Heteroploid gene transfers in Vaccinium, section Cyanococcus


Material Information

Heteroploid gene transfers in Vaccinium, section Cyanococcus
Physical Description:
viii, 76 leaves : ill. ; 28 cm.
Goldy, Ronald G., 1954-
Publication Date:


Subjects / Keywords:
Blueberries -- Breeding   ( lcsh )
Vaccinium   ( lcsh )
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )


Thesis (Ph. D.)--University of Florida, 1983.
Includes bibliographical references (leaves 68-75).
Statement of Responsibility:
by Ronald G. Goldy.
General Note:
General Note:

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 000381783
notis - ACC2315
oclc - 10358467
System ID:

Full Text

SECTION Cyanococcus







Appreciation is extended to Drs. Wayne Sherman, Professor, and

Gloria Moore, Assistant Professor, Fruit Crops Department for their

academic guidance and in the sharing of their academic experiences.

I also thank Drs. Mark Bassett, Associate Professor, Vegetable Crops

Department, and Ken Quesenberry, Associate Professor, Department of

Agronomy, for their role in furthering my knowledge of plant breeding

and genetics.

Special thanks go to Dr. Paul Lyrene, Associate Professor, Fruit

Crops Department, who not only served as committee chairman, and shared

his plant breeding expertise, but also through his unique personality

became a close and respected friend.

Appreciation is also extended to all the faculty, staff, and

graduate students of the Fruit Crops Department, especially those

students who occupied room 2123 for the past 3 years. They all

certainly contributed to making those 3 years an enjoyable and memorable


Finally, I am especially appreciative to my wife, Kathy, who for

the past 6 years has provided much in the way of spiritual and moral

support as well as exhibiting the patience and understanding necessary

to be the wife of a graduate student. This degree is as much hers as it

is mine and may our future adventures be as pleasant and successful as

this one has been.




LIST OF TABLES ... . . v

LIST OF FIGURES ... ............. vi

ABSTRACT ... ...................... vii



Polyploidy and Wide Hybridization . 6
Artificial Polyploid Induction . 14
Meiotic Pairing in Blueberries . .. 16

Vaccinium ashei (READE) x V. darrowi (CAMP)

Introduction ....... * 17
Materials and Methods ...... . 19
Results and Discussions .... ......... 20

Two Synezetic Knots and Nuclear Organizing
Regions (NOR), Figure 1-a . .. 21
Lagging Chromosomes at Anaphase I, Figure 1-b .. .21
Nonassociating Chromosomes at Telophase I,
Figure 1-c . . 25
Nonsynchrony of Meiosis II, Figure 1-d .25
More than 2 Nuclei at the End of Meiosis I,
Figure 1-e . 25
Nonparallel Spindles at Anaphase II,
Figure 1-f . . 26
Lagging Chromosomes at Anaphase II,
Figure 2-a . . 26
More than 4 Nuclei at Telophase II,
Figure 2-b ..... . 26
Extra Nucleoli at Telophase II,
Figure 2-c . . 29
Unreduced Gametes, Figure 3-a . .. 29


Mature Pollen Abnormalities; Polyspory,
Incomplete Tetrads, Figures 3-b and 3-c .




Introduction . .
Materials and Methods . .
Results and Discussion . .


Materials and Methods of Experiments 1-4 .

Experiment 1 . . .
Experiment 2 . . .
Experiment 3 . . .
Experiment 4 . . .

Results and Discussion of Experiments 1-4 .
Materials and Methods of Experiments 5-14 .

Experiment 5
Experiment 6
Experiment 7
Experiment 8
Experiment 9
Experiment 10 .
Experiment 11 .
Experiment 12 .
Experiment 13 .
Sirnn-4imnt 1

Detection of Polyploids and Ploidy
Determinations . .

Results and Discussion of Experiments 5-14 .




* .

. . .

. . .

. .


Table Page

1 Number of Pollen Grains Per Sporad and their
Apparent Chromosome Number . 32

2 Crossability and Fertility Data of 2 V. ashei x V.
darrowi Hybrids . .... 34

3 Comparative Fertility of 4x x 4x and 4x x 8x Crosses 44

4 Composition of Modified McCown and Lloyd's Woody
Plant Medium (WPM), Modified Anderson's Rhododendron
Medium (ARM), and Modified Knop's Medium (KM) 50

5 Growth of Highbush Blueberry Clone MHB on
Colchicine-Free Medium Following Treatment of
Explants with Various Concentrations of
Colchicine for 14 days on Solid Medium ... 59

6 Growth of Highbush Blueberry Clone MHB on
Colchicine-Free Medium Following Treatment
of Explants with Various Concentrations of
Colchicine for 48 Hours in Liquid Medium .. 59

7 Performance of 6 Highbush Clones on
Colchicine-Free Medium After 24 and
48 Hour in Vitro 0.1% Colchicine
Treatment in Liquid Arm .. 61

8 Vials Containing Visible Polyploids Following a
96 Hour Darkness or Cold Darkness Pretreatment,
7 Hours at Normal Incubation Conditions
and a 0.025% Colchicine Treatment for 24 or
48 Hours . . ... ...... 63

9 Vials Containing Visible Polyploids Following
7 Days of Darkness of Cold Darkness, 24-192
Hours at Normal Incubation Conditions and a
0.025% or 0.050% Colchicine Treatment for
24 Hours . . ... ...... 65


Figure rage

1 Meiotic irregularities observed in pollen mother
cells of V. ashei x V. darrowi hybrids ... 23

2 Meiotic irregularities observed in pollen mother
cells of V. ashei x V. darrowi hybrids ... 28

3 Irregularities observed in pollen of V. ashei x V.
darrowi hybrids . . .. 31

4 Meiosis and meiotic products of an 8x colchiploid
highbush blueberry . .... 42

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

Heteroploid Gene Transfers in Vaccinium
Section Cyanococcus


Ronald G. Goldy

August 1983

Chairman: Dr. Paul M. Lyrene
Major Department: Horticultural Science (Fruit Crops)

Two methods for interspecific gene transfers between 6x rabbiteye

(V. ashei Reade) and 4x highbush (V. corymbosum L.) blueberries were

investigated. One method involved an attempt to produce a synthetic

4x by crossing the 6x rabbiteye to the 2x V. darrowi Camp. This

synthetic 4x could then be crossed to 4x highbush. The other method was

an attempt to produce a synthetic 6x highbush by doubling 4x highbush

with colchicine to produce an 8x individual and then backcrossing it to

4x highbush.

Two 6x-2x hybrids and one 8x plant were evaluated for fruit set,

fruit weight, number of well developed seed/fruit and number of

seedlings/pollination. The two 6x-2x hybrids differed significantly in

all parameters except seeds/fruit when crossed as males onto 4x highbush

clones. The 8x clone was significantly less fertile than 4x controls in

all parameters. Tetrapolid x octoploid crosses produced one 6x hybrid

and 124, 4x seedlings from 739 pollinations. Chromosome numbers of

seedlings from the 6x-2x hybrids reciprocally crossed to each other and

4x highbush have not been determined.

Meiosis in PMC's of the 8x appeared normal except for the

occasional appearance of 2 extra nuclei at Telophase II. The 6x-2x

hybrids showed numerous cytogenetic irregularities including 60 somatic

chromosomes instead of the expected 48, 2 synezetic knots, 2 nucleolar

organizing regions, lagging chromosomes, nonassociating chromosomes,

meiotic nonsynchrony, micronuclei at Telophase I and II, nonparallel

spindles, extra nucleoli, increased percentage of unreduced gametes,

incomplete tetrads, and polyspory.

A technique for producing 8x plants from 4x highbush was developed

using tissue culture and colchicine. A 7 day pretreatment of cold (4C)

darkness followed by a growth phase of 48 to 96 hours enhanced the

effect of colchicine when 2-node cuttings were treated for 24 hours in

liquid media. Modified Anderson's Rhododendron Medium produced highbush

cultures with greater vigor than either modified McCown and Lloyd's

Woody Plant Medium or modified Knop's Medium. Shoot proliferation of

highbush was better with 10mg/1 2iP than 5mg/l, and as good as 20mg/l.




The blueberry, Vaccinium spp., is a variable and wide-ranging genus

(95). Many species are indigenous to North America (12) and the genus

is especially diverse in the southeastern United States, with Florida

alone having as many as 10 species (102). The basic chromosome number

of the genus is usually considered to be 12 (23), with the 26 (36) North

American species having one of 3 ploidy levels, 2x=24, 4x-48, or 6x=72.

Commercial blueberry production began in Florida with the

transplanting of wild plants to cultivated conditions in the late 1800's

(15). Controlled hybridizations were started in the early 1900's with

the first cultivars from controlled crosses being released in 1920 (69).

Breeding efforts have made considerable progress since these initial

crosses (21, 39) and several scientists are currently involved with

blueberry improvement.

Interspecific hybridization in Vaccinium has been responsible for

evolution of the 4x and 6x complexes (39). It has also formed the basis

for improvement in most breeding programs (70). The first hybrid

cultivars were a result of crosses between V. corymbosum L. and V.

australe Small (19). Coville (19) recognized the importance of

considering chromosome number when hybridizing. Others have

substantiated his observation, and it is generally accepted that only

weak crossing barriers exist between most Vaccinium species of equal

ploidy. Crosses between species of unequal polidy are generally more

difficult, with the resulting progeny having lowered fertility (19, 22).

All cultivated blueberries are either 4x or 6x. The 4x highbush

complex (V. corymbosum and V. australe) is cultivated from North

Carolina northward, while the 6x rabbiteye (V. ashei Reade) is limited

to North Carolina southward. Traits of horticultural merit exist at all

3 ploidy levels, with each level having beneficial characteristics not

found at the others (69). Because of this, interploid crosses have been

used in breeding despite the known low success rate.

The interspecific cross of greatest commercial interest would be

highbush x rabbiteye. This cross could combine low chilling require-

ment, vigor, high yield, and tolerance to drought, disease and heat of

rabbiteye with early ripening and high fruit quality of highbush

(11, 24). These crosses have been made and are one of the easiest

interploid crosses. However, the progeny are 5x and have reduced

fertility, complicating improvement at this level. Backcrossing to both

highbush and rabbiteye has been done, but most of the progeny are

noticeably lower in vigor than the parental species and the original Fl

population (Lyrene: personal communication). It was therefore

recognized that manipulation of the rabbiteye gene pool to create 4x

plants, or manipulation of highbush to create 6x plants, needed to be

done before the best qualities of each could be combined (24, 26).

Manipulation of the 6x rabbiteye gene pool to create 4x plants has

been pursued through 6x x 2x crosses. V. darrowi Camp, V. tenellum

Aiton, and V. elliottii Chap. have been the primary diploids used

(26, 84). Some success has been attained when using V. darrowi, with

the Florida blueberry breeding program releasing 3 tetraploid cultivars

which are complex hybrids of V. darrowi, V. ashei and V. corymbosum

(85, 86). For the most part the 6x-2x hybrids have been used only to a

limited extent in breeding cultivars.

Manipulation of the highbush to the 6x level is of more recent

interest and involves doubling 4x highbush to 8x through the use of

chemical agents. The 8x is then crossed to highbush, theoretically

creating a 6x plant. Since the genome of the 6x highbush consists

exclusively of highbush chromosomes, these may prove more useful in


Since 8x plants do not naturally occur in Vaccinium, potential

plant vigor and fertility cannot be determined conclusively from

observations on a few individuals. Several 8x clones should be produced

before evaluation of the 8x and the 6x highbush produced from them.

This study describes research aimed at facilitating 4x-6x gene

transfers, including 1) evaluation of the fertility of V. darrowi x V.

ashei hybrids, with meiotic investigation and crossability studies

between the hybrids and 4x highbush; 2) production of 6x highbush-types

from 4x x 8x crosses; 3) meiotic investigation of 8x colchiploids; and

4) development of a rapid and efficient method for producing and

screening 8x breeding lines from 4x plants.



The blueberry (Vaccinium spp.) probably served as a minor food

source for native North American peoples long before its domestication.

There is little doubt of its importance as a food source for wildlife

(especially birds) who aid in seed dispersion of the various species.

Orderly cultivation of blueberries began in Florida (15) with the

transplanting of V. ashei Reade plants dug, often indiscriminately from

the woods of northwest Florida (65). When these plants came into

production they averaged low in yield and fruit quality. During

the 1940's improved selections of the northern highbush blueberry

(V. corymbosum L. and V. australe Small) were widely planted and have

become commercially successful.

Since realization of their commercial potential, blueberries have

rapidly become one of the major temperate fruits in North America (39).

Research on blueberries has largely been conducted in the United States

and Canada, but interest is increasing in Europe (81), Australia

(Lyrene: personal communication), and New Zealand (79), as the fruit

and its potential become more familiar. Blueberry research in the

United States is carried on at both USDA and state experiment stations

and much has been determined about its anatomy (36), morphology (36, 88,

97), breeding (21, 39, 69, 70), cultural practices (7, 67, 68, 100),

propagation (67), taxonomy (12, 36, 39, 99, 102), fruit processing (94)

and marketing (78).

