

 

Enzyme Abbreviation E.C. No.
Aconitase ACO 4.2.1.3
.Aldolase ALD 4.1.2.13
Isocitrate dehydrogenase IDH(2) 1.1.1.42
Malate dehydrogenase MDH(Z) 1.1.1.37
MDH(3)
MDH(4)
6-phosphogluconate dehydrogenase 6PG(1) 1.1.1.44
Glutamate dehydrogenase GDH 1.4.1.3
Glutamate oxaloacetate transaminase GOT(3) 2.6.1.1
Phosphoglucose isomerase PGI(2) 5.3 1.9
Phosphogluc omutase PCM( 1) 2. 7 5. l

 

90

electrophoretically; Up to 40 embryos from partially

germinated seeds were analyzed from each female parent.

RESULTS

Full-sib interspecific hybrids were obtained from the
controlled pollinations, but only with Engelmann spruce as
the female parent. Much lower crossabilities--iJL, fewer
germinated seed--were observed for the hybrid crosses than
for the intraspecific crosses (Table 2). Means of
interspecific full-sib family viable seed yields ranged
from 0.00 to 1.74 percent, with an overall mean of 0.30
percent. Of a total 60 possible full-sib Engelmann x blue
spruce--fema1e x male--families, only 30 families produced
hybrid progeny. Sixteen of the 20 Engelmann spruce females
combined with 17 of the 20 blue spruce male parents to
produce a total of 158 viable hybrid progeny among the 30
full-sib families. The interspecific crosses also resulted
in a much higher proportion of abnormal--ungerminated-but-
full--seed (Table 2). This is not surprising, as many
abnormalities were observed among the viable hybrid
seedlots. These included a very high frequency of multiple
embryony, fused multiple embryos, and a very high incidence
of reverse germination. Up to six embryos germinated from a
single seed, with two or three embryos per seed very

common, while no multiple embryos were observed among the

91

intraspecific controlled crosses. Abnormalities and
electrophoretic analysis were used to verify the hybridity
of the interspecific crosses. Of a total 360 interspecific
full-sib replicates attempted in the study, nine replicates
were observed to be contaminated with very small amounts of
intraspecific pollen based on isozyme genotypes of the
progeny.

Variance component and narrow-sense heritability
estimates for the two germination traits are given in Table
3. For percent germination of the interspecific hybrids,
female general combining ability (GCA) variance was much
larger than the male GCA variance for this trait,
accounting for 20 and two percent of the total observed
variation, respectively. This difference in additive
variance estimates for the two species is also reflected in
the much larger narrow-sense heritability estimate for
Engelmann spruce--the female parent for all hybrid progeny.
Specific combining ability (SCA) variance accounted for 15
percent of the total variation in percent germination.

For percent ungerminated-but-full seed among the full-
sib interspecific families, both female and male GCA
variance estimates were very small, accounting for only two
and one percent of the total variation observed in this
trait, respectively (Table 3). The small narrow—sense
heritability estimates reflect the relatively small

influence additive sources of variation have on percent

92

 

 

 

 

 

 

Table 2. Mean values of percent germination (Z Germ) and percent
ungerminated-but-full (Z Ungf) seed on a total seed basis
for Open-pollinated (Open) and control-pollinated
(Biparental and Selfed) collections.

Open Biparental Selfed
Species ZGerm ZUngf ZGerm ZUngf ZGerm ZUngf
Blue spruce .1 4.9 11.9 2.2 4.8 1.0
Engelmann spruce .7 8.0 19. 3.6 9.5 2.3
Engelmann x blue .30 0.46
hybrids
Table 3. Variance component and narrow-sense heritability estimates

for production of viable and abnormal full-sib seed in
Numbers in parentheses
represent percentages of the respective variance components
as compared to the total observed variance for that trait.

Engelmann x blue spruce hybrids.

 

Variance Component

 

 

 

Female Male Femfle Ma£e
Traita GCA GCA SCA Error h h
Z Germ 0.079 0.007 0.061 0.254 0.80 0.09
(202) (22) (152) (63%)
Z Ungf 0.044 0.032 0.464 1.892 0.07 0.05
(22) (1%) (19%) (782)

 

a Z Germ '- percent germinated (viable) seed; 2 Ungf = percent
ungerminated-but-full (abnormal) seed.

93

ungerminated-but-full hybrid seed. SCA varaince accounted
for 19 percent of the observed variation in percent
ungerminated-but-full seed.

Of the 158 hybrid germinants, 95 survived (60
percent), and isozymes of dormant vegetative buds from
these seedlings were analyzed electrophoretically. Refer
to Ernst et al. (1985a) for zymograms and inheritance data
of the 11 loci analyzed in this study. Two loci--aldolase
(ALD) and Inalate dehydrogenase(2) (MDH(2))--were
monomorphic among all parents and their hybrid progeny.
Glutamate oxaloacetate transaminase--(GOT(3))--was
oppositely fixed among the blue and Engelmann spruce
parents; the GOT(3)-1 allele in Engelmann spruce and
GOT(3)-2 in blue spruce. Phosphoglucose isomerase--
PGI(2)--was fixed among the Engelmann spruce parents for an
allele not found in blue spruce--PGI(2)-l--and the blue
spruce parents were either homozygous or heterozygous for
alleles not found in Engelmahn spruce--PGI(2)-3 and -4.
Therefore, for both GDT(3)and PGI(2), hybrids exhibited
heterozygous phenotypes not observed among the parents--
GOT(3)-l/2, and PGI(2)-l/3 or PGI(2)-1/4.

