Title: Genetic variation in symptomology of slash pine in response to fusiform rust
CITATION THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00102787/00001
 Material Information
Title: Genetic variation in symptomology of slash pine in response to fusiform rust
Physical Description: Book
Language: English
Creator: Layton, Patricia Adlene, 1954-
Copyright Date: 1985
 Record Information
Bibliographic ID: UF00102787
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: ltuf - ACY0251
oclc - 13568920

Full Text












GENETIC VARIATION IN SYMPTOMOLOGY OF SLASH PINE
IN RESPONSE TO FUSIFORM RUST








BY


PATRICIA ADLENE LAYTON


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN
PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA


1985









This research is dedicated to my parents, James B. and

Rebecca Layton, who always "think I can." Their love and faith in

me have always provided a stable base in my life. Their help was

even more concrete in this study in that they gallantly volunteered

to help in data collection at the Rust Screening Center.















ACKNOWLEDGEMENTS


I am deeply grateful to W. Paul Shelley, Jr., Erin C. Shelley

and other members of the Shelley family who established the W. P.

Shelley, Sr., Graduate Forestry Fellowship which I received during

my last two years of graduate study.

I would like to thank the members of my supervisory committee

for their guidance and inspiration: Drs. Ray E. Goddard, Ramon C.

Littell, Donald L. Rockwood, Robert A. Schmidt, Anthony E. Squillace

and Charles J. Wilcox. Additionally I would like to thank Drs. Tom

Miller, George Blakeslee, Mitch Flinchum, Ron Dinus, Nancy Pywell,

Arnett Mace, Nick Comerford, Everett Hall and Bro Kinloch for their

advice and encouragement.

This project would not have been possible without the help and

support of the University of Florida Cooperative Forest Genetics

Research Program. I would like to thank the staff at the

University, which includes Greg Powell, Harm Kok, B. J. Rabe,

Charles Akins, and James Hayes, as well as the industrial members.

In addition, SFRC students Penny Parnell and Robert Richens were a

great help in collecting the data. Two of the industries, Brunswick









Pulp Land Company and Container Corporation of America, provided

materials for this study. I am especially grateful to James Hodges

and the research staff at Brunswick for their assistance in

collecting the field data.

Portions of the project were conducted at the US Forest

Service's Rust Screening Center in Asheville, North Carolina. I

appreciate the advice and help of the staff, especially Robert

Anderson and Carol Young.

I wish to thank Martin Marietta Energy Systems for helping in

the completion of this project.

I would like to thank my fellow graduate students for their

encouragement and support throughout this project. I would

especially like to thank Tim Meyer for his friendship and help.

Lastly, I want to thank Rick Cantrell, my husband, for his love,

patience and understanding.
















TABLE OF CONTENTS

Page

ACKNOWLEDGEMENTS . . . . . .... . . . iii

LIST OF TABLES . . . . . . .... . . . . vi

LIST OF FIGURES . . . . . .... . . . . .. x

ABSTRACT . . . . . . . . ... .. . . .. xi

INTRODUCTION . . . . . .... . . . ...... 1

LITERATURE REVIEW . . . . . . . .... . .. 4

MATERIALS AND METHODS . . . . .... . . . . 14

Field Tests--Experiment 1 . . . . . . . . 14
Artificial Inoculations . . . . . . . 19
Experiment 2--RSC Procedure . . . . . .. 19
Experiment 3--High Inoculum Concentration . . .. 21
Experiment 4--Inoculation of 1-0 Seedlings . .. 23
Analysis. . . . . . .... . . . . . 23

RESULTS AND DISCUSSION . . . . . .... . . . 31

Field Tests--Experiment 1 . . . . .... .. . 31
Artificial Inoculation . . . . .... . . . 54
Experiment 2--RSC Procedure . . . .... .. . 55
Experiment 3--High Inoculum Concentration . . .. 64
Experiment 4--Inoculation of 1-0 Seedlings .... 66
Research Perspectives . . . . . . .. 68

CONCLUSIONS . . . . ... . . . . . . .. 70

LITERATURE CITED . . . . . .... . . . . . 72

APPENDIX . . . . . .... . . . . . 79

BIOGRAPHICAL SKETCH . . . . . .... . . . . 90















LIST OF TABLES


Table Page

1. Location, year of establishment, site index, fifth
year height and rust incidence of Brunswick Pulp Land
Company's diallel progeny tests . . . . .... .15

2. Sets of slash pine crosses used in field and artificial
inoculation experiments . . . . . . .... .16

3. Percentage of trees infected with fusiform rust
for 48 slash pine families (four- to six-years-old)
planted in six progeny tests in southeast Georgia . . 17

4. University of Florida rust evaluations, standard
errors and number of tests used for evaluation of
slash pine parents . . . . . . .... . 18

5. Analysis of variance and expected mean squares
for a modified half-diallel as computed by DIALL .... . 24

6. Analysis of variance and expected mean squares
for a four-stage nested design using individual
tree data . . . . . . . . . . . . 26

7. Analysis of variance and expected mean squares
for Experiment 2 using mean number of symptoms
per tray . . . . . . . . . . . . 29

8. Analysis of variance and expected mean squares for
Experiment 4 using individual tree data . . . ... 30

9. Percentage of rust-associated mortality for 48 slash
pine families (four- to six-years-old) planted in six
progeny tests in southeast Georgia . . . . . .. 32

10. Percentage of trees dead and classified as 'rust bush many
cankers' for 48 slash pine families (four- to six-years-old)
planted in six progeny tests in southeast Georgia . . 33








Table Page


11. Results of analyses of variance of the proportion
of trees with different fusiform rust symptoms (arcsin
square root transformation) and the mean number
of these symptoms per tree for family and test
differences . . . . . . . .... .... .39

12. Significant replication (Reps), general combining
ability (GCA) and specific combining ability (SCA)
F values (5% level of significance) obtained from
analysis of variance using DIALL for number of
fusiform rust galls of several types/locations
on slash pine (18 analyses for each type) . . . .. 41

13. Significant replication (Reps), general combining
ability (GCA) and specific combining ability (SCA)
F values (5% level of significance) obtained from
analysis of variance using DIALL for the proportion
(arcsin square root transformation) of slash pine trees
with various fusiform rust symptoms (18 analyses for
each symptom) . . . ......... . . . 42

14. Significant site (S), replication (Reps), general
combining ability (GCA), specific combining ability
(SCA),and interaction (GCA x S) F values (5% level of
significance) obtained from analysis of variance
using DIALL for number of fusiform rust galls of
various types/locations on slash pine, planted at
several sites (nine analyses for each type) . . .. .45

15. Significant site (S), replication (Reps), general
combining ability (GCA), specific combining ability
(SCA),and interaction (GCA x S) F values (5% level of
significance) obtained from analysis of variance
using DIALL for the proportion (arcsin square root
transformation) of slash pine trees with various rust
symptoms planted at several sites (nine analyses
for each symptom) . . . . . . . . . . 46

16. Individual (hi) and family heritibilities (h )
and their standard errors (s) for rust incidence,
rust bush many cankers (RBMC), rust-associated mortality
(RAM) and the mean number per tree of 10 fusiform
rust gall types for slash pine, four- to
six-years-old . . . . . . . . . . . 50








Table Page


17. Genetic correlations between various fusiform
rust gall types and locations (mean number per
tree), rust incidence and mortality (RAM and RBMC)
for slash pine four- to six-years-old . . . ... 52

18. General combining ability estimates of the
proportion of trees with fusiform rust galls for
25 slash pine parents based on progeny performance
in six locations in southeast Georgia, aged
four to six years . . . . . . .... . 54

19. Percentage of seedlings in 20 slash pine families
with various reaction types six months after
artificial inoculation with fusiform rust and
the resistance index of the families . . . .... .56

20. Mean squares for the analyses of variance of the
proportion of slash pine seedlings per family
expressing various disease symptoms resulting from
artificial inoculation with fusiform rust at the
Resistance Screening Center, Asheville, North Carolina . 57

21. Individual heritabilities (h2) and their standard
errors (s) for artificial inoculation (AI) symptoms
by both proportion of trees with the symptoms and
mean number of symptoms per tree . . . . .... .59

22. Pearson product-moment correlations between the
mean number of artificial inoculation (AI) symptoms
per tree of slash pine seedlings inoculated with
fusiform rust, based on the progeny of 20 families . . 60

23. Pearson product-moment correlations between
field galls and artificial inoculation symptoms
(proportions) and the resistance index of 17 slash
pine families and fusiform rust . . . . .... .62


viii








Table Page


24. Results of artificial inoculation with high
concentrations of inoculum (2.0 x 106 spores/ml)
on 17 slash pine families, Resistance Screening Center,
Asheville, North Carolina . . . . . . .... .65

25. Comparison of family means for several fusiform rust
disease symptoms on artificially inoculated 1-0 slash
pine seedlings . . . . . . . .. . .67

26. Least squares estimates of the mean number of fusiform
rust galls per tree by gall type or location for 48
slash pine families (four- to six-years-old) planted
in six progeny tests in southeast Georgia . . .. 79

27. Least squares estimates of the mean proportion
of trees infected with fusiform rust galls by gall
type or location for 48 slash pine families (four-
to six-years-old) planted in six progeny tests in
southeast Georgia . . . . . . . . . 81

28. Variance components and their standard errors for
replication, general combining ability (GCA),
specific combining ability (SCA) and error effects
for the mean number of galls per tree for various
fusiform rust gall types or locations as computed
by the FORTRAN program, DIALL for each crossing
set and test . . . . . . . .... ..... .83

29. Specific combining ability for fusiform rust incidence
and fat galls based on the proportion of trees
diseased or having fat galls. Values reflect
performance of 48 families (four-to six-years-
old) in six progeny tests in southeast Georgia . . .. 88















LIST OF FIGURES


Figure Page

1. Gall types commonly found on field-grown slash
pine trees infected with fusiform rust . . . . .. 20

2. Galls found on slash pine seedlings six months
after artificial inoculation with fusiform rust . .. 22

3. Frequency distribution for slash pine trees
by the total number of fusiform rust galls, in six
progeny tests in southeast Georgia, at four-, five-,
and six-years-old . . . . . . . . . . 35

4. Frequency distribution for two slash pine families,
0098 x 0088 (resistant) and 0350 x 0354
(susceptible), for the total number of fusiform
rust galls at age five years, planted in a progeny
test in Camden County, Georgia . . . . . . .. 38
















Abstract of Dissertation Presented to the Graduate School of
the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy





GENETIC VARIATION IN SYMPTOMOLOGY OF SLASH PINE
IN RESPONSE TO FUSIFORM RUST

By

Patricia Adlene Layton

December,1985

Chairman: R. E. Goddard
Major Department: Forest Resources and Conservation

Genetic variation in the reaction of slash pine to fusiform

rust was investigated in field and greenhouse situations. In the

field, four- to six-year-old plantings of five crossing sets, each

involving five or six parents, were studied. Each crossing set was

planted at two or four sites. Rust incidence in the tests ranged

from 46% to 89%. Survival and stem form of each tree was

determined. Galls on infected trees were counted and classified as

to location and type.

Variances in traits measured were analyzed by crossing set,

crossing sets among sites and all crossing sets at all sites.

Significant differences among families were found for some gall

types, gall locations, total number of galls per tree, incidence,










mortality and bush stem form. Additive genetic effects were found

in many traits, but the number of fat galls had non-additive

effects. Total number of galls was the best field measure of

susceptibility, having high genetic correlations with mortality and

bush stem form.

Seedlings from 17 of the field-tested families were

artificially inoculated. After six months, there were differences

among families for the proportion of seedlings and number per tree

with seven resistance and susceptibility indicators. Although

heritabilities were often low, those for proportion of trees with

galls, purple spots and adventitious shoots were 0.23, 0.28 and

0.33, respectively.

Following artificial inoculation with high concentrations of

rust inoculum, families varied in the number of red spots at 10 and

48 days. Correlations with family field performance were low.

Galls developing on artificially inoculated one-year-old

seedlings resembled field gall types more than those on younger

seedlings. Families differed in percentage of trees with rust and

the mean number of fat galls. Three of four phenotypically

resistant males increased family resistance to a level equal to or

better than a rust-resistant orchard check.

These studies indicated genetic variation for specific disease

reactions in the field and confirmed such variation in artificial

inoculation studies. Proportion of trees infected, total galls per

tree, mortality, and bushstem form had the highest heritabilieies

and should be considered in screening for rust resistance.
















INTRODUCTION


The major pest of southern commercial forests is fusiform rust

(Cronartium quercuum (Berk.) Miyabe ex Shirai f. sp. fusiforme)

(Cqf). In a 1979 survey of five states (South Carolina, North

Carolina, Virginia, Georgia and Florida), more than one-third of the

acreage of slash (Pinus elliottii Engelm. var. elliottii) and

loblolly (P. taeda L.) pine had more than 10l of the trees infected

(Robert Anderson, personal communication 1985), representing an

annual loss of $32,894,000. Losses throughout the South are

expected to be much greater.

Disease incidence has been increasing over time (Schmidt et al.

1974; Griggs and Schmidt 1977). Management practices, such as

mismatching species and site, fertilization and intensive site

preparation (Powers et al. 1981), are implicated in this increase.

Another reason for the increase may be the abundance of

rust-susceptible Quercus spp. (alternate host of the fungus) in

southern forests (Squillace et al. 1978). Modifying these

management practices as well as utilizing more effective

silvicultural techniques for suppressing Quercus spp. may help










reduce the epidemic. Planting rust-resistant pines is an effective

way of significantly decreasing disease losses (Schmidt et al. 1981;

Schmidt et al. 1985).

The disease is caused by a heteroecious, macrocyclic rust

fungus. Pycnia and aecia occur on pine while uredia and telia occur

on oak. Basidiospores produced from telia on oak infect pine during

favorable environmental conditions. Most southern red oaks,

especially water oak (Q. nigra L), are susceptible to the fungus

(Dwinell 1974). Although much is known about the life cycle

(Dwinell 1977), knowledge is lacking about the genetics of the

fungus. There is a wide range of variation in the ability of Cqf to

infect pine (Snow and Kais 1970; Snow et al. 1975; Powers et al.

1977). Spores from galls on resistant families evidenced increased

virulence in comparison to other spore collections when used to

reinoculate the resistant families (Griggs and Walkinshaw 1982;

Powers et al. 1978; Snow and Griggs 1980).

Various review articles have summarized family resistance in

slash and loblolly pine (Goddard and Wells 1977; Powers et al.

1981; Schmidt et al. 1981). Certain loblolly pine geographic seed

sources also are resistant to infection by Cqf (Wells and Wakely

1966).

In the past, family resistance to Cqf has been evaluated by

percentage incidence in both field and greenhouse studies. This

method is effective in screening families for resistance, but does

little to identify resistance mechanisms or their inheritance. More

recent methods have evaluated disease resistance by examining










symptoms resulting from artificial inoculation (Carson 1984,

Framptom et al. 1983; Walkinshaw et al. 1980).

The objectives of this study were as follows:

1. To examine and describe qualitative characteristics

(symptoms) of the response of slash pine to Cqf, both

in the field under natural conditions and in the

greenhouse using artificial inoculation techniques;

2. To determine heritability and genetic correlations of

the traits; and

3. To correlate field and greenhouse observations and

suggest symptoms that may lead to a more precise

prediction of field response by greenhouse testing.














LITERATURE REVIEW


As fusiform rust Decame epidemic, tree improvement programs

incorporated disease resistance into their selection and breeding

programs. Although not a total solution, resistance is the primary

pest management strategy in intensively managed forests (Schmidt et

al. 1981; Schmidt et al. 1985). Original mass selections were

essentially random for rust resistance, out approximately 7.5% had

some degree of resistance when evaluated using artificial

inoculation techniques (Goddard and Arnold 1966). Goddard and

Arnold (1966) reported 0.199 as an estimate of heritability from

artificial inoculation. Estimates of heritability for percentage

rust incidence in field tests of slash pine varied greatly. Sohn

and Goddard (1979) stated that they varied with the incidence levels

in the tests examined. Individual heritabilities varied from 0.054

to 0.238 (Rockwood and Goddard 1973) in several progeny tests. Sohn

(1977) estimated individual heritabilities for incidence ranging

from 0.067 to 0.285 in other progeny tests. The relationship

between incidence and heritability in these studies supported Sohn

and Goddard's (1979) conclusion.










Various estimates of rust resistance are reported for loblolly

pine. Kinloch and Stonecypher (1969) reported family heritabilities

for incidence of 0.65 to 0.85 over four different environments.

Blair (1970) estimated heritabilities for number of galls per tree,

severity index and percentage incidence. In one field test, these

estimates were 0.29, 0.22 and 0.20, and in another they were 0.09,

0.04 and 0.12, respectively. In the same tests as Blair, but at an

older age, Barker (1973) estimated the heritabilities of the

severity index to be 0.11 and 0.35. These heritabilities supported

the concept that substantial gains in rust resistance were possible.

Screening selections for rust resistance became a priority.

Inoculation techniques were developed to screen families for

incorporation in breeding programs (Jewell 1960; Goddard and Schmidt

1971). Schmidt (1972) describes early techniques and their

deficiencies were identified by Laird and Phelps (1975) as

1. temperature and/or humidity not adequately controlled;

2. inoculation time not controlled;

3. inoculum concentration not controlled;

4. techniques not applicable to large numbers of

seedlings;

5. results not repeatable; and

6. results not quickly obtainable.

Laird and Phelps (1975) tested three methods of artificial

inoculation and determined that the concentrated basidiospore spray

system (CBS) (Matthews and Rowan 1972) was best for minimizing these

problems.










In 1973, the United States Forest Service established a

screening center (RSC) for rust resistance in loblolly and slash

pines from throughout the South at Asheville, North Carolina,

utilizing the CBS system. There, in a controlled environment,

seedlings could be inoculated with various pathogen sources at

optimum inoculum levels (Hubbard and Anderson 1980). The percentage

of trees galled at six to nine months of age was used to rank family

resistance. Resistant and susceptible checks were used in every

test.