The person initially responsible for blueberry improvement was the

USDA scientist F. V. Coville. He determined many of the cultural

requirements and propagation practices necessary for development of the

highbush blueberry industry (17, 18) and started the first breeding

program. He made the first crosses in 1911 and the first highbush

cultivars from controlled hybridization were released in 1920 (39).

Breeding of rabbiteye (V. ashei) and lowbush (V. angusifolium Ait.) has

received less attention, with the first hybrid cultivars released in

1950 (25) and 1975 (1), respectively.

Blueberries belong to the family Ericaceae, genus Vaccinium,

section Cyanococcus and are generally considered to have a base

chromosome number of x=12 (73). The diversity of the genus and rate at

which gene exchange occurs between species have created much taxonomic

controversy. Camp's (12) 1945 taxonomic treatment of the genus is the

most extensive. More modern studies have been done (36, 99, 102) but

are generally not well accepted, or are based largely on Camp's


Camp recognized 24 species in Cyanococcus: 9 diploids, 12 tetra-

ploids, and 3 hexaploids. Polyploid species tend to be more widespread,

better adapted, and of greater horticultural significance than diploids.

Blueberry cultivars are classified commercially into 3 types; lowbush,

highbush, and rabbiteye. Traits of horticultural merit exist in several

other species (39), and interspecific crossing is important in most

breeding programs (70).

Polyploidy and Wide Hybridization

Both wide hybridization and polyploidy play major roles in plant

evolution and crop improvement. The 2 processes are often interrelated:

in some cases polyploidy makes wide hybridization possible; in others

wide hybridization leads to polyploidy (32). Both processes are so

important to the development of modern crops that it would be hard to

imagine the type of plants that would be cultivated if the 2 processes

did not occur.

Stebbins', as cited in Goldblatt (40), estimated that 30-35% of the

angiosperms were polyploid. White's (103) estimate was 40% and Grant's

(42) 47%. Estimates differed primarily due to different ways of defin-

ing polyploidy and different estimation procedures (40). Stebbin's

estimate includes as polyploids those species having gametic chromosome

numbers that are multiples of the basic diploid number of their genus

(intrageneric polyploidy). White's 40% is based on the observation that

even haploid numbers exceed odd by 40% and he assumes that this number

is largely attributable to polyploidy. Grant postulated that species

with haploid numbers over 13 would mainly be polyploid, and those with

13 or less, predominantly diploid. Grant also stated that based on

17,138 species of angiosperms, 43% of the Dicotyledonae and 58% of the

Monocotyledonae were polyploid.

A 1980 discussion (40) suggests that almost all angiosperms with

haploid numbers above 9 and 10 probably have polyploidy in their

evolutionary history and those with 11 or higher almost certainly do.

However, plants of lower haploid numbers may also be derived from

polyploid ancestors. At least 70% to 80% of the monocots may in some

sense be polyploid. Lewis (61), in a similar survey of dicots,

concludes that 70% to 80% of all dicots are polyploid, indicating that

incidence of polyploidy in flowering plants has been underestimated, and

its significance in evolution is greater than previously assumed.

Polyploids are traditionally classified according to their assumed

mode of origin into autoploids or alloploids. These 2 concepts were

first proposed by Kihara and Ono (57). They describe autoploidy as a

doubling of the diploid genome and alloploidy as hybridization followed

by doubling of 2 different haploid genomes. These definitions were

sufficient for their time, but as polyploidy has become better under-

stood the definitions proved simplistic and often misleading, since most

natural polyploids fall somewhere between autoploidy and alloploidy


Polyploidy has been described as being a process rather than an

event (32). The evolutionary and genetic implications of polyploidy are

often discussed (31, 47, 55, 92) but modes of origin are not well

understood. DeWet (32) discussed 3 ways polyploids might originate:

zygotic chromosome doubling, meristematic chromosome doubling, and

gametic chromosome nonreduction.

Zygotic chromosome doubling was first proposed by Winge in 1917

(104). It was his belief that when zygotic chromosomes were suffi-

ciently different they failed to pair and the zygote died. However, if

the chromosomes of the hybrid zygote split longitudinally, each

chromosome had an homologous mate, permitting development of a hybrid

individual with double the compliment of parental chromosomes.

Supporting evidence for this method is lacking, and evidence exists

which contradicts it (32).

Meristematic chromosome doubling as a means of polyploid formation

is better supported by evidence (74) but because spontaneous somatic

chromosome doubling is rare (31), it is probably not a major contributor

to polyploid development.

Gametic chromosome non-reduction is currently thought to be the

mode by which most polyploids have arisen (32, 46). Polyploid formation

by this process involves a failure of chromosome reduction in Meiosis I,

or a failure of cytokinesis in Meiosis II. Occurrence of functional,

unreduced gametes has been reported in several plant families (32), with

2x hybridizations frequently resulting in polyploids (30).

Polyploids derived from non-reduced gametes are more commonly 3x

than 4x. This is because the probability of fertilization of a rare

unreduced female gamete by a rare unreduced male gamete is small. Also

2x pollen does not compete well with haploid pollen (32). The proposed

sequence for polyploidization according to this process is for an

unreduced gamete to be fertilized by a reduced gamete producing a

triploid (2x x x = 3x). The triploid then produces an unreduced gamete

which is fertilized by a normal gamete yielding a tetraploid

(3x x x = 4x). Another possible route is the natural 4x x 2x

interspecific hybridization with a 2n gamete from the 2x parent. This

may be more important than the 3x route since sympatric 4x and 2x

species are often represented by large populations whereas 3x plants are

usually rare. Higher ploidies can then be derived from the 4x plants.

The probability of survival of a newly arisen polyploid is small

and has been compared to that of a newly arisen mutant allele in a

population of self-fertilizing individuals (105). Few of these

polyploids survive beyond one generation and are usually eliminated in

competition with their parents. Success of new polypolids therefore

depends on their ability to compete with their parents in similar

habitats, in which case they will be sympatric, or to exploit areas not

favorable to either parent, exhibiting allopatry. The competition

polyploids face is further enhanced by the fact that they are generally

less fertile than their parents. DeWet (32) states that competition for

habitat has 2 components, ability of newly formed polyploid seedlings to

become established, and ability of these seedlings to produce adapted

offspring. He further states that such competition favors perennials,

since once established they have several years to form desirable gene

combinations capable of competing. Among modern genera, the highest

frequency of polyploids occurs among herbaceous perennials, lowest among

annuals, and woody perennials are intermediate (59).

The reduced fertility of newly produced polyploids compared to

their diploid parents is especially evident among autoploids which are

often characterized by chromosomal pairing irregularities and cytologi-

cally unbalanced gametes (32). Pairing irregularities in autopolyploids

result because formation of trivalents and quadrivalents is now possible

due to the pairing of homoeologous chromosomes. This is not the case in

alloploids since the genomes are dissimilar enough to reduce or elimi-

nate homoeologous chromosome pairing. Fertility of polyploids may be

improved by selection. DeWet (32) reports that Gillie and Randolph

noted a significant reduction of quadrivalents in tetraploid maize after

10 generations of sexual reproduction.

Polyploid plants, if successful, can offer several advantages over

their parental diploid's. Polyploidy can lead to a loss or decrease in

self-incompatability. Thus, some cross-fertilizing plants can become

self-fertilizing, making it possible for a single plant to reproduce

(105). Polyploidy also has the effect of buffering the shock of

absorbing foreign genomes as well as masking deleterious alleles. Also,

species that are genetically isolated as diploids may cross as auto-

tetraploids to produce amphidiploids.

Natural polyploids have been shown in some cases to have a wider

range of adaptability than their parents. Hagberg and Akerberg (44)

report an allotetraploid Galeopsis species having a distribution range 3

to 4 times greater than its parents. An increase in polyploids is seen

with increases in latitude and altitude. This suggests that polyploids

may be better adapted to withstanding severe climates. However, it is

possible that these environments select for characters associated with

polyploidy and not for polyploidy itself (53).

Allopolyploids are unique in that they maintain a state of

permanent heterosis (44), even in self-pollinating species. If strict

preferential pairing occurs in these polyploids heterozygosity is

maintained between the homoeologous chromosomes. Therefore, a plant

such as bread wheat (Triticum aestivum L.), even though inbred,

maintains a large amount of heterosis between the A, B, and D genomes.

Gene exchange through crossing over seldom occurs between homoeologous

genomes. Thus, heterosis is not only beneficial for species survival

but can also be exploited by the breeder.

Polyploids have several characteristics which make them

commercially desirable. The principal characteristics that make them

more attractive are that they generally have larger dimensions and

greater adaptability (105). Larger dimensions include increased size of

flowers, fruits, and seeds as well as roots, stems, and leaves. Larger

plant size may increase food and forage production as well as aesthetic

value. Adaptability is important because it allows the same crop to be

grown over a wide range of environments. Polyploids originating from

domesticated crops may have higher value than domesticated crops that

are polyploids since they may combine the advantage of first being a

desirable crop as a diploid with the advantages of polyploidy (105).

The success and importance of polyploids are indicated by the large

number of cultivated crops which are polyploid. The frequency of

polyploidy seems to be higher in cultivated plants than in wild species.

Also, in general where the species form a polyploid series, the species

with higher chromosome numbers often have the greatest agricultural

importance (43). Many agronomic crops such as wheat, tobacco, sugar

cane, oats, cotton, and alfalfa are polyploids (105). Horticultural

crops such as banana, many Brassicas, blueberries, sweet and Irish

potatoes, and many members of the Rosaceae are important polyploids (20,

105). Readers who wish more information on polyploidy may start with

the 1980 publication, Polyploidy: Biological Relevance, edited by

W.H. Lewis (60).

An evolutionary process that probably equals polyploidy in

importance is wide hybridization. Many cultivated crops are a result of

past wide hybridizations (4). Extensive review articles (43, 82, 91,

98) concerning the subject have been published. The scope of this

review will not be as broad as previous reviews, and readers who have a

special interest in this subject should refer to them.

As mentioned earlier, polyploidy and interspecific hybridization

are often related processes; one is successful because the other also

occurs. Progeny of wide hybridizations are often infertile due to

dissimilar genomes. However, if the genome undergoes a doubling,

fertility can be restored (2, 5, 6, 56, 83). In return, wide

hybridizations promote formation of cytologically unreduced gametes (the

primary cause of polyploid development) because many times the only

functional gametes produced from these hybrids are unreduced (32).

The most familiar examples of wide hybridizations are probably

those of Triticale (2n = 56), an alloploid between bread wheat (2n = 42)

and rye (2n = 14) (72) and Raphinobrassica (2n = 36) an alloploid of

radish (2-n 18) and cabbage (2n = 18) (13). Until recently both were

considered only curiosities and of no practical importance. Raphino-

brassica has been improved to the point of being useful as a forage crop

(66) and Raphinus Brassica crosses have been used by Bannerot et al.

(8) to obtain male-sterile cabbage. Within the last 10 years Triticale

has been improved to where it is commercially grown in Hungary, Spain,

Canada, United States and China (72).

The importance of wide hybridization in plant evolution and

speciation may best be demonstrated by the resynthesis of species from

their parental species. Tetraploid Galeopsis tetrahit was resynthesized

from its two parental species, G. speciosa and G. pubescens. Other

plants that have been resynthesized include wheat, rape, tobacco,

cotton, rutabaga, plum, and tart cherry (8, 13, 44, 77). Origins of the

many polyploid species of Vaccinium have been presumed (12) but actual

proof by resynthesis has not yet been done.

The role of wide hybridization in current breeding programs is

significant. For breeding purposes wide hybridization may be defined as

any hybridization in which the production of viable seed is difficult

when only traditional methods of pollination and seed germination are

used (98). Wide crosses include crosses between species within a genus,

crosses between cultivars or species within a genus that differ in

ploidy, crosses between cultivars or species from different genera

within a family, and crosses between cultivars or species from genera in

different families (98).

Uhlinger (98), in a recent review, gave 6 reasons for making wide

hybridizations: 1) they present a challenge; 2) to incorporate known

useful characteristics; 3) to develop new genetic combinations or to

permit expression of latent genes; 4) to broaden the genetic base or

germplasm pool; 5) to assist in taxonomic or phylogenetic studies; and

6) to provide a bridge between incompatible species. He also discusses,

as do others (42, 91), barriers to wide hybridizations and methods to

overcome them.

Wide hybridization as a breeding tool has been used most

successfully in clonally propagated plants (10). Cultivars of many

popular perennial herbs and shrubs (rhododendrons, irises, orchids,

cannas, dahlias, gladioli, roses, poppies, and violets) are of wide

hybrid origin and it is among ornamentals that wide hybridization is

most important (4).

After ornamentals, wide hybridization has been most important among

fruit and nut crops. It has been important in the evolutionary history

of these crops and is important as a breeding method for introgression

of desirable traits from wild species into cultivars (38, 48, 89).