Enough isozymic variation and parental combinations
existed at the remaining seven loci to perform segregation
tests on distributions of progeny genotypes, and the
results are given in Table 4. No significant deviation was
observed among cross-types--parental genotype

combinations--of three loci--aconitase (ACO), glutamate

94

.95 4?: as God 3.... 5:8: 8....

madame em.e ecu Aoewv oe.m .Aue_v ew.m

"~o>oH oucmomwwcwwm —o.o
"~m>o~ ouceomumcwme mo.o
"one Eovoouu mo noouwov one o~o>o~ oocuomwmcwmm mnemum> Ou mamocoomouuoo nonem> 0

 

 

 

.csoza uo: was mucuouozu use many eons ohmsds mundane anomaaoeon awesome someone scum wcmu~soou Scowoum m
Afivefigc Am.nve Am.nvm A m m\Nx~\~
A~Vwo.o AmVN Anve 0 ~ n\nx~\~
Anvnn.~ Am.~v~ Am.~v~ An.~v~ Am.~vm o N M\NRN\~
Advsw.o Am.e~vud Am.¢~vn~ an o ~\qu\~ Aevmnz
Advea.m Anavou Asmvcu em 0 N\~x~\_
ANvm~.e Am.~vc Amn.nve Amm.mvn m_ m N\~x~\~
Afivc~.o Am.nvn Am.mVe s N ~\_x_\~ Amvzaz
Advom.e Am.va<N Am.~mV~e no NH «\Nx~\~ Advzum
Agve~.m Aevm Acvm Nfi e N\~x~\~ :90
111111 A111V~ 111111 ~ g M\mx~\~
Amvwm.m Am.¢v¢ Am.eve Am.ev~ Am.evo w~ N m\~x~\~
Auvsm.o~ Am.eVH An.ovN~ m~ o N\~x_\~
Adeo.n Am.wvc Am.wvn~ n~ e n\~x~\~ Advumo
Afivo~.a an.mvo Am.nvn s e N\NNN\~ Auvan
A~v@~.~ Am.afivc~ Am.o~vm~ on w N\~x~\~ 00¢
nu mxu mfiw ~\~ «\N HRH hzwwmwm uoauouo somxu Auaoofiv
Awouoomxovvo>uocpontuomhuoco so cum .02 .oz noouo oszusm

 

.hcomouo omutafizw panama mafia
x cceegomam mo panama one cfl vomaouaxo nonzuocow oE»soe« mo nuuou =0muawouwon apocanofiwcmm

.e 623.

95

dehydrogenase (GDH) amd malate dehydrogenase(4) (MDH(4H.
Only one cross-type--l/2 x 2/2--deviated signifanctly for
MDH(3), and that was the result of a large deviation in
observed versus expected progeny genotypes in a single
full—sib hybrid family. Tests of the other two cross-types
for MDH(3) did not indicate any deviation from expected.
The progeny distributions for phosphoglucomutase--PGM(1)--
and isocitrate dehydrogenase--IDH(2)--deviated
significantly at the 0.05 and 0.01 levels, respectively.
Among the three cross-types for 6-phosphogluconate
dehydrogenase-—6PG(1)--progeny genotype distributions
deviated significantly for two of them. Therefore there is
some evidence for gametic selection or hybrid inviability
among the full-sib hybrid progeny based on isozyme
genotypes at three or four of the 11 loci. Unique hybrid
allozyme phenotypes--iJL, banding patterns not observed
among the intraspecific progeny (see Ernst et al. l985a)--
observed among the hybrid progeny are shown in Figure 1.
From the zone 3 Scotch Creek population, the putative
hybrid swarm area, 245 embryos from nine half-sib blue
spruce families and 357 embryos from 11 half-sib Engelmann
spruce families were analzyed electrophoretically. No
interspecific hybrids were observed among the embryos based

on electrophoretic phenotypes.

96

 

 

 

 

 

 

 

PGI(2) MDH(4) with MDH(3)
1.01
Rf 0.5-
‘ = i :I ”in D;
‘ E '.' Egg [:3 11:11 '3: 3'
. 3 3 '30 '80
t - a — I- -
OJ) ’
m 1/1 11011111 1/1 1/2 2/2 m m
Genotype 111111111 1/1 1/3 m m m
Figure 1. Unique isozyme phenotypes expresSed in bud tissue

of Engelmann x blue spruce full-sib progeny but'
not among intraspecific progeny. Heterodimers
are marked as 'H' and homodimers as 'D', and null
allozymes are represented by empty boxes.

97

DISCUSSION

Interspecific hybrids between blue and Engelmann
spruce were positively identifiable using isozyme analysis.
In a related study in the Dolores River drainage, the two
species exhibited several allelic differences among the 13
loci sampled (Ernst et al. 1985b), and genotypes of known
hybrids in this study were consistent with parental
genotypes across the 11 loci assayed. The locus best
suited for identification of interspecific hybrids in the
Dolores River drainage was PGI(2). Allele l was observed
only in Engelmann spruce--frequency >0.99--while alleles 3
and 4 were observed only in blue spruce--frequencies of
>0.9S and (0.01, respectively (Ernst et al. 1985b0. Allele
2 was observed in both species, but at frequencies of less
than 0.01 in Engelmann spruce and 0.04 in blue spruce.
Therefore, any individuals in the Dolores River drainage
which are heterozygous as PGI(2)-l/3 or -l/4 are very
strong candidates for interspecific hybrids. In
combination with genotypes at other loci which differ in
allelic frequency or composition between the two species--
GOT(3), IDH(2), 6PG(1), GDH, PGM(1), MDH(4), acid
phosphatase (ACPW2)), and diaphorase (DIA(2)) (Ernst et al.
l985b)--hybrids can be readily confirmed. How well these
species differences are maintained in other portions of the
ranges of blue and Engelmann spruce must await further

study.

