. I .. - hr». . h m . H , . , 32:.szJJVV—uJi-f-o4-ufioufif‘i‘rn.
.. V . , a.»1.i fa.9ov...s.
. augvv..cr.l

1. 1.1;;1 .tf'Q.
:{clOtLKQth-nivt.

£5.

7!. :05:
.1112}: t.
11!. . “.19...

 

 

RX...
3 ... A: z.lvl. I...“ 2 f,
.(..)v .h! 9:...1...

 

 

 

x... rp.v.!.
v.;l¢.r.lalv!l
1.7 L.II..

will. 314,1... ice...
. .1!“ o In... (It’ll

 

 

 

. 1 L...
(1’1). ...L

 

 

. ..\. ..
.11....13. y. u 7. a:

 

 

 

 

 

;:
n (\n. 7.: .tl.tr9 lull... v?t.lll.l§

 

..... 1. . . rt... .)

wimpy? s%.a..
.|..JJ.?.. . . :lv.

  

 

 

. 1.7... o. i?.. A... ...I..1. . .. . 2:3.
.3. 1.. . . . . .

.l . Viearfo... .. . . 5% {F

 

VHS“

WWWWWWW

3 1293 00910 1761

This is to certify that the

dissertation entitled

Introgression of Allium fistulosum into Allium cepa

 

 

through Hybridization

presented by

Uma Shanker Gupta

has been accepted towards fulfillment
of the requirements for

Ph. D.
degree in Plant Breeding and Genetics

 

 

cflde€€3wflfi

Major professor

Dm/AVZB

MSU is an Affirmative Action/Equal Opportunity Institution ' 0-12771

 

 

LIBRARY
Michigan State
; Unlverstty

1*

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.

DATE DUE DATE DUE DATE DUE

____m__l

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU Is An Affirmative Action/Equal Opportunity Institution
azuMpmS-pd

 

 

 

J

 

_____—____ _ _~_ M __

INTROGRESSION OF ALL] UM FIS TULOS UM INTO ALLI UM CEPA
THROUGH HYBRIDIZATION.

by

Uma Shanker Gupta

A DISSERTATION

Submitted to
Michi an State Universi
in partial ent of the requirements
for the degree of

DOCTOR OF PHILOSOPHY
Plant Breeding and Genetics Program

Department of Horticulture
1993

ABSTRACT

INTROGRESSION or ALLIUM FISTULOSUM INTO A. CEPA moucn
HYBRIDIZATION.
by

Uma Shanker Gupta

Allium cepa L. (onion) is an important vegetable crop which unfortunately lacks
resistance to many insects and diseases. Allium fistulasum (Japanese bunching
onion), on the other hand, has been widely recognized as the allied onion species
with resistance to many of the insects and diseases that attack the bulb onion. Plant
breeders all over the world have been trying to transfer the resistance and other
economic characters from A. fistulasum into A. cepa. To date, no published study
has shown development of a regularly pairing interspecific hybrid line with A. cepa

characteristics.

This study describes a classical genetic introgression experiment which used the A.
cepa-firmlosum allopolyploid, Beltsville Bunching, as the initial A. fistulosum
genome source. Morphological, cytogenetic, molecular and reproductive evidences
were collected and by the 5th generation, they show that the interspecific derivatives
externally resemble A. cepa, chromosome pairing is stabilizing, a DNA
polymorphism found in A. firtulosum has been carried over in the interspecific

backcrosses, and the reproductive barrier is broken.

The interspecific derivatives were screened for Fusarium and pinkroot resistance.
Some of the derivatives show strong resistance to both diseases. The data indicate
that the resistance in both cases could be a simply inherited dominant. It is possible

that the same gene is imparting the resistance / tolerance to both diseases.

DEDICATION

This study is respectfully dedicated to Mrs. Jo Ewart.

ACKNOWLEDGMENTS

I am grateful to many for their help and guidance in this study. I would like to thank
Dr. Jim Hancock, Dr. Bernie Zandstra and Dr. Joanne Whallon for their accurate
criticisms and suggestions in the preparation of this manuscript. I would also like to
thank people from Dr. Hancock’s lab, Pete, Jeff and Don, for their help in the
RAPD analysis.

This study could not be completed without the teaching, guidance and
encouragement of Lowell Ewart, my major professor. I could not have asked for a
better professor. He is a true friend, philosopher and guide. I have learned a great

deal from his experience as a plant breeder. Thanks Iowell.

Lastly I would like to thank my sons, Sonny and Manish, and wife, Kumud, for their

patience during the course of this study.

iv

Table of Contents

Introduction ........................... 1
Literature Review ......................... 3
A. cepa (Bulb Onion) ...................... 4

A. fistulosum (Japanese bunching onion) .............. 4
Inter-specific hybridization inAllium ............... 5
Materials and Methods ...................... 12
Hybridization ......................... 16
Pollination Procedure .................... 16
Disease screening ....................... 17
Inoculum Preparation ................... 18

Growing medium preparation ............... 19
Experimental design .................... 19

Cytoloy ........................... 20
Random amplified polymorphic DNA (RAPD) ........... 20
DNA Extraction ....................... 23
RAPD Procedure ...................... 24
Results and Discussion ....................... 25
Morphology .......................... 25
Cytogenetic Observations ..................... 29
RAPD Analysis ......................... 39
Fertility Improvement ...................... 45
Disease Screening . ....................... 49
Genetic Model ........................ 56
Summary and Conclusion ...................... 61
Appendices ............................ 64
Bibliography ............................ 68

Table 1:

Table 2:

Table 3:

Table 4:

Table 5:

Table 6:

Table 7:

Table 8:

Table 9:

Table 10:

Table 1 1:

Table 12:

Table 13:

Table 14:

Table 15:

Table 16:

List of Tables

Interspecific Allium Germplasm Studied and Developed at
Michigan State University in 1989.

Field numbers used after repotting the 1989 Allium
interspecific derivative plants.

Interspecific Allium Germplasm Studied and Developed at
Michigan State University in 1990.

Serpllences of the primer codes used in the RAPD analysis
in t e Allium interspecific derivative development study.

Various lines, accessions and crosses used in the RAPD
analysis used in the Allium interspecific derivative
development study.

Chromosomal pairing relationships in the parents used in
the Interspecific Allium Germplasm study.

Univalents in some of the interspecific Allium germplasm
metaphase I cells.

Seed set in the 5th generation (BC4 self) interspecific
germplasm development between A. cepa and A. fistwlosum.

 

Seed set in the 5th generation (BC5) interspecific
germplasm development between A. cepa and A. fistuloswn.

 

Means of Fusarium screening observations on selected
A. cepa and interspecific germplasm lines.

Means of pinkroot screening observations on selected
A. cepa and mterspecific germplasm lines.

The control screening observations on sclected A. cepa and
interspecific germplasm lines.

Overall germination of selected A. cepa and interspecific
lines used in the Fusan'um, pinkroot and control screening
experiments.

Possible model of a single dominant gene (P) for pinkroot
resistance in the Allium interspecific hybrid deve opment

study.

Possible model of a single dominant gene (F for Fusan'um
resistance in the Allium interspecific hybri development

study.

Possible model of a single dominant gene (R) for overall
resistance in the Allium interspecific hybrid development

study.

14

14

15

22

22

33

33

46

46

50

50

51

56

57

57

58

List of Figures

Figure 1: A. fisndosum and inters cific derivative plants. (a)
A. fistulosum. (b) interspecigiec derivative, 89-4-2.

 

Figure 2: Inflorescence of A. and interspecific derivative
plants. (a) A. fis osum var Odessa. (b) interspecific
derivatives.

Figure 3: Bulbs of A. cepa and interspecific derivative plants. (a)
A. cepa (b) interspecific derivative, 90-1 1.

Figure 4: Metaphase I chromosomes of (a) A. cepa showing
randomized chiasmata formation (b) A. firtulosum showing
localized chiasmata formation.

Figure 5: Chromosomes of the Allium interspecific derivatives,
showing meiotic structures in BC4 plants.

Figure 6: Metaphase I chromosomes in the Allium inters cific
derivative BC4 plant showing ring formation in 89-4- .

Figure 7: Metaphase I chromosomes in the Allium interspecific
derivative plants (a) 89-4-3 showin majority of chiasmata
being randomized type, and (b 90-12 showing two
univalents.

Figure 8: Metaphase I chromosomes in the Allium interspecific
derivative, 90-14, a BC self, showing complete bivalents
with combination of ran omized and localized chiasmata.

Figure 9: RW2 primer derived Ethidium bromide gel showing
polymorphic bands in A. cepa, A. finulosum and
mterspecific derivative DNA.

Figure 10: Schematic diagram of the stained agarose gel showing
amplified DNA banding patterns of Alltum gemplasm.

Figure 11: LE2 primer derived Ethidium bromide gel showing
polymorphic bands in A. cepa, A. firtulosum an
mterspecific derivative DNA.

Figure 12: Bar aph showing survival and germination percentages of
all nes used in the basal rot screening study using
interspecific derivative onion lines.

Figure 13: Bar graph showing survival and germination percentages of
all 'nes used in the pinkroot screening study using
interspecific derivative omon lines.