Screening by artificial inoculation (all methods) was plagued

by lack of adequate correlation with family field performance. The

University of Florida Cooperative Forest Genetics Research Program

began field testing families rather than relying solely on

artificial inoculation procedures. Short-term field progeny tests

were established to mass screen selections in their tree improvement

program (Goddard et al. 1972). Later, other tests were established

to retest the most resistant families in areas of high rust

incidence (Goddard et al. 1984). Many families with high degrees of

resistance were found with these methods. However, open-pollinated

families varied tremendously in their reactions. Powers and Zobel

(1978) found significant differences in infection occurred on

open-pollinated seedlings of the same clones produced in different

orchards. These results illustrate the strong influence of the

pollen parent in resistance. To understand the genetics of

resistance better, testing of full-sib families (artificial

inoculations and field tests) must be performed.










Artificial inoculation methods were useful in exploring host

variation in response to infection. Families were found that were

resistant to Cqf from some regions, but not to spores originating in

other areas (Snow and Kais 1970). Gall form (ratio of length to

width) varied by host family and inoculum source (Snow et al.

1982). Resistance was overcome in some families by particular

isolates of the pathogen (Griggs and Walkinshaw 1982; Powers and

Dwinell 1978). Selection for increased virulence in the pathogen

has been found in slash pine families (Dinus et al. 1975; Griggs and

Walkinshaw 1982; Snow and Griggs 1980). Snow et al. (1976) stated

that the greater virulence of inocula from resistant trees was

evidence of pathogenic specialization.

Virulence varies greatly in the fungus. Powers (1980) found as

much variation in 10 single aeciospore inoculations from one gall as

he did in stand- or county-wide collections of aecia. Powers and

Owinell (1978) reported that pathogen virulence had not increased

with time when comparing isolates from galls originating in 1945 and

1970; however, the 1970 isolates were more variable in their

response.

The fungus infects pine in a haploid (IN) state, while the

aeciospore, collected from pine for testing, is a dikaryon (N+N).

The biological mechanism for this change is not known. This was

studied in the related species C. ribicola by Hirt (1964), but he

found no conclusive evidence of sexual recombination. Until the

sexual processes of Cqf are understood, the exact nature of

pathogenic variation will be unknown.










Increasing the accuracy of assessment of host variability is a

continuing goal. Walkinshaw et al. (1980) developed a scoring

procedure for open-pollinated slash pine families tested by the RSC

to better predict resistance. They examined twelve traits.

Correlations between RSC performance and field evaluations were

increased by utilizing the proportion of trees in a family with 1)

smooth galls, 2) symptoms without swelling and 3) fat galls. These

variables accounted for 62% of the variation in field performance

evaluations. Predicting field performance with similar methods for

loblolly pine has not been reported. Walkinshaw and the staff at

the RSC continue to test methods for this species (C. H. Walkinshaw,

personal communication 1984).

Walkinshaw and Anderson (1983) described seven fusiform rust

symptoms found on seedlings in the greenhouse and eight on trees in

the field. In the greenhouse, the most resistant symptom found was

a purplish discoloration on the stem and/or a needle base, called

SYMNO. Fungal tissue was not found in the area of discoloration. A

short gall of less than 25 millimeters and galls with a rough

textured, slightly discolored surface were also indicative of

resistant reactions. More susceptible reactions included fat galls

(gall diameter is twice that of the stem below the gall), typical

galls (gall is fusiform shape and not abnormally large or small),

smooth galls (gall surface is smooth and generally green on

six-months-old slash pine seedlings) and baseball bat galls (base of

the gall is constricted, but there is a typical fusiform shape at

the top). Baseball bat, fat and typical galls were also found on










trees in the field. In addition, field symptoms included thin galls

(only slightly swollen), twisted galls (stem twists in the area of

the gall), sunken galls (a depression in the gall surface) and

witches' broom galls (a loss of apical dominance). The authors

proposed using these symptoms to increase the prediction

capabilities.

Prediction methods rely on the premise that resistant or

susceptible reactions in seedling progeny of a particular

open-pollinated family can predict response of other family members

grown in the field. Day (1974) advised that possible limitations of

this premise include the following:

1. The observed resistance or susceptibility may be a

feature of the juvenile tissues cotyledonss and

unsuberized stems).

2. Observed reactions may reflect test environment as

opposed to host genotype.

Other problems with relating field and artificial inoculation

tests are pathogen related. Composite inoculum from the area where

host families are to be outplanted eliminates (or complicates) many

of the problems of pathogenic variability. Inoculum concentrations

are extremely variable in the field, but are fairly constant in

artificial inoculations. Lundquist et al. (1982), utilizing

extremely high inoculum concentrations, found breakdowns in

resistance for some slash pine families--a product of both inoculum

concentration and host genotype. In their tests, families were

ranked by observations of pigmented spots on seedlings and the










rankings were similar to those achieved by the CBS method

(percentage galled). The authors suggested that two types of

resistance may exist: one that persists under high inoculum loads

and one that does not. They did not consider that high inoculum

concentrations significantly increase the probability of a virulent

spore's landing on a seedling. Further testing of this method is

needed.

Lundquist and Luttrell (1982) reported that pigmentation

patterns found in response to rust inoculation were similar to those

observed when seedlings were wounded or exposed to ultraviolet

light. They suggested this may have been a specious effect or a

reaction by-product. Timing of pigmentation varied with

host-parasite interaction and did not vary with other stress-related

reactions, suggesting it was related to a resistance mechanism.

Miller et al. (1976), utilizing different methods of

inoculation from Lundquist and Luttrell (1982), suggested there

might be four general types of host-parasite interactions at work.

The first interaction was the inability of germinated basidiospores

to penetrate and/or infect host epidermal cells. This was

considered total incompatibility of host and pathogen. The second

interaction, subliminal infection, resulted in infection and a

sparse spreading of mycelium with a few scattered haustoria. Host

cells did not show the distortion typically found in infected

trees. Miller et al. (1976) suggested that the pathogen in this

interaction was avirulent. The authors described three

hypersensitive reactions as a third interaction. Two of these










hypersensitive reactions resulted from stem infections and produced

darkly stained reaction zones. Purple stem lesions were visible for

both reactions. They differed primarily in size and time of

development. The third hypersensitive stem reaction resulted only

from primary needle infection. Internal reaction zones developed

and were stabilized from enlargement within three months. The

fungus was confined to the reaction zone and the seedlings recovered

from infection. There was no evidence of resistance in the needle

tissue; resistance occurred only after the fungus reached the

cambial area and began to spread. The fourth host-parasite

interaction was typical gall development commonly found with

susceptible pines and virulent Cqf.

Resistant zones similar to those identified by Miller et al.

(1976) were described by Jewell et al. (1982). Fungal tissues were

effectively isolated in resistant pines. In 1982, Jewell et al.

reported that wounding trees produced reactions similar to those

produced by fusiform rust infection. Resistant seedlings

differentiated tissues incompatible to rust thereby limiting the

spread of the fungus. Susceptible seedlings produced initial

resistant reactions, but were unable to produce incompatible tissues.

Walkinshaw (1978) found seedlings that had necrotic areas

surrounded by a periderm. His discussion did not point out

differences between resistant and susceptible seedlings, but

centered mainly on the number of necrotic, tannin-filled areas.

Lundquist and Miller (1984) described a relationship between

macroscopic pigmentation patterns and microscopic events in

resistant lesions. These lesions resulted from artificial










inoculation with high concentrations of inoculum. Developing

phellem cells formed a boundary that effectively checked the spread

of the disease and distorted infected cells, pushing them toward the

epidermis. Prior to or in conjunction with periderm formation, an

impermeable layer was formed.

Mullick (1977) discussed three non-specific processes that are

initiated during pathogen attack: 1) phellogen restoration, 2)

vascular cambium restoration and 3) sapwood conduction blockage.

The process triggered was dependent on depth of attack, with

processes one, two and three occurring with increasing depth of

injury. They may act independently or collectively depending on the

attack. The author used microscopic techniques not previously used

in observation of Cqf infections in pine tissues. His processes

appeared to describe the observations of Frampton et al. (1983),

Gray and Amerson (1983), Jewell et al. (1982), Lundquist and Miller

(1984), Miller et al. (1976), and Walkinshaw (1978). Mullick

(1977) proposed that host resistance or susceptibility is a function

of the successfulness of completing the periderm under the

pathogen's influence. During restoration, the pathogen is exposed

to a host of chemicals that might influence its ability to

successfully survive (e.g. phytoalexins). Virulence results from

the pathogen's 1) not triggering the host response, 2) blocking the

triggered response or 3) growing and spreading faster than phellogen

formation can respond.

The microscopic differences in host response as discussed by

the authors above should be represented by corresponding differences







13


in macroscopic symptoms. Carson (1984) found family differences in

macroscopic symptoms on artificially inoculated loblolly pine.

Walkinshaw and Anderson (1983) described different macroscopic

symptoms on slash pine in field studies. However, they did not

present evidence of family variation for these symptoms.














MATERIALS AND METHODS


Four experiments were conducted to examine fusiform rust

resistance variation in slash pine. Three experiments contained

full-sib families in five sets of crosses. The crossing sets,

belonging to Brunswick Pulp Land Company (BPL), were designed as

four- or five-parent half-diallels with an additional cross between

the females and a male from another geographic area (i.e. a wide

cross). Field tests of the crossing sets ranged from four- to

six-years-old. Seed from these sets were sent to the RSC for a

standard artificial inoculation study and a high inoculum

concentration technique study (Lundquist et al. 1982), Experiments 2

and 3, respectively. The fourth experiment involved artificially

inoculating a group of 1-0 nursery grown seedlings at RSC. These

seedlings were donated by Container Corporation of America (CCA) and

are part of a rust-resistant factorial mating design.


Field Tests--Experiment 1

Six progeny tests containing four diallel crossing sets each

were planted by BPL near Brunswick, Georgia. Progeny tests were

randomized complete block plantings with three to five blocks and

seven- to ten-tree row plots for each family. Crossing sets one,

two, four, five and nine were examined. Test sites were cleared,

windrowed and bedded prior to machine planting. Tests 34 and 35









Table 1. Location, year of establishment, site index, fifth year
height and rust incidence of Brunswick Pulp Land
Company's diallel progeny tests.


Crossing 5th Year Rust Site
Test Location/County Planted Set Mean Ht. Incidence Indexl/

(Georgia) (yr) (feet) (%)


34 Wayne 1978 1,3,4,5 14.5 89 60
35 Wayne 1978 1,3,4,5 14.8 61 55
36 Camden 1979 1,2,6,10 10.4 46 65
37 Camden 1979 1,2,6,10 12.1 58 60
40 Wayne 1980 2,4,5,9 -b/ 54 65
41 Camden 1980 2,4,5,9 -b/ 66 65


A/Average height of dominant and codominant trees at age 25.

Y/Fifth year heights were unavailable.


were disked before bedding. All tests sites were in the coastal

plain. Tests 34, 35 and parts of 37 were planted on well-drained

ultisols. Tests 36 and 40 were planted on somewhat poorly to poorly

drained spodosols. Test 41 and portions of Test 37 were planted on

somewhat poorly drained entisols. Site index at age 25 and fifth

year heights (Tests 34 to 37) are presented in Table 1, which

contains details of the tests and their locations. The parents and

crosses made are shown in Table 2. Not all crosses were represented

at all possible locations. A listing of crosses in each test may be

found in Table 3.

The 25 parents in the crosses were evaluated previously for

rust resistance in open-pollinated progeny tests by the University










of Florida Cooperative Forest Genetics Research Program. Parental

evaluations (Table 4) are expressed as weighted mean standard

deviations away from progeny test mean incidence levels. The

weighting factors were for incidence level (Sohn and Goddard 1979)


Table 2. Sets of slash pine crosses used
inoculation experiments.


Female


Set 1


0354
0350
0098


Set 2


0157
0146
0096


Set 4


0288
0352
0001


Set 5


0019
0141
0060


Set 9


0284
0270
0295


0350
1,2,3a_/




0146
1

1,3


0352
1



0141
1




0050
1
1,2,3


0098
1,2
1
1


0096
1
1



0001
1
1,2,3
1


0060
1,2,3
1



0295
1,2
1,2


in field and artificial


Males


0088
1,3
1,2
1,3


0065
1
1
1


0159
1,3
1,2,3
1


0287
1,2,3
1,2,3
1


0355
1
1,2
1


0048
1,2,3
1,2,3
1


0071
1
1



0050
1
1
1


0047
1,2,3
1
1


0064
1,2
1,2,3
1


a/1=Cross included in Experiment
Experiment 2; 3=Cross included


1; 2=Cross included in
in Experiment 3.










Table 3. Percentage of trees infected with fusiform rust for 48
slash pine families (four- to six-years-old) planted in
six progeny tests in southeast Georgia.

Least
Family Progeny Tests Squares
Female Male 34 35 36 37 40 41 Mean
----------------------------------%----------


0354 0350
0098
0088
0048

0350 0U98
0088
0048
0354


100.0
96.7
93.3
90.0

100.0
100.0
89.6


100.0
55.8
65.2
63.3

52.6
68.5
56.7


97.1
48.2
38.2
49.1

47.6
48.1
37.9
78.6


95.8
58.9
43.7
69.6

57.3
50.2
70.5
97.1


0098 0088 76.7 31.1 31.7 23.3
0048 63.3 13.3 17.3 22.3


0157 0146
0096
0065
0071

0146 0096
0065
0071

0096 0065
0146

0288 0352
0001
0159
0050

0352 0001
0159
0050

0001 0159
0050

0019 0141
0060
0287
0047

0141 0060
0287
0047


42.9
52.3
56.2
24.8


51.9
82.4
75.1
83.1


42.9
37.1 46.8
27.8 45.4


59.4
70.4
68.3
62.5


38.8
78.8
61.1
82.2


97.7
64.9
58.5
66.5

64.8
70.1
66.1
95.2

37.8
25.0

56.9
79.5
74.3
72.4


45.6
25.0 50.8 51.4
40.3 62.5 53.5


46.2 56.2 56.8 68.2 65.2
46.4 61.7 59.8


80.4
96.7
97.0
100.0

73.4
81.7
83.3

96.7


100.0
96.7
93.0
88.9


86.7
66.7
79.3
45.8

40.7
51.1
27.4

73.3
46.3

100.0
89.6
71.7


95.2


91.7


40.9
56.9
71.5
48.7

41.1
64.6
31.0

73.6


91.9
66.9
66.9
45.0

40.8
43.5
50.0


58.7
53.8
96.1
30.0


62.9
63.7
82.9
57.7


30.4 41.4
42.8 54.9
41.7


92.8
85.5
65.5
80.0


77.9
45.1

95.8
82.9
70.5
67.3


76.8
36.7 47.6
66.4










Table 3--continued.


Family
Female Male


Least
Progeny Tests Squares
34 35 36 37 40 41 Mean
---------------------,~6------f,----------------


0060 0287
0047

0284 0050
0295
0355
0064

0270 0050
0295
0355
0064

0295 0355
0064
0270


48.6
85.0 46.7


27.5
25.6

43.3
57.4
59.2
70.8

36.9
45.0
66.4
70.8

78.2
82.3
41.9


52.5 45.8
45.9


51.4
61.3
70.4
83.7

59.4
85.5
62.5
97.2

100.0
88.5
58.6


55.7
65.2
71.3
75.7

54.7
71.7
72.2
91.4

96.6
92.7
67.0


Table 4. University
and number
parents.


Florida rust evaluations, standard errors
tests used for evaluation of slash pine


No. of Rusti/ Standard/ No. of Rust Standard
Parent Tests Eval. Error Parent Tests Eval. Error
0157 9 .19 .12 0287 7 -.06 .35
0146 19 1.20 .14 0047 18 1.10 .17
0096 10 .86 .16 0284 13 .75 .18
0065 4 .82 .11 0270 7 .39 .14
0071 9 .64 .10 0295 8 .10 .18
0288 11 -.03 .18 0355 14 .31 .13
0352 10 .68 .18 0064 4 .53 .26
0001 10 .41 .13 0354 11 .12 .13
0159 14 .45 .20 0350 8 .49 .31
0050 12 1.07 .25 0098 12 1.15 .24
0019 10 -.41 .21 0088 9 .29 .14
0141 11 -.08 .09 0048 21 1.53 .17
0060 7 .07 .36 0621 14 .35 .18
0298 12 .31 .17 0618 18 .98 .14
0620 12 .60 .21


A/Rust Evaluation = Infection Weight x
Progeny Mean)/(Error Mean Square) /2).


Lot Weight x ((Test Mean -


b/Standard Error = (Variance of a family rating/number of tests)1/2.









and number of families in a test (Lewis 1973). The standard error

(Table 4) of an evaluation was based on the unweighted standard

deviations.

Data were collected from the tests in either March or June,

1984. Gall development was assumed not to change dramatically

during this time. However, each test was measured within a

three-day period and four of the six tests were measured during a

two-week period in June. Tests 36 and 37 were measured during a

one-week period in March. Data collected from each tree included

survival, stem form, and the number of 1) stem, 2) limb, 3) stem-

to-limb, 4) fat, 5) thin, 6) typical, 7) basally truncated fusiform

(BTF) (referred to as baseball bat by Walkinshaw and Anderson

(1983)), 8) witches' broom, 9) twisted, and (10) sunken galls. Gall

types are illustrated in Figure 1.


Artificial Inoculations

Experiment 2--RSC Procedures

Seventeen families (Table 2) were tested at RSC using routine

inoculation procedures described by Anderson et al. (1983). The

inoculum originated in the Nassau County, Florida, area which is

approximately 30-75 miles south of the field test sites. This same

inoculum source was used by Walkinshaw et al. (1980) in developing

their rust resistance prediction system. Two runs replicationss) of

24-60 seedlings from each family were inoculated with a

concentration of 25,000 spores per milliliter of solution. The

variable number of seedlings was due to germination and survival

differences. Three standard check lots were included with this
















r'-








Typical Fat Thin
















BTF Witches's Broom
















Twisted Rust Bush Sunken
Many Cankers

Figure 1. Gall types commonly found on field-grown slash pine trees
infected with fusiform rust.










test. Data collected from each seedling included the number of

1) symptoms without swelling (SYMNO), 2) galls present, 3) rough

galls, 4) fat galls, 5) galls less than 25 millimeters long (LT25M),

7) BTF galls and 8) sunken areas on the gall and the occurrence of

adventitious shoots in the galled area. Gall types are illustrated

in Figure 2. Data were collected with the assistance of an RSC

staff member who routinely collects data on seedlots tested there.