Seed propagated crops, which include most agronomic and vegetable

crops, have also benefitted from wide hybridization. As Briggs and

Knowles (10, pp. 334-335) point out, "Except for hybrid corn, no plant

breeding accomplishment has had more impact on agricultural production

in the United States than the transfer of genes for stem rust resistance

to bread wheat from other species." In seed propagated crops the

process is essentially one of backcrossing, where genes or small

chromosome segments are transferred so that the cultivated parent is

left substantially unaltered except for the desired trait (90). In

asexually propagated crops, the original hybrid or hybrids from

1 or 2 backcross generations may be economically useful.

Blueberries have greatly benefitted from the processes of

polyploidy and wide hybridization, interspecific hybridization in their

case. All cultivated blueberries are either auto- or allopolyploids.

All 15 polypolids present in Cyanococcus are presumed, but unproven, to

have originated from currently existing diploids (39). Wide hybridi-

zations are an important part of current breeding programs (70) since

several noncultivated species possess desirable horticultural traits

(38). Moore (70) lists several interspecific blueberry crosses which

have been made under controlled conditions. Artificial ploidy

manipulations to assist in wide hybridization in blueberries are also of

interest to present day breeders, as will be discussed in the next


Artificial Polyploid Induction

In the early and middle part of this century researchers believed

that if it were possible to polyploidize plants at will, it would

revolutionize breeding methods (28). Because of this thinking great

interest was aroused among plant breeders when Blakeslee announced the

discovery of colchicine as a powerful agent in inducing polyploidy (37).

Within 20 years after this announcement Allard (4) reported that the

idea of revolutionizing breeding methods through polyploidy has been

"thoroughly dispelled." Nevertheless, much literature has been

published concerning colchicine and its effects (27, 28, 37). Even

though artificial polyploid induction has not revolutionized plant

breeding, it remains a useful breeding tool. Other chemical (10) and

physical treatments (4) have been used to induce polyploidy, but because

of its physical properties and mode of action, colchicine is by far the

most effective (4).

Colchicine is an alkaloid derived from the autumn crocus, Colchicum

autumnale L. It appears to be effective only in rapidly dividing cells

and acts by upsetting normal nuclear division. The main action of

colchicine is interruption of spindle fibers so that karyokinesis occurs

but a cell plate is not formed. Therefore, cytoplasm does not divide

and the cell is left with a doubled chromosome number (37). Several

articles have been published concerning techniques of colchicine

application as well as enhancement of its effect (2, 9, 28, 76).

The most valuable use of colchicine has been in restoring fertility

in wide hybrids (2, 5, 6, 55, 83). Genome doubling often results in

homologous pairing and restored fertility. Also, plants that will not

cross as diploids may cross as induced tetraploids (32). Induced

,polyploidy can also be used in studies aimed at determining species

ancestory (32).

Levan (58) listed conditions which lead to successful

autopolyploidization: 1) number of chromosomes should be suboptimal;

2) chromosomes should not be too large; 3) species should be a

cross-fertilizer; and 4) the plant should be grown especially for its

vegetative parts. Zeven (105) listed several successful natural and

artificial polyploids some of which do not meet these conditions.

Even though blueberries do not meet all the conditions for

polyploid induction, induced polyploids have been used in breeding and

meiotic studies. Moore et al. (71) and Jelenkovic and Draper (50)

report on development and potential of a derived decaploid. Auto-

tetraploid blueberries have been produced and utilized by Draper et al.

(35) and Rousi (81). Autoploids are also being used by Lyrene (personal

communication) and his colleagues as a means of gene exchange between

diploids and tetraploids, and tetraploids and hexaploids.

Meiotic Pairing in Blueberries

After surveying existing literature on meiosis of blueberries one

can conclude that for a largely polyploid genus, meiotic pairing is

surprisingly regular. Few pairing abnormalities or fertility problems

are found at natural ploidy levels (16, 52, 62, 97) or in induced poly-

ploids (35, 81). Abnormalities that have been observed generally occur

in interspecific crosses or polyploids and include univalents (49, 97),

laggards (49, 50, 97), unequal chromosome distributions (49), polyspory

(62), and multivalent and secondary associations (16, 51, 57, 97).

The reason for regularity in meiotic pairing seems to be the small

size of the chromosomes. Fully constricted blueberry chromosomes range

from 1.1 to 2.3 microns (45). With frequency of multivalent associa-

tions depending on frequency of chiasmata and chromosome size (51), and

with chromosome size influencing chiasma frequency (87) the small size

of blueberry chromosomes effectively reduces meiotic pairing



(READE) x V. darrowi (CAMP) HYBRIDS


Interspecific hybridization has played an important role in

Vaccinium evolution and in most blueberry breeding programs. Several

cultivars, especially highbush types, have been developed from 1 or more

interspecific crossings. Since there are as many as 19 Vaccinium

species indigenous to eastern North America (12) great potential exists

for interspecific hybridization.

Of the indigenous species described by Camp (12), 8 are diploid

(2n 24), 8 tetraploid (2n 48), and 3 hexaploid (2n = 72). These

ploidy differences present a challenge to blueberry breeders. Several

workers (19, 39) have reported that hybrids between species of equal

chromosome number are generally easier to obtain than hybrids of species

differing in chromosome number.

Cultivated blueberries are primarily tetraploid highbushh complex;

V. corymbosum L. and V. australe Small and lowbush; V. angustifolium

Ait.) or hexaploid (V. ashei Reade). However, desirable horticultural

traits exist at all 3 ploidies, with each level possessing traits not

found at others (69). For this reason interploid crosses are actively

pursued despite low success rates.

Moore (70) lists several successful interspecific crosses made

under controlled conditions. Some of these crosses have been repeated

and progeny evaluated by several blueberry breeders. Intraploid hybrids

led to release of several cultivars (70), but only 3 interploid hybrids

have been released (85, 86).

Since commercial cultivation of blueberries is limited primarily to

tetraploids and hexaploids, crosses between these groups are of greatest

interest, especially between highbush and rabbiteye. Direct hybridiza-

tion between the 2 levels results in pentaploid (2n = 60) plants that

have not proven useful in further breeding. Therefore, 6x highbush-

types or 4x rabbiteye-types have been proposed to facilitate gene

exchange between the 2 groups (24, 26). The procedure that was first

pursued was development of 4x rabbiteye-types from 6x-2x crosses

(24, 26).

Darrow et al. (24, 26) and Sharpe and Sherman (84) have used

various 6x-2x crosses in breeding. Darrow et al. crossed V. ashei with

a southern lowbush 2x species, V. tenellum Ait., to produce 4x

rabbiteye-types which were then crossed to highbush. Tetraploidy of the

hybrids was determined by chromosome counts. Draper et al. (34) and

Sharpe and Sherman (84) both crossed V. ashei with 2x V. darrowi to

develop 4x plants. However, neither group confirmed that their hybrids

were 4x, and Draper et al. stated that their hybrids were sterile and

possibly 5x. Sharpe and Sherman continued to use their hybrids assuming

tetraploidy, eventually releasing 3 cultivars (85, 86). The purpose of

this section was to evaluate fertility and breeding potential of 2 V.

ashei x V. darrowi clones. Investigations included meiotic observations

as well as observations on the crossability between 4x highbush and

V. ashei x V. darrowi hybrids. Pollen tube mitosis was also observed to

determine ploidy levels of germinating pollen grains.

Materials and Methods

In February 1979 approximately 2,000 pollinations of V. ashei

were done with a mixture of half V. darrowi and half V. ashei pollen.

From these mentor pollinations approximately 3,000 seedlings were grown.

From these seedlings 2 V. ashei x V. darrowi hybrids were selected,

'Fla. 81-97' ('Powderblue' x V. darrowi) and 'Fla. 81-121' ('Climax' x

V. darrowi). Both had small leaves and were partially evergreen. None

of the other seedlings appeared to have any traits from V. darrowi and

were probably not V. ashei x V. darrowi hybrids. The 2 hybrids were

removed from the field, put in cold (70C) storage for 1 month and placed

in the greenhouse for crossing in February 1982.

To study meiosis of pollen mother cells, developing flower buds

were removed and fixed in 3:1 absolute ethanol:glacial acetic acid for

24 hours and then stored in 70% ethanol until studied. Anthers were

removed from the flowers, squashed in 1% acetocarmine, destined with

45% acetic acid, and observed using phase contrast at 1000X.

Pollen tube mitosis was studied using techniques outlined by

Stushnoff and Feliciano (96) except for an oven temperature of 25C

instead of 20*C. Percent pollen germination was obtained using the

methods of Goldy and Lyrene (41).

Indications of fertility and crossability were obtained by

comparing percent fruit set, seeds per fruit, and number of seedlings

per pollination. Hybrids were used as both females and males. To test

female fertility hybrids were reciprocally crossed and pollinated with

pollen from a highbush composite. Male fertility was tested by

pollinating 2 highbush clones with each hybrid. Data were analyzed by

chi-square (93) for percent fruit set and by t test (93) for seeds per


'Fla. 78-14', a 13 year old ramet of a V. darrowi x V. ashei

(2x x 6x) hybrid, was included in meiotic analysis but not in the

fertility study.

Results and Discussions

Meiotic analysis of the 3 V. ashei x V. darrowi hybrids revealed

several abnormalities, one of the most interesting of which was their

ploidy level. All 3 of the hybrids were 5x (2n = 60), not the expected

4x (2n = 48). They apparently resulted from fusion of an unreduced

V. darrowi gamete with a normally reduced V. ashei gamete. V. darrowi

has been shown to produce a significant amount of unreduced gametes

(16), but why no 4x hybrids were recovered is unclear. It may be that

2 V. darrowi genomes are necessary for intermediate phenotypic

expression. One genome may not be enough to make 4x hybrids

distinguishable from V. ashei x V. ashei hybrids. However, it is

possible that haploid V. darrowi pollen cannot successfully fertilize

V. ashei. Wherever V. ashei x V. darrowi hybrids appear in the

literature their ploidy level is either not discussed or is uncertain.

The only 6x-2x hybridizations that have produced documented tetraploids

are V. ashei x V. tenellum (26).

Other abnormalities in meiosis of the 3 hybrids, in their order of

occurrence were as follows.

Two Synezetic Knots and Nucleolar Organizing Regions (NOR), Figure 1-a

During meiotic prophase blueberry chromosomes form a structure

termed a synezetic knot consisting of condensing chromosomes and a

single NOR. Several meiotic figures were observed having 2 synezetic

knots and 2 NOR's. It appears that the V. darrowi and V. ashei genomes

are in some cases acting independently of each other. Outcome of this

type of action could be unequal numbers of chromosomes in the gametes

and/or polyspory.

Lagging Chromosomes at Anaphase I, Figure 1-b

Lagging chromosomes were evident in several Anaphase I structures.

Previous meiotic investigations of 5x blueberries have also revealed

lagging chromosomes during Anaphase I (50). However, previous 5x plants

resulted from 6x-4x crosses not 6x-2x. Anaphase I observations of 6x-4x

hybrids revealed 24 bivalents and 12 lagging univalents (50). This

indicates that either 2 genomes from the 4x parent pair with each other,

as do 2 from the 6x parent, leaving one 6x genome unpaired or that

2 genomes from the 4x parent pair with 2 homoeologous 6x parent genomes,

again leaving the 6x genome unpaired. In none of the Anaphase I

observations from these 3 hybrids were 12 univalents seen and 6 was the

highest number of laggards observed, as shown in Figure 1-b. Chromosome

pairing during meiosis of these hybrids should theoretically be similar

to pairing in 6x-4x pentaploids. V. darrowi genomes should either pair

with each other as should V. ashei genomes or 2 V. darrowi genomes pair

with 2 V. ashei genomes. Either case leaves 1 odd V. ashei genome.

With different species combinations it is difficult to determine how

Meiotic irregularities observed in pollen mother
cells of V. ashei x V. darrowi hybrids.
1-a. Leptotene showing 2 synezetic knots and
2 NOR's (arrows). 1-b. Lagging bivalents during
Telophase I. 1-c. Six nonassociating chromosomes
(arrows) during Telophase I. 1-d. Nonsynchrony
during Meiosis II. 1-e. Three areas of DNA at
Telophase I, 2 continuing with Meiosis II, the
other apparently through, 1-f. Nonparrallel
spindles during Anaphase II.

Figure 1.




pairing is actually occurring. Explaining 6 lagging bivalents, however,

is more difficult than explaining 12 lagging univalents.

The most interesting explanation is that homology between the

12 extra chromosomes is great enough to allow pairing. This theory

would imply that the base chromosome number of the genus is 6, not 12.

The theory also raises the question of why most previous researchers

have seen 12 univalents rather than 6 bivalents.

Stating that Vaccinium has a base number of 6 is contrary to

60 years of accepted blueberry cytogenetics. However, a search of the

literature reveals considerable evidence indicting a base number of 6.

Epacridaceae, a family closely related to Ericaceae, or blueberry

family, does have members with a base number of 6 and some taxonomists

believe that the families separated as a result of an ancient

polyploidization giving rise to the Ericaceae (80). Hall and Galetta

(45), in studying the blueberry genome, found 2 long, 2 short and

8 intermediate chromosomes. Members of each group were largely

indistinguishable from one another and the base genome could possibly be

1 long, 1 short, and 4 intermediate. Newcomer (73) and Ahokas (3)

suggested from their studies that blueberries may be secondary

polyploids. Lastly students of polyploid evolution have generally

believed that species with a base number of 10 or higher probably have a

polyploid ancestry (40). The findings of this study help to further

substantiate this theory. Definitive proof that x = 6 in Vaccinium

could be obtained if haploids obtained from a diploid species had

bivalent pairing.

