OVERDUE FINES ARE 25¢ PER DAY
PER ITEM

Return to book drop to remove
this checkout from your record.

 

9 .-..
“.1; u o 34::

"in
.' 1” ‘ufi‘;

”1.2

 

 

A PRELIMINARY EVALUATION OF THE

PICEA GLAUCA x PUNGENS HYBRID

 

By

Gregory M. Kudray

A THESIS

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

MASTER OF SCIENCE

Department of Forestry

ABSTRACT
A PRELIMlNARY EVALUATION OF THE

PICEA GLAUCA x‘g, PUNGENS HYBRID

 

by

Gregory M. Kudray

An evaluation of the white x blue spruce hybrid was undertaken
with emphasis on commercial potential as a Christmas tree or ornamental.
The average white x blue spruce cross yielded two seeds per cone with
seed weight and speed of germination values between parental rates. Fl
hybrid seedling heights were intermediate to white and blue spruce
seedlings but five to ten year heights were comparable or superior to
the faster growing parent, white spruce. F1 hybrids bore cones at early
flowering ages with intermediate dimensions and appearance to white and
blue spruce cones. Hybrid intermediacy was also noted for blue foliage
color, needle sharpness, and growing season frost susceptibility. Many
individual Fl hybrid trees had desirable combinations of fast growth
rate, blue color, and dull needles. A discriminant analysis using nine
characteristics identified bud length as the best single factor in
distinguishing species. Recommendations were made for continued

breeding efforts. Methods of producing hybrid trees for public release

were considered.

ACKNOWLEDGMENTS

I wish to express my gratitude to Dr. Hanover, my major professor,
Drs. Wright and Everson, committee members, and the Michcotip staff
for their guidance and encouragement.

I would also like to thank the many graduate students who gen-
erously gave of their time to assist in the technical details of my
project.

A special thanks goes to my wife, Kristi, for her patience, en-

couragement, and typing.

TABLE

LIST OF TABLES
LIST OF FIGURES
INTRODUCTION . . . .
MATERIALS AND METHODS
RESULTS AND DISCUSSION

Seed Characteristics

Growth and Reproductive Traits.
Morphological and Physiological
Multivariate Analysis

RECOMMENDATIONS

BIBLIOGRAPHY

OF CONTENTS

iii

Traits.

Table

10.

LIST OF TABLES

Seed yield summary of 1974-1978 controlled pollinations
involving P. glauca, P. pungens and their F1 hybrids
including 1978 germination rapidity ; , , , , , , , , ,

Mean seed weight of spruce species and hybrids derived
from controlled pollinations during 1976 at Kellogg
Forest 0 o o o o o o o o o o o o o o o o o o o o o o o 0

Mean height of 6-month and 12-month old P. glauca,,P.
pungens and their Fl hybrids grown in a greenhouse under
supplemental light . . . . . . . . . . . . . . . . . . .
Mean height of 6-month and 12-month old backcrosses
involving full-sib (P. gZauca x P. pungens) hybrid seed
parents with P. gZauca and P. pungens pollen grown in
the greenhouse under supplemental light .. . . . . . . .

Mean height and branch frequency for 8-monthoold spruce
seedlings grown in the greenhouse under supplemental
light 0 O O O O O O O O O O O I O O O O I O O O O O O 0

Correlation coefficients between 2-,5-, and 9—year height
measurements on blue and white spruces and their F1
hybrid O O I O O O O O O O O O O O O O O O O O O O O O 0

Length and width measurements of mature unopen spruce
cones from 1978 controlled pollinations at Kellogg
Forest 0 O O O O I O O I I I O O O O O O O O I O O I O O

Foliage color and needle sharpness of P. gZauca, P.
pungens, and the (P. glauca x P. pungens) hybrid spruce

Spring frost injury in the blue and white spruces and
their F} hybrid growing at East Lansing, Michigan in
1976 O . O O O O O I O O O O I O I O O I O I O O O O O O

Standardized discriminant function coefficients for
9-year-old P. glauca, P. pungens, and their F hybrid.
Relative ranking is in terms of disriminative efficiency.

iv

Page

12

14

15

22

26

28

31

35

Figure

1.

LIST OF FIGURES

Page

Mean heights for blue, white, and the white x blue

F1 hybrid spruce grown at Kellogg Forest. The vertical

bars at each age represent standard errors. All

measurements were taken in 1978. . . . , . . . . . . . . 19

Plot of discriminant score 1 (horizontal) vs. ~ egg ”_‘
discriminant score 2 (vertical) for ninesyearvold

spruce growing at Kellogg Forest. Each number

designates a tree's score, 1 2 white, 2 - blue, and

3 = white x blue F ~hybrid spruce. * indicates a

species' centroid.- . . . . . . . . . . . . . . . . . . . 37

INTRODUCTION

Blue spruce, Picea pungens Engelm., is a widely planted ornamental

 

throughout the North.temperate zone. Because blue spruce is not com-
monly a timber species, few genetic improvement programs other than

use of grafted varieties, have been undertaken (Cram, 1968; Dawson and
Rudolph, 1966; Hanover, 1975). Nevertheless, there are several economi-
cally important traits in blue spruce amenable to improvement.

The valued blue foliage color varies greatly with seed source
selection throughout the species' Rocky Mountain range (Bongarten, 1978).
A slow growth habit means high production costs and long-term land
commitments. Christmas-tree suitability is also diminished by extrem-
ely sharp needles making handling difficult for the grower and consumer,
although market value is high.

White Spruce, Picea glauca (Moench) Voss, is considered to be in

 

the same phylogenetic group but occupies a different range (Wright
1955b) than blue spruce. White spruce has a faster growth rate, soft
foliage, and lacks the blue color of blue spruce. Hybridization as a
method to combine valuable traits from these species was considered to
have potential. Within this framework the Michigan State University
spruce hybridization program began.

The first successful cross between white and blue spruce resulted
from controlled pollinations in 1967 (Hanover and Wilkinson, 1969).
Preliminary evidence of heterosis for early growth and seed germination

encouraged further research. Controlled pollinations have been made

annually since 1967 with the dual objectives of producing a superior
hybrid and increasing knowledge of evolutionary patterns in the genus
Kisse- .

The purpose of this study was to evaluate results and materials
from 11 years of spruce hybridization with emphasis on the development
of an improved hybrid for ornamental and Christmas tree use. Perfor-
mance of the promising white x blue spruce hybrid, which we suggest
designating the 'Spartan spruce,‘ will be compared with results from
similarily grown white and blue spruce. The effectiveness of hybridi-
zation in combining desirable traits from the two parent species into a
vigorous hybrid will be evaluated and recommendations for future study
will be made.

MATERIALS AND METHODS

 

Trees evaluated in this study were the result of annual controlled

pollinations in the spring from 1967-1978 involving Picea glauca, P,

 

pungens, and their hybrids. Seed parents were located at Michigan State
University or the Kellogg experimental forest in Kalamazoo County,
Michigan. Sources of pollen included trees at the Michigan State
University campus, Kellogg forest, a 1975 collection of blue spruce
pollen in Colorado, and blue spruce pollen from trees selected for blue
foliage in Cram's (1968) breeding program. Female strobili were iso-
lated with sausage casing bags when reproductive buds became apparent
and pollinated several days later when the scales were reflexed. Cones

were collected in late summer and seed was extracted, cleaned, blown

in an air column to eliminate empty seeds, counted, and weighed.
Measurement of cone dimensions were made after extraction.