98

Progeny genotypic distributions for at least three of
the 11 isozyme loci assayed in this study deviated from the
expected values, suggesting some selection may occur at
gametic or embryonic stages. This is not surprising based
on the hybrid inviability--abnormalities and poor survival
of hybrids--observed in this study and also in the study by
Fechner and Clark (1969). It is interesting to note that
for all loci where progeny distributions did deviate from
the expected, the imbalance was always towards the allele
more common to both species--IDH(2)-2, 6PG(1)-l and PGM(1)-
2 (Table 2)—-rather than the allele unique or more common
to only one species (see also Ernst et al. 1985b).

Based on the results from reciprocal interspecific
hybridizations among the 20 blue spruce and 20 Engelmann
spruce parents in this study, hybridization between the two
species is unidirectional and of very low crossability.
Viable seed was obtained only with Engelmann spruce as the
female parent, with an average of 0.30 percent germinated
seed on a total seed basis across all 20 Engelmann spruce
females. These results are similar to those obtained by
Fechner and Clark (1969), although their controlled crosses
were limited to one female parent of each species and two
blue spruce pollen parents and one Engelmann spruce pollen
parent. They reported viable seed only with Engelmann
spruce as the female parent and very low interspecific

crossability--mean viable seed yields of less than two

 

99

percent. Fechner and Clark (1969) also reported a high
frequency of hybrid abnormalities, subh as reverse
germination, termination of germination after emergence of
the radicle, and branched hypocotyls. However, they did
not report multiple embryony, which was very prevalent
among the viable hybrid seedlots of the present study.
Archegonia with multiple nuclei the size of the egg were
reported in blue spruce ovules pollinated with Engelmann
spruce pollen (Kossuth and Fechner 1973).

In an anatomical study of ovule development among
reciprocal interspecific crosses between blue and Engelmann
spruce, Kossuth and Fechner (1973) reported no viable seed
for the Engelmann x blue (female x male) spruce cross and
0.48 percent germination for the blue x Engelmann spruce
cross. They also used a limited number of parents, with
one female of each species and a two-tree mix of blue
spruce pollen and one Engelmann spruce pollen parent. They
did not observe any pollen tubes penetrating the nucellus
among the Engelmann x blue spruce ovules. In the
reciprocal cross, most pollen also died or pollen tubes did
not grow rapidly, but some pollen tubes did penetrate the
nucellus and archegonium, dead hybrid embryos were
observed, and a few viable seed were obtained. Kossuth and
Fechner (1973) also observed that incompatibility
breakdowns occurred primarily between nine and 30 days
after pollination and before fertilization, as evidenced by

termination of female gametophyte development and necrosis.

100

They attributed this breakdown to a lack of intraspecific
pollen rather than a result of incompatible pollen because
normal female gametophyte development is a response to the
presence of intraspecific pollen, even if ungerminated (see
also Mikkola 1969) .

The results presented by Kossuth and Fechner (1973)
indicate the interspecific cross between blue and Engelmann
spruce should also be possible with blue spruce as the
female parent--iJL, it is bidirectional. This contrasts
strongly with the results of this study, as none of the 20
blue spruce females crossed successfully with Engelmann
spruce-~a total of 60 blue x Engelmann full-sib families.
The parents were located in areas where both species were
present at least to a limited degree, and cross-
compatibility may not be so restricted among allopatric
populations of the two species. This could be easily

tested. This phenomenon has been documented in Drosophila

 

paulistorum, where incompatibility is stronger among
sympatric populations of several subspecies of 2:

paulistorum than among allopatric populations of the same

 

subspecies (Ayala et al. 1974; Ehrman 1969). A similar
situation was also observed in Eiliflr where sympatric
species exhibited stronger incompatibility than did
allopatric species of this same genus (Grant 1966).

The results from a related study of isozyme variation

among the blue and Engelmann spruce populations in the

101

Dolores River drainage suggests the elevationally
allopatric blue and Engelmann spruce subpopulations in this
drainage may be less divergent genetically than the
sympatric subpopulations (Ernst et al. 1985b). This may
indicate there is some selectin against interspecific
hybridization in the sympatric zone, and cross-
compatibility may be less restricted among the allolpatric
subpopulations. It may also be a function of the habitats
the subpopulations occupy and the extent of variability
within each of the subpopulations. In the studies by
Daubenmire (1972) and Taylor et al. (1975), less variation
was observed in morphological and phenolic characters in
the sympatric populations than in the allopatric
populations. Also, using a discriminant function composed
from morphological and terpenoid characters, Schaefer and
Hanover (1985c) found evidence of introgression among the
zone 3 Scotch Creek subpopulations of blue and Engelmann
spruce, while the zone 1 and zone 5 "pure species"
subpopulations were readily separable by the same
discriminant function. The putative hybrids they
identified resembled Engelmann spruce more strongly than
blue spruce, suggesting gene flow is favored towards
Engelmann spruce rather than blue spruce.

Differential cross-compatibility between sympatric and
allopatric populations of blue and Engelmann spruce may
exist and the interspecific cross may be bidirectional.