The reason that 6 bivalents are observed in these plants rather

than 12 univalents may be because of the unique relationship of the

species involved. Previous 5x blueberries were hybrids of V. corymbosum

L. x V. ashei and not V. ashei x V. darrowi. If homology exists between

V. darrowi and V. ashei genomes, it is possible to get pairing between

chromosomes of the respective species, and the 6 lagging bivalents may

be 6 V. ashei and 6 V. darrowi chromosomes and not 12 V. ashei.

Nonassociating Chromosomes at Telophase I, Figure 1-c

During Telophase I chromosomes generally form a tight group prior

to Metaphase II. Frequently the lagging chromosomes observed in

Anaphase I do not associate with the other chromosomes. The 6 non-

associating bivalents seen at each pole in Figure 1-c are the most ever

seen, but fewer than 6 have been observed. Whether or not these

chromosomes are included in the end products of meiosis would depend on

whether or not they continue to lag and/or where pollen walls form.

Nonsynchrony of Meiosis II, Figure 1-d

Several meiotic cells were observed where 1 end product of Meiosis

I was well into Meiosis II while the other had not begun Meiosis II, or

as shown in Figure 1-d, 1 set of chromosomes is uncoiling, apparently

indicating a failure of Meiosis II to occur. The latter situation would

lead to 2n gamete production which will be discussed later.

More than 2 Nuclei at the End of Meiosis I, Figure 1-e

More than 2 nuclei were observed in some Telophase I products.

Figure 1-e shows 1 nuclei with uncoiling chromosomes, indicating that

they are through dividing, and 2 areas beginning Anaphase II as

indicated by the presence of the spindle fibers. This situation again

could lead to unreduced gametes and/or polyspory.

Nonparallel Spindles at Anaphase II, Figure 1-f

Spindle fibers of Anaphase II normally align parallel to each other

and perpendicular to the Anaphase I spindle. This results in the

reduced/duplicated chromosomes going to their respective quadrants.

Nonparallel Anaphase II spindles are regularly observed as in

Figure 1-f. Sometimes they are end to end instead of side by side.

When nonparallel spindles occur, even though Meiosis II proceeds

normally, pollen wall formation may result in 2 sets of chromosomes

being included in the same microspore, again producing unreduced


Lagging Chromosomes at Anaphase II, Figure 2-a

The laggards observed at Anaphase and Telophase I continue to lag

as shown in Figure 2-a.

More than 4 Nuclei at Telophase II, Figure 2-b

The situation observed in Figure 2-b most likely resulted from that

in Figure 1-e. Ploidy level of the resulting pollen grains would vary

depending on how many nuclei were included. The pollen could either be

unreduced, or more than 4 pollen grains could result. The different

sizes of the nuclei in Figure 2-a indicate an unequal distribution of

chromosomes, something observed in the hybrids even when the normal

4 nuclei are present.

Meiotic irregularities observed in pollen mother
cells of V. ashei x V. darrowi hybrids.
2-a. Lagging chromosomes at Telophase II.
2-b. Six nuclei at Telophase II. 2-c. Extra
nucleoli at Telophase II.

Figure 2.


* *

Extra Nucleoli at Telophase II, Figure 2-c

Normal blueberry meiosis results in the formation of 4 nuclei, each

containing a single nucleolus. In some cases pollen nuclei from these

hybrids are observed having 2 nucleoli as in Figure 2-c. Figure 2-c

shows the normal 4 nuclei, 2 having a single nucleolus and 2 having

2 nucleoli each. Size of the nucleoli appears to vary with number; if a

single nucleolus is present, it is generally larger than if there are 2.

The origin of these 2 nucleoli may be connected to the condition

shown in Figure 1-a. Since nucleoli and nucleolar organizing regions

are closely associated, cells having 2 NOR's at Prophase could produce

pollen with 2 NOR's.

Unreduced Gametes, Figure 3-a

Unreduced gametes have been predicted for these plants as a result

of the meiotic figures shown in Figures 1-d, 1-e, 1-f and 2-b and as

Figure 3-a shows, they can be found. Frequency of unreduced gametes is

very high in these hybrids, with 25% of the sporads from 'Fla. 80-121'

containing unreduced gametes compared to 5% for 'Fla. 81-97' and 0 and

2% for V. ashei and V. darrowi, respectively (Table 1). Chromosome

number of these "unreduced" gametes could vary depending on location and

time of pollen wall formation. If walls formed prior to inclusion of

lagging chromosomes they could have 54 chromosomes. If wall formation

occurred after inclusion, they could have up to 60 chromosomes.

Mature Pollen Abnormalities; Polyspory, Incomplete Tetrads,
Figures 3-b and 3-c

From Figures 1-e and 2-b pentad and hexad sporads were predicted.

Pentad sporads were observed in 'Fla. 81-97' but not at the frequency in

Irregularities observed in pollen ov V. ashei x
V. darrowi hybrids. 3-a. A dyad containing
2 unreduced gametes. 3-b. Pentad sporad,
difference in size indicates a difference in
chromosome number. 3-c. SEM of pollen showing
shrunken, poorly filled tetrads.

Figure 3.







> 4)


o c

0 C


c w

0 m








4 4






4 0 0 0

-4 fid

&2 r-4

Ch -n -4 ca
CM. CM~ 0 Cen
C14 -r co c

CM m3 00 00
r-. 4 .0 0
-4 -4 CM'


Cu C
Cu '

;jl ;1









0 -

N1 fr*. 5

in in 0 CM

c4 -: 0 .o
-T CM4 -4

00 CM, -4 o
%0o cn r- V)

14 -4
"0r ON

which structures as in Figure 8 were observed. Instead of the expected

increase in polyspory, 'Fla. 81-121' had an increase in the number of

monads, dyads and triads, with many containing unreduced gametes.

Incomplete sporads probably result from a failure of pollen walls to

form around a critical amount of DNA, producing pollen that will abort

or be nonviable.

Much of the pollen formed by the hybrids was shrunken (Figure 3-b)

and not well filled. When pollen was shed it was not fine and granular,

but was released in groups that stuck together. Spherical bodies of

unknown origin were observed on the surface of the pollen and may be

responsible for the stickiness. They have been observed on pollen from

other interspecific blueberry hybrids but not pollen from normal plants.

Pollen germination was 31% and 8% for 'Fla. 81-97' and 'Fla. 81-121',


Crossing analysis found 'Fla. 81-97' to be more fertile than 'Fla.

81-121' both as a female and a male (Table 2). This is not surprising

considering the larger number of meiotic problems that occur in 'Fla.

81-121' as is shown by lower pollen germination and number of complete

tetrads. Meiotic problems observed in PMC's could also be occurring in

megaspore mother cells leading to reduced egg fertility.

Chi square analysis revealed that the 2 hybrids differed

significantly in both male and female fertility. When pollinated with a

highbush composite, 'Fla. 81-97' had 63% fruit set compared to 38% for

'Fla. 81-121'. When used as the pollen parent, 'Fla. 81-97' set 83% on

'Sharpblue', 'Fla. 81-121' set only 14%.

0 0% -- T 'l
00 0 0 I
. I
0 0 0 0

m o

r .









0 0

I m u
I -t

0 0 0 U 0 0 0
0 0 i -4 O I 0 0
c* *
-4 -4 -M -4

(n -4 00 '0 0 CV) IT
'.0 c 00 -4-














4 CM ,-

o 0 o

oo oo

r-4 -4
. 34o









0 CO


u O
cd .C
0 N

0 o

N >,

CM U, 1 CM 00 )
0 N. T- 0 r-a
4 r1 -4 -4

IC M I 0
00 CN C I4




A t test was performed on number of plump seeds per fruit data and

revealed a significant female difference but not a significant male

difference. When pollinated with the hybrids, 'Sharpblue' had fruit

containing an average of 15 and 5 seeds per fruit for 'Fla. 81-97' and

'Fla. 81-121', respectively. These values were not significantly

different at the 5% level. This lack of difference may be due to the

small sample sizes (6 for 'Fla. 81-121' and 23 for 'Fla. 81-97') and the

large variance values. Seed number per berry on 'Sharpblue' ranged from

1 to 36 seeds when 'Fla. 81-97' was the pollen parent and from 1 to 11

when 'Fla. 81-121' was the pollen parent.

'Fla. 81-97' gave more seedlings per pollination, both as a male

and as a female, than did 'Fla. 81-121'. As a female parent 'Fla.

81-97' produced 0.8 seedlings per pollination compared to 0.14 for 'Fla.

81-121'. As a pollen parent 'Fla. 81-97' produced 0.46 seedings per

pollination and 'Fla. 81-121' produced 0.12.

Seedlings from the crosses in Table 2 could have a range in

chromosome numbers. In hybrid x hybrid crosses it is possible that

2 unreduced gametes could combine giving 10x (2n = 120) plants.

However, if the 12 extra chromosomes are eliminated and 24 chromosome

gametes are produced, seedlings could have 48 (24 + 24) or 84 (24 + 60)

chromosomes depending on whether they combine with a reduced or

unreduced gamete. Also if the 6 laggards that are observed are lost,

"unreduced" gametes may only have 54 chromosomes, in which case a

78 (24 + 54) chromosome plant could be produced. Mitotic chromosome

analysis of these populations should be done to determine if they do

range in ploidy.

Analysis of pollen tube mitosis was unsuccessful in determining

ploidy of germinating pollen. The pollen of the hybrids germinated over

a long period of time, beginning after 2 hours and continuing to

20 hours after placement on the media. For doing mitotic studies it is

necessary to have a large number of cells undergoing division, something

not occurring in these hybrids since germination is not synchronized.

Usefulness of 6x-2x hybrids in transferring genes into cultivars is

something that has already been proven, since 3 cultivars exist that

have such an ancestry. Since these 3 cultivars are all 4x, the

pentaploidy of the initial hybrids must have been reduced with further

breeding. The 3 cultivars have not been well accepted by commercial

blueberry growers, mostly because of their susceptability to

Phytophthora cinnamomi (Rands), a root disease to which V. ashei

is tolerant. Since most V. ashei clones are highly tolerant to

P. cinnamomi, this approach to combining rabbiteye and highbush gene

pools should not be abandoned. However, for it to be successful,

breeders should concentrate on selecting hybrids possessing the good

traits of the species involved. This may not be easy due to the

difficulty in obtaining large numbers of 6x-2x hybrids.

Since the 2 hybrids in this study exhibited a difference in

fertility, to insure continued success in breeding, only those plants

such as 'Fla. 81-97' should be selected. 'Fla. 81-97' is not only more

fertile than 'Fla. 81-121', but also has larger, more attractive fruit,

'Fla. 81-97' also sets a large number of fruit for an interspecific, 5x

hybrid. Both plants, however, are interesting from a cytogenetic

standpoint because of their large number of meiotic problems.




Since 4x-6x Vaccinium crosses yield pentaploids of little breeding

value, previous efforts to combine 4x highbush (V. corymbosum L. and

V. australe Small) and 6x rabbiteye (V. ashei Reade) have been designed

to reduce the rabbiteye genome to 4x through 6x-2x crosses (24, 26, 84).

This technique incorporates genes of an unimproved diploid into the

breeding lines and a pure highbush-rabbiteye hybrid is not obtained.

Another approach to highbush-rabbiteye hybridization would be to

raise highbush to the 6x level. This could be done by doubling the

4x genome to 8x and then backcrossing to 4x, theoretically producing

6x plants of entirely highbush genes, which could then be crossed to the

6x rabbiteye. This technique has the advantage that the resulting

highbush-rabbiteye hybrids would only contain the genetic information

from the 2 gene pools. Also, 6 homoeologous genomes offer more

potential for heterosis than 4. One problem that may arise using this

method is the existence in blueberries of a 3x block, which keeps

3x plants from being obtained from 2x-4x crosses. Some researchers

theorize that 3x plants fail to form because of unfavorable genomic

ratios present in 2x-4x hybridizations (54). The genomic ratios of

4x-8x are the same as those of 2x-4x crosses and a 6x block analagous to

the 3x block could exist.

This study was conducted to determine the feasibility of using

8x colchiploids to obtain hybrids between highbush and rabbiteye.

Investigations included 1) meiotic analysis of the 8x colchiploid;

2) observation of pollen tube mitosis to determine ploidy of germinating

pollen; and 3) generation of synthetic 6x highbush plants through

4x-8x crosses.

Materials and Methods

Plant material consisted of a single 8x plant ('Fla 80-46')

produced by Chandler (14) from 120 colchicine treated seedlings of a

4x, 'Fla 3-8' x V. fuscatum Ait. cross. 'Fla 80-46' was selected as

having larger than normal guard cells and pollen tetrads, indicating

that at least the epidermal and gametic tissues were doubled. The plant

is also characterized by slow growth and small, misshapen leaves that

bear trichromes and often form rosettes.