Seedlings were grown in containers in a greenhouse under accelerated-
optimal-growth conditions utilizing Optimum fertilization, temperature,
and water regime with uninterupted light to ensure continuous growth
(Hanover et al., 1976). During some years seedling height was recorded
for various ages of trees.

Branch number counts were also made in one experiment involving a
comparison between P, glauca, P, pungens, F2 hybrids and backcrosses.
0n reaching plantable size, usually in 12 to 18 months, seedlings were
hardened off and then outplanted. Planting sites at Kellogg forest
underwent initial chemical weed control while the Michigan State
University site received periodic fertilization and irrigation.

Shoot damage was counted on 3-year-old seedlings after an early
spring frost in 1976. Measurements of foliage color, needle sharpness,
and height were made on all trees 5 years or Older in 1978. Correlation
coefficients between 2-, 5-, and 9-year-old heights of the 1967 pollin-
ation crop were calculated. Foliage color and needle sharpness were
qualitatively judged by the author on a numerical scale. Top whorl
branch samples from 9- and 10-year-old trees were removed and several
morphological features were measured for discriminant analysis. Measure-
ments included branch diameter, bud length and width, foliage color and
sharpness, and needle density, length, and angle to branch. A stepwise
discriminant analysis program was run on Michigan State University's

CDC 6000 computer using a subprogram routine (Klecka 1975).

RESULTS AND DISCUSSION

Seed Characteristics

 

Seed Yield In a hybrid breeding program, seed yield data are important

 

in defining the taxonomic relationship among species through genetic
compatibilities as well as in determining production feasibilities.

Capturing F hybrid vigor by mass generation of control pollinated seeds

1
has been successful for Hyun (1976) with pines in the 1950's, but to do
this economically, seed yields must be relatively high and labor inex-
pensive. Incompatibilities between species as reflected by zero or
reduced seed yields indicate genetic diversity induced by isolation
(Mikkola, 1969).

Table 1 presents a five year seed yield summary of controlled
pollinations in P3223. Bagged but unpollinated controls were employed
during this period and in the past to verify isolation of female stro-
bili from foreign pollen. Most Fl hybrid crosses were attempted using
white spruce maternal parents. Blue spruce does not commonly reach
reproductive maturity until it is 30 years old (Hanover, 1975). The
difficulty involved in pollinating these large trees and the small
quantities available locally necessitated greater emphasis on white
spruce seed parents. White spruce flowers early under cultivation
(Wright, 1964), are easily hand pollinated from the ground or a short
ladder, and many good genetic sources are available locally.

FI hybrid crosses, using white and blue spruce seed parents, yielded

an average of 1.7 and 1.9 seeds per cone, respectively. This represents

an 8-fold fecundity reduction from intraspecific crosses. A great range

5

Table 1. Seed yield summary of 1974-1978 controlled pollinations

 

 

 

involving P. glauca, P. girgens and their F1 hybrids
including 1978 germination rapidityui/
Filled 1973
Type of cross Females Males Crosses Successful Cones seeéifcne' Germination
. (no.) (no.) (no.) crdsses (no.) Mean Max rapidity
A
Female parent, P. glzucz
Unpcllinated 3 0 3 0 S 0.0 0.0
Open pollinated 40 - 40 100 1188 2.2 12.6
r P. glauca 39 15 7s 96 619 15.9 90.0 1' "a
r P. purgers 57 54 260 72 1734 1.7 27.0 15.9ac
: (P. glans: : pungens) 1 l 1 100 S 21.6 21.5
Female parent, P. gurgans
Ope. pollinated 6 - 6 67 79 1.0 3.3
c P. purgens 5 9 18 100 43 15.9 12.;
x P‘ glzuce 6 5 Female parent. $5. glaucé7: ; lggnsxl-
"npollinate 3 0 3 O S 0.0 0.0
Open pollinated 2 - 25 96 5J5 6.1 31.7
: P. 53::32 24 13 61 97 358 2a.6 93.5 1°.3b:
.:.' P. gurg‘vzs 19 31 55 64 325 2.7 35.‘ 17325::
.~.- (3. 3.: s.- - ;:..r..ge7:s/-3-/Z 10 33 100 375 3.3 37.5 19.9.
.~.- .9. pmgeus : gzgucazl/ 2 '~. 2 100 35 10.: 13.7 15.4.“:
Female parent, (P. gungans : glance)

: P. glaucs l l l 100 é 16.5 15.3 15.3ec
.r 9. ;'urg.ws 1 2 2 so 9 6.3 11.4 14.7dc
: (P. slauca x pungansj- 1 2 2 100 13 9.9 12.3 15.73:

 

Germination rapidity calculated as follows:

Rapidity (days) - Z seedlocs (days to germinate x number germinated seen)

I
(1"

l
\

Means followed by

 

n
:J
n
m
H
:1
C
8
0’
m
'1
O
H)
0")
ID
a
P
in
I 1
ID
1
U)
(D
«D
u.

.anover and Wilkinson, 1969).

the same latter are no: signifi:antly different at p a .

Both parents of these crosses have a common male parent.

in seed yields was experienced with individual parental combinations.
For example, 45% of the blue x white Spruce crosses were not fruitful
while crosses using a western source of white spruce pollen were parti-
cularily productive. The western white spruce pollen gave a maximum
yield of 17 seeds per cone in one cross and high yields with other blue
spruce females. A similar pattern occurred in white x blue and (white x
blue) x blue spruce crosses. Pollen from Pp 15 gave maximum seed yields,
90% higher than average in both instances, while some other pollens were
completely ineffective. In all cases described so far the superior
performance can be mostly attributed to pollen effects, generally good
results with other females were also achieved. Female parent and parental
combination effects have also been observed. The maximum 93.5 seeds per
cone yield for (white x blue) x white crosses was given by a pollen
source which rated second lowest in overall seed yield effectiveness.

While parental selection is a critical factor in influencing seed
yields, various other factors can modify results. Annual climatic
conditions, individual worker's pollination technique, and incomplete
crossing schemes confuse comparisons made between years. Fresh pollen
used in two successive years gave widely varying average seed yields.
Storage seemed to affect pollen differently, the best pollen one year
was only average the next when compared with others treated similarly.
Although results vary annually there are indications that some parents
are superior in producing high hybrid seed yields.