Figure 14: Euar .aph showing survival and germination percentages of

es used in the control screening study using
interspecific derivative onion lines.

vii

26

27

34

35

36

37

38

42

43

53

54

55

INTRODUCTION

Allium cepa L. is a diploid species (2n = 16) which has been cultivated since
prehistoric times. Historical references to onion grown for food date back to 3000
BC (Hanelt 1990). Although the genus Allium consists of hundreds of species, only a
few are commercially important food crops: A. cepa (onion), A. sativum (garlic),
A. ampeloprasum (great headed garlic, leek, kurrat), A. fistulosum (Japanese
bunching onion), A. schoenoprasum (chive), A. chinense (rakkyo) and A. tuberosum
(Chinese chive). None of the above species will cross freely with A. cepa by
conventional means. Experimental hybrids between A. cepa and A. fistulosum,
however, have frequently been reported (Emsweller and Jones 1935, Maeda 1937,
Levan 1941). An amphidiploid (2n = 32) hybrid, Beltsville Bunching (A. cepa x
A. firtulosum), and a triploid, Delta-Giant, (A. ascalonicum x A. fisndosum) have
been used as A. fistulosum genome sources in the ongoing attempts by breeders over
the past 50 years to introduce desirable characters from A. fistulasum (Jones et al.,
1963). Apart from some genetic studies, most researchers have abandoned such
work before the interspecific hybrids could be utilized in a breeding program. The
lack of progress is primarily attributed to the meiotic irregularities occurring in
A. cepa x A. firtulosum hybrids and their derivatives which cause sterility and
severely hamper transfer of genetic material between the two species (Dowker 1990,
Jones 1990).

Allium fistulosum possesses many economically important characters - resistance /
tolerance to pinkroot (Pyrenochaeta terrestrir; Porter and Jones 1933), smut
(Urocystis cepulae F.; Felix 1933), leaf rot (Botrytir squamosa; Maude 1990a), neck
rot (Botrytis allii; Maude 1990b), and thrips (Thrips tabaci; Jones et al., 1934, Soni
and Ellis 1990) as well as high soluble solid content and cold hardiness

1

2

(Cryder 1988). In short, a successful genetic introgression is very desirable and could

alleviate many of the short-comings associated with the cultivated bulb onion.

The current study evaluated an on-going A. cepa - A. fistulosum backcrossing
program. The major hypotheses tested are (a) if there is an actual genetic
integration from A. firtulosum into A. cepa, the recurrent parent. (b) if the
interspecifically derived hybrids are cytogenetically stabilized. (c) if the fertility of
the interspecific hybrids has significantly improved. (d) if the interspecific hybrid
lines carry resistance / tolerance to Fusarium and pinkroot, two important diseases
of the bulb onion. The purpose of this study was to evaluate the effectiveness of the
program / procedure and to assess the potential use of the developed plants in the

on-going onion breeding program at Michigan State University.

LITERATURE REVIEW

The botanical classification of Allium is still in controversy (Jones et al., 1963;
Cronquist 1968; Hanelt 1990). Initially, Allium was placed in the family Liliaceae,
then it was transferred into Amaryllidaceae on the basis of inflorescence structure,
and later Amaryllidaceae was included in Liliaceae. Thus, Allium was again included
in Liliaceae. Modern taxonomy, however, now puts Allium and its closest relatives

in a distinct family Alliaceae (Hanelt 1990). The following hierarchy is currently

accepted:
Class : Monocotyledones
Superorder : Liliifloreare
Order : Aspargalar
Family : Alliaceae
Tribe : Allieae
Genus : Allium

Allium is a large genus of about 600 species - mostly perennial and bulbous (Hanelt,
1986). It is widely distributed over the warm-temperate and temperate zones of the
northern hemisphere. It is estimated that the total world onion bulb production is
more than 250 million MT and that the value of this production approaches 5 billion
dollars annually (FAO, 1988).

In the Allium genus as a whole, there are three basic chromosome number groups,
x=7, x=8, and x=9. The x=7 species are largely confined to the New World, while
the majority of species (about 500) which are in the Old World are mainly x=8.
There are only a few species with x = 9 (Jones, 1990). All cultivated edible forms

have the same basic chromosome number of x = 8.

Allium cepa (Bulb Onion)

Allium cepa is the most widely known edible Allium. It has been in use for at least
3000 years. The species is an out-breeder and current improvement programs utilize
hybrid varieties, as well as open pollinated selections (Dowker et al., 1983). A. cepa
is a diploid and has a complement of eight pairs of metacentric and submetacentric
chromosomes (2n = 2x = 16). With the development of Giemsa c-banding
techniques, it is now possible to distinguish and name the eight chromosomes of the
basic set individually (Kalkman, 1984, de Vries, 1990). A. cepa has a nuclear DNA
value of 33.5 pg per 2C nucleus (Jones et al., 1968, Ranjekar et al., 1978). The
quality of DNA is such that the G + C base pairs comprise about 32% of the total
and up to 50% is in the form of repetitive sequences. There is no satellite DNA
(Jones, 1990). Meiosis is regular with eight bivalents. Chiasmata are formed mainly

in the distal and interstitial regions of the chromosome arms.

Allium fistulosum (Japanese bunching onion)

Allium fistulosum is a perennial, diploid out-breeder, widely used as an edible green
onion in Asia. The species exhibits high levels of resistance to several important
pests and diseases. A. fisnilosum has been observed to possess resistance / tolerance
to pink-root (Porter and Jones 1933, Corgan 1985), onion smut (Clarke and Mackay
1946), downy mildew (Henning and Mildenberger 1988, Weitsma and Kik 1989) and
thrips (Jones and Mann 1963).

The mitotic chromosome complement resembles that of A. cepa, although the

homeologues are smaller in both size and DNA amount. The nuclear DNA value is

5

26.3 pg, only 78.5% of the amount found in A. cepa. The difference is unevenly
distributed among the eight pairs of chromosomes (Jones et al., 1968, F iskejo, 1975).
The chiasmata are highly proximal (adjacent to the centromere) forming cruciform
shaped bivalents (Levan, 1933). It is assumed that as a result of the restricted
chiasmata location, there is an absence of crossing-over in the bulk of the

chromosome arms and the arms behave as super-genes.

Inter-specific hybridization in Allium

It is customary to look for resistance genes in the allied wild relatives of an
established crop. Domestic selections of A. fistulasum, however, retain many
valuable resistance characteristics generally attributed to wild species (Porter and
Jones 1933, Corgan 1985, Clarke and Mackay 1946, Henning and Mildenberger
1988, Weitsma and Kik 1989, Jones and Mann 1963).

Other Allium species have also been found to possess resistance characteristics but
to a lesser extent than A. fistulosum. A. roylei is resistant to downy mildew (Weitsma
and Kik 1989, Meer and de Vries 1990), pinkroot (Saini and Davis 1967) and to
thrips (Saini and Davis 1967). A. galanthum was found to be tolerant to pinkroot
(Saini and Davis 1967) and A. pskemense is tolerant to thrips and downy mildew
(Saini and Davis 1967).

Artificial inter-specific hybridization in Allium species was started in the early
thirties. The following hybridization attempts have been reported: A. fistulosum x
A. cepa (Emsweller and Jones 1935, Jones and Clarke 1942, Emsweller and Jones
1945, Clarke and Mackay 1946, Corgan 1985, Peffley 1987, Bino et al., 1988, Lu et
al., 1989), A. fistulosum x A. altaicum (McCollum 1982, Kokoreva et al., 1987), A.

6

cepa x A. galanthum (Saini and Davis 1967, McCollum 1971, McCollum 1982,
Wietsma and de Vries 1988), A. cepa x A. pskemense (Saini and Davis 1967,
McCollum 1971, Wietsma and de Vries 1988), A. cepa x A. oschaninii (McCollum
1974, McCollum 1982, Tarasova and Pole 1984, Wietsma and de Vries 1988), A.
roylei x A. cepa (Wietsma and de Vries 1988, Meer and de Vries 1990), A. cepa x
A. drobovii (Saini and Davis 1967), A. drobovii x A. galanthum (Saini and Davis
1967), A. drobovii x A. pskemense (Saini and Davis 1967), A. dmbovii x A. roylei
(Saini and Davis 1967), A. fistulosum x A. ascalonicum (Tashiro 1980, Tashiro 1981,
Tashiro 1984), A. cepa x A. vavilovii (Wietsma and de Vries 1988), and A. fistulosum
x A. mylei (Wietsma and Kik 1989). A. cepa crosses easily with A. galanthum, A.
dmbovii and A. pskemense (Saini and Davis 1967, McCollum 1971). Limited success
has been reported with A. cepa crossed with A. vavilovii and A. roylei (Wietsma and
de Vries 1988, Wietsma and Kik 1989). Few of the other crosses produced any
fertile hybrid plants. The success rate of A. cepa and A. fistulosum crosses is less
than 0.5% in a controlled environment (Emsweller and Jones 1935, Jones and

Clarke 1942, Emsweller and Jones 1945, Clarke and Mackay 1946, Corgan, 1985).