For study 'Fla. 80-46' was removed from the field in December 1980,

put in cold (7C) storage for 1 month, and then placed in the green-

house. It produced 50 flower buds in 1981, all of which were used for

crossing. In December 1981 it was again placed in cold storage for

1 month, and it was moved to the greenhouse in January 1982. The plant

produced 11 flower buds in 1982 which were used for meiotic analysis.

Further meiotic, pollen mitotic, and crossing studies were done in 1983.

By this time 8 ramets taken in 1980 came into flowering, and limited

flower supply was no longer a problem.

For meiotic analysis, flower buds at the appropriate state were

removed and fixed in 3:1 absolute ethanol:glacial acetic acid for

24 hours and stored until needed in 70% ethanol. Anthers were removed

and squashed in 1% acetocarmine, destined with 45% acetic acid and

observed using phase contrast at 1000X.

Pollen tube mitosis was studied using techniques outlined by

Stushnoff and Feliciano (96), except for using a 250C oven instead of

200C. Percent pollen germination was obtained using the method of Goldy

and Lyrene (41).

For fertility and crossability analysis five 4x plants were

pollinated with pollen from 'Fla. 80-46' and 'Sharpblue'. 'Sharpblue'

was pollinated with pollen from 'Fla. 80-46' and 'Avonblue'. Parameters

analyzed were 1) fruit set; 2) fruit weight; 3) number of well developed

seeds/fruit; and 4) number of seedlings/pollination. A 2-way analysis

of variance (93) was performed on each parameter and an F value was used

to test differences for significance.

Seeds from the crosses were removed from the fruit, disinfested

with 30% Clorox (1.6% sodium hypochlorite) for 15 minutes, rinsed twice

in sterile distilled water and placed in 50ml screw top vials containing

5ml of 0.45% agar. The agar had been autoclaved at 1.05 kg/cm2 for

15 minutes. Vials were placed on a window sill where they received

several hours of full sunlight each day. After germination, seedlings

were removed and rooted under mist in 100% peat. Root or shoot tips

from the seedlings were fixed and stored as described for flower buds.

In preparation for making chromosome counts, the tips were placed for

15 minutes in a solution of 0.03g pectinase and 0.03g cellulysin dis-

solved in 2ml water. After softening they were rinsed with water and

squashed in 1% acetocarmine. Slides were destined with 45% acetic acid

and observed using phase contrast at 1000X.

To test female fertility 10 flowers of 'Fla. 80-46' were selfed in

1981 and 10 were pollinated with pollen from 4x plants in 1981 and 1982.

Female fertility was again tested in 1983 using a composite of V.

elliottii Chap., V. corymbosum, V. ashei and 'Fla. 80-46' pollen.

Results and Discussion

Meiosis appeared relatively normal in 'Fla. 80-46' despite the

increased chromosome number. Anaphase I figures were observed with the

expected 48 chromosomes at each pole (Figure 4-a). Small amounts of

lagging DNA were occasionally observed at Anaphase II, but meiosis for

the most part appeared to proceed normally. Chromosomal connections

evident in Figure 4-a were often seen in meiosis of 'Fla. 80-46' and

have been reported in other plants (101). Whether or not they affect

fertility is unknown.

One abnormality evident in some Telophase II structures was

6 nuclei instead of the normal 4 (Figure 4-b). The 2 extra nuclei

always appeared smaller and may lead to the formation of pentads and

hexads instead of normal tetrads. The fact that they were smaller than

normal indicates that an extra reduction of some sort may have occurred.

Scanning electron micrographs (Figure 4-c) of pollen tetrads

revealed pollen walls that were somewhat concave, had a large plug in

each pore, and showed an increase in diameter of approximately 24% over

tetrads from tetraploids. Pollen germination was found to be 26%.

Pollen of 'Fla. 80-46' is sticky and released in groups. The

spherical bodies apparent on the pollen in Figure 4-b may be the reason

for the stickiness. These bodies are also observed in interspecific

crosses whose pollen is also sticky.

Meiosis and meiotic products of an 8x colchiploid
highbush blueberry. 4-a. Anaphase I showing
48 chromosomes at each pole and interchromosomal
connections. 4-b. Telophase II showing 6 nuclei.
4-c. SEM of tetrads showing pollen that is slightly
concave in the pore area and with each pore having
a large plug.

Figure 4.




Analysis of crossing data showed that 4x highbush plants were

significantly more fertile when pollinated with pollen from other

4x highbush than they were with pollen from 'Fla. 80-46', no matter what

fertility parameter was used (Table 3). Average fruit sets on 4x plants

pollinated with 8x and 4x pollen were, respectively, 12.5 vs. 72.4%,

average fruit weight was 1.12g vs. 1.68g, mean number of seeds per fruit

was 2.2 vs. 19.8 and the best indicator of fertility, seedlings per

pollination, was 0.15 vs. 2.07.

Although 125 seedlings were obtained from 735 highbush flowers

pollinated by 'Fla. 80-46', only 1 seedling had the expected chromosome

number of 72, and it was mitotically unstable. Mitotic chromosome

counts of shoot tips ranged from 48 to 168. The majority of the cells

however contained 72 chromosomes. The other 124 seedlings had

48 somatic chromosomes.

The simplest explanation for the origin of the 4x seedlings is that

they resulted from self pollinations. This may or may not have been the

case. Several factors from the 4x-8x pollinations indicate that they

may be hybrids rather than selfs. Average fruit weight was lower and

ripening time peaked 2 weeks later for 'Fla. 80-46' than for control

pollination. Smaller fruit and later ripening are characteristic of

both interploid crosses and selfs because of their lower seed set (18).

The strongest indication that the seedlings were hybrids was the fact

that seedling vigor appeared too great for selfs and that when planted

in the field, they showed traits characteristic of the pollen parent.

The question then arises as to how these "hybrids" may have formed.

As stated earlier some meiotic figures were observed with 6 instead of

4 nuclei in Telophase II. If the theory of a further reduction is



0 0
o.H o


(U i-
)W 0



*0 0


41 0,


3 U



z 44 0
0 -li-

0 0
-* -*


o -4

\o cn
* "0
-4 -t


'-> 0
* 0

-* 0

o oO








-4 xl






0 -at
SI *-4
00 .
.0 C) 0.
.4 rZ' W
o ou Cu
-4 0
g ^ )

o 0 0 0 -4

o o -r o o -4 r- 0 '0
. *
o o o 0 C O4 0 O

V- eCN^ 0 --T --T -3 e

o 0

-* .0 0 Um tN '. ^-. 0% '.0 -3

DO O U O U O .0 IO

o0 gO 00 0. oo (
CUI co -ca 0c
CB Cu Cu Cu Cu 0 Cu Cu *
-- -- > a-4
U) U) r 4 4 U) r- 4 .
-rnun FZ U

correct, these smaller nuclei may contain the previously normal number

of 24 chromosomes instead of the desired 48. Also, if a 6x block does

exist in these crosses, the screen for 24 chromosome gametes in 4x-8x

crosses might be as powerful as the screen for unreduced pollen in 2x-4x

crosses. Two tests were performed to determine the ploidy of germi-

nating pollen: analysis of pollen tube mitosis and pollinating

V. myrsinites Lam. with 'Fla. 80-46' pollen.

Mitotic analysis of pollen tubes was difficult using the techniques

of Stushnoff and Feliciano (96) and proved unsuccessful after repeated

attempts. Handling and observation of pollen tubes were difficult, and

when mitotic figures were observed, it was impossible to get accurate

counts because space limitations imposed by the pollen tubes resulted in

crowded, poorly spread chromosomes.

V. myrsinites was pollinated so that hybrid seedlings could be

phenotypically identified. V. myrsinites is a 4x lowbush blueberry with

small evergreen leaves. Tetraploid hybrids between 4x highbush and

'Fla. 80-46' would be difficult to determine phenotypically, but hybrids

between V. myrsinites and 'Fla. 80-46' should be readily distinguishable

from V. myrsinites selfs. Chromosome counts could then be performed on

the hybrids to determine if they are 4x or 6x. The crosses onto V.

myrsinites were done in February 1983 and results are not yet available.

The 6x hybrid obtained by pollinating 'Flordablue' with 'Fla.

80-46' resembled the pollen parent in several characteristics and was

definitely a true hybrid. As in the pollen parent the leaves are small,

misshapen and trichomatous. It was also slow growing and not nearly as

vigorous as other seedlings from 'Fla. 80-46' pollinations. As stated

earlier the plant is mitotically unstable with mitotic cells containing

48 to 168 chromosomes. This is a phenomenon that has been observed by

Lyrene (personal communication) in previously studied interploid

hybrids. However, since most cells contained 72 chromosomes the plant

may breed as a 6x. The mitotic instability that is observed in the

plant may be useful for developing breeding lines at other ploidy

levels. By putting the plant in tissue culture, it might be possible to

get a series of adventitious shoots with stable chromosome numbers

ranging from 4x to 14x. This opens the possibility of exploiting

previously nonexisting ploidy levels and making new interploid


Female fertility was not tested as extensively as male fertility

due to the small number of flowers. Because of the large number of

pollen grains compared to eggs in each flower, the most efficient use of

gametes is as males. Number of flowers available for crossing in 1982

and 1983 was further reduced by meiotic analysis.

Female fertility of 'Fla. 80-46' was not high. No fruit was set

from the self and 4x pollinations in 1981 and 1982. The composite cross

in 1983 set 2 berries from 15 pollinated flowers. Female fertility of

other colchiploids has been reported as questionable (81). The flowers

of 'Fla. 80-46' appear to produce the normal number of eggs and the

reason for low fruit set is unknown.

Usefulness of 8x plants and hybrids of 4x-8x crosses is difficult

to determine from a single plant. Although performance of 'Fla 80-46',

phenotypically and in production of 6x hybrids, was not as good as

desired, other 8x plants could perform better. Since 8x and 6x plants

are to be used only as bridges between highbush and rabbiteye, they need

not have the qualities of cultivars. Development of more 8x colchi-

ploids (as will be discussed in Section V) can play an important role in

improving this crossing technique. When more 8x plants become available

they can be intercrossed and selected for increased vigor and fertility.

This is also true for any subsequent 6x hybrids from 4x-8x crosses.

Usefulness of the 6x hybrid has yet to be determined since it has yet to

flower. It is hoped that it will produce at least some 3x gametes, but

with the mitotic instability it exhibits, this cannot be safely assumed.

With potential to do breeding at these 2 new ploidy levels it is

possible that highbush itself could be improved, given enough time and

concentrated effort. Development of more 8x and 6x lines containing

only highbush genes may pave the way for practical gene transfer from

4x, highbush and 6x rabbiteye.




The ability to artificially induce polyploidy in plants has long

been a desire of plant breeders who believed that if plants could be

polyploidized at will, breeding methods would be revolutionized (28).

Great interest was therefore aroused when colchicine, an alkaloid from

autumn crocus (Colchicum autumnale L.), was discovered to be a powerful

agent in inducing polyploidy. Although plant breeding was not

revolutionized, the ability to induce polyploidy remains an important

tool of plant breeders.

Woody species are more recalcitrant to the effects of colchicine

than are herbaceous species (29). One problem arising from treatment of

woody plants is the large number of cytochimeras which develop (29).

Induced polyploidy is often difficult to detect in early stages of

post-treatment growth. Also polyploid tissue often competes poorly with

normal tissue and becomes overgrown by normal tissue.

Due to poor competing ability of polyploid tissue it is necessary

to provide the most favorable environment possible for treatment,

growth,and selection. In vitro tissue culture has therefore been

proposed as a means of reducing stress imposed through colchicine

treatment (65). The purpose of this section is to report on preliminary

experiments leading to development of a rapid, efficient method for

producing and screening 8x blueberry breeding lines from 4x plants.

Many highbush blueberry clones have proven difficult to maintain in

tissue culture. The plants grow poorly, producing slow growing shoots

with minimal proliferation, many tending to be monopodial. Therefore,

before colchicine treatments could be performed it was necessary to

1) determine basic media for best performance; 2) discover media

alterations improving performance; and 3) select highbush clones with

rapid shoot growth and proliferation. Experiments 1 through 4 deal with

establishment of tissue cultures, while 5 through 14 deal with polyploid

induction. Unless otherwise indicated all experiments were conducted in

50ml screw top vials containing 10ml of media.