Average and maximum seed yields from F hybrids were generally

1

higher than white or blue spruce seed yields when using comparable

pollen. An exception to this effect is the low average seed yields with.
blue spruce pollen. While hybrid seed set with blue spruce pollen is
slightly higher than that of the white spruce females, it is much lower
than the intraspecific blue spruce cross. Blue spruce pollen has shown
strong incompatibility effects with any female other than blue spruce
throughout the study. Incompatibility is also strong in blue spruce
females pollinated with white spruce. The average seed yield value for
blue x white spruce crosses would be even lower, similar to results by
Hanover (1969), if the western white spruce pollen discussed earlier had
not been used. Backcrosses to white spruce averaged 9 times more sound
seeds than backcrosses to blue spruce. This suggests that inheritance
of reproductive behavior in the hybrid is influenced by either a white
spruce maternal or dominant effect resulting in an incompatibility with

blue pollen. Crosses involving F blue x white spruces show a lower

1
seed set with blue spruce pollen compared to white spruce pollen, sug-
gesting a maternal effect was not present, although only a small number
of these crosses have been made. A greater number of crosses involving
a diverse genetic background must be completed to clarify the method of
inheritance of reproductive behavior.

seed weight While seed weight is not generally considered to greatly
influence mature performance, Bingham et a1. (1956) found one and two
year progeny heights directly correlated with average seed weight in

hybrid pines. we have not investigated such a relationship in spruce

hybrids, but there is much genetic information available from seed

weight data. In a gymnosperm genus such as Pigga, the endosperm and
seed coat make up the greatest part of a seed and are.both of maternal
origin.with only the embryo containing chromosomes from both male and
female genomes. It could be expected that the main determinant of seed
weight would be the genetic constitution of the seed parent.

Blue spruce seeds usually weigh slightly more than white spruce
seeds (Cram, 1951). There can be confusion in determining accurate seed
weights unless a rigorous separation of sound seed from empty seed is
made. Seed weight data for parent species and hybrid combinations in
1976 are summarized in Table 2. 1976 data was chosen to represent
general trends because of the relatively high seed weight, indicating
rigorous separation of blind seed, and the wide range of crosses. Seed-
lots were small and quite variable, but some trends can be seen. Seed
from white spruce maternal parents was lightest, blue spruce heaviest,
and hybrid seed intermediate. Male parents had little effect except in
the hybrid, where white spruce pollen produced heavier seeds than blue
spruce pollen. This unexpected behavior may be due to the low number of
seeds available in each (white x blue) x blue spruce cross resulting in

an unreliable value.

Germination Germination rapidity values (Hanover and Wilkinson, 1969)

 

were calculated for the 1978 experimental sowing in the greenhouse
(Table 1). Seeds were sown in containers, covered with fine sand to a
depth of 1/2 inch, and watered regularly. Values for days to germinate

were higher than generally experienced with Picea and probably comparable

Table 2. Mean seed weight of spruce species and hybrids derived

from controlled pollinations during 1976 at Kellogg Forest.

 

 

Type of cross ' Crosses (no.) g/100 seeds

1/

P. glauca x glauca 27 .43a—
P. gZauca x pungens 38 .42a
(P. glauca x pungens) x pungens 12 .45a
(P. glauca x pungens) x glauca 32 .50b
P. pungens x pungens 17 .53b
P. pungens x gZauca . 5 .55b

 

l/

- Means followed by a common letter are not significantly different

at p = .05.

10

only within this trial. The slower rapidity rates may be attributed to
the tendency of the sand to form a hard crust when watered, subsequently
inhibiting emergence of the hypocotyl.

Hanover and Wilkinson (1969) reported a slight increase over both

parent species in germination rapidity for F hybrid seed. An inter—

l
mediate value of 15.8 days, not significantly different from white

spruce was recorded in 1978. Blue spruce seed germinated at the fastest
rate, 12.4 days. Other values reflected the genomic constitution of the

cross, i.e. F hybrids backcrossed to blue germinated faster than those

1
.backcrossed to white, although in many cases there were no significant
differences. The (white x blue) x (white x blue) spruce hybrid germinated
slowest of all Crosses with a rapidity value of 19.9 over 20% slower

than either the (white x blue) x (blue x white) or (blue x white) x

(white x blue) hybrids. While these rates may be confused due to inbreeding
with only one blue spruce available, it is noteworthy that the sexual

order of species in the F hybrid parent makes such a significant dif-

1

ference. Presumably if inbreeding affects germination speed, all control
pollinated F2 crosses would be subject because of the common limited
genetic base of their parents. The longest rate occurs only when white
spruce is the maternal parent and blue spruce is the pollen parent in
both Fl hybrid parents. If this pattern holds true with a wide variety
of hybrid crosses involving genetically diverse parentages, there would
be indications of a sexual effect. This phenomenon is similar to the

F hybrid white x blue spruce seed yield behavior. As noted earlier,

1

greatly increased compatibility with maternal pollen in F1 backcrosses

was noted.

Growth and Reproductive Traits

Greenhouse Growth All spruce seedlings were grown for at least 12

 

months in the greenhouse under accelerated-optimal-growth CAOG) condi-
tions utilizing a 24 hour photoperiod to ensure continuous growth
(Hanover, 1976). Height was measured periodically and a branch fre—
quency index, expressed as number of branches per cm. stem, was calcu—
lated in one experiment.

Height growth of white and blue spruce, their control pollinated
F1 and F2 hybrids, and backcrosses were compared in a greenhouse trial
(Table 3). Previous AOG experience indicated a pattern of relatively
slow white spruce early growth compared with the more rapid growth of
blue spruce. This response was verified after 6 months in the greenhouse;
blue spruce seedlings were 35% taller than white spruce, and significantly
taller than all other crosses. Hybrids generally were intermediate but
expressed some variation in the direction of their genomic constitution,
that is, blue spruce backcrosses initially grew faster than white spruce
backcrosses. By 12 months, differences were not significant between any ‘1
crosses with one exception which was a severe case of inbreeding depres-
sion in the F2 hybrid. The two F2 hybrid crosses tested were a self and
cross between full-sibs. Height growth rates for 6 and 12 month periods
were about 50% lower than the F1 hybrid.

This is an indication of deleterious recessive genes present in the

parents and illustrates the necessity of a wide genetic base for advance

generation hybrid breeding. A very slow growing cross such as this may,

11

12

Table 3. Mean height of 6-month and 12-month old P. glauca, P. pungens

 

 

 

and their 7 hybrids grown in a greenhouse under supplemental light.
Cross Height(cm)
6 months 12 months
P. gZauoa x gZauca 11.73;] 43.7a
P. gZauca x pungens 14.2ab 44.2a
P. pungens x pungens 18.1c 39.6a
(P. glauca x pungens) x gZauca 11.3a 45.4a
(P. gZauca x pungens) x pungens 15.0b 46.9a
(P. gZauca x pungens) x (P. gZauca x pungens)3/ 7.2d 21.5b

 

1/
Means in the same column followed by the same letter are not significantly
different at p = .05.

_2_/

This group includes a self and a cross between full—Sibsq

13

if the trend continues with_age, have potential as a dwarf horticultural
variety.

Full-sib F hybrid trees, (TT5559 x Ppl), were backcrossed with 7
pollen sources, four white spruce located in Michigan State University
Forest Genetic Plantation 1-60 and three blue spruce growing on campus.
Seeds were bulked by pollen source, sown in containers, and grown under
AOG conditions in the greenhouse. There were no significant height
differences between trees with different white spruce pollen sources
after 6 months. However, seedlings from the two tallest sources TTlOS49
and TT7850, retained their status at 12 months (Table 4). Blue spruce
pollen Pp7 produced the tallest 6 month progeny, about 20% superior to
Pp15 and Pp22 progeny, but this difference diminished after 12 months of
growth.