Presence of a pink blush on seedlings 10 days after inoculation

was also recorded. This symptom may be related to the red spots

reported by Lundquist and Luttrell (1982). It consisted of a faint,

irregular spot on the hypocotyl one to three weeks after

inoculation. The relationship of this symptom to resistance is

unknown.


Experiment 3--High Inoculum Concentration

Fifteen full-sib families from the BPL crosses (Table 2) were

inoculated following techniques of Lundquist et al. (1982). The

inoculum concentration of 2.0 x 106 spores per milliliter was

extremely high, allowing only 20 seedlings per family to be

inoculated with the limited quantity of available basidiospores.

Eleven of the families were used in Experiment 2 (Table 2).

Numbers and types of spots after 10 and 48 days were tallied for

each seedling in the test. Personnel at RSC transferred the

seedlings to different trays between the two measurement dates,

preventing observation of the increase in spots per tree over time.
























SYMNO


Thin

Thin


h


Gall with
Adventitious
Shoots


Gall with
Sunken Area


Figure 2. Galls found on slash pine seedlings six months after artificial
inoculation with fusiform rust.


LT25M


Rough










Experiment 4--Inoculation of 1-0 Seedlings

This experiment contained six full-sib families and two check

lots grown in the CCA nursery near Archer, Florida. The seedlings

were lifted in January 1984, transplanted in one gallon pots filled

with sterilized medium and transported to the RSC. There were two

seedlings per pot and 20 pots per family or check lot. The

seedlings were inoculated after their first flush of growth with the

CBS system using 35,000 spores per milliliter. The pathogen source

was from Louisiana as the trees flushed earlier than expected and

the Nassau County pathogen source was unavailable. The six families

were 1) 0298 x 2002, 2) 0298 x 2005, 3) 0298 x 2013, 4) 0620 x 2002,

5) 0618 x 2007 and 6) 0621 x 2005. The four male parents (2002,

2005, 2007 and 2013) were phenotypically rust resistant, selected in

stands exposed to very high levels of natural inoculum. Most trees

in these stands were severely infected. The male parents are

presumed likely to have genetic resistance (Goddard et al. 1975).

The female orchard selections and their rust evaluations are shown

in Table 4. The check lots were groups of seedlings from a

rust-resistant orchard and an orchard where there was no selection

for rust resistance. Observations on these seedlings were the same

as those in Experiment 1.


Analysis

To determine least squares estimates of family performance

across tests, the GLM procedure in SAS (SAS Institute Inc. 1982b)

was used. The model for the analysis was Y. = u + T. + F. +
ei where Ti as the effect of the test, F was the
e.., where T. was the effect of the i test, F. was the
I1 J 3










effect of the j tfull-sib family and e.i was the error

representing the family x test interaction. The analysis was

performed on the means of gall types and locations, and the

proportion of trees (arcsin square root transformation) having a

particular gall type or location.

Each diallel crossing set was analyzed using DIALL, a FORTRAN

program (Schaffer and Usanis 1969). This program allows computation

of general (GCA) and specific (SCA) combining ability, site,

replication and GCA by site effects (Table 5) by a least squares

approach to the analysis of variance. The computation method used

an abbreviated forward Doolittle solution rather than direct matrix

inversion. Estimates of the variance components were calculated by

inverting the matrix of coefficients of the expected mean squares


Table 5. Analysis of variance and expected mean squares for a
modified half-diallel as computed by DIALL.


Source df Expected Mean Squares


Sitea/ 1/ Ve + c9VGCAxS + clOVR + cllQS

Replication within Site 4 Ve + c7VGCAxS + c8VR

General Combining Ability 4 Ve + C4VGCAxS + c5VSCA + C6VGCA
(GCA)

Specific Combining Ability 4 Ve + c2VGCAxS + c3VSCA
(SCA)
GCA x Site 4 Ve + clVGCAxS

Error 36 Ve


./Analysis for one site follows a similar model except for site
and GCA x site effects.

-/The degrees of freedom here are an example of those generated
by the program.










and multiplying the computed mean squares for each variable by this

inverse. The standard error of a variance component was computed by

the equation (Anderson and Bancroft 1952)

std err = (i. (2a MS.2/DF +2))1/2

where a.= the coefficients of the linear combination of
1
the mean squares used to estimate the variance

component,

MS.= the ith mean square,

DFi= the degrees of freedom for the i effect.

In calculation of the F test, when the denominator mean square was

smaller than a lower order effect containing that variance component

in its expected mean square, the lower order effect was used as the

denominator. For example, in an analysis of one site, if the SCA

mean square was smaller than the error mean square, the error mean

square was used as the denominator in the F test computation for

GCA. DIALL did not adjust variance component computations to

reflect this problem. All VGCA were computed by subtracting the

mean square for SCA and dividing by the coefficient of VGCA'

Traits analyzed included total number of galls, number of galls

by location on the tree, and all six gall types found on trees in

the field. Plot means were used in this analysis. Additionally the

proportion of trees in a family having a certain gall type or gall

location was analyzed. Proportion data were transformed by the

arcsin square root transformation (Steel and Torrie 1960). GCA by

site interaction was tested by pairing only those diallel crossing

sets planted in the same year to avoid confounding year effects.










Estimates of heritability and variance components were computed

oy pooling all families at all sites as if they were in a nested

design (Table 6). This method made use of all data available to

compute the estimates. The procedure, while computationally

feasible, would oe expected to produce biased estimates of variance

components. The analysis accounted for the 18 crossing set-site

groups (CSS), males within CSS, females within males within CSS and

replications within females within males within CSS (Table 6). The

NESTED procedure in SAS (SAS Institute Inc. 1982b) was used to

analyze individual tree data. Incidence, rust-associated mortality

and rust bush many cankers were analyzed as binary data.


Table 6. Analysis of variance and expected mean squares for a
four-stage nested design using individual tree data.

Source df Expected Mean Squares


Crossing Set-Site (CSS) 17

Males (CSS) 58 Ve + 9.20VR(FMCSS) + 31.73VF(MCSS)+
59.23VM(CSS)

Female (Males, CSS) 75 Ve + 9.05VR(FMCSS) + 30.80VF(MCSS)

Replication (Females, 405 Ve + 8.28VR(FMCSS)
Males, CSS)

Error 4179 Ve









Individual tree heritabilities were calculated by the equation

(Wright 1976)

2 4VM(CSS)
I V + VR(FNCSS) + VF(MCSS) + VM(CSS)

where Ve = variance due to trees within plots,

VR(FMCSS) = variance due to replications nested
within females, males and CSS,

VF(MCSS) = variance due to females nested within
males and CSS,

VM(CSS) = variance due to males nested within CSS.
Family heritabilities were calculated similarly by the equation


S2 VM(CSS)
hF V V/59.33 + VR(FSS)/31.73 + V (CSS)/9.20 + V(CSS)

Standard errors(s) for individual heritabilities were

calculated by the equation (Wright 1976)

(1 h2/4) (1 + Kh2/4)
I =
[K(M 1)/2]1/2

and for family heritabilities by the equation (Wright 1976)

s = (1 t)(l + Kt)
[K(M 1)/2]1/2
where t = hJ/4
h+= individual heritability,

K = number of trees in a male half-sib family,

M = number of male half-sib families.

The NESTED procedure calculates a covariance analysis.

Utilizing the covariance and variance estimates of the male source









of variation, an estimate of 1/4 of the additive genetic variance,

genetic correlations were calculated by the equation

covxy
r 11___ __
rg (VX x Vy)I2


where covxy = the covariance between two traits for males
nested within CSS,

VX = VM(CSS) as described above for a trait,

Vy = VM(CSS) as described above for another trait.

General combining ability estimates for traits were computed

using the SAS MATRIX procedure (SAS Institute Inc. 1982b). The

computer program for this was developed by J. P. van Buijtenen

(personal communication 1983).

Data in Experiment 2 were analyzed using the GLM procedure in

SAS. Factors in the random model were runs, family, runs x family

and trays within runs x family was the error component (Table 7).

This model was used because the results from it would be similar to

results obtained by RSC. The RSC does not account for male and

female effects in their screening procedure analysis. Tray means

for the traits observed were used in the calculation.

Pink blush was analyzed with a one-way analysis of variance on

three to six tray means for each of 20 families.
Variance components were determined using the VARCOMP procedure

(MIVQUEO method) (SAS Institute Inc. 1982b) for females (VF),

males (VM), run x males (VRM), run x females (VRM), females x

males (VMF), trays within runs x females (VT(RF)), trays within

runs x males (VT(RM)) and error (V ). This model accounted for
T(RAI)










male and female effects and was a more complete model than the

previous one (Table 7).

Individual tree heritabilities were calculated by the equation


2 2(VF + VM)
I VE + VT(RM) + VT(RF)+ VMF + VRM VF + V



Table 7. Analysis of variance and expected mean squares for
Experiment 2 using mean number of symptoms per tray.



Source df Expected Mean Squaresa/


Run 1 Ve + c4 VR x F + c5 VR

Family_/ 19 Ve + c2 VR x F + c3 VF

Run Family 19 Ve + Cl VR x F

Trays(Run Family) 112 Ve


a/Due to the unbalanced design
expected mean squares were

b/Includes three checklots.


coefficients
unequal.


(c) in the true


and standard errors were calculated similarly. Correlations between

variables were calculated by the SAS CORR procedure (SAS Institute

Inc. 1982a).

Data from Experiment 3 were analyzed to determine within and

between family differences for the number of spots found 10 and 48

days after inoculation using the model for a one-way analysis of

variance for a completely randomized design. There were 15 families

with approximately 20 trees per family.










Data in Experiment 4 also were analyzed using the SAS GLM

procedure. Independent variables in the model were family and pots

within family (Table 8).


Table 8. Analysis of variance and expected mean squares for
Experiment 4 using individual tree data.

Source df Expected Mean Squares


Family 7 Ve + 1.8Vp(F) + 35.5VF

Pots(Family) 146 Ve + 1.9Vp(F)

Error 140 Ve















RESULTS AND DISCUSSION


Field Tests--Experiment 1

Rust incidence in the six field tests ranged from 46% to 89%

(Table 1). Test 34 had the highest percentage incidence and

rust-associated mortality (RAM) (Tables 3 and 9). All 30 trees of

family 0354 x 0350 were dead from rust and each had a rust bush

form. More than 30 cankers were observed on eacn of these trees

with a maximum of 75 on some trees. To facilitate measurements,

data were not collected from dead trees with a rust bush form and

many galls (RBMC), but a classification was established to account

for these trees. Total gall number was assumed to be 31 and the

number of distorted, stem, limb and stem-to-limb galls was set at

one. All other trees dead due to rust were classified as having one

stem gall; this was true on most tree remains found. Although

several of the other tests contained RBMC trees, Test 34 had the

highest percentage (Table 10). A comparison of Tables 9 and 10

indicates that as RAM increased, there was an associated increase in

the number of trees having a RBMC form.

The frequency distribution of number of galls per tree is found

in Figure 3. Distribution of galls per tree for individual families

did not always follow this same pattern.










Table 9. Percentage of rust-associated mortality for 48 slash pine
families (four- to six-years-old) planted in six progeny
tests in southeast Georgia.


Family Progeny Tests
Female Male 34 35 36 37 40 41
--- --- --- --- --- ---- -f- - - - - ------


Least
Squares
Mean


0354 0350
0098
0088
0048
0350 0098
0088
0048
0354
0098 0088
0048
0157 0146
0096
0065
0071
0146 0096
0065
0071
0096 0065
0146
0288 0352
0001
0159
0050
0352 0001
0159
0050
0001 0159
0050
0019 0141
0060
0287
0047
0141 0060
0287
0047
0060 0287
0047
0284 0050
0295
0355
0064
0270 0050
0295
0355
0064
0295 0355
0064
0270


100.0 30.0 17.2 9.1
32.1 0.0 16.0 3.3
53.3 10.3 6.7 0.0
40.0 10.0 17.6 0.0
36.7 7.1 5.9 4.2
41.4 20.7 8.0 3.7
34.5 3.3 3.1 8.7
15.2 14.3
10.3 0.0 13.3 0.0
3.8 0.0 3.3 0.0
14.3 10.3
3.1 0.0
10.0 0.0
0.0 0.0
5.3
5.7 9.1
5.0 12.5
0.0 0.0


45.8
56.7
50.0
54.5
50.0
28.6
30.0
48.3

86.7
53.3
48.3
37.9


20.0
23.3
10.3
8.3
10.3
10.3
6.9
20.0
11.5
11.1
9.1
7.1
26.7
0.0


45.8 3.8
0.0
14.3 20.0


27.7
2.8
5.2
5.1
3.0
7.1
2.3
10.2
0.3
0.1
7.3
2.8
4.5
2.2
5.9
5.8
4.7
2.8
6.5
6.2
11.1
8.4
7.5
5.7
4.6
3.3
9.8
6.7
16.3
7.3
5.0
7.7
1.3
4.8
4.4
2.2
1.7
6.5
7.1
6.5
4.8
6.5
4.8
4.8
8.4
8.4
4.8
4.8


2.7
2.7
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2.6
0.0
0.0
2.8
0.0
0.0
0.0
0.0
0.0
0.0










Table 10. Percentage of trees dead and classified as 'rust bush many
cankers' for 48 slash pine families (four- to six-years-old)
planted in six progeny tests in southeast Georgia.


Least
Family Progeny Tests Squares
Female Male 34 35 36 37 40 41 Mean
-----------


0354 0350
0098
0088
0048
0350 0098
0088
0048
0354
0098 0088
0048
0157 0146
0096
0065
0071
U146 0096
0065
0071
0096 0065
0146
0288 0352
0001
0159
0050
0352 0001
0159
0050
0001 0159
0050
0019 0141
0060
0287
0047
0141 0060
0287
0047
0060 0287
0047
0284 0050
0295
0355
0064
0270 0050
0295
0355
0064
0295 0355
0064
0270


)0.0 30.0 6.9
12.1 0.0 0.0
53.3 10.3 6.7
.0.0 10.0 2.9
36.7 5.1 0.0
1.4 20.7 0.0
31.0 3.3 3.1
9.1
6.9 0.0 6.7
3.7 0.0 0.0
5.7
0.0
0.0
0.0

5.7
0.0
0.0


45.8
56.7
40.0
50.0
46.2
21.4
23.3
37.9

86.7
46.7
34.5
34.5


20.0
23.3
10.3
8.3
10.3
10.3
6.9
20.0
3.8
7.4
4.5
7.1
26.7
0.0


45.8 3.9
0.0
14.3 13.3


4.6
0.0
0.0
0.0
0.0
3.7
4.3
10.7
0.0
0.0
6.9
0.0
0.0
0.0
5.3
6.1
0.0
0.0


10
3
5
4

4


0.0
0.0
0.0
0.0

0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0



2.9
0.0
0.0
0.0

0.0

0.0

0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2.9
0.0
0.0


2.7
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0


24.3
7.7
5.9
3.7
2.0
5.7
1.9
8.3
0.1
0.1
4.8
2.5
2.5
2.5
6.2
5.2
2.5
2.5
3.6
5.0
8.8
4.0
4.9
4.2
2.4
1.7
6.7
0.6
37.1
3.5
2.6
5.3
0.7
3.6
3.5
1.4
0.4
3.6
3.6
3.6
3.6
3.6
3.6
3.6
3.6
5.1
3.6
3.6






















Figure 3. Frequency distribution for slash pine trees by the total number of fusiform rust galls,
in six progeny tests in southeast Georgia, at four-, five-, and six-years-old.










2
1.9
1.8
1.7-
1.6
1.5
1.4
S1.3
UE 1.2
cL 1.1
E 101 -

S0.8
0.7
z
0.6
0.5
0.4
0.3
0.2 -
0.1 0 1 2479

0 1 2 3 4 5 6 7 8 9 10 11 31


GALLS PER TREE










Families relatively resistant to the disease had a skewed

distribution with few trees with multiple galls, whereas more

susceptible families often had many trees with many galls per tree.

Differences in distribution patterns are illustrated by two families

planted in Test 37 (Figure 4). Family 0098 x 0088 was fairly

resistant (average percentage rust incidence of 37.8) and exhibited

a strongly skewed pattern, whereas family 0350 x 0354 (average

percentage rust incidence of 95.3) followed a different pattern

(Figure 4). Griggs and Dinus (1977) found that distribution

patterns reflected resistance.

Results of the analyses of full-sib family means are presented

in Table 11. This analysis provided least squared means for

families for all traits. Twisted galls were infrequent and were

deleted from this and all further analyses. All variables analyzed,

with the exception of average number of sunken galls per tree and

proportion of trees having sunken galls, differed significantly

among tests. Sunken galls were rare with only 201 galls found.

Significant differences among families were found for proportional

data for gall location, the gall types witches' broom and typical,

and rust incidence. (In this and further discussion the term

incidence refers to the presence of any galls.) However, mean gall

count per tree varied among families only for witches' broom galls

and total galls per tree (Table 11). Families did not vary

significantly for most gall types. Except for witches' broom galls,

there did not appear to be any unique pattern of gall development

associated with families across all planting sites. It is notable



























Figure 4. Frequency distribution for two slash pine families, 0098 x 0088 (resistant) and
0350 x 0354 (susceptible), for the total number of fusiform rust galls at age five
years, planted in a progeny test in Camden County, Georgia.









14
13-
.12 -
11 V


w 9-/

-/ 8
0 7/
a 6-
S 5
2 \ \ \ \
4-/
3- \ \ \ -
2-



0 1 2 3 4 5 6 7 8 9 10 31
GALLS PER TREE
S0098x0088 = 0350x0354










Table 11. Results of analyses of variance of the proportion of
trees with different fusiform rust symptoms (arcsin
square root transformation) and the mean number of
these symptoms per tree for family and test differences.


Mean Squares
Galls Family Test


Proportion of trees:
Locations
Any 0.117** 0.723**
Stem 0.061** 0.980**
Limb 0.093** 0.799**
Stem-Limb 0.061** 1.370**
Types
Typical 0.088** 0.868**
Witches' Broom 0.077** 0.752**
BTFA/ 0.019 0.139**
Fat 0.014 0.183**
Thin 0.014 0.038*
Sunken 0.017 0.136
Mean number per tree:
Locations
Total 13.166* 561.985**
Stem 9.216 401.512**
Limb 0.862 13.933**
Stem-Limb 0.073 0.792**
Types
Typical 0.823 13.696**
Witches' Broom 0.016* 0.755**
BTFA/ 0.016 0.128**
Fat 0.006 0.075**
Thin 0.005 0.014*
Sunken 0.003 0.004


A/Basally truncated fusiform gall.