Materials and Methods of Experiments 1-4

Experiment 1

The objective of this experiment was to select highbush clones

which would grow well on modified Knop's medium (KM). In November 1980

approximately 400 seeds, from a 4x, 'Fla. 79-12' x Sharpblue', cross

were surface disinfested with 30% Clorox (1.6% sodium hypochlorite) for

15 minutes, rinsed twice in sterile, distilled water and placed in

40 vials containing a 0.45% agar solution which had been autoclaved at

1.05 Kg/cm2 for 15 minutes. The vials were then placed on a window sill

to receive several hours of full sunlight. After seedlings had

germinated and produced several leaves they were transferred to KM

(Table 4) containing 5mg/l 2iP (6-gamma-gamma-dimethyl-allyl amino

purine). The cultures were incubated between 22 and 320C with

Table 4

Composition of Modified McCown and Lloyd's Woody Plant
Medium (WPM), Modified Anderson's Rhododendron Medium
(ARM), and Modified Knop's Medium (KM)

Compound Modified WPM Modified ARM Modified KM

NH NO3 400 400 --
KNO3 -- 480 190
K SO 990 -- -

KH2PO4 170 -- 170
Ca(NO3) 4H20 556 -- 1,140
CaC12 2H20 96 440 --
MgSO4 7H20 370 370 370
NaH2PO H20 -- 380 --
Na EDTA 74.5 74.5 74.5
FeSO4 7H20 55.6 55.6 55.6
MnSO H20 22.3 16.9 22.3
ZnSO 7H20 8.6 8.6 8.6
H 3BO3 6.2 6.2 6.2
Na2MoO 2H 20 0.25 0.25 0.25
CuSO 5H20 0.25 0.025 0.025
KI -- 0.83 0.83
CoI2 6H 20 -- 0.025 0.025
Pyridoxine HC1 0.5 -- 0.5
Thiamine HC1 1.0 0.4 0.1
Nicotinic Acid 0.5 0.5
Myo-inositol 100 100 100
Adenine Sulfate -- 80 --
Glycine 2.0 -- 2.0
Casein Hydrolysate 1,000 1,000 1,000
Sucrose 30,000 30,000 30,000
Agar 4,000 4,000 4,000
2ip 10 10 10

pH adjusted to 5.7 with IN NaOH. Autoclaved at 1.05 Kg/cm2 for
15 minutes.

illumination of about IKlx for 16hr/day with Cool-White fluorescent

lamps. Genotypes that grew best were selected based on regeneration

time and vigor.

Experiment 2

This experiment was performed to determine the effect of 3 levels

of 2iP on in vitro shoot proliferation. One liter of KM was divided

into 3 portions. To each portion 2iP was added at the rate of 5, 10, or

20 mg/1. Each treatment was placed in 30 vials, 90 vials total. Three

2-node cuttings from 5 clones selected from Exp. 1 were placed in each

vial. Plants were placed under incubation conditions indicated in

Exp. 1. After 60 days plants were visually rated for shoot


Experiment 3

This experiment was done to determine shoot production potential

from 2 explant sources: hypocotyls and 1cm2 leaf sections placed on KM.

Leaf sections were obtained from 3 greenhouse-grown highbush sources.

Rapidly growing leaves were removed, disinfested in 10% Clorox (0.525%

sodium hypochlorite) for 20 minutes and rinsed in sterile, distilled

water. Sections were cut on all sides. Midrib tissue was excluded.

Hypocotyl sections were taken from a seedling composite obtained by open

pollination of several highbush plants. Disinfestation and germination

were the same as Exp. 1. Fifty vials for each of 2 explant types were

prepared: 1 leaf section per vial or 3 hypocotyls per vial. Explants

were incubated as for Exp. 1.

Experiment 4

This experiment was done to compare the effects of in vitro

multiplication of 3 growth media: Modified Knop's Medium (KM), modified

Anderson's Rhododendron Medium (ARM), and modified McCown and Lloyd's

Woody Plant Medium (WPM). Composition and preparation of media are the

same as that listed in Table 4. Each treatment was placed in 30 vials,

90 vials total. Three 2-node cuttings of 5 clones from Exp. 1 were

placed on the media. Plants were placed under the same incubation

conditions as Exp. 1. After 60 days plants were visually evaluated for

vigor and regeneration.

Results and Discussion of Experiments 1-4

From Exp. 1, 66 seedlings were selected and transferred to KM.

After incubation, 10 were selected for further evaluation and

transferred to fresh media, with a minimum of 10 replications.

Evaluation was repeated after a second 60-day incubation and 6 clones

were selected for superior growth in the shortest amount of time.

The experiment indicated that ability to habituate to in vitro

conditions depended on genotype. However, growth of the selections on

KM was still not as good as desired. Growth was more rapid but shoot

proliferation was minimal and monopodial cultures quickly outgrew the

vials. Since shoot proliferation is dependent on cytokinin activity

(33) and 2iP has proven effective on blueberries (63), Exp. 2 was

conducted to determine if levels other than 5 mg/l could induce more

shoots. A Omg/1 treatment was not included since it was known a priori

that 2iP was needed for shoot proliferation. Also since shoot

proliferation was not good at 5mg/l only higher levels were tested, with

5mg/l acting as the control.

Visual evaluation after 60 days revealed an apparent increase in

shoot proliferation on the 10 and 20mg/l treatments over the 5mg/l.

However, there was no apparent difference between 10 and 20mg/l

treatments. Therefore, 10mg/l 2iP was determined as the best level for

satisfactory shoot proliferation.

Experiment 3 was done to determine if chimerism after colchicine

treatment could be avoided through induction of adventitious buds, but

first it was necessary to determine if adventitious buds could be

obtained in vitro. Since leaf sections have been successful in other

plants and blueberries have the ability to produce adventitious buds

from hypocotyl tissue (75), both tissues were tried.

The 10% Clorox for 20 minutes gave adequate disinfestation of leaf

tissue with minimal damage, but shoot regeneration was never obtained

despite repeated experiments and various 2iP levels. Callus formation

did occur but cultures died after limited growth. Adventitious shoots

do form from juvenile leaf and stem tissue, but this is not helpful when

doubling of cultivars is desired.

Hypocotyl tissue also proved unsuccessful in forming in vitro

adventitious shoots. The tissue quickly entered a callus phase from

which shoots were never obtained. Further experiments with various

hormone additions and levels may be useful in obtaining shoots from such


Experiment 4 revealed that both ARM and WPM were superior to KM

after 60 days incubation. Out of 30 vials in ARM and WPM treatments,

73% and 60%, respectively, were rated as having medium to good growth,

while only 20% of those grown in KM received a similar rating.

As a result of these 4 experiments, 6 superior genotypes were

selected (213, 218, 221, 229, 236, and 238); 10mg/l 2iP proved better

than 5mg/l and as good as 20mg/l; ARM is better than WPM or KM; and

2-node stem cuttings of in vitro grown plant material were found to be

the best explant source. Once these factors were determined it was

possible to begin polyploid induction through colchicine treatment; and

unless otherwise noted the above conditions apply to the following


Materials and Methods of Experiments 5-14

Experiment 5

This experiment was done to determine the effect of colchicine

concentration on highbush blueberry. Colchicine at concentrations of

0.001, 0.010, 0.050, 0.100 and 0.200% was incorporated into solid KM.

Three cuttings of clone MRB (a clone that grows well on KM) were placed

in vials containing 5ml of media, 20 vials per treatment, 100 vials

total. Explants were left on the media for 14 days and then transferred

to fresh KM containing no colchicine. Plants were incubated as in Exp.

1. Cultures were visually evaluated for polyploidy every 30 days for

6 months.

Experiment 6

This experiment was performed to determine optimum exposure times

to 0.010% colchicine in solid media. Three cuttings per vial of clone

221 were placed in vials containing 0.010% colchicine. Explants were

left on the media for 24, 48, 96, 192, or 384 hours. They were then

transferred to fresh colchicine-free media. There were 20 vials per

treatment, 100 vials total. Evaluation was the same as for Exp. 5.

Experiment 7

This experiment was done to observe the effect of 3 colchicine

concentrations on highbush treated in liquid KM. Cuttings of clone MHB

were placed in vials containing 5ml liquid KM into which 0.0, 0.05,

0.10, or 0.20% colchicine had been added. The vials were then placed in

a rotating wheel at 3 rpm for 48 hours. The explants were triple rinsed

in autoclaved, distilled water and planted on fresh media, approximately

10 explants per vial. Evaluation was the same as for Exp. 5.

Experiment 8

The purpose of this experiment was to observe the effect of 0.1%

colchicine on 6 clones of highbush. Several cuttings of clones 213,

218, 221, 236, 238, and MHB were placed in 5ml of liquid ARM containing

0.1% colchicine and placed on a rotating wheel for 24 or 48 hours.

Further handling and evaluation was similar to Exp. 7.

Experiment 9

This experiment was similar to Exp. 8, except for the addition of

0.02g/1 ascorbic acid to determine if tissue browning due to colchicine

treatment could be reduced. Plant materials were clones 218, 236, 238,

and MHB.

Experiment 10

This experiment was an attempt to induce polyploidy using 0.05%

colchicine at exposure times of 72 and 96 hours. Several cuttings of

clones 218, 221, 236, and 238 were placed in 5ml liquid ARM containing

0.05% colchicine. Plants were then placed on the rotating wheel for 72

or 96 hours. Further handling and evaluation was the same as for

Exp. 7.

Experiment 11

The purpose of this experiment was to induce polyploidy using 0.050

and 0.025% colchicine in liquid ARM at exposure times of 24 and

48 hours. Cuttings of clones 218, 221, 236, and 238 were placed in

liquid media containing 0.050 or 0.025% colchicine. Further handling

and evaluation was the same as for Exp. 7.

Experiment 12

This experiment was done to determine if providing axillary buds

with a growth phase prior to colchicine treatment would facilitate

obtaining polyploids. Cuttings of clones 218, 221, 236, and 238 were

placed in vials containing 5ml of media. Vials were then placed on a

rotating wheel for 5 days, removed, and placed in a 0.1% colchicine

solution for 24 or 48 hours. Further handling and evaluation were

similar to Exp. 7.

Experiment 13

This experiment was performed to determine if cold and/or darkness

prior to colchicine treatment would aid in obtaining polyploids. Two

rapidly growing cultures each of clones 218, 221, 229, and 236 were

placed in darkness at 250C or at 4C in a refrigerator for 96 hours.

After pretreatment the plants were placed under normal incubation for

7 hours, cut into 2-node cuttings and treated 24 or 48 hours in 0.025%

colchicine. Handling and evaluation was the same as Exp. 7.

Experiment 14

This experiment combined cold treatment with a pretreatment growth

phase. Cuttings of clones 218, 221, 229, and 238 were placed in

100 x 15mm Petri dishes containing solid ARM.

The dishes were taped shut and placed in a 40C refrigerator for

7 days. The plants were then placed under normal incubation for 24, 48,

96, or 192 hours. After incubation, cuttings were removed and treated

for 24 hours in 5ml liquid ARM containing 0.025% colchicine. Further

handling and evaluation was the same as Exp. 7.

Detection of Polyploids and Ploidy Determinations

Initial detection of polyploidy was done visually according to the

methods of Lyrene and Perry (64). Those shoots which appeared to have

unusually large diameter were selected as "visible polyploids", removed

from culture and recultured on fresh media. After sufficient regrowth,

ploidy level of the plants was determined by removing actively growing

shoot tips, fixing them in 3:1 absolute ethanol:glacial acetic acid for

24 hours and storing until needed in 70% ethanol. For slide prepara-

tion, shoot tips were placed for 20 minutes in a solution of 0.03g

pectinase and 0.03g cellulysin dissolved in 2ml water, rinsed with

water, and squashed and stained in 1% acetocarmine. Slides were

destined with 45% acetic acid and observed using phase contrast at


Results and Discussion of Experiments 5-14

Experiment 5 indicated that colchicine levels between 0.01 and

0.05% were best for treatment. Higher levels usually resulted in

greater shock and death of the explants while lower concentrations had

essentially no effect (Table 5). Ideal concentration would be that

level giving minimum death with maximum polyploid production; therefore

0.01% was chosen for the next experiment.

Since the action of colchicine depends on cell division, maximum

effect occurs at times of rapid mitosis. Exp. 6 was designed to

determine a duration of exposure that was best to insure colchicine

incorporation during cell division. None of the durations were

effective and the explants remained in a nongrowing state when replanted

on colchicine-free medium. Few of them were killed by treatment, but

few of them recovered enough to grow within 6 months after treatment.

No visible polyploids were detected from either Exp. 5 or Exp. 6.

The explants appeared to be severely shocked by the colchicine

treatment; they neither died nor grew well after treatment. Treatment

on solid media, where the plants were not rinsed of residual colchicine,

gave poor results. Treatment in liquid media was therefore suggested.

Experiment 7 tested several procedures for liquid treatment. From

this experiment it was determined that colchicine levels between 0.05

and 0.01% gave the best results (Table 6). Shoot regeneration was

better and treatment shock was reduced, presumably due to rinsing off of

Table 5

Growth of Highbush Blueberry Clone MHB on Colchicine-Free Medium
Following Treatment of Explants with Various Concentrations
of Colchicine for 14 Days on Solid Medium

Number of Vials

% Colchicine Dead Greenz Growing

0.200 5 15 0

0.100 1 19 0

0.050 0 18 2

0.010 3 13 4

0.001 0 3 17

SGreen but lacking signs of growth.