Artificial inter-specific hybridization between A. cepa and A. fistulosum started in
1931 at the Agricultural Experiment Station in Davis, California (Emsweller and
Jones 1935). The major reason for the introduction of such a breeding procedure
was that A. fistulasum (var Nebuka), as noted earlier, was resistant to many
economically important diseases at the time. A. fistulosum as a non-bulbing item,
did not have much economic importance in America, and it was natural to try to
incorporate the resistance characteristics into the bulbous onion populations.
Yellow Globe Danvers (A. cepa), an inbred, was used as the female parent and
Nebuka (A. fistulasum) was used as the pollen parent. Only seven seeds were

secured after hundreds of pollinations. Only one seedling survived from these seven

7

seeds. Many more attempts were made to hybridize A. cepa and A. fistulosum in the
following years. In spite of obtaining some seeds, all the plants from such crosses

were found to be mostly self-sterile (Emsweller and Jones 1935).

The following are some of the characteristics of the early developed hybrids

(Emsweller and Jones 1935 Emsweller and Jones 1945, Levan 1941):

Bulbing: A. cepa is bulbing but A. firtulosum is non-bulbing. The hybrid
plants showed none or very slight bulbing. Appendix A shows the
bulbing inA. cepa, A. fistulosum and the F1 hybrid.

Life Cycle: A. firtulosum is a perennial, A. cepa is a biennial (bulbing one

year and setting seed the second year), the hybrids were perennials.

Date of Flowering: A. fistulosum flowers early, while A. cepa flowers late. The

hybrids were intermediate.

Inflorescence: In A. cepa, anthesis occurs over the entire inflorescence
throughout the flowering period, while in A. fistulosum, the terminal
flowers open first and anthesis proceeds progressively towards the
base. At the tip of the umbel, seed may be practically mature while
towards the base, buds are just opening. In the hybrids, the terminal
flowers open first, although this process is less clearly defined than in
A. firmlosum. They mature progressively towards the base with

anthesis over the entire inflorescence.

Flower Habit: In A. cepa, the flowers are fully expanded at anthesis, while in

A. firtulosum, the perianth remains erect. The hybrid resembles A.
fislulosum.

8

Fertility: Both A. cepa and A. fismlosum were fertile in both sexes. The
hybrids were usually self-sterile, but were found to be partially fertile
as a pollen parent. Two hybrids, Beltsville Bunching and Delta Giant,

have been released as commercial cultivars of green bunching onion.

Leaf Shape: A. jistulosum has circular leaves while A. cepa has semi-circular

leaves. The hybrids resembled A. cepa.

Vigor: The hybrids grow considerably taller than either parent and also

exhibited hybrid vigor in many other respects.

Chiasmata formation: A. fistulosum forms cruciform chiasmata located
closely on each side of the centromere. A. cepa forms randomized
chiasmata many of which are at the distal position of the arm. In most
cases, the hybrids showed a combination of both, though the random

form dominated.

Hybrids, when obtained, produced only a few F2 seeds. Emsweller and Jones (1935)
produced three F2 seeds from a White Persian x Nebuka cross. Maeda (1937)
obtained a few F2 plants from the cross of A. fistulosum var Hidanegi x A. cepa var
Yellow Globe Danvers. Levan (1941) grew several F2 plants from such a cross and
all had more than a diploid complement of sixteen chromosomes, but only one was
tetraploid. Emsweller and Jones (1935) obtained more seeds when the F1 hybrid
was used as the pollen parent in the backcross to A. fistulosum, than when the F1
was used as the pollen parent in the backcross to A. cepa. Levan (1941) never

succeeded in backcrossing his hybrid plants.

Cytological observation of hybrid cells showed variable numbers of bivalents. Most

studies reported 8 bivalents in more than 70% of the pollen mother cells (Emsweller

9

and Jones 1935, Maeda 1937). Univalents in the remaining cells varied in number
from two to eight. Chromatin bridges and fragments occurred frequently. In other
cases, the pairing was much more irregular, regular pairing being shown in less than

two percent of the division figures (Levan 1941).

Pairing between A. cepa and A. fistulosum chromosomes does occur in the diploid
hybrid (Emsweller and Jones 1935, Devan 1941, Maeda 1937, Cochran 1950).
Bivalent pairing occurs with a greater or lesser number of bivalents depending upon
the parents used in the cross (Emsweller and Jones 1945). Most studies show
variations in the frequency of chromosome pairing, number of chromosome bridges
and fragments, and heteromorphic chromosomal associations. Emsweller and Jones
(1935) found that chromosomes of unequal length paired and sometimes each had
long unpaired arm extending beyond the end of the other. Number of bivalents in
the adjacent cells was variable and occasionally bridges and fragments were found at
the anaphase I stage. Many studies found that the chiasmata in the interspecific
hybrid was a mixture of randomized and localized formations (Emsweller and Jones
1945, Cocharan 1950). Levan (1941) reported that complete bivalent pairings were
not typical and multivalents were frequently formed.

Several researchers have observed heteromorphic bivalents in the interspecific
hybrids (Emsweller and Jones 1935, Maeda 1937). Bridges and fragments in
anaphase I are reported and interpreted as a result of inversion (Emsweller and
Jones 1935, Emsweller and Jones 1938). Peffley (1986) reported formation of
multivalents suggesting occurrences of translocations in the two species
chromosomes. In a further study Peffley et al., (1989) presented evidence for three
independent reciprocal translocations between A. cepa and A. fistulasum
chromosomes by showing occurrence of tetra- and pentavalent associations. Ring

univalents and isochromosomes were also observed.

10

Sterility of the interspecific hybrids generally blocked progress towards integrating
A. cepa and A. fistulasum. Before efforts in this direction completely stopped,
however, nature provided a new hope. In the late thirties, Jones and Clarke (1942)
reported a natural amphidiploid between A. cepa var ’Australian Brown’ and
A. 17sz var ’Nebuka’. This amphidiploid occurred spontaneously by natural
doubling of the hybrid chromosomes. The amphidiploid was pollen fertile, uniform
in appearance, perennial in habit and more vigorous than either of the original
parents. The haploid chromosome number was 16 and meiosis was quite regular.
Fragments, chromatin-bridges and micro-nuclei were observed but not as frequently
as in the diploid interspecific hybrids. The pollen fertility of the amphidiploid gave
new hope to onion improvement programs. Polyploidy was considered the possible
route to follow. This idea and optimism, however, did not last for long as the
amphidiploids were found to be poor seed setters in comparison to either of the
parents (Davis 1955). Nevertheless, the amphidiploids form very slight bulbs and are
used as edible green onion. In 1950, a highly pinkroot resistant, amphidiploid, green
onion cultivar, Beltsville Bunching (A. cepa var ’White Portugal’ x A. fistulosum var
’Nebuka’) was released to the seedsmen by the United States Department of
Agriculture and California Agriculture Experiment Station (Jones and Mann 1963).

A similar amphidiploid was developed by Davis (1955) by artificially doubling the
chromosomes of the cross A. ascalonicum (shallot) x A. fistulosum. Shallot is very
closely related to A. cepa and is now considered to be just another form of A. cepa
(Jones and Mann 1963, Jones 1990). The two cross readily and exhibit normal
meiotic behavior. Shallot differs from onion mainly that they are small in stature,
have slender awl-like leaves, and have small, ovate-oblong angular bulbs that break

up into distinct sets that cohere at the base (Cochran 1950).

11

In the mid - fifties Delta Giant, a triploid shallot, was released. It was derived from
the A. fistulosum - shallot amphidiploid backcrossed to shallot. Delta Giant is
prolific, tolerates hot weather and can be grown both earlier and later than regular

shallot cultivars grown in the South, especially in Louisiana (Jones and Mann 1963).

New Mexico State University (NMSU) has been actively involved in the integration
of A. cepa and A. fistulosum genomes since the late seventies. NMSU has used the
triploid Delta Giant as an A. fisadosum source in this integration research (Corgan
1985, Peffley 1986) and has succeeded in obtaining trisornic backcross progenies to
A. cepa (Corgan 1985, Peffley 1986, Cryder et a1. 1989). Scientists at NMSU and
now at Texas Tech (Peffley et a1. 1985, Peffley 1986, Peffley 1987, Peffley et a1.
1989) are currently developing isozyme marker maps using the trisornic alien
addition lines. They are using cytological and isozymatic information to keep track
of the integration of the genetic materials. At the same time molecular markers
(isozyme) are identified which are found in the trisornic addition line. These
scientists hope to accomplish gene flow between A. cepa and A. firtulosum at the
diploid level using these addition lines (Peffley et a1. 1985, Peffley 1986, Peffley
1987, Peffley et a1. 1989). The isozyme markers can play a significant role in the
screening of the backcrosses from the selfs in an open pollinated crossing program

using A. cepa as the female parent.