* Significant at 0.05 level.

** Significant at 0.01 level.


that there was a significant tendency for some families to have a

higher incidence of stem or stem-to-limb galls than others, although

the number of galls at a specific location on the tree did not vary










significantly by family. Some bias in these results was possible

because the analysis used did not allow consideration of genetic

relationships among half-sib families.

An analysis considering all genetic relationships was conducted

using DIALL (Table 5). Crossing sets were analyzed individually at

each planting site. As each crossing set involved only five parent

trees, restricted and considerable differences in ranges of genetic

expression were to be expected. Results of the average number of

galls by type or location as well as the percentage of trees having

that characteristic are presented in Tables 12 ana 13. As each of

the five crossing sets were planted at two or four sites, there were

18 individual analyses for each gall type or location.

For all variables, one or more major sources of variation

(replication, general combining ability or specific combining

ability) in at least one of the analyses was significant (Tables 12

and 13). No gall type or location had a consistent pattern of

differences across all sets and planting sites. Significant

differences among replications were infrequent as expected because

the tests were designed with replications creating blocks to reduce

variation in growth resulting from soil-site variation and not to

block for differences in fusiform rust infection.










Table 12. Significant replication (Reps) general combining ability
(GCA) and specific combining ability (SCA) F values (5%
level of significance) obtained from analysis of
variance using DIALL for number of fusiform rust galls of
several types/locations on slash pine (18 analyses for
each type).


Source of Variation
Reps GCA SCA
Galls ---number of significant F values--


Location
Total 1 3 3
Stem 0 3 1
Limb 0 4 2
Stem-Limb 1 4 2
Types
Typical 3 2 3
Witches' Broom 0 2 3
BTFA/ 1 1 2
Fat 1 0 2
Thin 2 0 1
Sunken 0 0 1


./Basally truncated fusiform galls.


Average number of galls per tree of each type more often had

significant SCA effects than GCA (12 versus 5). Fat, thin and

sunken galls per tree had no significant GCA effects, hence no

expression of additive genetic variance. However, their SCA

effects were sometimes significant suggesting the presence of

nonadditive genetic effects. The average number of typical,

witches' broom and BTF galls per tree had both significant GCA and

SCA effects.

The mean number of galls per location had the opposite pattern

from gall types as there were more than twice as many significant

GCA effects (11) than SCA effects (five). Total number of galls

per tree had an equal number of significant GCA and SCA effects.









Taole 13. Significant replication (Reps) general combining ability
(GCA) and specific combining ability (SCA) F values (5%
level of significance) obtained from analysis of
variance using DIALL for proportion (arcsin square root
transformation) of slash pine trees with various
fusiform rust symptoms (18 analyses for each symptom).


Source of Variation
Reps GCA SCA
Galls ---number of significant F values--


Locations
All 1 8 0
Stem 0 3 2
Limb 0 3 1
Stem-Limb 0 4 0
Types
Typical 0 4 1
Witches' Broom 2 2 1
BTFI/ 0 1 3
Fat 0 0 3
Thin 1 1 0
Sunken 0 0 1
Mortality
RAM./ 1 4 1
RBMCC/ 0 2 2


A/Basally truncated fusiform gall.

b/Rust-associated mortality.

/Rust bush with many cankers.



The proportion of trees with each gall type had three more

significant GCA effects than the mean number of gall types. Two of

these were for the proportion of trees with typical galls and the

other for thin galls. Typical and witches' broom galls had more

significant GCA effects than SCA effects. There were nine

significant SCA effects for the proportion of gall types, and each










type had at least one except for thin galls. There was no evidence

of additive genetic effects for the proportion of trees with fat and

sunken galls .

The proportion of trees with each gall location (stem, limb and

stem-to-limb) had more than three times the number of significant

GCA effects than SCA effects. The proportion of trees with

stem-to-limb galls had no significant SCA effects. The proportion

of trees galled (rust incidence) also had no significant SCA

effects, but did have eight significant GCA effects. RAM had four

significant GCA effects while RBMC had two.

These results suggested that gall locations (mean or

proportion) were more controlled by additive genes than gall types

were, with the exception of typical and witches' broom galls. There

appeared to be more additive genetic effects for the incidence of a

gall type than for the average number of gall types per tree.

In crossing set two, there were significant GCA or SCA effects

for only four analyses, fewer than any other crossing set. In

contrast, crossing set one had significant GCA or SCA effects in 41

analyses. Such widely different results possibly are because of

differences in the extent of genetic variation within crossing sets

as only five parent trees were involved in each crossing set.

Substantial differences among sets were expected.

Lack of consistency for significant F values for the trait's

variance components implies that the variances are different among

the eighteen subpopulations. Each analysis can be considered a









subpopulation because at a test there are two to four crossing

sets. Also, at each test although the host genotype component is

controlled, the fungal component is not. The host-parasite

interaction is implicitly different, thereby producing

subpopulations.

Variance components and their standard errors for each crossing

set in each test as computed by DIALL (Table 5) are presented in

Table 28. There are discrepancies in the significance of a variance

component when judged by the F test or the standard error. Large

amounts of variation are inherent in the estimation of both variance

components and their standard errors. This variation, in addition

to the small sample size and unbalanced data, may have contributed

to the discrepancies. As an example, in crossing set 1, in Test 34

a highly significant F value was calculated for the GCA effect of

the mean number of stem galls per tree; however, the large standard

error of VGCA makes this value appear non-significant. The GCA

mean square (0.431) when divided by the SCA mean square (0.00796)

equaled 54.1, a highly significant F value (4 and 5 degrees of

freedom). Subtracting the SCA mean square from the GCA mean square

and dividing by the coefficient for VGCA (10.05) yielded 0.0421 as

an estimate of VGCA. The standard error of the estimate

calculated by the equation on page 25 was [(2(1/10.05)2(0.431)2 /

(4+2)) + (2(1/10.05) (0.00796)2 / (5+2))] /2 = 0.0245 which is

slightly more than half of VGCA'










Each crossing set was planted at two sites in a single year.

Year effects were not estimable as tests were not planted at the

same site in different years. Site effects, without confounding age

or year differences, were estimated by analyzing each crossing set

at two sites planted in the same year. In general, sites were

different for mean gall number per tree by type or location

(Table 14). GCA by site interaction (GCA x S) was significantly

different in 21 of 90 (10 traits in each of nine crossing set-site


Table 14. Significant site (S), replication (Reps), general
combining ability (GCA), specific combining ability
(SCA) and interaction (GCA x S) F values (5% level of
significance) obtained from analysis of variance using
DIALL for number of fusiform rust galls of various types/
locations on slash pine, planted at several sites (nine
analyses for each type).

Source of Variation
Site Reps GCA SCA GCAxS
Galls ------number of significant F values------


Location
Total 8 0 1 4 3
Stem 5 0 2 0 3
Limb 5 0 1 2 2
Stem-Limb 9 0 1 0 2
Types
Typical 9 0 1 3 3
Witches' Broom 5 0 1 0 2
BTFA/ 3 0 1 1 1
Fat 5 0 0 1 1
Thin 1 2 0 0 1
Sunken 2 1 0 0 4


A/Basally truncated fusiform gall.



combinations) different analyses for mean number of galls per tree (all

types and locations). Crossing set four had significant F-values for









GCA x S for every gall type or location in at least one analysis, which

is approximately half of the significant GCA x S effects. Confounding

pathogenic variation with environmental variation in the site effect

obscured the cause of the significant GCA x S effect.

Site effects were significant in 40 of the 90 analyses of the

proportion of trees having the various gall types or locations

(Table 15). Two-thirds of the analyses for proportion of trees with



Table 15. Significant site (S), replication (Reps), general
combining ability (GCA), specific combining ability (SCA)
and interaction (GCA x S) F values (5% level of
significance) obtained from analysis of variance using
DIALL for the proportion (arcsin square root
transformation) of slash pine trees with various rust
symptoms planted at several sites (nine analyses for
each symptom).

Source of Variation
Site Reps GCA SCA GCAxS
Galls -----number of significant F values------


Location
All 7 1 4 0 2
Stem 6 0 1 0 2
Limb 6 1 4 1 1
Stem-Limb 8 0 2 0 1
Types
Typical 6 0 4 1 1
Witches' Broom 7 2 1 2 2
BTFI/ 2 0 0 1 1
Fat 4 0 0 1 1
Thin 0 1 0 1 1
Sunken 1 0 0 0 3
Mortality
RAMb/ 5 0 1 1 2
RBMC_/ 3 0 1 0 2


A/Basally truncated fusiform gall.
b/Rust-associated mortality.

c/Rust bush with many cankers.










limb galls had significant GCA effects. Proportion of trees with

stem-to-limb galls was influenced more by site effects than by GCA

or SCA. Major effects on the proportion of trees with typical galls

were site and GCA, but there was significant interaction between

these effects in only one analysis. GCA x S interactions were

significant in only 11 of the 90 analyses of proportion of trees

with a particular gall type or location.

There were no significant GCA effects for thin, sunken or fat

galls (mean or proportion) and only one significant GCA effect for

the mean number of BTF galls. These results as well as those in

Tales 12 and 13 confirm that there was no additive genetic

variation for sunken and fat galls in this study. There was limited

evidence of additive genetic variation in thin and BTF galls.

When the analyses were examined to compare significant effects

for gall types with those for gall locations, it was found that gall

types had more significant SCA effects than locations. GCA x S

significant effects were more prevalent among gall types than gall

locations. There were, however, several significant GCA x S effects

for gall location. The impact of pathogenic variability and

environmental variation on family performance was evident in these

results. However, these tests were not designed to determine which

of these aspects of site were most important.

The lack of consistency of significant differences found in

this study was reflected in Blair's (1970) analysis of loblolly

pine. He found significant differences among half-sib families in

his 1963 plantings but not the 1964 plantings. Disease incidence in

the 1963 plantings averaged 50% and in 1964, 41%. Although he










analyzed the number of galls per tree, he did not present the

results of that analysis. He stated that the degree of additive

genetic control (percentage incidence, galls per tree and severity

index) ranged from weak to moderate and depended on the trait

measured, site and year of planting.

SRockwood and Goddard (1973) found that in some progeny tests

there were significant differences among half-sib families for

percentage rust incidence while in others there were none.

Differences among families were highly significant when progeny

tests were combined. Barker (1973) also found significant and

non-significant differences among families for his disease severity

index depending on test site. When data from sites were combined

there were significant differences among families.

Sohn (1977) found significant differences among half-sib

families of slash pine planted in several locations in two different

years. Kinloch and Stonecypher (1969) found significant differences

among control- and wind-pollinated families of loblolly pine at all

sites they analyzed for both average number of galls per tree and

percentage rust incidence.

Barker's (1973) combined analysis for severity index had a

highly significant interaction between sites and crossing sets, but

a nonsignificant interaction between sites and families within

crossing sets. He found little or no evidence of genotype x

environment interaction.

Sohn (1977) found significant family by site interactions for

percentage incidence with a binominal analysis of individual trees










and an analysis of plot means, but not with transformed plot means.

In addition, the significant differences occurred only in tests

planted in 1971, suggesting that planting years were important.

Sohn (1977) found rust incidence varied at some sites from year to

year, and genetic variances increased with increases in incidence.

Conditions favorable to rust infection varied from year to year,

resulting in variation in expression of genetic effects.

Results obtained from this study confirm the mixture of

previous results. Families were different in their resistance to

fusiform rust infection. Some symptoms of the disease including

thin, fat, BTF and sunken galls showed little or no additive genetic

effects. Rust incidence was affected by site, year of planting and

families planted. Witches' broom, typical, total number, RAM, RBMC,

stem, limb and stem-to-limb galls manifested variation because of

family and site. Analysis of transformed plot means reduced the

number of GCA x S interactions just as Sohn (1977) found.

Individual tree heritabilities for rust incidence, RBMC and RAM

as well as mean number of total, typical, BTF, fat, sunken, stem,

limb and stem-to-limb galls were calculated (Table 16) using the

model found in Taole 6. Individual and family heritabilities for

fat, thin and sunken galls were estimated as zero because variance

components for males were zero or negative. This was expected based

on the results of the diallel analysis. Estimates of h2 for

percentage rust incidence and total number of galls per tree are

comparable to those of Blair (1970) for loblolly pine. His

estimates of h2 for percentage rust incidence were 0.20 and
I










Table 16.


Individual (hi) and family heritabilities (h?)
and their standard errors (s) for rust incidence, rust
bush many cankers (RBMC), rust-associated mortality (RAM)
and the mean number per tree of 10 fusiform rust gall
types for slash pine, four- to six-years-old .


Heritability

2 2
Trait h s h s
I-- -F-s


Rust Incidence!/

RBMCa/

RAM!/

Location of Galls

Total

Stem

Limb

Stem-Limb

Types of Galls

Typical

BTF F

Witches' Broom

Fat

Thin

Sunken


0.196

0.346

0.240



0.227

0.088

0.055

0.157



0.046

0.004

0.128

0.0

0.0

0.0


0.681

0.809

0.751


0.695

0.519

0.397

0.680


0.306

0.048

0.643

0.0

0.0

0.0


a/Binomial analysis of variance.

./Basally truncated fusiform gall.










0.12 for two planting years. Additionally his estimates of h2

for total number of galls per tree were 0.29 and 0.09 for the same

years.

Estimates of percentage rust incidence in slash pine are also

similar to those obtained in this study. Rockwood and Goddard

(1973) estimated h2 to range from 0.035 to 0.277 in different

progeny tests. Sohn (1977) estimated h2 for percentage rust

incidence to range from 0.115 to 0.393, but her combined analysis

estimates were 0.280 and 0.217 for two planting years. Sohn's

combined estimate of family heritability for percentage rust

incidence was 0.78, with a range of 0.389 to 0.752.

The highest estimates of heritabilities in this study, both

individual and family, were for rust incidence, RBMC, RAM and total

galls per tree. These heritabilities suggested that selection

against death and malformation because of Cqf is possible. Gall

locations had moderate heritability values. Heritabilities of gall

types, with the exception of witches' broom and typical, were very

low or zero. Witches' broom and typical galls therefore, were the

only gall types with evidence of additive genetic effects based on

heritability estimates and the diallel analysis.

Genetic correlations from the model in Table 6 are found in

Table 17. Total gall number is highly correlated with RAM (0.99)

and RBMC (0.98). The correlation between total gall number and rust

incidence was lower at 0.51. Rust incidence was correlated to a

lesser degree with RAM and RBMC. The average number of typical

galls was positively correlated with incidence (0.71) and negatively















Table 17.


Genetic correlations between various fusiform rust gall types and locations (mean number
per tree), rust incidence and mortality (RAM and RBMC) for slash pine, four- to


six-years-old.


Stem- Witches Rust
Gall Types/ Stem limb Limb Typical Thin BTF Sunken Broom Fat Incidence RAM RBMC


Total
Stem
Limb
Stem-Limb
Typical
Thin
BTF
Sunken
Witches' Broom
Fat
Rust Incidence
RAM


0.60 -0.86 -0.43
0.32 0.25
0.87


-1.05
0.28
0.96
0.97


-b/ -2.61
-0.83
3.33
1.63
3.12


0.24
0.86
1.18
0.84
1.24


0.51
- 0.88
- 0.75
- 0.67
- 0.71


0.99
0.52
-0.95
-0.52
-1.08


2.33 0.62 -2.96


- 0.94 -0.03


0.98
0.37
-0.96
-0.59
-1.11
-l l

-2.82

-0.12


0.25 0.19
1.00


b/Undefined, at least one variance component was zero.


a/BTF = Basally truncated fusiform gall; RAM =Rust-associated mortality; RBMC =Rust bush, many
cankers.










correlated with RAM (-1.08) and RBMC (-1.11). Some genetic

correlations were larger than one or smaller than negative one.

Although these values are conceptually impossible they are

computationally feasible. Genetic correlations are usually subject

to large sampling errors and are therefore, seldom very precise

(Falconer 1960).

On the basis of genetic correlations, field counts of total

number of galls or the number of typical galls would provide more

information about potential RAM and malformation than simply

recording presence or absence. Disease severity estimates such as

Blair's (1970) severity index incorporate some of these features.

General combining ability estimates (gca) obtained from

full-sib progeny tests are an indication of parental value.

Table 18 contains gca estimates of parental values for several gall

types. Parents 0047, 0048, 0098, 0287 and 0352 were the best

general combiners for rust resistance and parents 0019, 0064 and

0350 were the worst on the basis of the traits in Table 18.

It is interesting to note that University of Florida (UF)

evaluations (Table 4) were high (greater than 1.0) for 0047, 0048

and 0098. However, the evaluation for 0352 was 0.68 and 0287 had a

slightly below average (-0.06) evaluation. The UF evaluation for

0287 had a standard error of 0.35 representing a range of -1.28 to

1.54 for its individual test evaluations. Progeny performance in

tests reported here indicate parental resistance levels to be higher

than previously assessed and, particularly in the case of 0287, may

reflect differences in crossing partners as suggested by Powers and

Zobel (1978).










Table 18. General combining ability estimates of the proportion of
trees with fusiform rust galls for 25 slash pine parents
based on progeny performance in six locations in
southeast Georgia, aged four to six years.