However, the unidirectional crossability reported by

102

Fechner and Clark (1969) and in this study, and evidence
for unidirectional gene flow in morphological and terpenoid
characters (Schaefer and Hanover 1985c) suggest the cross
is compatible only with Engelmann spruce as the female
parent. The results of the study by Kossuth and Fechner
(1973) indicate the reciprocal cross may be feasible as
well. The many instances of gametophytic breakdown and
irregularities in the archegonia in conjunction with low
seed set support their conclusion. It is also possible the
hybrids obtained by Kossuth and Fechner (1973) were
intraspecific contaminants. The hybridity of the progeny
could not be confirmed because they were accidentally
destroyed (Fechner, personal communication). Four
intraspecific contaminanted seedlings were observed in
three different blue x Engelmann spruce family replicates
in this study and were documented as such only by
electrophoretic analysis. The reduced viable seed yields--
low percent germination--of these families suggested the
four seedlings were interspecific hybrids, but they did not
possess isozyme genotypes characteristic of the respective
parents. The seedlings apparently resulted from very
slight pollen contamination, and because of the strong
hybridization barriers between blue and Engelmann spruce
the few blue spruce pollen grains that were present in
these bags were manifest.

In the Engelmann x blue spruce cross, both additive

and nonadditive sources of genetic variation appear to

103

exert an influence in the production of viable hybrid seed.
Additive sources of variation in the female parent,
Engelmann spruce, are apparently very important in the
production of viable interspecific seed while they are not
in the male parent, blue spruce. Maternal effects may be
confounded in the large female additive variance estimate.
The maternal effects varaince could not be separated
because it could not be assumed the additive genetic
variance was equivalent in both species. However, maternal
effects were not present in intraspecific crosses of
Engelmann spruce (Ernst et al. 1985). Certain Engelmann
spruce females crossed much better than others with a
variety of blue spruce pollen parents; 9£Lq six Engelmann
spruce females produced viable hybrid seed, and in greater
amounts, with all three blue spruce pollen parents they
were crossed to, while two Engelmann spruce female parents
produced viable seed in crosses with only two blue spruce
males, eight Engelmann spruce females crossed with only one
blue spruce male, and four Engelmann spruce females did not
produce any viable hybrid seed. Whether these differences
are truly genetic or environmental--maternal--in origin
must await further testing. The influlence of maternal
effects in the Engelmann 1: blue spruce cross can be tested
by replicating the same crosses in seed orchards of
equivalent genotypic composition located at two or more

sites. The influence of nonadditive sources of variation

104

is supported by the fact that certain full-sib families
produced much larger quantities of hybrid seed--three to
six times the overall mean.

No mature Fl hybrids were observed in the Dolores
River drainage based on morphological and isozymic
phenotypes of 56 blue spruce and 76 Engelmann spruce
individuals (see also Ernst et al. 1985b). The sampled
individuals included the 40 parents used to make the
interspecific crosses in this study. One blue spruce
individual from the lower elevation, pure blue spruce zone
of the Dolores River was heterozygous for GOT(3)-l/2 (see
Ernst et al. 1985b). However, this individual possessed
isozyme genotypes characteristic of blue spruce at all
other loci and resembled blue spruce morphologically.
Therefore, based on the location of the individual and its
resemblance to blue spruce morphologically and in isozymic
composition, the presence of GOT(3)-l in blue spruce
probably represents a rare allele in this species rather
than a result of interspecific gene flow (introgression).

Because of the very low crossability and extent of
hybrid inviability observed in this and previous studies
(Fechner and Clark 1969; Kossuth and Fechner 1973) under
the best of conditions--artificial pollination with no
intraspecific pollen competition and a controlled
germination environment-~it was not expected that mature F1
hybrids would be found. Based on isozyme genotypes there

was no evidence of introgression among these individuals or

105

their progeny either. Therefore, if natural hybrids do
exist, they must be very rare--one tree in several million
at best--and introgression very localized and "diluted" if
the hybrids are indeed fertile. The hybrids produced
artificially in this study will be grown to maturity to
determine if they are fertile and can be backcrossed
successfully to the two pure species parents.

No interspecific F1 hybrids between blue and Engelmann
spruce were observed when 602 open-pollinated embryos from
single-tree cone collections of nine blue spruce and 11
Engelmann spruce parents in zone 3 of Scotch Creek were
analyzed electrophoretically. Embryos from seed of blue
and Eng elmann spruce trees in zone 3 of Scotch Creek were
assayed because there was a much greater probability for
natural interspecific pollination at that site, and the
embryos had not been subjected to the environment of
germination and seedling establishment in the field.
However, tens of thousands of embryos must be screened in
the hopes of finding a natural hybrid embryo because of
intraspecific pollen competition and low interspecific
crossability. The screening of embryos in this study
represents a small fraction of the required sample size.
It may be easier to analyze electrophoretically only those
embryos which show abnormal germination characteristics
similar to those exrpessed by the hybrids produced

artificially in this study.

106

The concept of low and unidirectional crossability
between blue and Engelmann spruce is consistent with both
species having maintained their species identities with
little if any evidence of natural hybridization and
introgression. Because air flow in the mountain valleys
occupied by the two species is generally from high to low
elevation, the predominant direction of pollen flow will be
from Engelmann spruce to blue spruce. For the two species
to maintain their species identities, selection pressure is
predominately upon blue spruce as the female parent because
blue spruce is more likely to be in the presence of
Engelmann spruce pollen than the reverse scenario.
Therefore, while crossability is expected to be very low
between blue and Engelmann spruce, genetic barriers to
hybridization may be even stronger--or complete--with blue

spruce as the female parent.