Height and branch frequency measurements were recorded for 8-month-
old spruce seedlings (Table 5). Significant height differences were
apparent only for the (white x blue) x (blue x white) spruce hybrid.
Inbreeding depression is probable because the same blue spruce was used
in both crosses. The (white x blue) x (white x blue) spruce hybrid
performance may also not be indicative of potential because of inbreeding

due to using full-sibF hybrid parents.

1
A branch frequency index was formulated to express the dense
branching habit of some hybrids. Blue spruce seedling form under AOG

conditions in the greenhouse, has been Observed to be very Open with few

branches. This was exemplified by the low (.2) branch frequency value

Table 4.

14

Mean height of 6vmonth and 12'month-old backcrosses
involving full-sib (P. glauca x P. pungens) hybrid
seed parentsl/ with P. gZauca and P. pungens pollen

grown in the greenhouse under supplemental light.

 

Pollen parent

Height (cm)

 

 

 

6 months 12 months
P. gZauca
2 4/

LL14S49— 11.0a- 50.1a

TTlOS49 12.2a 58.3b

TT 7350 12.1a 57.6ab

UU18SSO 10.0a 51.2ab
P. pungens

Pp 7-3-/ 17.8b 59.21:

Pp 15 15.3c 55.0ab

Pp 22 15.0c 53.9ab
y

Cross number TT5859 x Ppl located in Michigan State University

Forest Genetic Plantation 21-70 in Kalamazoo County.

Location number in Michigan State University Forest Genetic

Plantation 1-60.

Location number on Michigan State University Campus

Means in the same column followed by the same letter are not
significantly different at p = .01.

15

Table 5. Mean height and branch frequency for 8-month old spruce
seedlings grown in the greenhouse under supplemental light.

 

 

Species Mean Branchg/
height(cm) frequency
1/
P. glauca 13.5a- .7a
P. pungens 12.2a .2b
P. gZauca x (P. glauca x pungens) 13.3a .6a

(P. glauca m pungens) x

(P. glauca x pungens)§/ 11.9a l.lc
(P. glauca x pungens) x
(P. pungens x glaucalil 8.9b .6a

 

'l/ Means followed by a common letter are not significantly different
at p = .05.

2/

-' Branch frequency = No. of branches per cm. stem.

3/

Parents are full sibs.

'4/ Both Fl hybrid parents of the cross had the same blue spruce parents.

16

(Table 5). White and hybrid spruce had fuller form, but the (white.x
b1ue).x (white x blue) spruce cross was very dense with nearly twice as
many branches on a given length of stem.

It is not known if the dense branching form of the (white x blue) x
(white x blue) spruce hybrid is a consequence of inbreeding, a reaction
to the 24 hour photoperiod, or a peculiarity of the cross. If this
juvenile character is retained as the tree matures, it may be of some
value. Densely branched trees of good form are desirable for ornamental
and Christmas tree use.

The (white x blue) x (blue x white) spruce hybrid was quite differ-
ent in it's branching pattern with a value of .6, similar to white
spruce and backcrosses to white spruce. Unfortunately, the number of
crosses involved in these F2 hybrids is too small to verify the cause of
this behavior; it may be a manifestation of inbreeding or due to parental
selection or order.

Field Growth Reports of heterotic height growth responses in inter-

 

specific conifer hybrids are numerous, (Bingham et al., 1956; Critchfield,
1963; Duffield et al., 1958; Hyun, 1976; Keiding, 1968; Langner, 1959;
Rouland, 1971; wright, 1955a; and Wright, 1959). Unfortunately, only
early growth data is available in most cases. Duffield et a1. (1958)

has stressed the importance of long-term testing through the growing
cyclerJ_Differences in species' growth curves may cause the hybrid to

exhibit a heterotic response during one period of development, but

perhaps not at another phase. The oldest F1 hybrid material we now have

17

available in quantity is 10 years old. While not yet mature enough_to
evaluate as pulpwood material, we can consider its value for Christmas
tree and ornamental use.

Figure 1 shows height growth of white, blue and the F1 hybrid
spruce growing at the Kellogg experimental fewest. It should be noted
that all measurments were taken in 1978 and the age values represent
different annual pollination groups rather than one planting measured
for several years. Each group was exposed to slightly different plant-
ing environments, cultural treatments, and growing conditions. Valid
comparisons can only be made within a year.

Generally, white x blue spruce grew at similar rates to white
spruce; there were no significant height differences in any given year.
Both, however, were significantly faster growing than blue Spruce when
compared within a year, an increase of 30% was common. White spruce
parents of hybrids and white spruce controls grown at Kellogg were
selected from a variety of plantations, mostly located elsewhere in
Kellogg Forest. Many sources were available, usually phenotypically
superior and often from good genetic origins. Blue spruce pollen
parents were few and of unknown genetic constitution. The importance of
parental selection in hybrid progeny performance has been recognized
'(Bingham et al., 1956; Duffield,¢l958; Hyun, 1976; Langner, 1956; Little

gland Trew, 1976; and Vidakovic, 1965). The differences in reported

. X“. _M”\

‘5.

hybrid growth rates have been large enough to obscure hybrid potential

if poor combining parents were used.

I8

Figure 1. Mean heights for blue, white, and the white x blue F hybrid
spruce grown at Kellogg Forest. The vertical bars at each

age represent standard errors. All measurements were taken
in 1978.

HEIGHT (CM)

360

32o

280

240

200

160

120

80

............... P. a LA DC A
P.GLAUCA x PUNGENS
____ P PUNGEN_S_ 1

 

6 7 9 1O

8
AGE (YRS)

20

Mean.height differences of over 20% in 5-year-old hybrid progeny of
the same white spruce seed parent pollinated with different blue spruce
sources have been observed. 502 height differences were recorded in a
larger S—year-old trial varying white Spruce seed parents while retaining
a constant blue spruce pollen source. Although this information was
compiled from relatively small quantities of trees, parental selection
can be seen to be crucial in producing a fast growing hybrid.

Another important variable to be considered in hybrid evaluation is
planting site conditions. One explaination of heterosis asserts that a
hybrid habitat, different from that of the parent Species, was available
and could be utilized best by the hybrids (Anderson, 1953). While this
hypothesis usually refers to natural hybridizationjaartificial hybrids
often Show variable performance when tested in a variety of areas
(Bingham et al., 1956; Hyun, 1976; Little and Trew, 1976). They may be
heterotic in one location but not in another. The genetic diversity
‘fcontained in a hybrid may enable more efficient physiological use of
9 good conditions or greater adaptation to rigorous environments.

J”

“I White, blue and F hybrid Spruce were also grown in nursery beds on

1
the Michigan State University campus to the age of 5 years. Cultural
practices included fertilization, irrigation and weeding when necessary.
The parentage of the hybrid was more diverse than the Kellogg forest
Spruce plantation described earlier, i.e., a larger number of blue
spruce pollen parents were available. F1 hybrids grew vigorously, 20%
taller than white spruce controls planted in the same beds. Only a few

21

blue spruce progenies were available for comparison. .Many hybrids were
particularily vigorous, up to 2.5 times taller than average trees and
exhibiting desirable morphological qualities such as good form and blue
color.

This superior performance may be attributed to the wider genetic
base in the hybrids or their increased ability to utilize Optimal
conditions. Probably both of these contributed to the heterotic
behavior.