MATERIALS AND METHODS

In 1966, Q. P. van der Meer at the Institute of Horticultural Plant Breeding (IVT),
Wageningen, the Netherlands, began making interspecific crosses between A. cepa
and A. fistulosum, using Beltsville Bunching (allopolyploid) as the source for the
A. fistulosum genome. Parents and backcross (BC) generations up to BC3 were
saved and maintained over the next 20 years. In 1986, L. Ewart of Michigan State
University made a comprehensive cytogenetic study of the available A. cepa-
fistulosum germplasm, furthered the backcross generation to BC4 and made some
other crosses with the available materials at IVT (Ewart 1986). The current study is
a continuation of the experimentation started in 1986 with the germplasm brought
from IVT. The following studies were planned with the interspecific derivative

materials:

i. Advancement of the backcross generation: This study was initiated to
determine if the barriers of fertilization, which were initially present,
were broken by the current generation. Since the material was
brought from an extra long day period region, this required
application of a flowering synchronization procedure as described

later in the hybridization sub-heading.

ii. Cytogenetic observations: Such observations were deemed necessary
to determine whether initially observed chromosomal aberrations still
persisted or were significantly decreased, and to see if the materials

were comparable to regular diploid A. cepa.

iii. Genome integration: RAPD markers were used for this study to verify

if the interspecific genome was carrying A. fismlosum DNA.

12

13

iv. Disease screening: this study was initiated to determine if the
introgressed lines carried resistance / tolerance to Fusarium and

pinkroot, two important onion diseases.

Interspecific plants, 1253-1, 1253-2, 1253-3 and 1255, all in the backcross 4
generation were used in the study. The pedigree of the above plants is described in
the Appendix B. The following MSU A. cepa inbreds were used to make backcrosses
with the above interspecific materials - 611, 1731, 1849 and 2399.

Table 1 shows the pedigree of the materials used in 1989. The selfs are designated
as 9 throughout the manuscript. As is customary, the first pedigree listed in a
cross is the female parent. Table 2 shows the new designations given to the old
plants when their bulbs were repotted after they were used once in the crossing

process. Table 3 shows the crosses and selfs successfully made in 1990.

14

Table 1: lnterspeciflc Allium Germplasm Developed at Michigan State

 

 

University in 1989.
Cross Field #
611 x 1253-1 901
1253-1 6 90-2
1253-2 6 90-3
1253-3 0 90-4
1253-3 a 90-5
1255 a 90-6

 

Table 2: Field numbers used after repotting the 1989 Allium interspecific

 

 

derivative plants.
Original Pedigree New Field #
1253-1 89- 3
1253-2 89- 12
1253-3 89- 14

1255

89- 4

 

15
Table 3: Interspecific Allium Germplasm Studied and Developed at Michigan

 

 

State University in 1990.
Cross Field #1
894-2 0 90-1 1
894-3 0 90-12
8944 a 90-13
89-12-4 0 90-14
611 x 89-3-2 90-15
611 x 89-4-2 90-16
611 x 894-4 9017
611 x 89-12-4 90-18
2399 x 89-3-1 90-19
2399 x 89-4-1 90-20
2399 x 89-4-2 90-21
1731 x 89-3-2 90-22
1731 x 8944 90-23
89-4-1 x 1849 90-26

 

z The field # given in table 3 is the digree assigned to the seed when it was
mu in 1990 and 1991. Some of ese plants were used for cytogenetic and
D investigations.

Hybridization

A. cepa, as well as interspecific material seedlings were vernalized in a growth
chamber and were grown under environmental conditions designed to decrease the
seed-to-seed generation from two years to about nine months by bypassing the
bulbing stage. The generation reduction procedure was based on the results of the
extensive physiological studies done by Brewster (1977a, 1977b, 1983, 1985a, 1985b),
which was reported and used by Bacher (1989). Three major factors were
controlled: First the nutrition was drastically reduced and the photoperiod was
increased to 24 hours. Second, the plants were grown initially at 17°C for the first
100 days to initiate growth and then at 9°C for 40 days to vernalize the plants.
During these two temperature periods, the plants were grown under cool white light
(200 micromole per sq meter per second). This procedure by-passes the bulbing

phase allowing the plants to initiate flowering instead.

Pollination Procedure: The onion inflorescence is an umbel bearing 50 to more than
1000 flowers, though 300 - 500 is a more usual situation. A single umbel may
continue to flower over 3 to 4 weeks (Currah and Ockendon 1978). Though the
number of flowers opening initially is small, it rapidly increases with a peak between
4 and 10 days in which 50 or more flowers open in a single day (Currah and
Ockendon 1978). The opening of the flowers is strongly affected by temperature and
day-length (Masuda 1956). Stigrnatic exudates, though essential for pollen capture,
were not found to play an essential role in pollen germination (Currah and
Ockendon 1978). Stigmata have been found to be most receptive between the 3rd
and 5th day of opening of the flower (Moll 1954), and stigma receptivity is strongly
influenced by temperature (Chang and Strukmeyer 1976a and 1976b).

16

17

The umbels that were going to be used as seed parents were bagged as soon as the
first flowers started to open in the greenhouse. In most crosses, the interspecific
derivatives were used as pollen parents and were allowed to flower freely There
was one exception in the second year when 89-4-1, an interspecific derived plant was
used as a female and was bagged. A. cepa plants were used as females and were
emasculated twice daily and rebagged. Because of a limited number of heads, the
same heads were used for crossing and selfing in many cases. Crossing was
accomplished either by using a camel brush for pollination or by placing male and
female heads in a single bag along with flies which did the pollination. Some selfing

also was done in the same way when a single flower head could be used.

Disease Screening

Seedlings from selfs of the interspecific derived germplasm were screened for two
major onion diseases, basal rot and pinkroot. Basal rot, caused by Fusarium
oxysporum f. sp Cepae and pink root, caused by Pyrenochaeta terrestrii, are two very
devastating soil-bom onion diseases. Basal rot has long been recognized as a serious
soil-bom disease in the midwestem and eastern United States (Abawi and Lorbeer
1971a and 1971b). Pink root has long been a severe yearly problem in the southern
and southwestern United States, and it may be severe on midwestem organic soils
during periods of relatively high temperature (Gorenz 1949 and Walker 1961). Both
basal rot and pink root develop best under controlled conditions at soil
temperatures of 26-300C (Walker and Tims 1924, Gorenz 1949, Kehr et al., 1962).

18

Inoculum Preparation:

Pink-root: PR-4 and PR-6 strains of Pyrenochaeta terrestris were used in this
screening study. These inocula were originally developed at the University of
Wisconsin by Dr. Clint Peterson, a USDA onion breeder. In this study, they were
obtained from Dr. Mel Lacy of the Department Botany and Plant Pathology,
Michigan State University, who had obtained some of this material and prepared

the inoculum for the screening.

The inoculum is kept as infected muck soil at 4°C. When needed, it is incubated in
Czapek’s sucrose nitrate solution unshaken for 6-9 day. Then it is diluted to 1 / 20.
One ml of inoculum is used per gram of white silica sand (Dr. Lacy’s laboratory

protocol).

Basal-rot: F -593 (University of Wisconsin), F110A and F156B (from New York)
strains of Fusarium oxyspomm were used in this study. The inoculum was also
obtained from Dr. Mel Lacy of the Department Botany and Plant Pathology,
Michigan State University.

Ten days prior to seeding, the inoculum was placed on potato dextrose agar and
grown at 26°C. Three days prior to seeding, 7-9 mm discs were cut from the agar
and placed in 250 ml flasks containing 50 ml potato dextrose broth, one disk per
broth flask. Then the inoculum was increased on a rotary shaker at 24-26°C. After
72 hours of growth, the broth was mixed for two minutes in a blender at low speed.
Then it was centrifuged at approximately 3400 RPM for 10 minutes. The
supernatant was discarded, and the spores were resuspended in distilled water. The
spores were counted with a hematocytometer to determine the concentration so that

20,000 spores per gm of sand could be used.

l9

Growing medium preparation: Though the inocula were developed differently, the
rest of the screening procedure was much the same for both of the diseases. The
procedure was first developed for pinkroot (Krueger 1986) and currently is used for
both Fusarium and pinkroot (Gabelman 1988). In both cases, a spore concentration
of 20,000 per gram of sand was used. Two 50 x 20 x 11 cm stainless steel pans filled
with 20 kg of white sand were used for each of the inocula. The pans were sterilized
in an autoclave. A smaller pan was used as a control and as an indicator for normal
seed germination without any inoculum for each disease. The other two pans were
inoculated with the appropriate amount of spore suspensions prepared by adding
the necessary amount to distilled water. About 1 cm deep grooves were made 3 cms
apart across the sand, and were seeded with A. cepa and the interspecific derivative
seeds. Owing to limited supply of seeds from the interspecific derived germplasms,
only 10 seeds were planted per row. The pans were covered with aluminum foil and
placed in temperature-controlled water baths set at 26°C for pink root and at 24°C
for basal rot. The foil was removed after the seedlings had emerged and the
seedlings were grown under high intensity discharge lights using a 15-16 hour
photoperiod. When the cotyledons reached 1 cm in height, the sand temperature for
basal rot was increased to 26-28°C. The optimum temperature of growth for
Fusarium oxysporum is 26-30°C (Walker and Tims 1924, Davis and Henderson 1937,
Abawi and Lorbeer 1972). Plants were watered with distilled water as needed and
were fertilized with 20:86:16.6 N :PzK at 100 ppm after two weeks. The number of

seedlings were counted on the 10th and the 21st day after planting.