Table 6

Growth of Highbush Blueberry Clone MHB on Colchicine-Free Medium
Following Treatment of Explants with Various Concentrations
of Colchicine for 48 Hours in Liquid Medium

Number of Vials

% Colchicine Dead Greenz Growing

0.20 12 17 6

0.10 2 11 12

0.05 4 16 10

0.00 1 0 20

ZGreen but lacking signs of growth.

residual colchicine. However, this method requires considerably more

handling of the explants introducing a greater possibility of


Experiments 8-11 attempted to further define the liquid treatment

method. Experiment 8 revealed variation among genotypes with respect to

sensitivity to treatment (Table 7) and that generally more shoots were

regenerated from an exposure time of 24 hours than from 48. Addition of

ascorbic acid to the media in Exp. 9 did not appear to reduce tissue

browning. Due to the large amount of death in Exps. 8 and 9 the 0.10%

colchicine treatment was concluded to be too strong. Treatment was

repeated in Exp. 10 using 0.05% colchicine for 72 and 96 hours. This

treatment again proved too harsh and many of the regenerated shoots were

adventitious. Treatment apparently killed the axillary buds in which

colchicine was hoped to have its greatest effect. Shoots arising from

adventitious buds were probably not affected by colchicine, since the

explants were rinsed to remove residual colchicine before adventitious

buds arose. Adventitious buds gave rise to many shoots, none exhibiting

signs of induced polyploidy. Because of these findings lower concen-

trations and/or exposure times were investigated.

Experiment 11 revealed that a 0.050% treatment for both 24 and

48 hours gave good regeneration (up to 60%) but that most of the shoots

were again adventitious. The 0.025% treatment gave lower shoot

regeneration; however, many of the shoots were from axillary buds. The

0.025% treatment gave lower regeneration than 0.050%, probably because

axillary buds grew at this level, suppressing adventitious growth.

Neither the 0.025% nor the 0.050% treatment produced visible polyploids.

Table 7

Performance of 6 Highbush Clones on Colchicine-Free Medium
After 24 and 48 Hour in Vitro 0.1% Colchicine
Treatment in Liquid Arm

Number of Vials

Clone Time(hrs) Dead Growing %Growing

213 24 11 2 15
48 11 1 8
218 24 10 3 23
48 11 2 15
221 24 5 7 58
48 13 4 24
236 24 14 2 12
48 13 6 32
238 24 7 11 61
48 10 4 29
MHB 24 11 0 0
48 11 0 0

Findings of the previous induction experiments led to a dilemma:

when shoots were treated with low concentrations, no change occurred; if

treated with higher concentrations, the tissue was damaged beyond

regeneration. Highbush appears to be more sensitive to colchicine than

other blueberries similarly treated, and at the same time appears to be

more resistant to its chromosome doubling effects. These conclusions

led to Exps. 12-14. Since varying colchicine concentrations and

exposure times had little effect, it was decided that treatments

predisposing the plant material might be beneficial when treating at

lower levels. Because colchicine is effective only in dividing cells,

several treatments to enhance cell division prior to colchicine

treatment were investigated.

In Exp. 12 plants were cut into the usual 2-node cuttings but were

allowed to grow for 5 days on colchicine-free medium. The goal was to

give axillary buds a chance to begin growth, thereby increasing the

mitotic index. A colchicine concentration of 0.10% was used, but again

axillary buds were killed and only adventitious growth occurred.

Pretreatments for Exp. 13 were cold and/or darkness. The idea

behind these treatments was to stop growth for an extended period of

time and then provide optimum growing conditions prior to colchicine

treatment. With the extra handling necessary for pretreatment,

contamination became a serious problem. However the treatments were

successful in inducing polyploid shoots.

From Exp. 13 a total of 14 vials containing visible polyploids were

selected, 7 from the cold/dark treatment and 7 from the darkness

(Table 8). The 2 pretreatments appeared equally effective, with 15% of

the cold/dark treated vials containing visible polyploids and 13% of the

Table 8

Vials Containing Visible Polyploids Following a 96 Hour Darkness or Cold
Darkness Pretreatment, 7 Hours at Normal Incubation Conditions
and a 0.025% Colchicine Treatment for 24 or 48 Hours

Total # Vials with
Pretreatment Clone Time(hrs) of Vials Fat Shoots

darkness 218 24 0 0
48 7 5
221 24 12 0
48 10 0
229 24 7 1
48 6 0
238 24 6 0
48 5 1

total 53 7

cold darkness 218 24 8 2
(70C) 48 4 2
221 24 8 0
48 7 1
229 24 8 1
48 0 0
238 24 7 1
48 6 0

total 48 7

dark treated vials with polyploids. The 48 hour colchicine treatment

appeared best for polyploid induction following darkness, while 24 hour

exposure was best for induction following cold/darkness.

Experiment 14 again revealed that lower concentrations (0.025%)

appeared more effective than higher (0.050%) (Table 9). There was also

a slight increase in visible polyploids in Exp. 14 over the refrigerator

treatment in Exp. 13, 17.5% and 15%, respectively. Experiment 14 also

indicated that growth periods between 24 and 48 hours after pretreatment

may enhance the effect of colchicine. Further refinement of the

technique could increase polyploid induction.

Not all of the shoots that were selected as visible polyploids

proved to have doubled chromosome numbers after microscopic observation.

From the 24 shoots visually selected, half had the desired 96 chromo-

somes and the other half had 48.

Experiments of the sort described in this section are difficult to

analyze statistically. Strict control of total numbers and replications

is difficult due to the type of plant material, difficulty of defining

the experimental unit, and losses due to contamination. However, these

experiments are necessary and important for directing further research

even though the assessment of experimental results is necessarily

somewhat subjective. Conclusions from these experiments are:

1) highbush is difficult, but not impossible to propagate in vitro;

2) highbush is more sensitive to the toxic effects of colchicine, and

more resistant to its chromosome doubling effects than other

blueberries; 3) problems associated with propagation and polyploid

induction can be overcome with proper treatment.

Table 9

Vials Containing Visible Polyploids Following 7 Days of Darkness
or Cold Darkness, 24-192 Hours at Normal Incubation Conditions
and a 0.025% or 0.050% Colchicine Treatment for 24 Hours

Total # Vials with
% Colchicine Clone Time(hrs)ZY of Vials Fat Shoots

0.025 218 24 6 3
48 2 1
96 3 1
192 4 1
221 24 6 0
192 6 0
238 24 5 1
48 4 0
192 4 0

total 40 7

0.050 218 24 7 0
48 6 1
96 5 1
192 10 1
221 24 9 0
48 9 0
96 8 0
192 10 0
229 24 9 0
96 7 0
238 192 9 0

total 40 3

ZDuration of growing phase prior to colchicine treatment.

YMissing treatments were lost due to contamination.



Interspecific hybridization will no doubt remain an important

aspect of blueberry breeding. This method holds much promise for

improvement of bush performance and/or fruit quality. Visualizing the

types of plants desired from these crosses is easy, but obtaining them

is often difficult due to the crossing barriers that exist between

species with different ploidy levels.

Currently the most promising way of combining the rabbiteye and

highbush gene pools is still through the proven way of developing 4x

rabbiteye-types from 6x-2x crosses and crossing these to 4x highbush.

This research has shown that the initial hybrids are not 4x but 5x.

These plants have numerous meiotic problems which result in lower

fertility and an increased percentage of 2n gametes. However, the 5x

plants, through further breeding, eventually eliminate the extra genome

to become 4x.

Because of the difficulty in obtaining hybrids from 6x-2x crosses,

extra care must be taken in using them for further breeding.

Approximately 1 hybrid seedling results from 1,000, 6x-2x pollinations,

and as a result breeders are reluctant to discard them even though they

may have serious flaws. This can lead to development of plants lacking

the desirable traits of the parents. Therefore, selection should be not

only for hybrids but those hybrids possessing good fertility and the

desired parental traits.

Evaluation of combining rabbiteye and highbush gene pools through

developing 6x highbush-types should not be based on the performance of

1 plant. More 8x plants need to be developed, and possibly their

fertility increased through intercrossing prior to backcrossing to the

4x level. The production of 8x plants has proven difficult but not

impossible. With further refinement of the techniques used in this

research, large numbers of 8x highbush plants should be obtainable. The

fact that one 6x plant was obtained from a poorly growing, semifertile

8x and from the limited efforts of 1 person should inspire the efforts

of more people as soon as the 8x germplasm has been developed and



1. Aalders, L.E., A.A. Ismail, I.V. Hall, and P.R. Hepler. 1975.
Augusta lowbush blueberry. Can. J. Plant Sci. 55:1079.

2. Ackerman, W.L., and H. Dermen. 1972. A fertile colchiploid from
a sterile interspecific camellia hybrid. J. Hered. 63:55-59.

3. Ahokas, H. 1971. Notes on polyploidy and hybridity in Vaccinium
species. Ann. Bot. Fennici. 8:254-256.

4. Allard, R.W. 1960. Principles of plant breeding. John Wiley and
Sons, Inc., New York.

5. Arisumi, T. 1973. Morphology and breeding behavior of colchicine-
inducted polyploid Impatiens spp. L. J. Amer. Soc. Hort. Sci.

6. 1975. Phenotypic analysis of progenies of artificial
and natural amphiploid cultivars of New Guinea and Indonesian
species of Impatiens L. J. Amer. Soc. Hort. Sci. 100:381-383.

7. Ballinger, W.E. 1966. Soil management, nutrition, and fertilizer
practices, p. 132-178. In: P. Eck and N.F. Childers (eds.).
Blueberry culture. New Brunswick, Rutgers Univ. Press.

8. Bannerot, H.L., L. Loulidard, Y. Cauderon, and T. Tempe. 1974.
Eucarpia Cruciferae conference. Scottish Hort. Research
Institute, Invergoarie, Dundee, Scotland. pp. 52-54.

9. Blakeslee, A.F., and A.G. Avery. 1937. Methods of inducing
doubling of chromosomes in plants by treatment with colchicine.
J. Hered. 28:393-411.

10. Briggs, F.N., and P.F. Knowles. 1967. Introduction to plant
breeding. Reinhold Publishing Co., New York.

11. Brightwell, W.T., 0. Woodard, G.M. Darrow, and D.H. Scott. 1955.
Observation on breeding blueberries for the southeast. Proc. Amer.
Soc. Hort. Sci. 65:274-278.

12. Camp, W.H. 1945. The North American blueberries with notes on
other groups of Vacciniaceae. Brittonia 5:203-275.

13. Cauderon, Y. 1977. Allopolyploidy, p. 131-145. In: E. Sanchez-
Monge and F. Garcia-Olmeda (eds.). Interspecific hybridization in
plant breeding. Proc. 8th Eucarpia Cong., Madrid, Spain.

14. Chandler, C.K. 1980. Guard cell length and leaf thickness as
indicators of induced polyploidy in Vaccinium. MS Thesis.
University of Florida, Gainesville.

15. Cochran, H. 1961. Blueberries, with special reference to Florida
culture. Bull. no. 33, New Series, Fla. St. Dept. Agric.,
Tallahassee, Fla.

16. Cockerham, L.E., and G.J. Galetta. 1976. A survey of pollen
characteristics in certain Vaccinium species. J. Amer. Soc.
Hort. Sci. 101:671-676.

17. Coville, F.V. 1910. Experiments in blueberry culture. USDA
Bureau of Plant Industry Bul. 193.

18. 1921. Direction for blueberry culture, 1921. USDA
Bul. 974.

19. 1937. Improving the wild blueberry, p. 559-574.
In: USDA yearbook of agriculture. 1937. United States govern-
ment printing office Washington.

20. Darrow, G.M. 1949. Polploidy in fruit improvement. Proc. Amer.
Soc. Hort. Sci. 54:523-532.

21. 1960. Blueberry breeding: past, present, and
future. Amer. Hort. Mag. 39:14-33.

22. and W.H. Camp. 1945. Vaccinium hybrids and the
development of new horticultural material. Bul. Torrey Bot. Club

23. W.H. Camp, H.E. Fischer, and H. Dermen. 1944.
Chromosome numbers in Vaccinium and related groups. Bul. Torrey
Bot. Club 71:498-516.

24. H. Dermen, and D.H. Scott. 1949. A tetraploid blue-
berry from a cross of diploid and hexaploid species. J. Hered.

25. and D.H. Scott. 1966. Varities and their character-
istics, p. 94-110. In: P. Eck and N.F. Childers (eds.).
Blueberry culture. New Brunswick,
Rutgers Univ. Press.

26. D.H. Scott, and H. Dermen. 1954. Tetraploid blue-
berries from hexaploid x diploid species crosses. Proc. Amer.
Soc. Hort. Sci. 63:266-270.

27. Dermen, H. 1938. A cyctological analysis of polyploidy.
J. Hered. 29:211-229.

28. Dermen, H. 1940. Colchicine polyploidy and technique. Bot.
Rev. 6:599-635.

29. ., and H.F. Bain. 1944. A general cytohistological
study of colchicine polyploidy in cranberry. Amer. J. Bot.

30. Den Nija, T.P.M., and S.J. Peloquin. 1977. 2n gametes in potato
species and their function in sexual polyplodization. Euphytica

31. deWet, J.M.J. 1971. Polyploidy and evolution in plants. Taxon

32. 1980. Origins of polyploids, p. 3-15. In:
W.H. Lewis (ed.). Polyploidy: Biological relevance. Plenum
.Press, New York.