The necessity of understanding the genotype-environment interaction
has been recognized and these hybrids will be outplanted at a variety of
Sites throughout Michigan. Their performance under a wide range of
conditions will be evaluated for pulpwood, Christmas tree, and ornamen-
tal uses.

Cbrrelation Coefficients Length of improvement programs can be shor-
tened if early selection of juvenile trees is possible. For this
technique to be feasible, early performance rankings must correlate very
closely with rotation age rankings. This will enable the breeder to
eliminate poor material early and concentrate on higher potential trees,
increasing overall program efficiency.

An important selection parameter is height growth rate. Correla-
tion coefficients were calculated between 2-, 5-, and 9-year height

measurements on white, blue and the F white x blue spruce hybrid (Table

1

6). There was an indication of increased stability with age, as corre—

wwmilations between 5- and 9- year heights were higher than those between 2-

22

Table 6. Correlation coefficients between 2-, 5-, and 9-year height

 

 

 

 

measurements on blue and white spruces and their Fl hybrid.
Age(yr3)
A88(YTS) Species
5 9
1/
2 P. glauca .70— .55
P. pungens .84 .70
P. gZauoa x pungens .65 .50
5 P. glauca .80
P.pw@aw .90
P. glance x pungens .74
I/

r.01; 40 d.f. ‘ '49

23

and 5-year values. The relationship between 2- and 9-year—heights,

althoughfigignificant, was too low to consider early selection possibilities.

v
436‘"
K'

. ..
we

Blue spruce heights gave the highest correlations followed by white

spruce and the F hybrid. The relatively low correlation coefficient of

1

the F1 hybrid indicates variability in the early growth curve. This

stresses the importance of testing throughout the rotation age to

guarantee hybrid performancelm__

i.

FloweringResQonse Breeding work in trees is often limited by late

 

flowering ages leading to long generation times. Blue spruce, which
does not usually reach reproductive maturity in nature until 30 years of
age (Hanover, 1975), is difficult to work with for this reason. Addi-
tionally, the female strobili are normally borne high in the crown,
making controlled pollinatons difficult. There is evidence that hybrid-
ization can stimulate precocious flowering. Some pine hybrids have
produced cones earlier than their comparably grown parent species
(Critchfield, 1963; Fowler and Heimburger, 1958;), Others flowered at
similar dates to the parent with the earlier age of sexual maturity
(wright, 1955a; and Wright et al., 1969). This may imply dominance of

early fruiting behavior in certain F hybrids (wright, 1955a).

1
White Spruce can reach reproductive maturity in 4 years (Sutton,
1969; Young and Hanover, 1976) with significant cone crops on trees 6-tO

lO-years-Old in breeding orchards (Wright, 1964). There is evidence

that F1 hybrids have retained the early flowering characteristics of

white spruce in some instances, although relevant data is available for

24

only a few years. The hybrid crop of sufficient age to examine contains
two crosses, TTSSS9 x Ppl, and R4-l4 x Ppl, both having the same male
parent but exhibiting different flowering responses. After 6 years, 50%
of TTSSS9 x Ppl Offspring had produced female strobili with.7Z bearing
male strobili. 892 had flowered by the age of 10 years with heavy, 30
per tree, cone crops. The 112 that had not responded were below average
trees in growth and quality. Male strobili production was widespread by
this time although precise data are not available.

Conversely, only 52 of the R4—l4 x Ppl offspring bore flowers at
age 8, their earliest flowering age, with 20% productivity by age 10.
Male strobili generation was similarily delayed. R4-l4 x Ppl, in addi-
tion to a later age Of reproductive maturity, also had other blue Spruce
characteristics such as: sharper needles, shorter height, and bluer
foliage when compared to TT5859 x Ppl.

Unfortunately, flowering data is not available from the white
spruce maternal parents for comparison. Even with the same male parent,
physiological and morpological trait expression in the hybrid was greatly
influenced by the combining ability of the P, glauca parent. This
emphasizes the importance of including good combining parents as well as
selecting on the basis of external characteristics.

Flower initiation age at Kellogg forest can probably be considered
typical for an untreated breeding orchard. Flowering may possibly be
increased with fertilization, girdling, or cultural treatments.

Cone Characteristics Cone characteristics are used extensively in

hybridization studies because of their many distinguishing features and

25

resistance to environmental effects (Khalik, 1974; Weaver, 1963), Blue
and white spruce cones are readily differentiated, the latter being
shorter, more narrow and lighter in color. White spruce.cones also have
thin orbicular scales with rounded or slightly emarginate apex compared
to the elongated erose apex Of the blue spruce scales (Sargent, 1922).
F1 hybrid cones are darker than white spruce cones and tinted orange or
light brown. The cone scales are intermediate with slightly erose,
undulating apex.

A small sample of mature unopen cones from controlled pollinations
at Kellogg Forest was measured in 1978 (Table 7). F1 hybrid cones were
Significantly longer and wider than white spruce cones.

This reflects an intermediate value because blue spruce cones are
commonly 5-10 cm long (Den Ouden and Boom, 1965) and over 20 mm wide
(personal observation). Flowering hybrids were all from one cross
(TT5859 x Ppl), previously identified as having relatively more white
spruce morphological characteristics. This particular combination could
also disproportionately resemble white spruce in cone features.

MOrphological and Physiological Traits

Blue Cblor The ornamental value of blue spruce resides in it's unique

 

blue foliage coloration. This highly valued trait is considered to be a
result of surface wax deposition (Hanover and Reicosky, 1971). The
degree of foliage color depends primarily on the quantity and physical
structure of the wax and is influenced by a number of factors both
genetic and environmental (Hanover and Reicosky, 1971; Reicosky and

Hanover, 1978).

26

Table 7. Length and width measurements of mature unopen spruce cones
from 1978 controlled pollinations at Kellogg Forest.

 

 

Species Cones(no.) Length(cm.) Width(mm.)
1/
P. glauca 126 4.5- 13.1
P. glauca x pungens 129 5.4 13.5

 

l‘ Means in the same column are significantly different at p = .05.

27

Nurserymen have commonly grown blue spruce using seed from geographic
origins known to yield trees of good color. Hanover (1975) and Bongarten
(1978) identified several of these areas. Cram (1968) has had some
success in a selection program for blue color but the inheritance pattern
does not seem to be simple.

Moisture and nutritional regimes affect expression of genetic
capability for blue foliage color (Hanover and Reicosky, 1971). Ferti-
lizer is commonly recommended to enhance ornamental color. Surface
waxes are also influenced by weathering, first year needles are more
glaucous than older foliage, with best color on newly flushed needles.

White spruce also possesses surface waxes similar to blue spruce but
in much lesser amounts (Hanover and Reicosky, 1971). Some variation in
color can be seen although white spruce foliage never reaches the degree
of glaucousness present in blue spruce.

Foliage color ratings for 6 to 10 year old Spruce indicate good
retention of blue color in the F1 white x blue Spruce hybrid (Table 8).

A subjective evaluation revealed a 5 year mean of 2.3 (4.0 8 very blue
foliage, 0.0 - green foliage) for the hybrid, slightly lower than blue
spruce (2.6), but much higher than white spruce (1.2).