Experimental Design: Owing to lack of adequate seeds, two of the interspecific
derivative lines were replicated only once in the Fusarium (89-4-3 self and 89-4-4
self) and pinkroot (89-4-4 self) screening experiments. Completely randomized

design (CRD) does not impose any restriction on the number of replications per

20

treatment (Steel et al., 1980); and so CRD was chosen for the planning of the
experiment. However, the inclusion of singly replicated lines diminishes greatly the
resolving power of an analysis and the inequality of replications does not allow for
the separation of block effects. All the observations of an experiment were

randomized in a single pool for each experiment, as required by the design.

For the pinkroot screening, two replications of 89-4-2 and 89-4-3 selfs and one
replication of 89-4-4 self with four replications of each of three A. cepa inbred
checks, were used. For the Fusarium screening experiment, three replications of 89-
4-2 self and one replication each of 89-4-3 and 894-4 selfs with four replications of
each of three A. cepa inbred checks, were used. Each experiment had seventeen

total experimental units.

Cytology

Flower bud samples were collected in the morning fiom plants and were fixed in
6:3:1 ethanol:chloroform:glacial acetic acid (Peffley 1986) solution for about 24
hours. They were transferred to 70% ethanol for storage in a refrigerator. The
refrigerated samples were used as needed. The meiocytes were stained with aceto-
carmine. Metaphasel and anaphasel were the stages examined under a light
microscope. At least three slides were prepared for each sample and the whole field

of each slide was viewed for appropriate cells.

Random amplified polymorphic DNA (RAPD)

Two types of molecular markers have been found useful for developing detailed

genetic maps in plants - restriction fragment length polymorphisms (RFLPs) and

21

random amplified polymorphic DNAs (RAPDs). A majority of previous studies
have used RFLPs, which can detect genetic changes as base substitution and
insertions / deletions. RFLPs are often codominant but are useful only in the region
of low or single copy sequences (Burr et al., 1983; Tanksley et al., 1989). In 1990, a
new technique, RAPD, was developed (Williams et al., 1990). This technique relies
on the differential enzymic amplification of small DNA fragments using the
polymerase chain reaction (PCR) with arbitrary oligonucleotide primers (usually 10-
mers). Polymorphism results either from insertion / deletion in the amplified
regions or from base changes that alter primer binding causing the region not to
amplify. The procedure is rapid, requires only a small quantity of DNA which does
not need to be of high quality; and best of all, it does not require the use of
radioactive probes. Since no hybridization is involved, it can be used to detect
polymorphism in regions of highly repeated sequences, which is not possible with
the RFLP approach. RAPD provides dominant markers, since, polymorphism is

detected as the presence or absence of bands.

A RAPD study was done using arbitrary DNA primers to determine if there was a
polymorphism in A. cepa, A. fistulosum and the interspecific derived germplasm.
Total DNA was obtained from samples of each of the materials (A. cepa,
A. fistulasum and the interspecific germplasms). Six different lines of A. cepa, two
different sources of A. fistulosum genome, Odessa (A. firmlosum cultivar) and
Beltsville Bunching (amphidiploid) and three different interspecific hybrid lines
were used in the study. Two oligonucleotide primers (10-mers) were used to amplify
the DNA obtained from the above sources. Table 4 shows the dNTP sequence of

the oligonucleotide primers and Table 5 shows the plant materials used in the study.

22

 

 

 

 

 

Table 4: Sequences of the primer codes used in the RAPD analysis in the
Allium lnterspecific derivative development study.
Code Sequence
LE2 5’-GCACGATTI‘G-3’
RW2 5’-GGCCACCGTC-3’
Table 5: Various lines, accessions and crosses used in the RAPD analysis used
in the Allium interspecific derivative development study.
A. cepaz A. fistulosum Interspecific Hybrid
420 Odessa 90-26-6
611 Beltsville Bunching? 90-12-18
1731 90-11-9
1849
2399
8155

 

Inbreds used: 420 is a Univ. of Wisconsin line; 611, 1731, 1849, 2399
and 8155 are MSU inbred lines. All these lines were used in this
study, for making backcrosses as well as disease screening.

Beltsville Bunching is an amphidiploid containing 2n A. cepa and Zn
A. fistulosum genomes.

23

DNA Extraction: Samples were taken from newly emerging leaf tissue (5 g fresh
weight) of three interspecific germplasm, one A. firtulosum, one Beltsville Bunching
and six A. cepa inbreds. The tissue was immediately frozen in liquid nitrogen and

processed the same day.

Total genomic DNA was obtained by grinding tissue to a fine powder with a mortar
and pestle in the presence of liquid nitrogen. Approximately 30 m1 of extraction
buffer (15 ml of 100mM Tris.HC1, pH 8; 200mM Na2EDTA pH 8; 500 mM NaCl;
10 mM B-mercaptoethanol) was added to the cold mortar for each sample. The
ground leaf tissue samples were filtered through four layers of cheese cloth into 30
ml oakridge tubes (Du Pont). The samples were incubated at 60°C for an hour.
After cooling the tubes to room temperature, the tubes were filled with equal
volumes of chloroformzisoamyl alcohol (24: 1 v/v). The tubes were centrifuged at
10,000 RPM for 10 minutes. The aqueous layer of each sample was put in another
oakridge tube to which 10 ml of chloroformzisoamyl alcohol was added and then
centrifuged at 10,000 RPM for 10 minutes at 20°C. The top layer of each sample
was pipetted off into a clean corex tube (Corning), and 1/ 10 volume of 3M Na-
acetate and 3 /4 vol of ice cold isopropanol was added. DNA was precipitated by
gently inverting the tubes. The tubes were kept at -20°C for an hour and then
centrifuged at 10,000 RPM for 10 minutes at 4°C to pellet the DNA. The
supernatant was poured off and pellets were air dried until no odor of isopropanol
was apparent. The pellets were dissolved in 400 pl of 1 TE and the content was
transferred to Eppendorf tubes (Perkin Elmer). 20 ug of RNAase were added and
the tubes were incubated for an hour at 37°C. 40 ul of 3M Na-acetate and 800 pl of
ice cold 100% ethanol was added to the tubes. The tubes were placed on ice for half
an hour. The tubes were shaken to mix the layers and to precipitate the DNA. The
tubes were centrifuged at 13,000 RPM for 12 minutes at 4°C to pellet the DNA.

24

The supernatant was poured off and the tubes were filled with 70% ethanol and
again centrifuged for 5 minutes at 5,000 RPM. After discarding the supernatant
liquid, the pellets were air dried until no odor of ethanol was noted. 400 [11 0.1 TB
was added to the tube to resuspend the DNA. The tubes were stored at -20°C.

RAPD Procedure: Amplification reactions were performed in volumes of 25 ul
containing 10 mM Tris-Cl, pH 8.0, 50 mM KCl, 2 mM MgC12, 0.001% gelatin, 100
uM each of dATP, dCl'P, dGTP, dTTP, (Pharmacia), 0.2 M primer, 25 ng of
genomic DNA and 0.5 units of tag DNA polymerase (Perkin Elmer Cetus).
Amplification was performed in a Perkin Elmer Cetus DNA Thermocycler
programmed for 47 cycles of 1 minute at 94°C, 1 minute at 35°C, 2 minutes at
72°C, using the fastest available transitions between each temperature (Williams et
al., 1990). The amplification products were analyzed by electrophoresis in 1.8%
agarose gels at 45 volts for 20 hours, and detected by staining with ethidium
bromide. lambda / HindIII and lambda / DraI fragments were used as size

markers.

RESULTS AND DISCUSSION

Allium cepa is a biennial crop and normally takes 2 years to complete its life cycle.
A. cepa and the interspecific derivatives were grown under the cultural procedure
outlined in the Materials and Methods section. The seed to seed generation time
was effectively reduced to approximately nine months in most cases. Thus, the

crosses (and selfings) could be made in 1989 and 1990 consecutively.
Morphology

A. fistulosum differs from A. cepa in having a bunching (many) stem, round leaves,
short stature, oval inflorescence, non-bulbing stem (Figure 1a) and perennial habit;
on the other hand, A. cepa has a non-bunching (much fewer) stem, flatter or ovate
leaves, larger stature, spherical inflorescence, a bulbing stem and a biennial life
cycle. The BC3 plants had slight characteristics of A. fistulosum (bunching stem,
slightly formed bulbs, and rounder leaves), where as, the BC4 generation plants had
all the observable characteristics matching A. cepa (Figure 1b). Moreover, the
backcross generation plants had a distinct vigor and were easily distinguishable from
either of the parents. Figures 1 to 3 show the various morphological characters of

A. cepa, A. firtulosum and the interspecific germplasms.

The morphological development of the interspecific germplasm has proceeded as
expected in a backcross breeding program. Initially, bunching type stems with very
slight bulbing were common (up to BC3, Ewart, 1986). The BC4 plants were non-
bunching (most had a single stem) and visually it was not possible to distinguish
between derived bulbs and original A. cepa bulbs (Figure 3). These plants have
become biennial since BC4, while up to the BC3 generations, they had slight but
unmistakable properties of a perennial nature (Ewart, personal communication).