CHAPTER VI

RECOMMENDATIONS FOR FUTURE STUDY

Based on the results of this study, a variety of

research directions can be pursued to further assess the

degree of natural hybridization and introgression between

blue and Engelmann spruce. The following represents a

partial list of such studies as suggested by this author.

No

1.

attempt was made to prioritize the suggestions.

Determine the extent of maternal effects in blue
spruce. This can be accomplished by replicating given
crosses in seedling or clonal seed orchards over a
variety of sites.

Determine the extent of isozyme variability in blue,
Engelmann and white spruce throughout their respective
ranges. This would be best accomplished by sampling as
many individuals as possible (e.g., 50 single-tree
collections) from a variety of locations throughout the
range of each species.

Increase the number of enzyme systems which can be
assayed in blue and Engelmann spruce, and determine the
inheritance of the isozymes in each enzyme system.
Conduct further interspecific crosses between blue and
Engelmann spruce to determine if the cross is truly
reciprocal as suggested by Kossuth and Fechner (1973).
These crosses should include parents of blue and

Engelmann spruce from areas of allopatry.

107

108

Conduct studies in other unique populations of blue,
Engelmann and white spruce; emh, outlying populations
in New Mexico, Arizona, Wyoming and Montana, as well as
locations where both blue and Engelmann spruce (and
possibly white spruce) occur in sympatry and cross-
pollination is probable.

Compare morphological, anatomical, biochemical and
physiological traits of blue and Engelmann spruce when
grown in common garden experiments on a variety of
sites. A majority of the studies comparing such
traits in blue and Engelmann spruce have sampled from
individuals in situ, and environmental influences
greatly interfere in making accurate genetic
comparisons.

Investigate the physiological and genetic basis for
incompatibility between blue and Engelmann spruce.
Essentially no studies have been conducted on
compatibil ity/incompatibil ity mechanisms in
gymnosperms, yet these 'naked seed' plants offer

the simplest of conditions because there no
intermediary tissues between the pollen grain and
ovule. Such an understanding may also shed light on
the mode of speciation in blue and Engelmann spruce.
Compare the viability and growth, morphology, anatomy,
biochemistry, physiology, cytogenetics, fertility and

crossability of the F1 Engelmann x blue spruce hybrids

109

generated in this study relative to the pure species.
The question of hybrid viability and fertility is of
utmost importance in evaluating possible modes of
speciation and adaptation in blue and Engelmann spruce.
Incorporate best linear unbiased prediction (BLUP)
techniques into the breeding evaluation procedures of
forest trees and other plant species. BLUP and
associated selection index techniques allow
unprecedented flexibility for unbalanced data
situations, a common occurrence in plant breeding

experiments.

LIST OF REFERENCES

110

LIST OF REFERENCES

Adams, VLTu, and S. Coutinho. 1977. Isozyme genetic
markers useful for studies of the Pinus rigida x Pinus
taeda hybrid. New Hampshire Agri. Expt. Sta. Scient.
Cntr. No. 847.

Adams, W.T., and R.J. Joly. 1980. Genetics of allozyme
variants is loblolly pine. J. Hered. 71: 33-40.

Allen, 6.3., and J.N. Owens. 1972. The life history of
Douglas-fir. Cat. No. Fo42-4972, Information Canada,
Ottawa.

Anand, I.J., and B.R. Murty. 1969. Serial analysis of
combining ability in diallel and fractional diallel
crosses in linseed. Theor. Appl. Genet. 39: 88-94.

Ayala, FHJ. 1975. Genetic differentiation during the
speciation process. Evol. Biol. 8: 1-78.

. 1982. Population and evolutionary genetics: a
primer. Benjamin/Cummings Publ. Co., Inc., Menlo Park,
CA.

, M.L. Tracey, D. Hedgecock and R.C. Richmond. 1974.
Genetic differentiation during the speciation process
in Drosophila. Evolution 28: 576-592.

 

Banks, B.D., I.L. Mao and J.P. Walter. 1985. Robustness of
the restricted maximum likelihood estimator derived
under normality as applied to data with skewed
distributions. J. Dairy Sci. 68: 1785-1792.

Bonga, J.M. 1981. Vegetative propagation of mature trees
by tissue culture. p. 191-196 I_r_1 A.N. Rao (ed.), Proc.
COSTED Symp. on Tissue Culture of Economically
Important Plants. Singapore, 1981.

Bramlett, D.L., T.R. Dell and W.D. Pepper. 1983. Genetic
and maternal influences on Virginia pine seed
germination. Silvae Genet. 32: 1-4.

111

Cecich, R.A. 1979. Ovule development and abortion in Pinus
banksiana. p. 33-40 In Proc. Joint IUFRO Symp. on
Flowering and Seed Development in Trees, May 15-18,
1978, Mississippi State Univ., Starkville, MS.

 

Chaudhary, B.D., S.N. Kakar and R.K. Singh. 1977.
Estimation of genetic parameters in barley (Hordeum
vulgare L.). II. Partial diallel analysis. Theor.
Appl. Genet. 49: 153-156.

Cheliak, W.M., and J.A. Pitel. 1984. Genetic control of
allozyme variants in mature tissues of white spruce
trees. J. Hered. 75: 34-40.