A wide range in yearly means of foliage color for blue spruce was
observed, extending 732 of the overall mean, compared to 252 for the
hybrid and white spruce. This may reflect the smaller sample size and
the small number of blue spruce parents used in several years. Often

within a year, the blue and F hybrid white x blue spruce means were not

1
significantly different. While hybrid means did not vary extensively,

28

Table 8. Foliage color and needle sharpness of P. glauca,
P. pungens, and the (P. gZauca x P. pungens) hybrid Spruce.

 

 

 

Foliage colorl/ Needle sharpnessg/
Species Trees(no.) 5 yr. mean Range in 5 yr. mean Range in
yearly means yearly means
P. gZauca 738 1.2 1.1 - 1.4 0.6 0.2 - ,7
P. pungens 142 2.6 1.6 - 3.5 2.6 2.3 - 2.7
P. glauca x .
pungens 915 2.3 2.0 - 2.6 1.5 1.0 - 1.8

 

1]

Color rating as follows: 0 = no blue, 4 very blue

'- Needle sharpness rating as follows: 0 soft, 3 = sharp

29

means of individual crosses within a year.ref1ected much variability.
Values ranged from .5 to 3.5 depending on the parental combination.
Diversity within a cross was also considerable, it was common to find
foliage color ratings from maximum to minimum when there were numerous
members in the cross.

Foliage color ratings were also scored for S-year-old hybrids grown
in nursery beds on the Michigan State University campus. Cultural
conditions were excellent with fertilization and irrigation, promoting
full expression of foliage color. The F1 hybrid spruce average color
rating was 2.3, even higher than the value for a small sample of blue
spruce present. Many individual hybrids had very good blue color. This
higher overall rating may reflect good cultural conditions and choice of
blue spruce pollen sources, which included pollen received from Cram,

and resulting from a selection program for blue color (Cram, 1968).

Needle Sharpness Blue spruce are often difficult for nurserymen to

 

handle and unpopular as Christmas trees because of their extremely sharp
foliage. A tempering of this quality could lead to increased use and
value for the species.

Needle sharpness is dependent on three factors: rigidity, acute-
ness of tip, and angle of the needles to the branch. Blue spruce has
very rigid, acute needles extending at an angle of 75° to 90° from the
branch. White spruce needles are more appressed to the branch and are
usually soft and flexible. Intermediacy for this trait occurs in the F

1
hybrid although somevariation is present.

30

A subjective scoring of needle sharpness was undertaken for spruce
5 years and older. Individual tree foliage.was rated in late.summer
from 0, very soft, to 3, very sharp (Table 8). The F1 hybrid 5-year
mean.needle sharpness value was 1.5, almost exactly intermediate to the
parent species. The relatively small range in yearly mean values for
blue and white Spruce reflect the consistency of the trait. Hybrid
yearly means ranged slightly larger. Values within a year and within a
cross generally showed little variation except in the hybrid with
slightly larger deviations.

Five-year-old spruce grown in nursery beds gave similar results to
field plantations indicating increased moisture and nutritional levels
have little effect on foliage sharpness.

Correlation coefficients between needle sharpness and blue color
indicate a positive correlation between the two blue spruce character-
istics, sharp needles and blue color. However, r is only .6 implying a
relatively weak correlation. Simultaneous selectiOn for blue foliage
color and duller needles should be possible in a breeding program.
Frost Injury Growing season frost is the most critical limiting factor
when northern species are grown south of their native range (Wright,
1976). Frost injury can result in growth retardation and malformed
trees which is especially damaging for ornamentals or Christmas trees.
White spruce, with its predominantly boreal range, is susceptible when
grown in lower Michigan or farther south (Wright, 1977). Blue spruce,
native to the Rocky Mountains, is not usually injured by growing season

frosts.

31

 

 

Table 9. Spring frost injury on the White and blue spruces and their
F:Lhybrid growing at East Lansing, Michigan in 1976.
Percent damaged
Spec1es No. of trees shoots per tree
1/
P. gZauca 61 41.6-
P. pungens 62 3 . 5
P. gZauoa x pungens 50 12.4

 

All means are significantly

different at p =

.05.

32

In 1976 a late spring frost caused extensive damage to the campus
nursery spruce planting. The percentage of damaged shoots on each
three-year-Old tree was recorded (Table 9). White spruce, as expected,
experienced the most severe injury, 41.62 of the shoots were damaged.
Only 3.5% of the blue spruce shoots were damaged. Hybrid frost resis-
tance was intermediate with 12.42 of the shoots damaged. Individual F1
hybrid trees showed great variability, some injured as much as white
Spruce, whileothers were as resistant as blue spruce.

Multivariate Analysis

Discriminant Analysis Discriminant analysis is a multivariate technique

 

which can be used to statistically distinguish related groups of taxa.
Discriminant analysis weights and combines various characters to best
"discriminate" between groups by means of a series of linear equations
(Fisher, 1936). These equations are of the form:

Di 3 dilzl + d1222+ ... + dipr

where Di is the score on discriminant function i, d's are weighting
coefficients, and 2's are the standardized values of the p discrim-
inating variables used (Klecka, 1975).

A discriminant analysis can be separated into two objectives.
Effectiveness of the equations and the involved variables in discri-
minating between the groups is calculated first. Next, the relative
classification of the groups is established enabling unknowns to be
assigned to groups (Kleca, 1975).

First applied in forestry to distinguish between clones of locust

(Hopp, 1941), discriminant analysis has proved useful in introgression

33

and hybridization studies (plifford and Binet, 1954; Dancik and Barnes,
1975; Ledig et al., 1969; Mergen et al., 1966; Mergen and Furnival,
1960; Namkoong, 1966; Sharik and Barnes, 1971). Discriminant analysis
was recognized as being particularly efficient in selecting discerning
characteristics (Dancik and Barnes, 1975) and in separating Species
(Namkoong, 1966). Mergen and Furnival (1960) saw special benefits for a
beginning hybridization breeding program. Individual features and their
relationships can be examined and the best diagnostic characteristics to
isolate the hybrid can be identified.

A particularily useful form of discriminant analysis uses a step-
wise procedure to add or delete variables on the basis of their dis-
criminating power. During the process, previously selected variables
may lose their discriminatory power with the addition of a new variable
and be removed from the equation. Often a relatively small number of
variables will define the species with sufficient accuracy and the
entire model need not be used.

Stepwise discriminant analysis was performed on a CDC 6000 computer

for 9- and 10-year-old white, blue, and F hybrid spruce growing at

1
Kellogg forest. Results were kept separate by age class. Variables
entered were as follows: height, twig diameter, bud length and width,
foliage color and sharpness, and needle density, length and angle

to branch. All branch measurements were taken from t0p whorl samples.

Two discriminant functions were necessary to define horizontal and

vertical dimensions. The functions, while using the same variables,

34

weigh them differently with coefficients to maximize distance between
groups that was not explained by a previous function. The importance of
the two equations is expressed as the relative percentage of trace they
contain. With the spruce data, 92.3% of trace was covered by the first
function, indicating the second equation is of little value. Stand-
ardized coefficients are computed to weigh the comparative value of each
variable's discriminating power in the equation.

Table 10 lists standardized discriminant function coefficients for
samples of 9-year-old spruce including their relative rank. The data
for 9-year-old spruce was chosen to typify this discriminant analysis
because a greater variety of crosses were available. The only difference
in the coefficients for 10-year-old spruce was a juxtaposition of two
variables in rank for the first discriminant function.