The leaf type has changed to the morphology of A. cepa.

25

(b)

 

Figure 1: A. fistulosum and inters ecific derivative (BC ) lants. (a)
A. fistulosum var Odessa. (b3 interspecific derivative, 9-4- (BC4 .

(a)

 

(b)

 

Figure 2: Inflorescence of A. firtulosum and interspecific derivative (BC4)
plants. (a) A. fistulosum var Odessa. (b) interspecific derivative, 89-4-
. The diameter of the interspecific inflorescence is more than twice

that of A. fistulasum and similar to A. cepa.

(a)

(b)

 

Figure 3: Bulbs of A. cepa and interspecific derivative (BC4 self) plants. (a)
A. cepa (b) interspecific derivative, 90-11.

Cytogenetic Observations

A. cepa and A. fistulosum have eight pairs of chromosomes each and Beltsville
Bunching, the amphidiploid, has sixteen pairs. The major difference between the
A. cepa and A. fistulosum chromosomes are size and chiasmata type. On an average,
A. cepa chromosomes are larger than A. fistulasum chromosomes and A. cepa
genome contains 33% more DNA than A. fistidosum genome. Randomized
chiasmata are characterized by the random nature of the crossing-over between the
homologous chromosomes. Such chiasmata show pairing closer to the end of the
arms forming characteristic oval (semi-circular) shapes. Localized chiasmata, on the
other hand, are formed by pairing of homologous chromosomes closer to the
centromere. Because of the tension near the centromere, chromosome arms are
stretched towards the opposite direction, developing the characteristic cruciform
structure. It is not entirely known that localized chiasmata are formed by non-
random (localized) crossing-over. If they are, the genetic combination in the
chromosome arms will remain constant from generation to generation and the arms
will act as super genes in the meiotic recombination. A. cepa form randomized
chiasmata (Figure 4a) while A. fistulosum form localized chiasmata (Figure 4b). The
Beltsville Bunching shows a mixture of both. Table 6 summarizes the chromosome

pairing relations as found in the parents used in this study.

Stickiness, multivalents, heteromorphic association and partial pairing were quite
common in the early generations of the interspecific germplasm development
(Ewart 1986). Most lines in the BC4, BC4 self and BC5 generation still show
occasional stickiness ring formation and multivalents at the metaphase I and
anaphase I chromosome configurations. Because of poor separation of

chromosomes, it was not possible to make exact chromosome counts in such cells.

29

30

The cells with reasonable separation of chromosomes, however, showed that the
chromosome number was 16. Figures 5 to 8 show metaphase I or anaphase I
chromosome configurations in various generations of the interspecific germplasm
development. Figure 5 (a) shows the anaphase I stage of the line 1253-1. The pairing
shows sticky chromosomes and the chromosome pairs are starting to separate.
Unusual pairing formation, some multivalents as well as univalents are observed.
Figure 5 (b) shows stickiness, multivalents, ring formation and cruciform chiasma at
metaphase I stage of 1255. Both BC4 plant cells, 1253-1 and 1255, could contain 8
pairs of chromosomes but it is extremely difficult to be absolutely positive because
of the overlapping and intermingling of the chromosomes. Figure 6 shows a
relatively distinct metaphasel chromosome configuration of 89-4-1, a BC4 plant.
Multivalent chromosomes, ring and cruciform formations are observed. Figure 7 (a)
shows a metaphase I chromosome configuration of 89-4-3, another BC4 plant. This
cell shows predominantly randomized chiasmata with various unusual pairing
configurations. One bivalent is paired only at one end of the chromosome arm
(chained chromosomes). Stickiness and univalents are also observed. Figure 7(b) is
a metaphase I chromosomal configuration of a BC4 self plant, 90-12. This cell
contains two univalents. Pairing between most chromosomes is partial and only a
few have complete pairing. Figure 8 is the metaphase I configuration seen in the
BC4 self plant, 90-14. The chromosome pairing of this cell is closest to a normal
pairing chiasmata configuration seen in the interspecific derivative materials
including BC5 plants. The cell shows both randomized as well as localized

chiasmata.

Multivalents, ring formation and stickiness are the results of interspecific meiotic
irregularities resulting from translocations, inversions and duplications (Peffley et

al., 1989). Plants used in this study were in the BC4 generation. These meiotic

31

formations are still carried over in the BC4, underscores the genetic exchange
between initial A. cepa and A. fistuloswn genomes. The severity of the irregularity,

however, is decreasing with the increase of generations.

The stickiness and multivalent pairing indicates common DNA in the non-
homologous chromosomes. Since Beltsville Bunching cells do show some bridges
and fragments (Jones et al., 1942, Jones et al., 1963), it is reasonable to assume
translocations and genetic exchanges have occurred between A. cepa and A.
fistulasum genomes in the Beltsville Bunching nucleus. Thus, the interspecific
triploid F1 developed by crossing Beltsville Bunching with A. cepa inherited one A.
cepa genome which already had genetic integration from the A. firtulosum genome.
Depending on the extent of A. fistuloswn DNA integration, the F1 can be fertile as
the male parent. When this triploid F1 is backcrossed to A. cepa, the two A. cepa
genomes, probably incompletely paired, recombined to reduce the chromosomal
mutational effects and fertilization could be accomplished. It is further envisioned
that the chromosomal architectural alterations were gradually reduced every
generation as the interspecific DNA recombined with the normalA. cepa DNA. The
stickiness and partial pairing observed in the current generations, will probably take
many generations of selfing before little or none is observed. Nevertheless, if the
pairing is good enough not to warrant a significant decrease in the fertility, one may

consider such meiotic chromosomes as stabilizing.

The deleterious effects of the chromosomal mutations in the crosses between
normal A. cepa and A. fistulasum were strong enough to affect fertility to a great
extent. Early integration efforts (Emsweller et al., 1935, Emsweller et al., 1945,
Cocharan 1950, Kehr 1957) did not succeed because of the sterility of the crosses.
The sterility was due to a high level of meiotic irregularities caused by strong

chromosomal mutations experienced in the interspecific hybrids. Since Beltsville

32

Bunching is an amphidiploid, the genetic integration between its genomes would be
expected to be gradual and in small doses. Thus, the development of fertile
interspecific hybrids using Beltsville Bunching was possible because of the presence
of the original A. cepa genome in the Beltsville Bunching nucleus. If the speculation
about gradual genetic integration between Beltsville Bunching genomes is correct,

the A. fistulosum genome should also have reciprocal integration from A. cepa.

The stickiness was severe up to the BC3 generation, it still persists, though the
severity has decreased with the increase of generations. Some cells, however, were

found to have near normal pairing ( 90-14, in figure 8).

Two to four univalent chromosomes were found in some of the metaphase I cells of
the interspecific plant materials. Occurrence of univalents does not necessarily
mean that the chromosomes were unpaired, since the chromosomes may have
separated prematurely (Swanson et al., 1981). Univalents, however, do increase the
chances of nondisjunction and should be considered unfavorable structures to a
normal diploidy development. Cells with univalent metaphase I and anaphase I
stage chromosomes were counted in the BC4, BC4 self and BC5 generations. Table
7 summarizes the observations. The results, by generation, show that 26% of the
BC4 cells had two to four univalents, 15% of the BC4 self cells had two univalents
and only 7% of the BC5 cells had two univalents. It seems genetic recombination
(from selfing and crossing with A. cepa) is decreasing the incidence of univalents
probably by increasing homology between the interspecific chromosomes. This
directly translates to progress towards development of diploids with regular meiotic

behavior in the currently developed Allium germplasm.

33

Table 6: Chromosomal pairing relationships in the parents used in the
Interspecific Allium Germplasm study.

 

 

Plant Paired chromosomes Chiasmata
A. cepa 8 pairs randomized
A. fistulosum 8 pairs localized
Beltsville Bunching 16 pairs mixture

 

Table 7: Univalents in some of the interspecific Allium germplasm
metaphase 1 cells.

 

 

# of # of
Line # Generation Cells Univalents
1253 BC4 5 0
2 2
1 4
1255 BC4 6 0
7 2
894-1 BC4 3 0
1 2
1 4
894-2 BC4 9 (2)
1
89-4-3 BC4 6 0
2 2
90-3 BC4 0 Z (2)
90-12 BC4 9 g 3
90-14 BC4 o 5 0
90-18 BC5 6 0
90-23 BC5 8 g

 

 

(a)

 

(b)

 

 

Figure 4: Metaphase I chromosomes of (a) A. cepa showing characteristic distal
, chiasmata (b) A. firmlosum showing localized (proximal) chiasmata.

35

(a)

(b)

 

Figure 5: Chromosomes of the Allium interspecific derivatives, showing meiotic
structures in (a) 1253-1 at anaphase I (b) 1255 at metaphase I. CC =
chained chromosomes, CF = crucr orm chiasmata, RF = ring
formation, M= multivalent, SS = sticky spot.