Clayton, J.W., and D.N. Tretiak. 1972. Amine-citrate
buffers for pH control in starch gel electrophoresis.
J. Fish. Res. Board Can. 29: 1169-1172.

Cram, W.H. 1984. Some effects of self-, cross-, and open-
pollinations in Picea puflens. Can. J. Bot. 62: 392-
395.

 

Daubenmire, R. 1972. On the relation between Picea pungens
and Picea engelmannii in the Rocky Mountains. Can. J.
Bot. 50: 733-742.

 

Dowling, T.E., and W.S. Moore. 1984. A program for
estimating genetic variability within and between
populations. J. Hered. 75: 34-40.

Eckert, R.T., R.L. Joly, and D.B. Neale. 1981. Genetics of
isozyme variants and linkage relationships among
allozyme loci in 35 eastern white pine clones. Can. J.
For. Res. 11: 573-579.

Ehrman, L. 1965. Direct observation of sexual isolation
between allopatric and between sympatric strains of the
different Drosophila paulistorum races. Evolution 19:
459-464.

 

El-Kassaby, Y.A. 1981. Genetic interpretation of malate
dehydrogenase isozymes in some conifer species. J.
Hered. 72: 451-452.

, E.C. Yeh and O. Sziklai. 1981. Inheritance of
allozyme variants in coastal Douglas-fir (Pseudotsuga
menziesii var. menziesii). Can. J. Genet. Cytol. 24:
325-335.

 

 

 

Ernst, S.G., J.W. Hanover and D.E. Keathley. 1985a.
Inheritance of isozymes in seed and bud tissues of blue
and Engelmann spruce. In preparation.

112

. 1985b. Allozyme variation of blue and Engelmann
spruce in southwestern Colorado. In preparation.
. 1985c. Assessment of natural hybridization and
introgression of blue and Engelmann spruce. In
preparation.

, and I.L. Mao. 1985. Genetic variation and control
of intraspecific crossability in blue and Engelmann
spruce. In preparation.

Falconer, [LS. 1960. Introduction to quantitative
genetics. Ronald Press, New York.

Fechner, G.H. 1979. The biology of flowering and
fertilization. p. 1-24 In Proc. Joint IUFRO Symp. on
Flowering and Seed Development in Trees, May 15-18,
1978, Mississippi State Univ., Starkville, MS.

, and ILW. Clark. 1969. Preliminary observations on
hybridization of Rocky Mountain spruces. In Proc. of
the Comuu on Forest Tree Breeding in Canada 11: 237-
247.

Finnerty, Vt, and G. Johnson. 1979. Post-translational
modification as a potential explanation of high levels
of enzyme polymorphism: xanthine dehydrogenase and
aldehyde oxidase in Drosophila melanogaster. Genetics
91: 695-722.

 

Fowler, ILP., and L. Roche. 1975. Genetics of Engelmann
spruce. USDA For. Serv. Res. Pap. WG-30.

Franklin, ELC. 1970. Survey of mutant forms and inbreeding
depression in species of the family Pinaceae. USDA

Futuyma, ILJ. 1979. Evolutionary biology. Sinauer Assoc.,
Inc., Sund erl and , MA.

Gordon, G.H. 1980. A method of parental selection and
cross prediction using incomplete partial diallels.
Part I: a simulation study. Theor. Appl. Genet. 56:
225-232.

Grant, V. 1966. The selective origin of incompatibility
barriers in the plant genus Gilia. Amer. Nat. 100: 99-
118.

. 1981. Plant speciation (Second edition). Columbia
Univ. Press, New York.

113

Greathouse, 12E. 1966. Inheritance of germinative energy
and germinative capacity in Douglas-fir. p. 60-62 In
Joint Proc. Second Genetics Werkshop Soc. Am. For. 33d
Seventh Lake States For. Tree Improv. Conf., USDA For.

serVo Res. Pap. NC-Go

Gupta, D., and S.L. Basak. 1983. Genetics of germination
and seedling growth of flax (Linum usitatissimum).
Seed Sci. and Technol. 11: 251-256.

 

 

Guries, R.P., and F.T. Ledig. 1978. Inheritance of some
polymorphic isoenzymes in pitch pine (Pinus rigida

 

. 1981. Genetic diversity and population structure in
pitch pine (Pinus rigida Mill.). Evolution 36: 387-402.

 

Habeck, J.R., and T.W. Weaver. 1969. A chemosystematic
analysis of some hybrid spruce (Picea) populations in
Montana. Can. J. Bot. 47: 1565-1570.

Hallauer, A.R. 1981. Selection and breeding methods. p.
3-55 I_n K.J. Frey (ed.), Plant Breeding II, Iowa State
Univ. Press, Ames, IA.

Hamrick, J.L., J.B. Mitton and Y.B. Linhart. 1981. Levels
of genetic variation in trees: influence of life
history characteristics. p. 35-41 In Proc. of the
Symp. on Isozymes of North Amer. Forest Trees and
Forest Insects, USDA For. Serv. Gen. Tech. Rep. PSW-48.

Hanover, J.W. 1975. Genetics of blue spruce. USDA For.
Serv. Res. Pap. WG-28.

, E. Young, W.A. Lemmien and M. Van Slooten. 1976.
Accelerated-Optimal-Growth: a new concept in tree
production. Michigan State Univ. Agric. Exper. Sta.
Res. Pap. No. 317.

Harris, H., and D.A. Hopkinson. 1972. Average
heterozygosity in man. Ann. Human Genet. 36: 9-20.