Witches' Stem- Rust
Parent Typical Thin BTFA/ Sunk Fat Broom Stem Limb Limb Incidence
-------------------TT r -7 r-----% -------------------~ r-Trr ------
1 41.5 2.8 2.3 3.8 1.5 43.0 22.1 45.7 24.7 52.0
19 95.6 6.7 8.5 5.8 1.1 62.4 56.6 96.2 46.1 106.3
47 24.1 2.1 5.5 7.3 4.6 10.7 9.0 22.2 7.3 34.6
48 28.7 1.1 4.2 0.4 -0.3 1.5 -2.3 31.5 6.6 29.9
50 39.4 -1.4 3.3 0.6 5.2 30.6 19.5 32.6 17.6 47.3
60 51.2 6.2 6.6 7.3 0.7 12.8 32.3 46.4 18.3 62.9
64 99.4 1.7 22.2 12.3 16.1 56.9 35.9 89.9 68.7 109.2
65 47.2 0.6 1.3 1.3 0.7 36.8 18.9 48.5 25.0 64.3
71 52.6 -2.7 3.6 2.1 0.5 29.0 23.5 37.0 34.3 66.8
88 33.3 -0.5 3.5 0.1 1.2 16.3 3.6 39.9 5.2 35.7
96 57.0 2.8 5.9 1.6 3.1 24.6 21.9 57.1 20.7 70.4
98 25.9 5.1 4.4 1.6 4.8 -7.2 -2.0 16.0 2.9 31.3
141 72.1 -0.2 3.1 3.1 3.9 48.1 49.8 72.4 45.7 86.4
146 35.2 3.9 3.1 0.1 1.2 37.0 27.0 33.8 23.2 36.1
157 69.6 3,7 12.8 3.8 9.0 30.9 26.7 69.8 31.0 82.1
159 78.7 4.2 0.1 9.1 2.5 49.6 40.0 67.0 30.2 89.2
270 31.9 8.9 2.2 9.0 -4.8 27.6 37.6 38.7 22.1 59.3
284 38.5 1.0 3.8 6.8 -4.7 7.7 41.2 31.1 16.4 50.2
287 12.3 3.1 -1.9 0.2 3.0 8.5 1.2 6.9 13.3 24.0
288 61.4 8.9 6.6 7.5 2.2 50.6 45.1 64.9 36.9 78.0
295 67.8 3.0 4.2 5.8 9.9 46.4 28.7 64.7 53.1 82.5
350 96.1 4.1 5.6 0.8 6.8 57.9 50.8 92.4 53.0 100.8
352 26.9 0.3 7.9 0.9 2.5 15.9 13.4 26.0 11.3 33.8
354 82.4 -0.02 12.2 1.7 7.1 65.7 57.7 81.9 55.6 93.5
355 79.0 3.2 0.7 14.0 3.9 52.2 39.6 71.2 63.5 96.1


A/Basally truncated fusiform gall.



Specific combining ability of a particular cross (sca) is

defined as the family deviation from the mean of the two parents.

Specific combining abilities of families for percentage rust

incidence and fat galls are presented in Table 29. Specific

combining ability values for the proportion of trees with fat galls

ranged from -8.00 (0284 x 0064) to 3.84 (0295 x 0064). Crosses with

0064 had the greatest positive and negative deviations from parental

means (sca) for the proportion of trees with fat galls. Parent 0064










had the highest estimate of gca for this trait as well as for

proportion of trees with sunken galls and percentage rust

incidence. Specific combining ability estimates for families in

which 0064 was a parent varied greatly for these last two traits

also.

Specific combining ability may be an influence in the

relatively good performance of parent 0287. Crosses between this

parent and 0019 or 0060 have sca's of 5.31 and 2.25, respectively,

for percentage rust incidence. When crossed with 0141, however, it

has a sca of -7.63. Parent 0141 has a gca estimate midway between

those of parents 0019 and 0060. Further crosses with 0287 and 0141

and other parents may disclose the reasons for this apparent good

combination.

In several of the analyses for fat galls SCA effects were

significant. This fact, along with sca values for fat galls in some

crosses, suggested that nonadditive genetic effects were important

in this trait.


Artificial Inoculation

Experiment 2--RSC Procedure

Rust incidence (gall formation) following the RSC procedure

ranged from 14.4% to 88.7% among the 20 seedling lots tested (17

full-sib families and 3 check lots). Proportion of galled seedlings

was higher in all families than in the two resistant checks, and

only two families had lower galled proportions approaching that of

the resistant checks. Seven of the families had a higher properties

of galled seedlings than the susceptible check. Means for all

traits measured are indicated in Table 19 by seedling lot. A










Table 19.


Percentage of seedlings in 20 slash pine families with
various reaction types six months after artificial
inoculation with fusiform rust and the resistance index
of the families.


Family Pink Incidence Symptom Less than Resistance
Female Male Blush Healthy of Rust No Swelling Rough 25 mm Index
----------------------- --%---------------------

30001/ 15.8 24.4 66.2 9.4 11.5 25.9 209.4
400Q0/ 5.0 41.7 31.7 29.2 5.0 11.7 195.2
500c/ 1.8 57.6 14.4 31.7 3.3 9.4 144.8
0354 0350 1.6 18.0 63.9 18.0 11.7 36.1 159.3
0354 0088 2.7 15.3 48.9 38.0 10.9 30.2 122.2
0354 0048 4.2 15.5 62.9 28.2 0 25.9 140.4
0350 0088 8.3 20.0 56.0 30.0 8.0 26.0 102.3
0350 0048 19.4 33.8 53.0 15.9 4.6 20.5 171.2
0352 0001 7.3 38.8 34.9 27.2 5.3 9.3 197.1
0352 0159 3.2 34.2 34.9 30.9 11.1 5.3 163.1
0270 0050 41.0 42.6 38.5 13.9 7.8 14.6 183.3
0019 0060 6.5 8.9 88.7 2.4 17.9 23.2 61.3
0019 0287 16.9 18.3 70.7 15.5 6.4 37.9 132.9
0019 0047 44.3 38.0 49.4 17.3 4.7 23.3 157.5
0141 0287 13.3 28.3 58.3 16.7 1.7 33.3 166.6
0270 0355 31.9 16.0 79.8 5.1 17.2 25.7 128.4
0284 0295 12.2 19.6 77.6 4.3 7.2 23.6 108.4
0284 0064 13.4 20.5 70.4 9.1 8.8 25.2 112.5
0270 0295 28.6 13.6 82.2 4.2 23.8 22.7 142.1
0270 0064 39.1 8.9 88.7 1.8 22.7 16.7 149.7


A/ RSC susceptible check lot, Georgia Slash

b/ RSC resistant check lot, FA-2

/ RSC resistant checK lot, LA11


resistance index value was calculated for the seedling lots tested

based on the RSC index equation (Robert Anderson, personal

communication 1984) (Table 19). A high resistance index value

indicates a high occurrence of resistant reactions. The high index

value for the susceptible check was unexpected.

Families were significantly different for all traits analyzed

(Table 20) (see model, Table 7). Differences among families for









Table 20. Mean squares for the analyses of variance of the
proportion of of slash pine seedlings per family
expressing various disease symptoms resulting from
artificial inoculation with fusiform rust at the
Resistance Screening Center, Asheville, North Carolina.


Source of Variation
Disease Symptom Run Family


Healthy 0.165** 0.088**
SYMNQa/ 0.006 0.122**
Incidence of Rust 0.306** 0.400**
Rough Gall 0.001 0.029**
< 25 mm Gall 0.134** 0.110**
Thin Gall 0.100** 0.042**
Fat Gall 0.164** 0.011*
BTF Gallk/ 0.002 0.001*
Adventitious Shoots 0.103** 0.137**
Sunken Areas 0.006 0.056*


a/Symptom, no swelling.

P/Basally truncated fusiform gall.

*Significant at the 5% level.

**Significant at the 1% level.


mean number of fat and BTF galls and galls with sunken areas were

significant at the 5% level while other traits were highly

significant (1% level) (Table 20).

Carson (1984) analyzed five of these traits (galled, SYMNO,

rough, LT25M ana thin) on loblolly pine. With a composite inoculum

source, there were significant differences among families for four

of these but not for the fifth, LT25M.

Individual heritabilities were calculated using the model

described on page 28 for the proportion of seedlings with the

various artificial inoculation responses and the mean number of










responses per tree (Taole 21). Proportion of trees with SYMNO and

galls with adventitious shoots had fairly high individual

heritabilities, 0.28 and 0.33, respectively. Mean number of galls

per seedling had the highest heritability with a value of 0.38.

These results indicated that several resistant and susceptible

characteristics are heritable traits.

Correlations between variables in the greenhouse produced

several significant observations (Table 22). The number of healthy

seedlings in a family had significant negative correlations with

mean number of galls, rough galls and galls with adventitious

shoots, and was negatively related to other reactions except SYMNO.

In general, it should be expected that symptoms suggesting

resistance (such as SYMNO) should be positively correlated with

other resistance symptoms and negatively correlated with symptoms

suggesting susceptibility. Galls with adventitious shoots had a

highly significant negative correlation with healthy and SYMNO and

highly significant positive correlations with number of galls and

rough galls. Galls with adventitious shoots appeared to be a

susceptible reaction type. Sunken areas were negatively correlated

with SYMNO and positively correlated with BTF, rough, and mean

number of galls.









Table 21. Individual heritabilities (h2) and their standard
errors (s) for artificial inoculation (AI) symptoms
by both proportion of trees with the symptoms and
mean number of symptoms per tree.


Proportion Mean

AI Symptomsa/ h2 s h2 s


Healthy 0.08 0.08 -

SYMNO 0.28 0.18 0.32 0.20

Incidence of Rust 0.23 0.16 0.38 0.23

Rough Gall 0.18 0.13 0.11 0.09

LT25M Gall 0.10 0.09 0.08 0.08

Thin Gall 0.06 0.07 0.06 0.07

Fat Gall 0.02 0.05 0.01 0.04

BTF Gall 0.05 0.06 0.05 0.06

Adventitious Shoots 0.33 0.21 0.32 0.20

Sunken Areas 0.11 0.09 0.09 0.08


A/SYMNO =symptom, no swelling; LT25M =gall less
BTF =basally truncated fusiform gall.


than 25 mm long;


The correlation between SYMNO and healthy was small (less than

0.4) and not significant. Healthy seedlings may be escapes or may

have one of the two interactions Miller et al. (1976) described as

either total incompatibility (i.e., no germination or penetration by

basidiospore) or subliminal infection possibly due to an avirulent

pathogen. SYMNO appeared to be a hypersensitive reaction.











Table 22. Pearson product-moment correlations between the mean
number of artificial inoculation (AI) symptoms per tree of
slash pine seedlings inoculated with fusiform rust, based
on progeny of 20 families.


AI Symptoms
AI Symptom, Incidence Adventitious Sunken
Symptoms no swelling of Rust Rough LT25M-/ Thin Fat BTF / Shoots Area
--------------------Galls---------------------------


Healthy C

Symptom, no
swelling

Incidence
of Rust

Rough Galls

LT25Ma/ Galls

Thin Galls

Fat Galls

BTFI/ Galls

Adventitious
Shoots


).35 -0.86** -0.52*

-0.70** -0.54*


-0.49

0.01


-0.54* 0.52*


-0.22

-0.41


-0.25

0.11


0.53* 0.14


-0.07 0.01

0.61**


-0.13

0.10

0.13


-0.33

-0.48


0.45


0.59*

0.08

0.28

-0.01


-0.65**

-0.79**


-0.37

-0.54*


0.78** 0.45*


0.67**

-0.03

0.13

0.09

0.42


0.70**

0.06

0.18

-0.39

0.63**


0.38


a/Gall less than 25 mm long.

b/Basally truncated fusiform gall.

* Correlation significant at the 5% level.

**Correlation significant at the 1% level.










There were significant negative correlations between percentage

rust incidence in the field and proportion of healthy and SYMNO

seedlings in the greenhouse (Table 23). Field rust incidence had

significant positive correlations with proportion of seedlings

infected with any gall type, rough galls and galls with adventitious

shoots. The proportion of trees with stem, limb, stem-to-limb and

typical galls in the field were significantly negatively correlated

with the proportion of healthy seedlings in the greenhouse. There

were no significant correlations between RAM and RBMC and any of the

artificial inoculation symptoms.

The positive correlations between the proportion of trees

galled in the field and rough (significant) and LT25M

(nonsignificant) galls in the greenhouse were surprising as both

symptoms are considered resistant (Charles Walkinshaw, personal

communication 1984). Walkinshaw and others (1980) found a negative

correlation between University of Florida rust evaluations and the

proportion of seedlings with smooth galls. Therefore, they included

this variable as well as proportions of trees with fat galls and

SYMNO in their index. Rough galls, seemingly the opposite of smooth

galls, were categorized by Carson (1984) and by RSC for their

current index (Robert Anderson, personal communication 1985).

LT25M had a positive relationship with percentage rust

incidence and total number of galls per tree in the field study.

Seedlings with LT25M galls nine months after inoculation in the 1984

study of Griggs and others were free of gall symptoms and active











Table 23. Pearson product-moment correlations between field galls
and artificial inoculation symptoms (proportions) and the
resistance index of 17 slash pine families and fusiform
rust.


Artificial Inoculation SymptomsA/

Incidence Adventitious Resistar
Fiela Galls Healthy SYMNO of Rust Rough LT25M Fat Shoots Inde)

Proportion
of Trees:
Typical -0.54* -0.27 0.50* 0.32 0.35 0.38 0.44 -0.41
Thin -0.32 0.64** 0.57* 0.63** 0.03 -0.19 0.52* -0.2!
BTFb/ -0.20 -0.01 0.14 -0.03 0.04 0.47 0.27 -0.1;
Sunken -0.24 -0.70** 0.55* 0.64** -0.22 0.02 0.61** -0.1;
Fat -0.25 -0.25 0.01 0.02 0.12 0.57* 0.07 -0.0O
Witches' Broom -0.29 -0.01 0.14 0.38 0.36 -0.24 -0.03 0.11
Stem -0.52* -0.48 0.58* 0.46 0.42 -0.06 0.48 -0.3;
Limb -0.60* -0.18 0.48 0.34 0.39 0.30 0.36 -0.4;
Stem-Limb -0.54* -0.48 0.60* 0.46 0.42 -0.12 0.42 -0.1:
Inciaence
of Rust -0.63** -0.51* 0.69** 0.54* 0.37 0.21 0.59* -0.4!
RAMC/ -0.16 -0.03 0.09 0.13 0.34 -0.11 0.00 0.0(
RBMCc/ -0.13 0.11 -0.01 0.06 0.41 -0.16 -0.13 0.0!

lean number
of galls:
Typical -U.34 -0.07 0.22 -0.01 0.16 0.68** 0.09 -0.3(
Thin -0.18 -0.61** 0.48 0.50 0.06 -0.28 0.45 -0.1
BTFR/ -0.14 0.29 -0.03 -0.22 0.09 0.57* -0.08 -0.1
Sunken -0.30 -0.75** 0.61** 0.76** -0.24 0.07 0.68* -0.1l
Witches' Broom -0.24 -0.05 0.08 0.34* 0.25 0.01 -0.06 0.1'
Fat -0.16 0.39 -0.11 0.13 0.19 0.56* -0.32 -0.0;
stem -0.20 0.02 0.08 0.11 0.44 -0.20 -0.05 -0.0;
Limb -0.27 0.20 0.08 0.01 0.12 0.68** 0.05 -0.3(
Stem-Limb -0.53* -0.48* 0.62** 0.30 0.38 0.08 0.42 -0.1;
Total -0.29 -0.05 0.14 0.12 0.44 0.05 -0.02 -0.0(


V/SYMNO = symptom, no swelling, LT25M = gall less than 25 mm long.

k/Basally truncated fusiform gall.

c/RAM = rust-associated mortality, RBMC = rust bush many cankers.

*Correlation significant at the 5% level.

**Correlation significant at the 1% level.










mycelium 23 months later, and therefore, could be considered a

resistant symptom in that study.

There were no significant correlations between field symptoms

and the RSC resistance index in this study. Calculation of the

correlations did not include the check lots. The resistance index

did not successfully reflect family field performance in this study

for any field symptom.

Griggs and Dinus (1977) found extremely low correlations

between artificial inoculation and field tests at young ages, but

the correlations increased and were significant at age 12. They

implied that the correlation was improved after incidence had

increased over time. This suggested that artificial inoculation

predicted performance better on sites with high incidence than on

sites with moderate incidence. Because field tests in this study

had moderate to high incidence levels it was expected that

artificial inoculation would predict performance more accurately.

Pink blush was observed on the seedlings 10 days after

inoculation. Families were significantly different in the

percentage of seedlings expressing this character. Percentage of

seedlings with a pink blush ranged from 1.6 to 44.3 (Table 19). In

some cases the blush was a faint, small pink spot on the stem just

above the cotyledons. In other cases the spot was larger, sometimes

encircling the stem, or varied in color to a rich red. Pink blush

was negatively correlated with SYMNO (r=-0.53) and essentially not

correlated with percentage healthy trees (r=-0.04). It was










positively related to percentage of galled seedlings (r=0.30). Pink

blush does not appear to be a resistant symptom. However, a more

intensive study is needed to conclude this.


Experiment 3--High Inoculum Concentration

Numbers of spots on each seedling were counted 10 and 48 days

after inoculation (Table 24). Red spots observed 10 days after

inoculation were generally small and confined to the stem. After 48

days, some spots were very large and numerous spots were present on

the primary needles. Large spots in some cases appeared sunken.

Color varied from red to purple. There were highly significant

differences among families for spots both 10 and 48 days after

inoculation. Lundquist and Luttrell (1982) stated that most

families became pigmented within 48 days. Therefore, the results at

48 days for proportion of trees with spots were expected.

Average number of spots per seedling 10 and 48 days after

inoculation was correlated to field performance of the same families

using Pearson correlations (SAS Institute Inc. 1982a). Significant

negative correlations were found between proportion of trees with

spots after 10 days and proportion of trees with typical galls, limb

galls and rust incidence (r > -0.59).

Lundquist et al., (1982) found that under high inoculum

concentrations resistance was ineffective in family FA-2, a RSC

resistant check lot. This family (4000) was included as a check lot

in both this experiment and Experiment 2. Very little pigmentation

occurred in the test, supporting the conclusion of Lundquist et al.

(1982), that resistance of this family was not effective under high









Table 24.


Results of artificial inoculation with high
concentrations of inoculum (2.0 x 106 spores/ml) on
17 slash pine families, Resistance Screening Center,
Asheville, North Carolina.