 

Figure 6: Metaphase I chromosomes in the Allium interspecific derivative giant,
89-4- (BC4) showinfiFmeiotic formations at metaphase I. =
cruciform chiasmata, = ring formation, M = multivalent.

 

37

(a)

0))

Metaphase I chromosomes in the Allium interspecific derivative BC4
and self plants. (a) 89-4-3 showing majority of chiasmata being
randomized type. and Nib) 90-12 s owing two univalents. CC =
chained chromosomes, = multivalent, U = univalent.

38

 

Figure 8: Metaphase I chromosomes" in the Allium interspecific derivative, 90-
4, a BC4 self, sh owing complete 8 bivalents. with a combination of
randomized and localized chiasmata.

RAPD Analysis

The original parents of the interspecific derivative lines (Beltsville Bunching and
A. cepa) were from a European collection. Neither the actual parental plants nor
their varietal representative plants were available for this study. An accession of
Beltsville Bunching and an A. firmlosum cultivar Odessa, were used as the
A. fistulosum DNA source. Six MSU inbreds, all of which were used in the disease
evaluation experiment, were employed as the A. cepa DNA source. The assumption
was that RAPDs common in A. firmlosum but unique to A. cepa could be used to
document introgression in the interspecific lines. The populations used in this study
are BC4 selfs and theoretically carry over 96% A. cepa DNA.

The absence of a parental RAPD in an interspecific hybrid does not necessarily
mean that the hybrid is not carrying the non-recurrent parental genetic materials;
however, the presence of it does indicate that the hybrid contains a portion of the
non-recurrent genome. In the process of genetic recombination, the region
producing a specific RAPD may itself get recombined or get lost. But the likelihood
of a region similar to the one present in the non-recurrent parent developing de
novo in the hybrid is very small. So, if the interspecific germplasms display RAPDs
from A. fistulosum, there is a very high probability that the germplasm also carries a
portion of the A. firtulosum genome.

Figure 10 is a schematic drawn based on Figure 9 which shows the ethidium
bromide gel for the primers RW2. This primer uncovered polymorphism in the
parental DNA (A. cepa and A. fistulasum genome sources) and the unique RAPDs
present in A. fistulosum (Odessa and Beltsville Bunching) were also found in the
interspecific hybrids. Because the primer RW2 showed polymorphism between

39

40

A. cepa and A. fistulosum DNA, it shows strong potential as a genetic fingerprint of

interspecific Allium germplasm.

In total, 17 RAPD marker loci were identified. Most of the A. cepa lines showed 6
unique RAPDs (3,5,8,10,12 and 13), although line 611 had RAPD 4 which was also
present in A. fistulosum and the hybrid lines. Odessa, the other theoretical parental
line (A. fistulosum ), had 9 unique RAPDs. Interestingly, all of the RAPDs found in
A. fistulosum, were missing in A. cepa with the exception of 2 and 14, which were

present in both A. cepa and A. fistulasum lines.

Beltsville Bunching, the amphidiploid, carries both A. cepa and A. firtulosum
genomes. As expected, it showed a high number of RAPDs, 12 out of the 17 found
in the diploid species. Not surprisingly, 5 of its RAPDs were the ones found in

A. cepa, another 5 were the ones found in A. fisndosum and two were common to

both.

The interspecific lines, like A. cepa lines, showed very little polymorphism. They all
contained 9 to 11 RAPDs. Lines 90-26-6 and 90-12-18 carried RAPDs 2, 3, 4, 5, 6, 8,
9, 12, 13, 14 and 16. The third interspecific line 90-11-9 did not have RAPDs 6 and
16.

Beltsville Bunching had 5 of the 9 unique RAPDs found in Odessa. The interspecific
line also carried 5 of the 9 unique Odessa RAPDs. The interspecific derivative lines
are BC4, which is an advanced backcross generation. The presence of 5 out of 9
unique RAPDs in the BC4, highlights the homogeneous integration of the
A. fistulasum DNA into the A. cepa genome. Although the lines examined were not
among the original parents, it is possible that the RAPDs appearing to be species
specific in these studies would also be found in both species in a broader germplasm

survey.

41

LE2, the other primer used in this study (Figure 11), produced only a single RAPD
specific to the A. fistulasum genome. Unfortunately, not all lines of A. cepa were

amplified by this primer.

The RAPDs are amplified 200-2000 base pair long DNA sequences. Such sequences
may happen to be in the proximity of a morphological locus. The possibility that the
amplified region may fall close to or within the sequence of an economically
important gene provides us with an opportunity that the RAPD can be used to select
the linked economically important genotype in a backcross program. This is
specially valuable when the gene being introduced is a recessive one. Traditionally
such a situation requires the breeder to self the cross (BC1, BC2 etc) in order to
identify the individual carrying the gene. Linkage, closer than 25 cM has been shown
to be identified by the amplification (Michelrnore et al., 1991). Indirect selection
using a closely linked RAPD will accelerate the breeding process by cutting the

number of generations needed in half by avoiding the selfing generation.

42

 

Figure 9:

RW2 primer derived ethidium bromide gel showing polymorphic
bands in A. cepa, A. tulosum and interspecific derivative DNA. e
left most lane IS mar er lambda DNA. Next six rows are interspecific
derivatives (1-2:90-26-6, 3-4:90-12-18, 5—6z90-11-9), next two rows
Beltsville Bunching (7-8 , next two rows Odessa(9-10) and last fifteen
rows are A. cepa DNA 11-1228155, 13-14:1731, 15:611, 16-17:420, 18-
19:2399, 20-21:1849, 22-23:611a, 24:Water, 25:611). Left lane arrows
show A. fistulosum s ecific RAPDs while right lane arrows show
A. cepa specific RAP s. Bidirectional arrows show non-segregating
(commom to both) RAPD loci.

43

 

 

 

 

90-26 90-1 1 Odessa I731 420 1849
90-12 BB. 8155 611 2399 611a

l -- l
2------------2
3---- -------3
4----- - 4
s---- -------5
6-- - 6
7 -- 7
3---- -------8
9--- - 9
10 -------10
ll - ll
12-—-- -------12
l3---- -------13
14--------_---l4
l5 - 15
l6-- -- 15
17 -- 17
Hybrid Lines jar] Allium cepa Lines

 

 

Figure 10: A schematic showing the banding pattern shown by RW2 primer in
the RAPD analysis of the interspecific Allium germplasm. Only one
replication of the two replications r sample is shown. Lane 611a is
another DNA sample of ' e 611. ' e preparing the schematic, very
faint bands were not considered.

 

Figure 11:

LE2 primer derived ethidium bromide gel showing pol orphic bands
in A. cepa, A. firtulosum and interspecrfic derivative NA. The right
most lane is marker lambda DNA. Next six rows are interspecrfic
derivatives (1-2:90-26-6, 3-4:90-12-18, 5-6:90-11-9), next two rows
Beltsville Bunching (7-8), next two rows Odessa(9-10) and last six
rows are A. cepa DNA (11-12:1731, 13-14:2399, 15-16:1849). The
arrow shows the A. fistulosum specific RAPD.

Fertility Improvement

Because of low fertility in the interspecific F1, F2 and BC1 generations of earlier
A. cepa and A. fisttdosum integration projects, they were abandoned without
utilizing the developed materials in a breeding project. These interspecific hybrids
greatly suffered from the sterility due to the meiotic irregularities of the crosses. The
level of seed set in the earlier projects ranged from 0 to 0.1% in a controlled
pollinated environment (Emsweller et al., 1935, Emsweller et al., 1945, Cocharan
1950, Kehr 1957 and Corgan 1985). The seed set even in the Open pollinated
environment would reach only up to 2% (Emsweller et al., 1945). In addition, in the
course of the current research, various A. cepa lines were used to cross with
A. fistulosum (var Odessa) in an attempt to obtain interspecific F1 seeds. In spite of

attempting hundreds of pollinations, no viable seeds were obtained.

For generation advancement, the interspecific germplasms were not only
backcrossed but also selfed. In general, selfing produced more seeds than did
crosses. Tables 8 and 9 show the seed set result of BC4 selfs and BC5 respectively. It
seems by the BC3 generation, that seed set begins to increase significantly. The
point to note here is that the severity of the barriers which were expressed in the
initial interspecific F1, BC1 and F2 generations, are greatly decreased. In onions,
traditionally artificial (emasculated) crossing does not yield a high percent of
success. The crossing results with the current interspecific germplasms have become
encouraging by the BC4 generation. The seed set of BC4 selfs, though not quite as

good as for the A. cepa selfs, has shown much improvement.

45

46

Table 8: Seed set in the 5th generation (B sell) interspecific germplasm
development between A. cepa and A. .

 

 

# of florets
Cross pollinated # seeds % success
89-4-20 33 73 73.7
89-4-3 o 33 49 49.9
8944 0 33 39 39.4
89-12-4 0 17 4 7.8
Mean - - 47.4

 

Table 9: Seed set in the 5th generation (BC ) interspecific germplasm
development between A. cepa and A. osum.