Heit, C.E. 1961. Laboratory germination and recommended
testing methods for 16 spruce (Picea) species. Proc.
of the Assoc. of Official Seed Analysts of North Amer.
51: 165-171.

Henderson, C.R. 1973. Sire evaluation and genetic trends.
p. 10 In Proc. of the Animal Breeding and Genetics ‘
Symposium in honor of Dr. J.L. Lush, ASAS, ADSA,
Champaign, IL.

114

Jones, J.R., and N.T. Bernard. 1977. How to tell Engelmann
from blue spruce in the Southwest. USDA For. Serv.

Kearsey, MJL 1965. Biometrical analysis of a random
mating population: a comparison of five experimental
designs. Heredity 20: 205-235.

Kempthorne, CL, and ILN. Curnow. 1961. The partial diallel
cross. Biometrics 17: 229-250.

King, J.P., P.O. Rudolf, R.M. Jeffers and H. Nienstaedt.
1970. Effects of varying proportions of self-pollen in
seed yield, seed quality and seedling development in
Picea glauca. Paper presented and distributed at the
Meeting of the Working Group in Sexual Reproduction of
Forest Trees, IUFRO Sec. 22, Varparanta, Finland, May,
1970.

 

Kossuth, S.V., and G.H. Fechner. 1973. Incompatibility
between Picea pungens Engelm. and Picea engelmannii
Parry. Forest Sci. 19: 50-60.

 

 

Kudray, G.M., and J.W. Hanover. 1980. A preliminary
evaluation of the Spartan spruce (Picea glauca x‘g;
pungens). Michigan State Univ. Agri. Expt. Sta.

Res. Pap. 405.

Leigh Brown, A.J., and G.H. Langley. 1979. Reevaluation of
level of genic heterozygosity in natural populations of
Drosophila melanogaster by two-dimensional
electrOphoresis. Proc. Natl. Acad. Sci. USA 76: 2381-
2384.

 

Lewontin, Etc. 1974. The genetic basis of evolutionary
change. Columbia Univ. Press, New York.

Lundkvist, K. 1975. Inheritance of acid phosphatase
isozymes in Picea abies. Hereditas 79: 221-226.

 

Mao, I.L. 1982. Modeling and data analysis in animal
breeding: notes for an internordic post-graduate
course. Dept. of Animal Breeding and Genetics,
Sweedish Univ. of Agric. Sci., Uppsala, Sweden.

Mayr, E. 1963. Animal species and evolution. Harvard
Univ. Press, Cambridge, MA.

Millar, C.L. 1983. A steep cline in Pinus muricata.
Evolution 37: 311-319.

 

115

Mitton, J.B., and R. Andalora. 1981. Genetic and
morphological relationships between blue spruce, Picea
un ens Engelm. and Engelmann spruce, Picea engelmannii
Parry 1n the Colorado Front Range. Can. J. Bot. 59:
2088-2094.

Mitton, J.B., Y.B. Linhart, J.L. Hamrick and J.S. Beckman.
1977. Observations on the genetic structure and mating
system of ponderosa pine in the Colorado Front Range.
Theor. Appl. Genet. 51: 5-13.

Mitton, J.B., Y.B. Linhart, K.B. Sturgeon and J.L. Hamrick.
1979. Allozyme polymorphisms detected in mature needle
tissue of ponderosa pine. J. Hered. 70: 86-89.

Morgenstern, ELK. 1974. A diallel cross in black spruce,
Picea mariana (Milld B.SJL Silvae Genet.23: 67-70.

 

Morris, R.W., and P.T. Spieth. 1978. Sampling strategies
for using female gametophytes to estimate
heterozygosity in conifers. Theor. Appl. Genet. 51:
217-222.

Muller, G. 1977. Short note: cross-fertilization in a
conifer stand inferred from enzyme gene-markers in
seeds. Silvae Genet. 26: 223-226.

Murty, B.R., V. Arunachalam and I.J. Anand. 1967. Diallel
and partial diallel analysis of some yield factors in
Linum usitatissimum. Heredity 22: 35-41.

 

Namkoong, G., and J.H. Roberds. 1974. Choosing mating
designs to efficiently estimate genetic variance
components for trees. Silvae Genet. 23: 43-53.

Neale, D.B., and W.T. Adams. 1981. Inheritance of isozyme
variants in seed tissues of balsam fir (Abies
balsamea). Can. J. Bot. 59: 1285-1291.

Neale, D.B., J.C. Weber and W.T. Adams. 1984. Inheritance
of needle tissue isozymes in Douglas-fir. Can. J.
Genet. Cytol. 26: 459-468.

Nei, M. 1972. Genetic distance between populations. Amer.

. 1977. F-statistics and analysis of gene diversity
in subdivided populations. Ann. Human Genet. 41: 225-
233.

. 1978. Estimation of average heterozygosity and
genetic distance from a small number of individuals.
Genetics 89: 583-590.

116

, and Aux. Roychoudhury. 1974. Sampling varianices of
heterozygosity and genetic distance. Genetics 76: 379-
390.

Newton, ILJ. 1979. A gene which alters the electrophoretic
mobilities of maize mitochondrial malate dehydrogenase
isozymes (Abstract). Genetics 91 (Suppl.): $88-$89.

Nguyen, ILTu, and ILA. Sleper. 1983. Genetic variability
of seed yield and reproductive characters in tall
fescue. Crop Sci. 23: 621-626.