10 Days after Inoculation 48 Days after Inoculation
Family No. of % of Trees No. of % of Trees
Female Male Spots Mean With Spots Spots Mean with Spots


3000Q -- 0 0 0 7 0.4 25.0
4000o -- 0 0 0 4 0.2 20.0
0354 0088 1 0.05 5.0 299 14.9 100.0
0098 0088 12 0.63 42.1 281 14.8 94.7
0354 0048 1 0.05 5.0 110 5.5 90.0
0352 0001 10 0.5 45.0 412 20.6 95.0
0288 0159 0 0 0 130 6.5 100.0
0019 0047 1 0.05 5.0 154 7.7 100.0
0350 0048 1 0.5 5.0 404 20.2 100.0
0270 0050 70 3.5 85.0 362 18.1 100.0
0141 0287 4 0.2 15.0 152 7.6 95.0
0350 0354 0 0 0 154 7.7 100.0
0019 0060 5 0.25 10.0 148 7.4 85.0
0096 0146 34 1.7 65.0 325 16.2 95.0
0019 0287 1 0.05 5.0 79 3.9 80.0
0352 0159 13 0.93 64.3 134 9.6 92.9
0270 0064 6 0.3 25.0 230 11.5 100.0


./RSC susceptible check lot, Georgia Slash.

./RSC resistant check lot, FA-2.



inoculum loads. Lundquist and Miller (1984) compared their results

with Mullick (1977) and other researchers. They reported that the

pigmentation seen in their study signified that a host had reacted

to prevent the spread of the fungus. They further stated that

pigmentation was not the complete mechanism of resistance, but was

symptomatic of physiological and anatomical resistance mechanisms.

Further research is needed to understand these mechanisms before


breeders can exploit the variation that is present.










Experiment 4--Inoculation of 1-0 Seedlings

Galls developing on 1-0 seedlings more closely resembled galls

on trees in the field than those on the seedlings in Experiment 2.

This suggested that host response varied because of maturation state

as concluded by Day (1974). Families were significantly different

for rust incidence, total number of galls per tree, fat galls, thin

galls, and typical galls. Galls were located on the growth flush

that had occurred just prior to inoculation. Many infected trees

had stem-to-limb galls that probably would result in a RBMC if the

trees were allowed to mature. Goddard and Schmidt (1971) also found

differences among families in rust incidence when they inoculated

1-0 seedlings, and their results were correlated with the results of

inoculation of the same seedlings at age six-to-ten weeks.

The susceptible check had significantly greater rust incidence

than the other families (Table 25). Family 0621 x 2005 had the next

highest infection level. Many galls on these trees were short and

fat and did not encompass the entire stem circumference. Orthogonal

contrasts revealed that for total galls per tree the susceptible

check was significantly different from 0621 x 2005 and 0298 x 2005

and that these families also differed. In addition, the contrasts

disclosed differences in rust incidence between 0621 x 2005 and

0298 x 2005. Male 2005 was apparently susceptible.

This small sample size and lack of field data limit conclusions

from this study. Field performance in Florida of these families

should not be predicted from these results since the pathogen was

not local. However, it can be seen that seedlings from the










Table 25.


Comparison of family means for several fusiform rust
disease symptoms on artificially inoculated 1-0 slash pine
seedlings.


Family Disease Symptoma/

Female Male % Rust Incidence Total Fat Thin Typical
--------galls per tree--------


Susceptible Checkb/ 72.5A 1.95A 0.03B 0.2A 1.7A

0621 2005 52.5B 1.38A 0.18A 0.08 1.05B

0298 2005 22.5C 0.458 0.038 0.0B 0.38C

Resistant Check'/ 28.9C 0.32B 0.0B 0.0B 0.29C

0618 2007 17.5C 0.18B 0.08 0.OB 0.18C

0620 2002 17.4C 0.17B 0.08 0.OB 0.13C

0298 2013 11.4C 0.17B 0.OB 0.0 0.17C

0298 2002 10.5C 0.138 0.05B 0.OB 0.05C


a/Means with the same letter are not significantly different.

k/Checks were Container Corporation of America's resistant and
nonresistant orchard progeny.


resistant orchard check and some crosses with phenotypically

resistant males had significantly less infection than the

susceptible check. The study also demonstrates that older seedlings

may be effectively artificially inoculated with Cqf at RSC. This

may be useful in future studies with slash pine and Cqf.










Research Perspectives

Disease results from the interaction of a host, a pathogen, and

their environments. Although fusiform rust disease of slash pine

has been studied intensively for the last 20 years, many questions

remain unanswered.

It is known that the fungus varies in its virulence. Carson

(1984) found disease symptoms varied within loblolly pine families

when different pathogen sources were used to inoculate seedlings.

It can be concluded, therefore, that disease symptoms may vary

because of the pathogen. However, additional research is necessary

to corroborate these findings for slash pine and other species.

Environmental differences, such as site or year, cause

variation in rust incidence by affecting sporulation and infection.

However, the impact of environment on variation in disease symptoms

is unclear. Field disease symptoms in this study varied by site,

but the fungal component was confounded with environment. No clear

statement was possible of the effect of environment on disease

symptoms. One method of addressing environmental effects may be to

inoculate 1-0 seedlings as in Experiment 4 and then outplant them at

different sites.

Future studies could include development of the relationship

between microscopic processes described by Miller et al. (1976) and

Mullick (1977) and gall types. Griggs et al. (1984) observed that

LT25M galls disappeared over time. What of other gall types? Do

they differ or change over time? Does the environment affect gall







69


development? Research aimed at answering these questions may add

valuable information to the knowledge of disease resistance in

slash pine.














CONCLUSIONS


Additive genetic variation of slash pine in response to natural

infection with fusiform rust exists for many symptoms: number of

stem, limb, stem-to-limb, total, typical and witches' broom galls,

the proportion of trees with these galls, incidence of rust, RAM and

RBMC. However, this variation is not consistent. Little or no

genetic variation is apparent for thin or sunken galls. Fat galls

have evidence of specific combining ability, an expression of

nonadaitive genetic effects. The existence of genetic variation

appears to depend on site, planting year, inoculum and the sample of

families in a test.

Individual tree and family heritabilities for the field

traits--rust incidence, RAM, RBMC, total galls per tree,

stem-to-limD galls and witches' broom galls--were relatively high

and, where comparisons were possible, similar to those found in

other studies. Individual heritabilities were 0.196, 0.346, 0.240,

0.227, 0.157 ana 0.128, respectively. Family heritabilities were

0.68, 0.81, 0.75, 0.70, 0.68 and 0.64, respectively.

KAM and RBMC were the most severe reactions to rust infection

and were the most heritable in this study. Selection against

families with high incidence of these traits is imperative in

breeding for increased rust resistance.










Individual heritabilities of artificial inoculation symptoms

were best for proportion of trees with SYMNO (0.28), rough galls

(0.18), galled (any type) (0.23) and galls with adventitious shoots

(0.33). Variation among full-sib families existed for all traits

observed. Healthy, SYMNO and galled (any type) were correlated with

several field symptoms but were more often significantly correlated

with the proportion of trees with a field symptom than with the mean

number of symptoms. Rough galls, considered to be a resistant

symptom, had a significant positive correlation with percentage rust

incidence in the field. The RSC resistance index did not

successfully predict field performance of the full-sib families in

this study.

The high inoculum concentration technique was not successful in

predicting field performance. However, it may be useful in

screening families whose resistance could be overcome in the field

with high inoculum concentrations. Significant differences among

families existed for their response to this inoculation procedure.

One-year-old progeny of phenotypically rust-resistant males

were successfully inoculated in this study. Families with three of

the four males as a parent were as resistant as a rust-resistant

check. The fourth male was apparently not resistant. Selection of

phenotypically rust-resistant slash pines, therefore, continues to

be a viable method of adding resistant selections to a breeding

program.













LITERATURE CITED


Anderson, R. L., and T. A. Bancroft. 1952. Statistical Theory in
Research. McGraw-Hill, New York. 399 p.

Anderson, R. L., C. H. Young and J. D. Triplett. 1983. Resistant
screening center procedures manual: A step-by-step guide used
in the operational screening of southern pines for resistance
to fusiform rust. USDA Forest Serv Forest Pest Manag Rep
83-1-18. 55 p.

Blair, R. L. 1970. Quantitative inheritance of resistance to
fusiform rust in loblolly pine. Ph D Thesis, North Carolina
State Univ, Raleigh. 87 p.

Barker, J. A. 1973. Location effects on heritability and gain
predictions for ten-year-old loblolly pine. Ph D Thesis, North
Carolina State Univ, Raleigh. 105 p.

Carson, S. D. 1984. Indirect screening of loblolly pine for
fusiform rust resistance through controlled inoculation. Ph D
Thesis, North Carolina State Univ, Raleigh. 57 p.

Day, P. R. 1974. Genetics of Host Parasite Interaction. W. H.
Freeman and Company, San Francisco, CA. 288 p.

Dinus, R. J., G. A. Snow, A. G. Kais and C. H. Walkinshaw. 1975.
Variability of Cronartium fusiforme affects resistance breeding
strategies. In Proceedings 13th Southern Forest Tree
Improvement Conference, p 193-196. National Technical
Information Service, Springfield, VA. 262 p.

Dwinell, L. D. 1974. Susceptibility of southern oaks to Cronartium
fusiforme and Cronartium quercuum. Phytopathology 64:400-403.

Dwinell, L. 0. 1977. Biology of fusiform rust. In Management of
Fusiform Rust in Southern Pines (R. J. Dinus and R. A. Schmidt
eds) p 18-24. Symp Proc Univ Florida, Gainesville. 163 p.










Falconer, D. S. 1960. Introduction to Quantitative Genetics. The
Ronald Press Company, New York. 365 p.

Frampton, L. J., Jr., H. W. Amerson and R. J. Weir. 1983.
Potential of in vitro screening of loblolly pine for fusiform
rust. In Proceedings 17th Southern Forest Tree Improvement
Conference, p 325-331. National Technical Information Service,
Springfield, VA. 375 p.

Goddard, R. E., and J. T. Arnold. 1966. Screening select slash
pines for resistance to fusiform rust by artificial
inoculation. In Breeding Pest Resistant Trees (H. D. Gerhold,
E. J. Schreiner, R. E. McDermott, and J. A. Winieski, eds) p
431-435. Pergamon Press Ltd, London. 505 p.

Goddard, R. E., D. L. Rockwood and H. R. Kok. 1984. Cooperative
Forest Genetics Research Program. 26th Prog Rep, Univ Florida,
School Forest Resourc Conserv Rep No 35. 26 p.

Goddard, R. E., and R. A. Schmidt. 1971. Early identification of
fusiform rust resistant slash pine families through controlled
inoculation. In Proceedings 11th Southern Forest Tree
Improvement Coinference, p 31-36. National Technical
Information Service, Springfield, VA. 284 p.

Goddard, R. E., R. A. Schmidt, R. K. Strickland and H. R. Kok.
1972. Cooperative Forest Genetics Research Program. 14th Prog
Rep Univ Florida, School Forest Resourc Conserv Rep No 20. 26
P.

Goddard, R. E., R. A. Schmidt and F. Vande Linde. 1975. Effect
of differential selection pressures on fusiform rust resistance
in phenotypic selection of slash pine. Phytopathology
65:336-338.

Goddard, R. E., and 0. 0. Wells. 1977. Susceptibility of
southern pines to fusiform rust. In Management of Fusiform
Rust in Southern Pines (R. J. Dinus and R. A. Schmidt, eds),
p 52-58. Symp Proc Univ Florida, Gainesville. 163 p.

Gray, D. J., and H. W. Amerson. 1983. In vitro resistance of Pinus
taeda embryos to the fusiform fungus, Cronartium quercuum f.
sp. fusiforme: Ultrastructure and histology. Phytopathology
73:1492-1493.

Griggs, M. M., and R. J. Dinus. 1977. Patterns of fusiform rust
increase and their implications for selection and breeding. In
Proceedings 14th Southern Forest Tree Improvement Conference,
p 43-52. National Technical Information Service, Springfield,
VA. 352 p.










Griggs, M. M., R. J. Dinus and G. A. Snow. 1984. Inoculum source
and density influence assessment of fusiform rust resistance in
slash pine. Plant Dis 68:770-774.

Griggs, M. M., and R. A. Schmidt. 1977. Increase and spread of
fusiform rust. In Management of Fusiform Rust in Southern
Pines (R. J. Dinus and R. A. Schmidt, ed.), p 32-38. Symp Proc
Univ Florida, Gainesville. 163 p.

Griggs, M. M., and C. H. Walkinshaw. 1982. Diallel analysis of
genetic resistance to Cronartium quercuum f. sp. fusiforme in
slash pine. Phytopathology 72:816-818.

/Hirt, R. R. 1964. Cronartium ribicola: Its growth and
reproduction in the tissues of eastern white pine. St Univ Col
Forest Syracuse Univ, NY, Tech Pub 86. 59 p.

Hubbard, S. D., and R. L. Anderson. 1980. The resistance screening
center. USDA Forest Serv Gen Rep SA-GR 16. 1 p.

Jewell, F. F. 1960. Inoculation of slash pine with Cronartium
fusiforme. Phytopathology 50:48-51.

iJewell, F. F., D. C. Jewell and C. H. Walkinshaw. 1982.
Histopathology of anatomical mechanisms for resistance to
fusiform rust in slash pine. In Resistance to Diseases and
Pests in Forest Trees (H. M. Heybroek, B. R. Stephan and K. von
Weissenberg, eds.) p 110-118. Centre Agric Publ Document,
Wageningen, The Netherlands. 503 p.

Kinloch, B. B., Jr., and R. W. Stonecypher. 1969. Genetic
variation in susceptibility to fusiform rust in seedlings from
a wild population of loblolly pine. Phytopathology
59:1246-1255.

Laird, P. P., and W. R. Phelps. 1975. A rapid method for mass
screening of loblolly and slash pine seedlings for resistance
to fusiform rust. Plant Dis Rep 59:238-242.

Lewis, R. A. 1973. Quantitative assessment and possible
biochemical indicators of variation in resistance to fusiform
rust in loblolly pine. Ph D Thesis, North Carolina State Univ,
Raleigh. 94 p.

Lundquist, J. E., and E. S. Luttrell. 1982. Early symptomatology
of fusiform rust on pine seedlings. Phytopathology 72:54-57.

Lundquist, J. E., and T. Miller. 1984. Development of stem lesions
on slash pine seedlings infected by Cronartium quercuum f. sp.
fusiforme. Phytopathology 74:514-518.









Lundquist, J. E., T. Miller and H. R. Powers, Jr. 1982. A rapid
technique for determining resistance of slash pine to fusiform
rust. Phytopathology 72:613-615.

Matthews, F. R., and S. J. Rowan. 1972. An improved method for
large-scale inoculations of pine and oak with Cronartium
fusiforme. Plant Dis Rep 56:931-934.

Miller, T., E. B. Cowling, H. R. Powers, Jr. and T. E. Blalock.
1976. Types of resistance and compatibility in slash pine
seedlings infected by Cronartium fusiforme. Phytopathology
66:1229-1235.

Mullick, D. B. 1977. The nonspecific nature of defense in bark
and wood during wounding, insect, and pathogen attack. In The
Structure, Biosynthesis, and Degradation of Wood. Vol 1T
Recent Advances in Phytochemistry (F. A. Loewus and C. C.
Runeckles, eds) p 395-441. Plenum Press, New York. 521 p.

Powers, H. R., Jr. 1980. Pathogenic variation among single-
aeciospore isolates of Cronartium quercuum f. sp. fusiforme.
Forest Sci 26:280-282.

Powers, H. R., Jr., and L. D. Dwinell. 1978. Virulence of
Cronartium stable after 25 years. Plant Dis Rep 62:877-879.

Powers, H. R., Jr., F. R. Matthews and L. 0. Dwinell. 1977.
Evaluation of pathogenic variability of Cronartium fusiforme on
loblolly pine in the southern USA. Phytopathology 67:1403-1407.

Powers, H. R., Jr., F. R. Matthews and L. D. Dwinell. 1978. The
potential for increased virulence of Cronartium fusiforme on
resistant loblolly pine. Phytopathology 68:808-810.

Powers, H. R., Jr., R. A. Schmidt and G. A. Snow. 1981. Current
status and management of fusiform rust on southern pines. Ann
Rev Phytopathol 19:353-371.

Powers, H. R., Jr., and B. J. Zobel. 1978. Progeny of specific
loblolly clones vary in fusiform rust resistance according to
seed orchard of origin. Forest Sci 24:227-230.

Rockwood, D. L., and R. E. Goddard. 1973. Predicted gains for
fusiform rust resistance in slash pine. In Proceedings 12th
Southern Forest Tree Improvement Conference, p. 31-37. National
Technical Information Service, Springfield, VA. 352 p.

SAS Institute Inc. 1982a. SAS User's Guide: Basics, 1982
edition. SAS Institute, Inc, Cary, North Carolina. 923 p.










SAS Institute Inc. 1982b. SAS User's Guide: Statistics, 1982
edition. SAS Institute, Inc, Cary, North Carolina. 584 p.

Schaffer, H. E. and R. A. Usanis. 1969. General least squares
analysis of diallel experiments: A computer program DIALL.
Genetics Dept Res Rep No 1, North Carolina State Univ,
Raleigh. 61 p.

Schmidt, R. A. 1972. A literature review of inoculation techniques
used in studies of fusiform rust. In Biology of Rust
Resistance in Forest Trees (R. T. BTngham, R. J. Hoff and G. I.
McDonald, eds). USDA Forest Serv Misc Publ 1221:341-356.

Schmidt, R. A., R. E. Goddard and C. A. Hollis. 1974. Incidence
and distribution of fusiform rust in slash pine plantations in
Florida and Georgia. Instit Food and Agric Sci, Univ Florida
Bull 763. Gainesville. 21 p.

Schmidt, R. A., R. C. Holley and M. C. Klapproth. 1985. Results
from operational plantings of fusiform-rust-resistant slash and
loblolly pines in high-rust-incidence areas in Florida and
Georgia. In Proceedings of the Rusts of Hard Pines Working
Party Conference (J. Barrows-Broaddus and H. R. Powers, Jr.,
eds) p 33-41. The Georgia Center for Continuing Education,
Univ Georgia, Athens. 331 p.

Schmidt, R. A., H. R. Powers, Jr. and G. A. Snow. 1981.
Application of genetic disease resistance for the control of
fusiform rust in intensively managed southern pine.
Phytopathology 79:993-997.

Snow, G. A., R. J. Dinus, and A. G. Kais. 1975. Variation in
pathogenicity of diverse sources of Cronartium fusiforme on
selected slash pine families. Phytopathology 65:170-175.

Snow, G. A., R. J. Dinus, and C. H. Walkinshaw. 1976. Increase in
virulence of Cronartium fusiforme on a resistant slash pine.
Phytopathology 66:511-513.