33. Dodds, J.H., and L.W. Roberts. 1982. Experiments in plant tissue
culture. Cambridge University Press, Cambridge.

34. Draper, A.D., G.J. Galetta, W.T. Brightwell, J.M. Spiers,
W.B. Sherman, and G. Jelenkovic. 1976. Interspecific
hybridization in Vaccinium. Fruit Var. J. 30:27-28.

35. A.W. Stretch, and D.H. Scott. 1972. Two tetraploid
sources of resistance for breeding blueberries resistant to
Phytophora cinnamomi Rands. HortScience 7:266-268.

36. Eck, P. 1966. Botany, p. 14-44. In: P. Eck and N.F. Childers
(eds.). Blueberry Culture. New Brunswick, Rutgers Univ. Press.

37. Eigsti, O.J., and P. Dustin. 1955. Colchicine. Iowa State
College Press, Ames, Iowa.

38. Ferwerda, F.P., and F. Wit (eds.). 1969. Outlines of perennial
crop breeding in the tropics. H. Veenman, and Zonen N.V.,
Wageningen, The Netherlands.

39. Galetta, G.J. 1975. Blueberries and cranberries, p. 154-196.
In: J. Janick and J.N. Moore (eds.). Advances in fruit breeding.
Purdue Univ. Press, West Lafayette, Ind.

40. Goldblatt, P. 1980. Polyploidy in angiosperms: Monocotyledons,
p. 219-239. In: W.H. Lewis (ed.). Polyploidy: Biological
relevance. Plenum Press, New York.

41. Goldy, R.G., and P.M. Lyrene. 1983. Pollen germination in
interspecific Vaccinium hybrids. HortScience 18:54-55.

42. Grant, V. 1963. The origin of adaptations. Columbia Univ. Press,
New York.

43. Hadley, H.H., and S.J. Openshaw. 1980.
generic hybridization, p. 133-159. In:
(eds.). Hybridization of crop plants.
Sci. Soc. Amer., Madison, Wisconsin.

Interspecific and inter-
W.R. Fehr, and H.H. Hadley
Amer. Soc. Agron. Crop.

44. Hagberg, A., and E. Akerberg. 1962. Mutation and polyploidy in
plant breeding. Heinemann Educational Books Ltd., London.

45. Hall, S.H., and G.J. Galetta. 1971. Comparative chromosome
morphology of diploid Vaccinium species. J. Amer. Soc. Hort. Sci.

46. Harlan, J.R., and J.M.J. deWet. 1975. On 0 Winge and a prayer:
The origins of polyploidy. Bot. Review 41:361-389.

47. Jackson, R.C. 1976. Evolution and systematic significance of
polyploidy. Ann. Rev. Ecol. Syst. 7:209-234.

48. Janick, J., and J.N. Moore (eds.). 1975. Advances in fruit
breeding. Purdue Univ. Press, West Lafayette, Indiana.

49. Jelenkovic, G., and A.D. Draper. 1970. Fertility and chromosome
behavior of a derived decaploid of Vaccinium. J. Amer. Soc.
Hort. Sci. 95:816-820.

50. and A.D. Draper.
pentaploid interspecific hybrids
vocarslvo. God. VII, br. 25-26,

1973. Breeding value of
of Vaccinium. Jugoslovensko
st. 237-244.

and Harrington. 1971. Non-random chromosome
association at diplotene and diakenesis in a tetraploid clone
of Vaccinium australe Small. Can. J. Gen. Cyt. 13:270-276.

and L.F. Hough. 1970.
in the first meiotic division in three
Vaccinium corymbosum. L. Can. J. Gen.

Chromosome associations
tetraploid clones of
Cyt. 12:316-324.

53. Johnson, A.W., J.G. Packer, and G. Reese. 1965. Polyploidy,
distribution and environment. In: H.E. Wright and D.G. Frey
(eds.). Quaternary of the United States. Inter Assoc. Quatern.
Res. Publications, Yale University Press, New Haven, Connecticut.

54. Johnston, S.A., and R.E.
endosperm balance number
diploid Solanum species.

Hanneman, Jr. 1982. Manipulation of
overcome crossing barriers between
Science 217:446-448.

55. Jones, K. 1970. Chromosome changes in plant evolution. Taxon

56. Kamemoto, H., and K. Shiado. 1964. Meiosis in interspecific and
intergenetic hybrids of Vanda. Bot. Gaz. 125:132-138.

57. Kihara, H., and T. Ono. 1926. Chromosomenzahlen and systematische
gruppiering der Rumex arten. Zeitschr. Zelf. Mikrork. Anat.

58. Levan, A. 1945. The present state of plant breeding by induction
of polyploidy. Sveriges Utsades-forenings Tidskrift.

59. Levin, D.A., and A.C. Wilson. 1976. Rates of evolution in seed
plants. Net increase in diversity of chromosome numbers and
species number though time. Proc. Nat. Acad. Sci. USA.

60. Lewis, W.H. (ed.). 1980. Polyploidy: Biological relevance.
Plenum Press, New York.

61. _. 1980. Polyploidy in angiosperms:
Dicotyledons, p. 241-263. In: W.H. Lewis (ed.). Polyploidy:
Biological relevance. Plenum Press, New York.

62. Longley, A.E. 1927. Chromosomes in Vaccinium. Science

63. Lyrene, P.M. 1978. Blueberry callus and shoot-tip culture.
Proc. Florida State Hort. Soc. 91:171-172.

64. and J.L. Perry. 1982. Production and selection of
blueberry polyploids in vitro. J. Hered. 73:377-378.

65. __, and W.B. Sherman. 1979. The rabbiteye blueberry
industry in Florida 1887 to 1930 with notes on the current
status of abandoned plantations. Economic Botany 33:237-243.

66. MacNaughton, J.H. 1976. Eucarpia Cruciferae Newsletter, 1:23.

67. Mainland, C.M. 1966. Propagation and planting, p. 111-131. In:
P. Eck and N.F. Childers (eds.). Blueberry culture. New
Brunswick, Rutgers Univ. Press.

68. Marucci, P.E. 1966. Insects and their control, p. 199-235. In:
P. Eck and N.F. Childers (eds.). Bluberry culture. New
Brunswick, Rutgers Univ. Press.

69. Moore, J.N. 1965. Improving highbush blueberries by breeding
and selection. Euphytica 14:39-48.

70. 1966. Breeding, p. 45-74. In: P. Eck and N.F.
Childers (eds.). Bluberry culture. New Brunswick, Rutgers
Univ. Press.

71. D.H. Scott, and H. Dermen. 1964. Development of a
decaploid blueberry by colchicine treatment. Proc. Amer. Soc.
Short. Sci. 84:274-279.

72. Muntzing, A. 1980. Problems of allopolyploidy in Triticale,
p. 409-426. In: W.H. Lewis (ed.). Polyploidy biological
relevance. Plenum Press, New York.

73. Newcomer, E.H. 1941. Chromosome numbers of some species and
varieties of Vaccinium and related genera. Proc. Amer. Soc.
Hort. Sci. 38:468-470.

74. Newton, W.C.F., and C. Pellew. 1929. Primula kewensis and its
derivatives. J. Genetics 20:405-467.

75. Nickerson, N.L. 1978. In vitro shoot formation in lowbush blue-
berry seedling explants. HortScience 13:698.

76. North, C. 1979. Plant breeding and genetics in horticulture.
The MacMillan Press Ltd., London.

77. Olden, E.J. and N. Nybom. 1968. On the origin of Prunus cerasus
L. Hereditus 59:327-345.

78. Perkins, F.A. 1966. Economics and Marketing, p. 302-319. In:
P. Eck and N.F. Childers (eds.). Blueberry culture. New
Brunswick, Rutgers Univ. Press.

79. Powell, C.L. 1982. The effect of the Ericoid Mycorrhizal Fungus
Pezizella ericae (Read) on growth and nutrition of seedlings of
blueberry (Vaccinium corymbosum L.) J. Amer. Soc. Hort. Sci.

80. Raven, P.C. 1975. The basis of angiosperm phylogeny: Cytology.
Ann. of the Missouri Botanical Garden. 62:724-764.

81. Rousi, A. 1966. Cytological observations on some species and
hybrids of Vaccinium. Zuchter 36:352-359.

82. Sanchez-Monge, E., and F. Garcia-Olmeda. 1977. Interspecific
hybridization in plant breeding. Proc. 8th Eucarpia Congress.
Madrid, Spain.

83. Semnuik, P. 1978. Colchiploidy in Exacum. J. Hered.

84. Sharpe, R.H., and W.B. Sherman. 1971. Breeding blueberries for
low chilling requirement. HortScience 6:3-5.

85. 1976. 'Flordablue' and 'Sharpblue': Two new blue-
berries for central Florida. Fla. Agric. Expt. Sta. Cir. 2-240.

86. Sherman, W.B. and R.H. Sharpe. 1977. 'Avonblue' blueberry.
HortScience 12:510.

87. Schulz-Schaeffer, J. 1980. Cytogenetics. Springer-Verlag,
New York.

88. Shutak, V.G., and P.E. Marucii. 1966. Plant and fruit develop-
ment, p. 179-198. In: P. Eck and N.F. Childers (eds.).
Blueberry culture. New Brunswick, Rutgers Univ. Press.

89. Simmonds, N.W. (ed.). 1976. Evolution of crop plants. Longman,

90. 1979. Principles of crop improvement.
Longman, London.

91. Stebbins, G.L. 1958. The inviability, weakness, and sterility
of interspecific hybrids. Adv. Genet. 9:147-215.

92. 1971. Chromosomal evolution in higher plants.
Addison-Wesley Publ. Co., Reading, Mass.

93. Steele, R.G.D., and J.H. Torrie. 1960. Principles and procedures
of statistics. McGraw-Hill, New York.

94. Stiles, W.C., and D.A. Abdalla. 1966. Harvesting, processing,
and storage, p. 280-301. In: P. Eck and N.F. Childers (eds.).
Blueberry culture. New Brunswick, Rutgers Univ. Press.

95. Sturtevant, E.L. 1972. Sturtevant's edible plants of the world.
Dover Press, New York (reprint).

96. Stushnoff, C., and A.J. Feliciano. 1968. A simple technique
for observing mitotic division of the generative nucleus in
pollen tubes of Vaccinium spp. HortScience 3:174.

97. and B.F. Palser. 1970. Embryology of five
Vaccinium taxa including diploid, tetraploid and hexaploid species
or cultivars. Phytomorpholgy 19:312-321.

98. Uhlinger, R.D. 1982. Wide crosses in herbaceous perennials.
HortScience 17:570-574.

99. Vander Kloet, S.P. 1980. The taxonomy of the highbush blueberry,
Vaccinium corymbosum. Can. J. Bot. 58:1187-1201.

100. Varney, E.H., and A.W. Stretch. 1966. Diseases and their
control, p. 236-279. In: P. Eck and N.F. Childers (eds.). Blue-
berry culture. New Brunswick, Rutgers Univ. Press.

101. Viinikka, Y., and S. Nokkala. 1981. Interchromosomal connections
in meiosis of Secale cereale. Hereditas 95:219-224.

102. Ward, D.B. 1974. Contributions to the flora of Florida 6,
Vaccinium (Ericaceae). Castanea 39:191-205.

103. White, M.J.D. 1952. The chromosomes, ed. 2. Methuen, London.

104. Winge, 0. 1917. The chromosomes. Their numbers and general
importance. Compt. Rend. Trav. du Lab. de Carlsberg 13:131-275.


105. Zeven, A.C. 1980. Polyploidy and domestication: The origin
and survival of polyploids in cytotype mixtures, p. 385-407.
In: W.H. Lewis (ed.). Polyploidy: Biological relevance.
Plenum Press, New York.


Ron Goldy was born in West Branch, Michigan, on September 30, 1954.

He received his Bachelor of Science degree from Eastern Michigan

University in April, 1977.

He was admitted by Michigan State University as a graduate student

in the Horticulture Department in June, 1977, under the direction of Dr.

Robert Andersen who was responsible for introducing him to fruit

breeding. He was awarded the degree of Master of Science in

horticultural science in June, 1980. He was then admitted to the

University of Florida in September, 1980, as a graduate student in the

Fruit Crops Department under the direction of Dr. Paul Lyrene. He was

awarded the degree of Doctor of Philosophy in August, 1983, in

horticultural science (Fruit Crops).

He is married to Kathy (Garner) Goldy and is a member of the

American Society for Horticultural Science.

I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.

Paul M. Lyrene Chairman
Associate Professor of
Horticultural Science

I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.


Wayne BJ Sherman
Professor of Horticultural

I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.

Mark J. Bass
Associate Pr fessor of
Horticultural Science

I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.

....- CL. 'i ?i: ?.t_ C
Gloria M. Moore
Assistant Professor of
Horticultural Science

I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.

Ken H. Quesenberry
Associate Professor of


This dissertation was submitted to the Graduate Faculty of the College
of Agriculture and to the Graduate Council, and was accepted as partial
fulfillment of the requirements for the degree of Doctor of Philosophy.

August, 1983

Dean College

of Agric ture

Dean for Graduate Studies and

I 1262 08554 0069
3 1262 08554 0069