Nienstaedt, EL, and A. Teich. 1971. The genetics of white
spruce. USDA For. Serv. Res. Pap. WO-lS.

O'Malley, D.M., F.W. Allendorf and G.M. Blake. 1979.
Inheritance of isozyme variation and heterozygosity in
Pinus ponderosa. Biochem. Genet. 17: 233-250.

 

O'Malley, D.M., N.C. Wheeler and R.P. Guries. 1980. A
manual for starch gel electrophoresis. Univ. of
Wisconsin Dept. of Forestry Staff Pap. No. 11.

Pederson, D41 1972. A comparison of four experimental
designs for the estimation of heritability. Theor.
Appl. Genet. 42: 371- 377.

Pryor, ALJ. 1974. Allelic glutamic dehydrogenase isozymes
in maize--a single hybrid isozyme in heterozygotes?
Heredity 32: 397-401.

Reed, A.N.F., and J.W. Hanover. 1983. Geographic variation
in cortical monoterpenes of blue spruce, Picea pungen .
Paper presented at the 7th North American Forest
Biology WOrkshop, 1982, Univ. of Kentucky, Louisville.

Ridgway, G.J., S.W. Sherburne and R.D. Lewis. 1970.
Polymorphisms in the esterases of Atlantic herring.
Trans. Am. Fisheries Soc. 99: 147-151.

Rogers, J.S. 1972. Measures of genetic similarity and
genetic distance. Studies in Genetics VII. Univ.
Texas Publ. 7213 145-153.

Rudin, EL, and I. Ekberg. 1978. Linkage studies in Pinus
sylvestris L. using macro gametophyte allozymes.
Silvae Genet. 27: 1-12.

 

Sarich, VLM. 1977. Rates, sample sizes, and the neutrality
hypothesis for electrophoresis in evolutionary studies.
Nature 265: 24-28.

117

Scandalios, J;G. 1969. Genetic control of multiple
molecular forms of enzymes in plants: a review.
Biochem. Genet. 3: 37-79.

Schaefer, PJL, and J.W. Hanover. 1985a. A morphological
comparison of blue and Engelmann spruce in southwestern
Colorado. Silvae Genet. (in press).

. 1985b. Monoterpene concentrations in cortical
oleoresin of blue and Engelmann spruce. Forest Sci.
(in press).

. 1985c. Evidence of natural hybridization between

blue and Engelmann spruce in southwestern Colorado. In
preparation.

Schaeffer, IuR. 1976. Maximum likelihood estimation of
variance components in dairy cattle breeding research.
J. Dairy Sci. 59: 2146-2151.

Shaw, D.V., and R.W. Allard. 1981. Analysis of mating
system parameters and population structure in Douglas-
fir using single-locus and multilocus methods. p. 18-
22 In Proc. of the Sympu on Isozymes of North American
Forest Trees and Forest Insects. USDA For. Serv. Gen.
Tech. Rep. PSW-48.

Slaughter, C.A., D.A. Hopkinson and H. Harris. 1975.
Aconitase polymorphism in man. Ann. Hum. Genet. 39:
193-202.

Sokal, R.R., and F.J. Rohlf. 1969. Biometry. W.H.
Freeman, San Francisco.

Taylor, R.J., S. Williams and R. Daubenmire. 1975.
Interspecific relatinships and the question of
introgression between Picea engelmannii and Picea
pungens. Can. J. Bot. 53: 2547-2555.

Tigerstedt, ELM.A. 1974. The application of ecological
genetics principles to forest tree breeding. Silvae
Genet. 23: 62-66.

Timmis, R., and G.A. Ritchie. 1984. Progress in Douglas-
fir tissue culture. p. 37-46 In Proc. Intl. Symp. of
Recent Advances in Forest Biotechnology, Michigan
Biotechnology Institute, Michigan State Univ., East
Lansing, MI.

Wehner, 15C. 1984. Estimates of heritabilities and
variance components for low-temperature germination
ability in cucumber. J. Amer. Soc. Hort. Sci. 109:
664-667.

118

Wendel, J.F., and C.R. Parks. 1982. Genetic control of
isozyme variation in Camelia japonica L. J. Hered. 73:
197-204.

 

Wheeler, N.C., and R.P. Guries. 1982. Population
structure, genic diversity, and morphological variation
in Pinus contorta Dougl. Can. J. For. Res. 12: 595-
606.

 

Wright, J3W. 1955. Species crossability in spruce in
relation to distribution and taxonomy. Forest Sci. 1:
319-349.

. 1976. Introduction to forest genetics. Academic
Press, New York.

Wright, S. 1951. The genetical structure of populations.
Ann. Eugen. 15: 323-354.

. 1965. The interpretation of population structure by
F-statistics with special regard to systems of mating.
Evolution 19: 395-420.

Yeh, F.C., and Y.A. El-Kassaby. 1979. Enzyme variations in
natural populations of Sitka spruce (Picea sitchensis
(Bong.) Carr.). I. Genetic variation patterns in ten
IUFRO provenances. p. 25 In Proc. 52nd Northwest Sci.
Assoc. Meeting. ‘——

 

Yeh, F.C., and D. O'Malley. 1980. Enzyme variations in
natural populations of Douglas-fir, Pseudotsuga
menziesii (Mirbg) Franco, from British Columbia. I.
Genetic variation patterns in coastal populations.
Silvae Genet. 29: 83-92.

 

 

Zouros, E. 1976. Hybrid molecules and the superiority of
the heterozygote. Nature (London) 262: 227-229.

IES

1111111