Bud length was the best single factor in differentiating white,

blue, and the F hybrid spruce. Other variables also significantly

I
contributed with the exception of needle density, which was removed from
' the analysis during the stepwise procedure. The coefficient rankings
were different for the second discriminant function, as expected, but
the small amount of trace explained renders this set relatively unime
portant.

The function equations defined individual tree values and expressed
them in a 2-dimensional plot (Figure 2). The greater amount of discri-

mination possible with the first function, expressed as horizontal dis-

tance, is readily apparent. Small vertical differences reflect the

35

Table 10. Standardized discriminant function coefficients
for 9-year-old g; glauca, g, pungens,and their
F1 hybrid. Relative ranking is in terms of
discriminative efficiency.

 

 

First discriminant function Second discriminant function
Variable Rank Coefficient Rank Coefficient
Bud length 1 -.977 2 -.768
Needle sharpness 2 -.897 5 -.542
Bud width 3 .853 3 .741
Needle length 4 -.625 1 .872
Height 5 .582 7 -.424
Needle angle 6 -.507 4 .698
Blue color 7 .483 8 -.227
Twig diameter 8 .197 6 -.536
Needle density1 9 -.111 9 -.139

 

1Not significant for inclusion in stepwise discrimant analysis by
change in Raos V at P = .05

36

Figure 2. Plot of discriminant score 1 (horizontal) vs. discriminant
score 2 (vertical) for nine-year-old spruce growing at
Kellogg Forest. Each number designates a tree's score,
1 - white, 2 = blue, and 3 3 white x blue F1 hybrid spruce.
* indicates a species' centroid.

0cm.< como_ cma.ql can.:|

ocaowc

cocoa-

ocmgca

accent

acoonu

canon-

"C
m
(‘0

camomn

r)
c

9":
(‘1

pk?
F")
C.
:0

C

N

osmo_ ccmou

oaoon

cco.n

o:m.c

c:
I—oHHFMMHHH—HHHHHHHHHv—fi—‘H—l—u—nr—nr—wHHv—uHHr—nmv—cr—av—nH—tH—qu—q—HHP-nr-l—fl
GI!

H
HHI-ip-IHH HH—Q—vHHp—u—nl—IHHHHHNP‘HHH-flHt—{HHMHHHHHHI—QF‘HMu—cu—qb—IHHHHu—q
O

assoc acooo

o=:.¢ cocon o . coo.n: oso.ct

38

minor contribution of the second function, predominantly useful in
increasing distance between the hybrid and the pure species. The plot
for discriminant analysis of 10-year-old spruce, not shown, is similar
with the exception of a slightly closer relationship expressed between
the hybrid and white spruce.

Prediction results were calculated on the success of discriminant
group assignations compared to known parentage: 1002 of white and blue

spruce and 902 of the F hybrids were correctly identified in the

1
analysis of 9-year-old spruce. 6.7% of the hybrids were incorrectly
classified as white spruce and 3.3% as blue spruce. Prediction results
from ten-year-old Spruce were similar, but incorrectly classified hybrids
were catergorized as white spruce.

Discriminant analysis presents a clear pictorial representation of
hybrids relative to parental species. Patterns in hybrid inheritance
for traits studied can be easily recognized. Discerning characteristics

of high value are identified and can be used in further research.

RECOMMENDATIONS

 

Nurserymen and Christmas tree growers have already expressed inter-
est in obtaining hybrid spruces. Unfortunately, up to the present, all
hybrids produced have been needed to continue research and breeding
work. The generation of hybrids for public release may be accomplished
through several methods.

Mass production of F1 hybrid pine seed by controlled pollination

was Successfully done by Hyun (1976) in the 1950‘s, but this approach is

39

economically unfeasible with our low seed yields and expensive labor.
Another approach,.vegetative reproduction by grafting or rooted cuttings
has the advantage of perpetuating preferred genotypes, thereby elimin—

ating variability present in F hybrids.

1
Varieties of blue spruce are commonly grafted, but this approach is
expensive with variable success rates (Wells, 1953). Similar problems
may be expected in grafting hybrids. Vegetative propagation by rooted
cuttings has been undertaken for hybrid Spruce in Europe (Rouland, 1971)
with large clonal variations in performance. Some easy rooting clones
gave 82% success rates while others were difficult to root. Mist
chamber rooting of Spruce at Michigan State University has given good
results with some Species (personal observation). Rooting ability

decreases with the Stock age, but a small scale F hybrid production may

1
be possible if young stock is used. This method is likely to be of
limited usefulness due to expense and topoPhySis effects. At present, F1
hybrid availability will probably be confined to the Michigan State
University research program and initial testing by large growers in the
Michigan Cooperative Tree Improvement Program.

Greenhouse seedling growth of backcrosses has been vigorous under
AOG conditions. Backcrosses may have commercial potential if this
growth rate continues. Unfortunately backcross production must be by
controlled pollination, encountering the same difficulties as Fl hybrid
mass production.

The most workable approach to mass production of hybrids may be an

F2 hybrid seed orchard. F1 hybrids would be isolated from sources of

40

contaminant pollen and allowed to wind.pollinate. Unfortunately there
is no guarantee of hybrid vigor extension beyond the F1 generation. Al-
though wright (1955a) and Hyun (1976) have reported encouraging early
results with F2 hybrids, our data is preliminary. The only Fl

Spruce presently flowering are of Similar parentage and the intercrossed

hybrid

progeny have been severely stunted in growth trials in the greenhouse.

In the next few years F hybrids of diverse parentage will be flowering,

l

enabling F crosses of sufficiently wide genetic base to be made. If

2

these F2 hybrids Show early vigorous growth and retain the desirable

features of the F1 hybrid, a grafted F2

considered. By this time good combining F

hybrid seed orchard should be

1 hybrid parents could be

identified. Grafting would ensure that the superior combining performance

of the selected Fl hybrids would continue into the F2 seed orchard.

Intensive culture would aid good growth, and early flowering would
probably ensue with significant seed production within 10 years.
An alternative to this approach would involve orchard establishment

with untested F1 hybrids. Mild selection could be practiced with

combining ability judged after F hybrid production and growth. The

2
savings in time with this method may be offset by the unknown performance

of F2 hybrids.

Present work Should concentrate on ensuring a diverse genetic base

for F'

1 hybrids, and making a wide variety of F crosses. Selection of

2
superior parents has been emphasized earlier. A selection score has
been computed for all spruce examined, on the basis of growth rate, blue

color, and needle sharpness. Phenotypically superior trees can be used

in further crosses.

41

Blue Spruce pollen should be collected in the native range from
areas known for high genetic value. If phenotypically desirable blue
spruce can be found locally, their value would be high for control
pollination work. More blue spruce females should be pollinated with

white Spruce to learn about parental sex effects in inheritance.

BIBLIOGRAPHY

i.

BIBLIOGRAPHY

L

J"

,5

(Anderson, E. 1953. Introgressive hybridization. Biol. Rev. Cambridge

Phil. Soc. 28:280-307.

Bingham, R. T., A. E. Squillace, and R. F. Patton. 1956. Vigor,
disease resistance, and field performance in juvenile progenies
of the hybrid Pinus monticola Dougl. x Pinus strubus. L.
Zeitschrift F. Forstgenet. 5:104-112.