Snow, G. A., and M. M. Griggs. 1980. Relative virulence of
Cronartium quercuum f. sp. fusiforme from seven resistant
families of slash pine. Phytopath Medit 19:13-16.

Snow, G. A., and A. G. Kais. 1970. Pathogenic variability in
isolates of Cronartium fusiforme from five southern states.
Phytopathology 60:1730-1731.

Snow, G. A., W. L. Nance and E. B. Snyder. 1982. Relative
virulence of Cronartium quercuum f. sp. fusiforme on loblolly
pine from Livingston parish. In Resistance to Diseases and
Pests in Forest Trees (H. M. Heybroek, B. R. Stephan and K. von
Weissenberg, eds), p 243-250. Centre Agric Publ Document,
Wageningen, The Netherlands. 503 p.










Sohn, S. 1977. Heritability of resistance to Cronartium fusiforme
in Pinus elliottii as affected by disease incidence and a
comparison of breeding procedures. Ph 0 Dissertation, Univ of
Florida, Gainesville. 92 p.

Sohn, S., and R. E. Goddard. 1979. Influence of infection percent
on improvement of fusiform rust resistance in slash pine.
Silvae Genetica 28:173-180.

Squillace, A. E., R. J. Dinus, C. A. Hollis and R. A. Schmidt.
1978. Relation of oak abundance, seed source, and temperature
to geographic patterns of fusiform rust incidence. USDA Forest
Serv Res Pap SE-186. 20 p.

Steel, R. G. D. and J. H. Torrie. 1960. Principles and Procedures
of Statistics. McGraw-Hill Book Company, Inc, New York. 481 p.

Walkinshaw, C. H. 1978. Cell necrosis and fungus content in
fusiform rust-infected loblolly, longleaf, and slash pine
seedlings. Phytopathology 68:1705-1710.

Walkinshaw, C. H., and R. L. Anderson. 1983. Fusiform rust:
Illustration of different symptoms in the greenhouse and
field. USDA Forest Serv Forest Pest Manag Rep 83-1-21. 13 p.

Walkinshaw, C. H., T. R. Dell and S. D. Hubbard. 1980. Predicting
field performance of slash pine families from inoculated
greenhouse seedlings. USDA Forest Serv Res Pap SO-160. 6 p.

Wells, 0. 0., and P. C. Wakeley. 1966. Geographic variation in
survival, growth, and fusiform rust infection of planted
loololly pine. Forest Sci Monogr 11. 40 p.

Wright, J. W. 1976. Introduction to Forest Genetics. Academic
Press, New York. 463 p.




















APPENDIX











Table 26.


Least squares estimates of the mean number of fusiform rust galls per tree by gall type or location
for 48 slash pine families (four- to six-years-old) planted in six progeny tests in southeast
Georgia.


Gall Type Gall Location
Witches'
Family Typical Thin BTFa/ Sunken Broom Fat Stem Limb Stem-Limb Total
--------------------------------mean number of galls per tree-----------------------------


10050
10159
190047
190060
190141
190287
600047
600287
960065
960146
980048
980088
1410047
1410060
1410287
1460065
1460071
1460096
1570065
1570071
1570096
1570146
2700050
2700064
2700295


0.945
2.535
1.221
1.789
1.769
1.156
0.817
0.780
1.149
2.141
0.457
1.106
1.441
1.534
0.603
1.223
1.492
0.975
1.523
1.420
1.781
1.382
0.776
1.706
0.913


0.037
0.034
0.096
0.105
0.046
0.024
0.215
0.097
0.025
0.154
0.037
0.039
0.045
0.031
0.043
0.063
0.012
0.058
0.074
0.024
0.055
0.029
0.034
0.069
0.071


-0.036
0.016
0.177
0.097
0.021
0.079
0.383
0.003
0.109
0.119
0.076
0.115
0.022
0.172
0.036
0.061
0.054
0.041
0.099
0.154
0.139
0.114
0.053
0.196
0.035


-0.008
0.087
0.067
0.069
0.045
0.059
0.074
0.034
0.029
0.027
0.020
0.001
0.052
0.079
0.014
0.011
0.018
-0.019
0.063
0.063
0.033
0.044
0.044
0.180
0.136


0.295
0.900
0.401
0.411
1.085
0.675
0.300
0.261
0.402
0.575
0.041
0.191
0.469
0.503
0.346
0.534
0.505
0.423
0.503
0.421
0.377
0.461
0.364
0.534
0.550


0.089
0.018
-0.005
-0.003
0.026
0.062
0.019
0.037
0.041
0.034
0.025
0.065
0.080
0.049
0.008
0.041
0.037
0.032
0.063
0.063
0.090
0.066
0.008
0.064
0.022


0.960
3.623
3.759
3.157
6.624
2.359
0.703
1.908
2.508
3.083
-1.100
-0.357
2.954
1.733
2.994
3.325
2.554
3.589
2.564
2.575
2.537
3.443
3.008
3.148
3.084


0.855
2.739
1.353
1.738
1.909
1.349
1.201
0.765
1.271
2.186
0.598
1.270
1.496
1.624
0.523
1.444
1.418
1.306
1.632
1.387
1.873
1.396
0.816
1.806
0.890


0.368
0.522
0.261
0.297
0.514
0.376
0.197
0.246
0.215
0.348
0.041
0.179
0.270
0.283
0.267
0.273
0.436
-0.003
0.281
0.397
0.306
0.281
0.206
0.528
0.473


2.183
6.884
5.373
5.192
9.047
4.084
2.101
2.920
3.994
5.617
-0.461
1.092
4.720
3.640
3.785
5.042
4.409
4.892
4.477
4.359
4.714
5.210
4.030
5.382
4.447


---------------











Table 26--continued.


Gall Type Gall Location
Witches'
Family Typical Thin BTFA/ Sunken Broom Fat Stem Limb Stem-Limb Total
--------------------------------mean number of galls per tree-------------------------------


2700355
2840050
2840064
2840295
2840355
2880001
2880050
2880159
2880352
2950064
2950270
2950355
3500048
3500088
3500098
3500354
3520001
3520050
3520159
3540048
3540088
3540098
3540350


0.935
1.606
1.263
1.027
1.013
1.523
1.091
1.960
1.310
2.115
0.655
1.629
1.820
2.705
2.085
2.674
0.817
1.601
0.964
1.972
0.901
1.911
1.798


0.084
0.029
0.015
0.007
0.040
0.075
0.024
0.123
0.046
0.038
0.161
0.046
0.048
0.015
0.079
0.010
0.015
0.041
0.011
0.021
0.006
0.041
0.046


0.022
0.075
0.131
0.054
0.049
0.075
0.067
0.035
0.105
0.160
0.077
0.052
0.091
0.256
0.132
0.156
0.132
0.081
0.040
0.262
0.090
0.184
0.157


0.121
0.068
0.068
0.098
0.120
0.077
0.090
0.090
0.040
0.084
0.056
0.118
0.015
0.019
0.003
0.011
0.058
0.013
0.043
0.009
0.010
0.056
0.021


0.509
0.497
0.441
0.265
0.408
0.830
0.614
0.648
0.693
1.129
0.413
0.632
0.382
0.434
0.314
0.902
0.462
0.395
0.595
0.395
0.450
0.266
0.867


0.022
0.073
0.036
0.028
0.008
0.048
0.043
0.043
0.029
0.205
0.050
0.067
0.133
0.187
0.076
0.132
0.010
0.085
0.062
0.079
0.063
0.188
0.110


3.094
3.119
3.106
3.064
3.114
5.041
3.492
3.162
4.034
3.066
3.054
3.521
1.808
3.565
1.901
5.212
3.298
0.885
1.567
2.647
3.860
1.155
9.145


0.996
1.596
1.152
0.767
0.822
1.915
1.367
2.008
1.583
2.505
0.696
1.355
1.763
3.026
2.233
2.483
1.044
1.706
1.198
1.928
1.032
1.807
1.982


0.435
0.286
0.417
0.377
0.437
0.330
0.324
0.418
0.288
0.733
0.354
0.770
0.476
0.292
0.319
0.756
0.103
0.307
0.249
0.506
0.177
0.571
0.620


4.524
5.001
4.675
4.207
4.372
7.287
5.183
5.587
5.905
6.303
4.104
5.646
4.047
6.883
4.454
8.451
4.444
2.898
3.014
5.081
5.069
3.534
11.747


a/Basally truncated fusiform gall.










Table 27.


Least squares estimates of the mean proportion of trees infected with fusiform rust galls by gall
type or location for 48 slash pine families (four- to six-years-old) planted in six progeny tests in
southeast Georgia.


Gall Type Gall Location
Witches'
Family Typical Thin BTFa/ Sunken Broom Fat Stem Limb Stem-Limb
---------------------------------------------------------------------------------------


10050
10159
190047
190060
190141
190287
600047
600287
960065
960146
980048
980088
1410047
1410060
1410287
1460065
1460071
1460096
1570065
1570071
1570096
1570146
2700050
2700064
2700295
2700355


30.57
69.67
55.46
74.94
84.71
55.87
34.16
36.11
50.44
52.53
19.36
34.77
55.91
59.24
35.93
42.01
47.00
37.30
59.23
57.96
67.20
50.74
40.69
77.12
47.19
49.16


1.13
2.94
5.57
7.32
3.93
2.29
1.52
7.90
0.42
1.33
1.92
2.72
2.45
1.50
0.88
4.16
0.22
5.21
1.52
0.91
4.76
2.47
1.03
4.80
5.49
7.33


0.06
0.71
8.37
7.72
2.73
4.74
5.24
0.06
4.86
9.28
2.26
4.99
3.76
7.69
1.38
2.29
1.32
0.50
5.70
10.27
7.35
9.19
2.55
14.63
1.36
0.25


0.02
7.33
6.54
5.09
3.41
5.60
7.57
2.45
1.71
1.51
1.40
0.32
5.00
7.78
0.34
0.51
1.12
0.01
2.47
2.89
2.53
2.20
3.92
16.63
6.99
10.72


30.84
49.37
31.27
28.95
66.84
37.78
19.02
14.91
29.06
33.43
3.75
16.97
27.41
27.53
21.69
35.19
30.98
33.59
37.20
31.96
24.08
32.34
28.22
36.07
42.81
43.74


3.044
1.494
0.297
0.635
2.502
4.885
2.247
2.251
1.687
2.213
0.710
3.591
7.154
2.598
0.259
2.121
1.776
0.233
3.862
3.862
8.186
4.920
0.259
4.145
1.378
1.353


21.13
29.63
32.96
41.34
52.78
32.23
19.93
17.07
18.89
28.96
1.30
5.45
29.95
44.52
21.81
21.32
26.75
23.29
26.43
24.08
23.27
25.78
26.32
40.47
33.19
38.58


33.52
65.59
53.67
69.84
90.08
52.74
29.97
36.77
49.26
48.52
20.01
33.54
57.21
55.12
28.35
44.81
36.84
40.81
59.02
52.01
68.61
48.27
37.07
71.84
49.26
54.54


9.92
35.68
23.90
30.14
55.08
25.31
14.79
23.03
20.15
26.27
2.99
11.84
27.32
24.87
26.61
27.08
28.15
18.01
27.80
33.22
28.28
24.38
19.74
48.96
39.28
42.08











Table 27--continued.

Gall Type Gall Location
Witches'
Family Typical Thin BTFa/ Sunken Broom Fat Stem Limb Stem-Limb
----------------------------------- --%-----------------------


2840050
2840064
2840295
2840355
2880001
2880050
2880159
2880352
2950064
2950270
2950355
3500048
3500088
3500098
3500354
3520001
3520050
3520159
3540048
3540088
3540098
3540350


48.24
61.40
56.28
53.92
51.30
44.37
69.48
50.97
79.64
42.27
84.49
64.12
65.25
61.47
86.42
34.75
34.78
43.89
61.76
52.09
56.43
89.28


1.47
0.88
0.55
2.42
6.27
2.26
8.32
3.89
3.37
8.55
1.58
3.25
0.82
5.67
0.14
1.23
1.64
1.06
1.10
0.27
2.20
3.17


6.10
10.64
3.89
2.28
5.80
5.76
3.31
5.21
13.21
4.11
3.64
4.15
7.15
4.22
6.92
6.97
5.19
4.55
11.01
4.29
10.14
9.88


6.51
4.35
9.65
9.40
5.27
4.09
8.85
3.94
8.34
3.53
11.63
0.72
1.33
0.38
0.86
3.91
0.81
3.53
0.57
0.59
2.60
1.49


32.35
29.26
17.68
29.05
49.43
38.42
46.30
36.53
60.77
34.29
46.36
26.97
29.46
16.64
67.49
29.66
18.96
33.52
29.72
36.25
18.88
75.18


2.650
3.405
1.780
0.259
3.554
2.181
1.629
2.797
16.843
3.259
4.352
5.125
3.745
5.258
10.228
1.059
3.054
3.723
3.139
3.792
7.534
2.664


36.13
37.58
34.63
35.81
30.10
28.37
49.09
30.01
29.48
31.64
38.78
20.75
26.94
20.10
54.76
22.25
16.35
21.57
27.75
26.29
23.99
61.81


41.36
53.19
50.98
45.91
54.95
43.07
62.88
54.54
77.21
45.59
73.56
59.53
64.83
56.11
85.76
32.59
29.63
40.34
62.87
56.67
45.25
90.42


28.22
37.93
31.97
36.16
32.59
26.40
29.98
26.64
62.02
33.17
62.75
31.47
26.83
22.51
49.16
19.16
15.37
16.02
31.16
24.83
28.70
65.48


A/Basally truncated fusiform gall.












Variance components and their standard errors
replication, general combining ability (GCA),
combining ability (SCA) and error effects for
number of galls per tree for various fusiform
types or locations as computed by the FORTRAN
DIALL, for each crossing set and test.


for
specific
the mean
rust gall
program,


X-ing Variance Component
Set Test Gall Replication GCA SCA Error


1 34 Total
Stem
Limb
Stem-Limb
Typical
Witches' Broom
BTFA/
Fat
Thin
Sunken

35 Total
Stem
Limb
Stem-Limb
Typical
Witches' Broom
BTFI/
Fat
Thin
Sunken

36 Total
Stem
Limb
Stem-Limb
Typical
Witches' Broom
BTFI/
Fat
Thin
Sunken

37 Total
Stem
Limb
Stem-Limb
Typical
Witches' Broom
BTFI/
Fat
Thin
Sunken


0.53 + 0.81
-0.001 + 0.001
0.29 + 0.55
0.03 + 0.03
0.09 + 0.38
0.002 + 0.006
0.009 + 0.01
0.01 + 0.01
0.001 + 0.001
-0.0002 + 0.00006

0.11 + 0.16
0.0007 + 0.001
0.1 + 0.13
0.003 + 0.004
0.11 + 0.13
0.007 + 0.009
0.003 + 0.002
-0.001 + 0.0003
0.0 + 0.00002
0.0005 + 0.0003

-0.01 + 0.06
-0.0007 + 0.002
-0.001 +-0.03
0.0002 + 0.013
-0.02 + 0.01
-0.01 + 0.004
0.001 + 0.0008*
-O.000oT + 0.007
0.0003 + 0.0005
-0.0001 + 0.0002

0.004 + 0.07
0.008 + 0.015
-0.04 + 0.01
-0.006 + 0.01
-0.045 + 0.02
-0.007 + 0.016
0.003 + 0.004
-0.0001 + 0.002
-0.0001 + 0.0004
-0.001 + 0.001


-1.07 + 1.31
0.04 + 0.02**
-0.99 + 1.05
0.01 + 0.04
-0.94 + 0.85
0.2 + 0.12**
-0.01 + 0.01
-0.01 + 0.01
-0.0002 + 0.0004
-0.0001 + 0.0004

2.11 + 1.25*
0.02 + 0.01*
1.26 + 0.75**
0.04 + 0.03*
1.02 + 0.63**
0.09 + 0.06
0.01 + 0.007*
-0.0007 + 0.0004
0.00001 + 0.00003
0.001 + U.001

0.45 + 0.37
0.01 + 0.01
0.17 + 0.13
0.02 + 0.02
0.11 + 0.09
0.02 + 0.02
0.0009 + 0.0006*
0.004 + 0.003
-0.00002 + 0.0002
-0.0001 + 0.0002

0.85 + 0.56*
0.03 + 0.03
0.18 + 0.12*
0.11 + 0 08*
0.35 + 0.23*
0.09 + 0.07
0.0001 + 0.005
0.0003 T 0.002
0.0008 + 0.0006
0.001 + 0.002


5.28 + 3.92*
-0.005 + 0.003
4.46 + 3.22*
0.11 + 0.08*
3.59 + 2.69*
0.004 + 0.014
0.05 + 0.04*
0.05 + 0.04*
0.0001 + 0.001
0.001 + 0.001

-0.17 + 0.16
0.003 + 0.003
-0.16 + 0.11
0.01 + 0.01
-0.03 + 0.14
0.05 + 0.04*
-0.002 + 0.004
-0.001 + 0.002
-0.00001 + 0.0001
0.003 + U.002*

0.29 + 0.27
-0.003 + 0.004
0.09 + 0.09
0.006 + 0.03
0.05 + 0.06
-0.003 + 0.02
-0.0006 + 0.0004
-0.02 + U.007
-0.0003 + 0.0006
-0.0001 ; 0.0005

0.16 + 0.2
0.004 + 0.02
0.003 + 0.06
0.012 + 0.03
0.014 + 0.08
0.04 + 0.06
0.012 + 0.01
-0.0009 + 0.003
-0.0006 + 0.0006
-0.003 + 0.002


5.87 + 1.87
0.02 + 0.007
4.58 + 1.45
0.12 + 0.04
4.16 + 1.32
0.06 + 0.02
0.07+ 0.02
0.06 + 0.02
0.006 + 0.002
0.002 + 0.0006

1.1 + 0.35
0.01 + 0.003
0.83 + 0.26
0.03 + 0.01
0.72 + 0.23
0.06 + 0.02
0.02 + 0.007
0.009 + 0.003
0.0003 + 0.0001
0.003 + 0.001

0.86 + 0.21
0.04 + 0.009
0.34 + 0.08
0.18 + 0.04
0.27 + 0.07
0.14 + 0.03
0.004 + 0.001
0.1 + U.02
0.005 + 0.001
0.004 + 0.001

1.0 + 0.23
0.17 + 0.04
0.51 + 0.12
0.21 + 0.5
0.64 + 0.14
0.31 + 0.07
0.04 + 0.009
0.033 + 0.007
0.008 + 0.002
0.03 + 0.007


Table 28.













Table 28--continued.


X-ing Variance Component
Set Test Gall Replication GCA SCA Error


2 36 Total
Stem
Limb
Stem-Limb
Typical
Witches' E
BTFa/
Fat
Thin
Sunken


37 Total
Stem
Limo
Stem-Limb
Typical
Witches' Broom
BTFa/
Fat
Thin
Sunken


-0.003 + 0.02
-0.002 + 0.001
-0.004 + 0.01
-0.0002 + 0.001
-0.012 + 0.006
Room 0.006 7 0.007
0.0 + 0.002
-0.0 o 0.0001
-0.0063 + 0.0002
-0.0004 + 0.0001


0.02 + 0.05
0.002 + 0.005
-0.008 + 0.02
-0.01 + 0.005
0.008 + 0.03
0.002 ; 0.01
0.0001 + 0.001
-0.00004 + 0.0008
-0.00005 + 0.0002
0.0005 + 0.001


40 Total 0.03 + 0.04
Stem 0.001 + 0.001
Limb 0.03 + 0.04
Stem-Limb -0.0002 + 0.001
Typical 0.07 + 0.06*
Witches' Broom -0.005 + 0.005
BTFa/ 0.0004-+ 0.001
Fat -0.0 + 0.00005
Thin -0.00T + 0.004
Sunken -0.00003 + 0.0001


41 Total
Stem
Limb
Stem-Limb
Typical
Witches' Broom
BTFa/
Fat
Thin
Sunken


-0.09 + 0.05
-0.001 + 0.0007
-0.06 + 0.05
0.0 + 0.004
-0.07 + 0.04
0.3 + 0.3
0.001 + 0.001
0.001 T 0.001
0.002 + 0.005
0.0002 + 0.0002


0.008 + 0.03
-0.001 + 0.005
0.005 T 0.01
0.0005 + 0.0007
0.003 + 0.01
-0.003 + 0.01
0.0006 + 0.0007
0.0001 0.0001
0.0002 + 0.0002
0.0006 + 0.0003

0.07 + 0.06
0.007 + 0.007
0.04 + 0.03
-0.002 + 0.007
0.04 + 0.04
0.002 + 0.001
0.002 + 0.002
0.0006 + 0.0004
0.0002 + 0.0002
0.004 + 0.003

0.04 + 0.09
-0.0004 + 0.001
0.05 + 0.08
0.0003 + 0.002
-0.002 + 0.08
0.001 ; 0.002
0.0008 + 0.001
0.0001 7 0.0002
-0.00005 + 0.005
0.0001 + 0.0004

0.05 + 0.15
0.0004 + 0.003
-0.02 + 0.06
0.03 7 0.03
0.03 7 0.08
U.005 + 0.01
-0.001 + 0.001
-0.0001 + 0.0004
0.001 + 0.004
0.0001 + 0.0001


0.01 + 0.05
0.006 + 0..8
-0.006 + 0.02
-0.002 T 0.001
-0.006 + 0.019
0.017 T 0.02
0.0003 + 0.0007
-0.0003 7 0.0002
-0.0005 + 0.0003
-0.0007 + 0.0002

-0.045 + 0.05
-0.0003+ 0.006
-0.04 + 0.02
-0.004 + 0.02
-0.02 + 0.03
-0.03 + 0.008
-0.0002 + 0.002
-0.002 + 0.0006
-0.0004 + 0.0002
-0.0003 + 0.002

0.1 + 0.1
0.007 + 0.002
0.075 + 0.1
0.002 + 0.004
0.16 + 0.13*
-0.02 + 0.007
0.0005 + 0.002
0.0001 + 0.0002
-0.0005 + 0.01
0.0004 + 0.0006

0.09 + 0.2
0.003 + 0.005
-0.04 + 0.1
0.02 + 0.02
0.002 + 0.1
-0.01 + 0.02
0.003 + 0.004
0.0002 + 0.001
-0.001 + 0.007
-0.0001 + 0.0001


0.18 + 0.06
0.016 + 0.005
0.12 + 0.03
0.01 7 0.003
0.1 + 0.03
0.03 + 0.01
0.002 + 0.0007
0.002 T 0.0005
0.003 + 0.0008
0.003 + U.0008

0.47 + 0. 1
0.04 + 0.01
0.3 + 0.08
0.12 + 0.03
0.25 + 0.07
0.13 + 0.04
0.01 + 0.003
0.009 + 0.003
0.003 + 0.0007
0.01 + 0.003

0.3 + 0.1
0.00 + 0.002
0.32 + 0.09
0.02 + 0.006
0.22 + 0.06
0.09 + 0.03
0.01 + 0.003
0.001 + 0.0003
0.06 + 0.02
0.002 + 0.001

0.99 + 0.3
0.02 7 0.005
0.9 + 0.3
0.04 + 0.01
0.8 + 0.2
0.16 + 0.05
0.01 + 0.003
0.004 + 0.001
0.04 + 0.01
0.001 + 0.0002












Table 28--continued.


X-ing Variance Component
Set Test Gall Replication GCA SCA Error


4 34 Total 0.44 + 0.9
Stem -0.001 + 0.001
Limb 0.29 + 0.67
Stem-Limb 0.005 + 0.02
Typical 0.27 +-0.63
Witches' Broom 0.02 + 0.005
BTFa/ -0.002 + 0.002
Fat -0.0004 + 0.003
Thin -0.0003 + 0.0003
Sunken 0.0005 + 0.002

35 Total -0.06 + 0.06
Stem 0.002 + 0.003
Limb -0.03 + 0.08
Stem-Limb -0.006 + 0.003
Typical 0.02 + 0.07
Witches' Broom -0.02 + 0.01
BTFa/ 0.001 + 0.002
Fat -0.00005 + 0.001
Thin 0.0001 + 0.0002
Sunken -0.0003 0.0002

40 Total 0.07 + 0.1
Stem -0.0003 + 0.001
Limb 0.07 + 0.1
Stem-Limb -0.0003 + 0.001
Typical 0.01 + 0.05
Witches' Broom 0.03 + 0.03
BTFa/ 0.001 + 0.001
Fat 0.001 + 0.0009*
Thin -0.0005 + 0.0006
Sunken -0.0001 T 0.0002

41 Total -0.02 + 0.02
Stem -0.0003 + 0.0007
Limb -0.002 + 0.02
Stem-Limb -0.004 + 0.005
Typical 0.02 + 0.03
Witches' Broom -0.005 + 0.003
BTFa/ 0.001 + 0.001
Fat -0.0002 + 0.0006
Thin 0.009 + 0.008*
Sunken -0.0003 + 0.0004


2.28 + 1.98
0.022 + 0.016
1.2 + 1.2
0.8 + 0.6
0.89 + 1.3
0.08 + 0.05
0.0001 + 0.005
0.001 + 0.002
0.0004 + 0.004
0.005 + 0.004

0.07 + 0.3
0.01 + 0.01
0.07 + 0.15
-0.01 + 0.01
0.006 + 0.1
0.01 + 0.03
0.0003 + 0.0009
-0.0005 + 0.0004
0.00003 + 0.0003
0.0005 + 0.0004

0.1 + 0.1
0.00T + 0.001
0.1 + 0.1
0.001 + 0.001
0.06 + 0.05
0.01 + 0.04
-0.001 + 0.002
-0.001 + 0.001
0.0002 + 0.0006
0.0004 i 0.001

0.5 + 0.3
-0.01 + 0.02
0.4 + 0.3*
-0.001+ 0.009
0.3 + 0.2
0.04 + 0.02*
-0.006 + 0.02
0.001 + 0.0007
0.002 +0.003
0.00005 + 0.002


0.19 + 1.6
0.006 + 0.007
0.08 + 1.2
-0.02 + 0.02
0.08 + 1.6
-0.01 + 0.02
0.004 + 0.008
-0.005 + 0.006
0.008 + 0.006**
-0.002 + 0.004

0.57 + 0.56
0.01 + 0.01
0.02 + 0.2
0.04 + 0.04
0.2 + 0.2
0.02 + 0.06
-0.001 + 0.002
-0.001 + 0.001
0.0004 + 0.0007
-0.0008 + 0.0006

-0.1 + 0.1
-0.0004 + 0.001
-0.1 + 0.1
-0.005 + 0.001
-0.08 + 0.06
0.05 + 0.05
0.003 + 0.004
0.004 + 0.003**
-0.001 + 0.001
0.001 + 0.001

0.02 + 0.08
0.03 T 0.02*
-0.04 + 0.02
-0.002 + 0.01
-0.04 + 0.02
-0.01 + 0.005
0.03 + 0.02
-0.001 + 0.0008
-0.002 + 0.004
0.001 + 0.002


6.2 + 2.19
0.014 + 0.005
4.9 + 1.7
0.16 + 0.06
4.6 + 1.6
0.12 + 0.04
0.02 + 0.009
0.03 + 0.01
0.005 + 0.001
0.02 + 0.007

1.2 + 0.4
0.02 + 0.005
1.2 + 0.4
0.08 + 0.03
0.6 + 0.2
0.2 + 0.07
0.01 + 0.004
0.009 + 0.003
0.002 + 0.001
0.004 + 0.001

1.1 + 0.3
0.01 + 0.002
1.0 + 0.3
0.02+ 0.005
0.5 + 0.2
0.1 + 0.05
0.01 + 0.003
0.003 + 0.001
0.009 + 0.003
0.003 + 0.001

0.2 + 0.09
0.007 + 0.003
0.13 + 0.05
0.05 + 0.02
0.15 + 0.06
0.04 + 0.02
0.005 + 0.001
0.005 + 0.002
0.02 + 0.007
0.004 + 0.002













Table 28--continued.


X-ing Variance Component
Set Test Gall Replication GCA SCA Error


5 34 Total
Stem
Limb
Stem-Limb
Typical
Witches'
BTFa/
Fat
Thin
Sunken


-0.3 + 0.3
0.006 + 0.01
-0.34 + 0.3
0.005 + 0.01
-0.2 + 0.2
Broom 0.01 + 0.02
0.01 + 0.04
-0.003 + 0.002
-0.02 + 0.03
-0.0005 + 0.0005


35 Total 0.07 + 0.1
Stem -0.01 T 0.003
Limb 0.08 + 0.1
Stem-Limb 0.0007 + 0.006
Typical 0.13 + 0.11*
Witches' Broom 0.02 T 0.04
BTFa/ -0.001-+ 0.01
Fat 0.0001 + 0.0008
Thin 0.0006 + 0.0007
Sunken 0.0006 + 0.001

40 Total 0.02 + 0.04
Stem -0.002 + 0.001
Limb 0.02 + 0.04
Stem-Limb -0.001 + 0.001
Typical 0.006 ; 0.02
Witches' Broom -0.0005 + 0.003
BTFa/ -0.0005 + 0.0004
Fat 0.0004 7 0.0004
Thin 0.0007 + 0.001
Sunken 0.0001 + 0.0004

41 Total -0.1 + 0.9
Stem -0.02 + 0.006
Limb -0.05 + 0.07
Stem-Limb 0.0001 + 0.003
Typical -0.06 + 0.06
Witches' Broom 0.001 + 0.01
BTFal -0.001 0.0007
Fat -0.0003 + 0.0004
Thin 0.002 + 0.002
Sunken 0.0006 + 0.0005


-0.1 + 0.5
-0.04 + 0.06
0.4 + 0.3
0.05 + 0.03
-0.02 + 0.3
0.1 + 0.08
0.01 + 0.05
0.001 + 0.001
0.001 + 0.05
0.002 + 0.001

0.4 + 0.4
0.02 + 0.02
0.17 + 0.2
-0.01 + 0.04
0.3 + 0.3
0.01 + 0.06
-0.002 + 0.02
0.0002 + 0.0006
0.0003 + 0.0008
0.001 + 0.001

0.23 + 0.2
0.001 + 0.002
0.1 + 0.1
0.004 + 0.004
0.05 + 0.08
0.02 + 0.02
0.0003 + 0.0004
-0.0003 + 0.0003
-0.0003 + 0.0006
0.0003 + 0.0004

0.3 + 0.9
0.008 + 0.009
0.1 + 0.7
0.01 + 0.005
0.3 +-0.4
-0.03 + 0.08
0.003 + 0.002
-0.0006 + 0.0009
0.002 + 0.001
0.0001 + 0.0002


-0.4 + 0.8
0.08 + 0.08**
-1.0 + 0.5
-0.02 + 0.01
-0.2 + 0.5
-0.01 + 0.01
-0.02 + 0.07
-0.006 + 0.003
-0.02 + 0.08
-0.002 + 0.001

0.5 + 0.4*
-0.00i1 + 0.02
0.2 + 0.2
0.1 T 0.08**
0.2 + 0.2
0.06 + 0.1
0.01 + 0.03
-0.001 + 0.001
0.0008 + 0.001
-0.002 + 0.001

0.17 + 0.16
0.002 + 0.004
0.1 + 0.1
O.OT + 0.004
0.12 + 0.1*
0.03 + 0.02*
-0.001 + 0.001
0.0009 + 0.0008
-0.001 + 0.002
-0.0002 + 0.0006

1.04 + 0.98*
-0.002 + 0.009
0.9 + 0.8*
-0.006 + 0.003
0.3 + 0.4
0.12 + 0.1*
-0.003 + 0.001
0.0006 + 0.001
-0.001 + 0.001
0.0 + 0.0003


3.7 + 1.5
0.08 + 0.03
3.7 + 1.5
0.07 + 0.03
2.1 + 0.9
0.08 + 0.03
0.3 + 0.1
0.02 + 0.009
0.3 + 0.1
0.005 + 0.002

0.5 + 0.2
0.08 + 0.3
0.4 + 0.1
0.06 + 0.02
0.2 + 0.09
0.3 +0.1
0.1 + 0.04
0.007 + 0.002
0.002 + 0.0009
0.007 + 0.003

0.4 + 0.1
0.02 + 0.04
0.3 + 0.1
0.02 + 0.006
0.2 + 0.07
0.05 + 0.01
0.01 + 0.003
0.002 + 0.0005
0.01 + 0.004
0.005 + 0.001

0.4 + 0.2
0.03 + 0.1
0.3 + 0.1
0.02 + 0.007
0.4 + 0.2
0.05 + 0.02
0.007f+ 0.003
0.003 + 0.001
0.003 + 0.001
0.001 + 0.0004













Table 28--continued.


X-ing Variance Component
Set Test Gall Replication GCA SCA Error

9 40 Total 0.09 + 0.07* 0.04 + 0.06 0.07 + 0.09 0.3 + 0.07
Stem -0.001 + 0.0004 0.0007 + 0.0009 -0.002 + 0.002 0.02-+ 0.04
Limb 0.05 + 0.05 0.03 + 0.04 0.02 +-0.06 0.3 + 0.07
Stem-Limb 0.005 + 0.004* 0.009 + 0.005* -0.004 + 0.002 0.02 + 0.006
Typical 0.04 +-0.03** 0.04 +-0.03 0.01 + 0.03 0.1 +- 0.03
Witches' Broom 0.003 + 0.005 0.006 + 0.008 0.004 + 0.01 0.05 + 0.01
BTFa/ 0.0006 + 0.001 0.002 + 0.002 -0.001 + 0.002 0.016 + 0.004
Fat 0.0001 + 0.0003 0.00002 + 0.0003 0.0 + 0.001 0.005 + 0.001
Thin -0.001 + 0.0004 -0.0007 + 0.0008 -0.0003 + 0.002 0.01 + 0.004
Sunken 0.0004 + 0.0007 0.0004 + 0.0009 0.0008 + 0.001 0.007 + 0.002

41 Total 0.07 + 0.08 0.7 + 0.4* 0.12 + 0.16 0.6 + 0.1
Stem 0.0002 + 0.003 0.0004 + 0.001 -0.007 + 0.003 0.04 + 0.01
Limb 0.01 + U.04 0.42 + 0.27* 0.11 + 0.13 0.4 +-0.1
Stem-Limb 0.005 + 0.008 0.04 + 0.02* -0.002 + 0.01 0.07 + 0.02
Typical -0.01 + 0.02 0.23 +0.16 0.07 + 0.1 0.4 + 0.1
Witches' Broom 0.04 + 0.05 0.08 + 0.07 -0.008 + 0.07 0.4 + 0.1
BTFa/ 0.000T + 0.0009 0.001 + 0.0007 -0.003 0.001 0.01-+ 0.003
Fat 0.002 + 0.002 0.005 +0.004 0.0007 + 0.003 0.01 0.003
Thin 0.003 + 0.002* 0.001 + 0.0006* -0.002 +-0.0006 0.009 + 0.002
Sunken 0.002 0.002 -0.0007-+ 0.002 0.002 T 0.004 0.014 + 0.004

a/Basally truncated fusiform gall.

*F test for the effect significant at the 5% level.

**F test for the effect significant at the 1% level.



















Table 29. Specific combining ability for fusiform rust incidence and fat galls based on the
proportion of trees diseased or having fat galls. Values reflect performance of 48
families (four- to six-years-old) in six progeny tests in southeast Georgia.




Specific Combining Ability
-........------------------------------------------------- Males-----------------------------------------------------------
Female UU01 0047 00U48 0050 0060 0064 0065 0071 0088 0096 0098 0141 0146 0159 0270 0287 0295 0350 0352 0354 0355

Percentage of Trees Infected With Rust


UUU1
UUbU
UObU

UU98
U141
Ul4b
ulbl


u288
U295
U3bU
u3b2

0354


-3.13
-2.8


b.88


b.87


-1.46


4.82


-4.51


-2.18


2.15


1.38
6.97
-4.92


1.13


7.1b
-4.04


-7.32


6.57


4.2b


-6.47 2.01
1.06 -2.06


-3.19


1.83


-7.65
3.24


-1.24


-6.14 2.55


5.31
2.25


-7.63


-2.19


0.78
-1.13


-3.93


-6.57


6.97


-1.94


-5.47
-1.89

7.32


0.58




University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - Version 2.9.7 - mvs