 

 

Bongarten, B. 1978. Genetic variation in blue spruce. Ph.D. Thesis.
Michigan State University.

Clifford, H. T., and F. E. Binet. 1954. A quantitative study of a
presumed hybrid swarm between Eucalyptus elaeophora and E.
goniocalyx. Aust. J. Bot. 2:325-336.

Cram, W. H. 1951. Spruce seed viability: dormancy of seed from four
Species of Spruce. For. Chron. 27:1-9.

Cram, W. H. 1968. ‘13: Summary report on spruce breeding at Indian
Head Nursery, p. 12. Can. Dept. Reg. Econ. Expansion.

Critchfield, W. B. 1963. The Austrian x red pine hybrid. Silvae

Dancik, B. P., and B. V. Barnes. 1975. Multivariate analyses of
hybrid populations. Naturaliste Can. 102:835-843.

Dawson, D. H., and P. O. Rudolph. 1966. Performance of seven seed
sources of blue spruce in central North Dakota. USDA For. Serv.
Res. Note NC-S, 4 p. North Central Exp. Stat, St. Paul, MN.

Den Ouden, P., and B. K. Boom. 1965. Manual of cultivated conifers.
Martinus Nijhoff, The Hague, Netherlands.

Duffield, J. W., and E. B. Snyder. 1958. Benefits from hybridizing
American forest tree Species. J. For. 56:809-815.

Fisher, R. A. 1936. The use of multiple measurements in taxonomic
problems. Ann. Eugenics 7:179-188.

Fowler, D. P., and C. Heimburger. 1958. The hybrid Pinus peuce
Griseb. x Pinus strobus L. Silvae Genet. 7:81-86.

 

 

42

43

Hanover, J. W. 1975. Genetics of blue Spruce. USDA For. Serv. Res.
Paper WG-28, 12 p.

Hanover, J. W., and D. A. Reicosky. 1971. Surface wax deposits on
Picea_pungens and other conifers. Am. J. Bot. 58:681-687.

 

Hanover, J. W., E. Young, W. A. Lemmien, and M. Van Slooten. 1976.
Accelerated-Optimal-Growth: a new concept in tree production.
USDA Ag. Exp. Sta. Res. Rpt. 317, 16 p.

Hanover, J. W., and R. C. Wilkinson. 1969. A new hybrid between blue
Spruce and white Spruce. Can. J. Bot. 47:1693—1700.

Hopp, H. 1941. Methods of distinguishing between shipmast and common
forms of black locust on Long Island, N.Y. USDA Tech. Bull. 742. 24 p.

‘\§ Hyun, S. K. 1976. Interspecific hybridization with the special reference
‘t; to Pinus rigida x taeda. Silvae Genet. 25:183-191.

 

Keiding, H. 1968. Investigation of inbreeding and outcrossing in larch.
Silvae Genet. 17:159-165.

Khalik, M. A. K. 1974. Genetics of cone morphology in white spruce
(Picea glauca). Can. J. Bot. 52:15-21.

 

Klecks, W. R. 1976. Discriminant analysis, p. 434-467. lE.N- H. Nie,
C. H. Hull, J. G. Jenkins, K. Steinbrenner, and H. H. Bent [ed.],
Statistical package for the social sciences. McGraw-Hill, N.Y.

Langner, W. 1959. Ergebnisse einiger Hybridisierungsversuche zwischen
Picea stichensis (Bong) Carr, und Picea omorika (Pancic) Purkyne.
Silvae Genet. 8:138-143.

 

 

\y Ledig, F. T., R. W. Wilson, J. W. Duffield, and C. Maxwell. 1969. A
‘ discriminant analysis of introgression between guercus prinus L.
and Quercus alba L. Bull. Torrey Bot. Club 96:156-163.

 

 

. Little, 3., and I. F. Trew. 1976. Breeding and testing pitch x
“~; loblolly pine hybrids for the Northeast. Proc. 23rd Northeast
Tree Improv. Conf., p. 71-85.

Mergen, F., and G. M. Furnival. 1960. Discriminant analysis of Pinus
thunbergii x Pinus densiflora hybrids. Proc. Soc. Amer. For.,

 

 

Mergen, F., D. T. Lester, G. M. Furnival, and J. Burley. 1965. Dis-

criminant analysis of Eucalyptus cinerea x Eucalyptus maculosa
hybrids. Silvae Genet. 15:148-154.

44

Mikkola, L. 1969. ObServations on interspecific sterility in Picea.

Namkoony, G. 1966. Statistical analysis of introgression. Biometrics
22:488-502.

Reicosky, D. A. and J. W. Hanover. 1978. Physiological effects of sur-
face waxes. I. Light reflectance for glaucous and nonglaucous
Picea pungens. Plant Physiol. 62:101-104.

 

Roulund, H. 1971. Observations on spontaneous hybridization in Picea
amorika (Pancic) Purkyne. Tree Improvement 2:3-17.

Sargent, C. S. 1923. Manual of the trees of North America (Exclusive
of Mexico). Second ed. (Dover Reprint, 1965). Houghton—Mifflin
Company, Boston. 2 vol.

Sharik, T. L., and B. V. Barnes. 1971. Hybridization in Betula
allgghaniensis Britt. and B, lenta L.: a comparative analysis
of controlled crosses. For. Sci. 17:415-424.

 

Sutton, R. F. 1969. Silvics of white spruce (Picea glauca [Moench]
Voss). Canada Dept. Fisheries and For., For. Br. Publ. 1250, p. 57.

 

Vidakovic, M- 1965. Some characteristics of the needle structure and
growth in hybrids between Austrian pine and Japanese red pine.
Silvae Genet. 15:155-159.

Weaver, T. W. 1963. Variation in the spruce complex of the northern
Rocky Mountains Picea glauca, Picea pungens, and Picea engelmannii.
MA Thesis, University of Montana.

 

Wells, J. S. 1953. Propagating Koster spruce. Amer. NurS. 98:48-53.

Wright, J. W. 1955a. Cone characteristics and natural crossing in a
population of F1 pine hybrids. Zeitschrift f. Forstgenet 4:45-48.

Wright, J.W. 1955b. Species crossability in Spruce in relation to
distribution and taxonomy. For. Sci. 1:319-349.

Wright, J. W. 1959. Species hybridization in the white pines. For.
Sci. 5:210-222.

Wright, J. W. 1964. Flowering age of clonal and seedling trees as a
factor in choice of breeding systems. Silvae Genet. 13:21-27.

Wright, J. W. 1976. Forest Genetics. Academic Press, N. Y., 463 p.

v1 “x

v

45

Wright, J. W., W. A. Lemmien, and D. S. Canavera. 1969. Abundant natural

hybridization between Austrian and Japanese red pines in southern
Michigan. For. Sci. .15:269-274.

Wright, J. W., H. Nienstaedt, W. A. Lemmien, J. N. Bright, M. W. Day,
and R. L. Sajdak. 1977. Better white spruce for Michigan. USDA
Ag. Exp. Sta. Res. Rpt. 316, 8 p.

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

STRTE UNIV. L IBRRR

IIIII IIII III IIII I IIIIIIIIIIIILII IIIII |II2I|IIII|III|