 

 

# of florets

Cross pollinated # seeds % success
611 x 89-3-2 10 2 6.7
611 x 89-4-2 10 2 6.7
611 x 89-4-4 10 1 3.3
611 x 89-12-4 10 12 40.0
2399 x 89-3-1 10 3 10.0
2399 x 89-4-1 10 14 46.7
2399 x 89-4-2 10 13 43.3
1731 x 89-3-2 10 1 3.3
1731 x 89-4-4 10 5 16.7
89-4-1 x 1849 10 11 36.7

Mean - - 21.3

 

47

The BC4 self seed set ranged from 8% to 73% with a mean at 47%. There was a
variation for seed set in different lines. In general, lines with the 89-4 pedigree
produced more selfed seeds than line 89-12-4 which had only an 8% selfing success

rate. Line 89-4-2 with a 73% success rate was the highest self seed producer.

The BC5 generation was developed using A. cepa as the female parent. Only one
cross was made with 89-4-1, a BC4 plant, used as the female. The seed set rate for
the crosses ranged from 3.3% to 46.7% with the mean at 21.3%.

Four A. cepa inbred lines were used in the BC5 development. The mean success
rate of the A. cepa lines was quite variable. The lines could be divided into two
distinct groups. Lines 1849 and 2399 were among the high seed set group with 1849
yielding 36.7% and 2399 close behind at a 33.33% success rate. The low yielding
group consisted of the 611 and 1731 inbreds. Line 1731 was last with a seed set rate
of only 10% while 611 was not too far off at a 14.17% success rate. The high seed set
level of the lines 1849 and 2399 indicate a possibility that the A. cepa lines used in
early backcrosses (up to BC3) may have had similar genetic background as these
two A. cepa lines. It could partially explain why lines 1849 and 2399 scored
dramatically better than the other two A. cepa lines used in the study.

In general, interspecific lines with the 894 background produced the best results,
not only in selfing but also in crossing. The mean selfing success rate of these lines
was 54.20% while the mean crossing success rate was 25.56%. Interspecific line 89-
4-1 was not used for selfing but was used in crossing reciprocally. In either crossing,
it had a high crossing success rate, 36.7% as the female and 46.7% as the male
parent. It seems reasonable to assume that lines developed from 89-4 pedigree
should have a relatively low level of aneuploidy and chromosomal mutation. Figures

5(b), 6 and 7(a and b) show chromosomal configuration of such individuals at

48

metaphase I or anaphase 1. Though there seems to be some stickiness between non-
homologous chromosomes, probably the chromosome architecture in the BC4
generation has recombined and pairing properties have improved enough to provide

the level of fertility obtained in this study.

Lines related to the 1253 pedigree did not do as well as lines with the 89-4 pedigree.
Only one line of the 1253 pedigree, 89-12-4 was selfed and produce only 8% selfs.
The mean crossing rate of such lines was 15 %. Line 89-12—4 scored poorly in selfing
but scored high when it was used as the male parent with a 40% seed set rate. This
indicates the possibility that the plant 89-12-4 may be a trisornic and / or may have a
significant chromosomal mutation which severely hampers its performance as the
female gamete producer. Figures 5(a) and 8 show the metaphase I chromosomal
configurations of some plants of 1253 pedigree. Some individuals of this pedigree
show stickiness of non-homologous chromosomes and at the same time others show
almost regular pairing structures (Figure 8). It seems some lines of this pedigree, as
indicated by the their seed set, still have a high level of aneuploidy and / or
chromosomal mutation that can affect their performance as female fertile gamete

producers.

Disease Screening

Because of an insufficient number of seeds, only the selfed seeds of 89-4-2, 89-4-3
and 8944 could be screened for disease resistance. The A. cepa lines, 6701, 1849
and 611 were used as checks in the Fusan'um disease screening. In the pink root
screening, the A. cepa lines 420, 8155 and 611 were used as inbred checks. Tables
10, 11 and 12 show the result of the Fusarium, pinkroot and control (without
inoculum) screening respectively. The survival percent of the seedlings was adjusted
for the germination percent assuming that germination is independent of inoculum
effects. Thus, the survival is the ratio of the number of surviving seedlings on the
21st day to the number of seeds sown, and the adjusted survival is the ratio of the
number of seedling surviving on the let day to the number of seedlings germinated
on the 10th day.

All the analyses related to the Fusan'um and pinkroot screening were performed on
arcsine transformed data (Steel and Torrie 1980). Treatments with single replication
were not included in the AN OVA calculation. Such treatments, however, are

presented in the tables to facilitate comparison with other treatment means.

Table 12 shows the germination, survival and the adjusted survival percentage
observation of all the A. cepa lines and a single interspecific derivative 89-4-3 self
used in the study. The values are the observed values and not means as only a single
replication of the control treatments was used due to too few interspecific seeds. No
definite conclusions can be made because of lack of adequate replication, however,
there was variation in the germination percent between lines. Even in the absence of
any disease inoculum, some seedlings died after germination. An unexpected
observation was that the interspecific derivative 89-4-3 self which consistently
germinated 100% in the disease environment, germinated only 60% in the

controlled benign environment.

49

50

 

 

 

Table 10: Means of Fusarium screening observations on selected A. cepa and
interspecific germplasm lines!.

Line #Rows Germtn. %Y Survival%¥ Adj.Surv%Y

6701 4 82.50 50.00 b 60.60 b

1849 4 77.50 2.50 c 3.23 c

611 4 90.00 80.00 a 88.88 a

894-2 0 3 83.33 70.00 a 84.00 ab

89-4-3 a" 1 100.00 80.00 80.00

8944 e" 1 40.00 10.00 25.00

I The calculations were done on arcsine transformed data and pairwise

comparison was done using s
were not included in the AN 0 A calculation.

y Means followed
probabili
x not inclu ed in the AN OVA calculation.

level.

cific lsd for the pair. Single row treatments

by the same letter are not significantly difl'erent at the 0.05

Means of pinkroot screening observations on selected A. cepa and

 

 

Table 11:
interspecific germplasm linesz.

Line # Rows Germtn.% Survival%¥ Adj. Survival%
420 4 92.50 ab 92.50 a 100.00
8155 4 37.50 c 27.50 c 73.33
611 4 92.50 ab 85.00 ab 91.90
89-4-2 o 2 75.00 b 60.00 b 80.00
89-4-3 e 2 100.00 a 90.00 ab 90.00
89-4-4 0" 1 50.00 30.00 60.00

 

The calculations were done on arcsine transformed data and pairwise
comparison was done usin s cific lsd for the pair. Single row treatments
were not included in the O A calculation.

Means followed by the same letter are not significantly different at the 0.05
probabili level.

not inclu ed in the AN OVA calculation.

51
Table 12: The control screening observations on selected A. cepa and

 

 

interspecific germplasm lines.
Line # Seed Sown Germination% Survival% Adj. Sur%
420 10 80.00 70.00 87.50
611 10 100.00 80.00 80.00
1849 10 90.00 60.00 66.67
6701 10 90.00 60.00 66.67
8155 10 50.00 50.00 100.00
894-3 9 10 60.00 60.00 100.00

 

Tables 10 and 11 show Fusarium and pinkroot screening respectively. The data are
presented as a mean percent germination, survival and adjusted survival of the lines
in both the environments. For Fusarium it does not matter whether the data are
adjusted or not. The ranking of the lines / derivatives in their expression of
resistance remains the same. The lines 611, 89-4-2 self and 894-3 self showed the
highest level of resistance. The line 6701 which was considered the best resistant
line available, showed less resistance and lines 1849 and 894-4 self showed the
lowest level of resistance. The interspecific line 89-4-4 probably does not carry the

resistance gene.

The pinkroot screening result is affected by the germination adjustment. If the
unadjusted data is used, lines 420 and 89-4-3 self performed the best, lines 611 and
89-4-2 self were next and as expected the susceptible line 8155 performed the worst
as did the 89-4-4 self. Overall, the A. cepa lines behaved as expected. Nevertheless,
if the survival is adjusted for germination, two things happen. The experimental
error became higher and the range of means narrowed down. The test did not find a

significant difference between the means. The adjusted survival percent data,

52

therefore, suggest that there was no difference in the expression of resistance to
pinkroot between any of the lines / interspecific derivatives used in the study. The
germination percent was closely reflected by the unadjusted survival percentage, and
the lines showing better resistance also showed better germination. The original

screening data is graphically presented in the figures 11-13.

Table 13 shows the overall germination percentage of the pooled observations from
the three experiments. There was a significant difference in the germination
percentage. The interspecific derivative selfs 89-4-2, 89-4-3 were among the best and

the third 89-4-4 was among the worst germinators.

In conclusion, interspecific selfed lines 89-4-2 and 894-3 show excellent potential
for resistance / tolerance to Fusarium. The interspecific derivative, 89-4-3 was one
of the high tolerant lines to pinkroot and 894-2 showed tolerance equivalent to the
one showed by the normal A. cepa line. When their score, however, was adjusted

for germination, there was no significant difference between scores.

53

Screening Results
Fusarium

 

 

 

 

80 - """" ‘I"‘ ‘
,.. .,.,
.... ...
... ....
.... ...
... ....
.... ...
... .,,.
‘I‘l‘l‘ I I‘I‘l‘l
- s o s s \ \ \ \ \ \ ‘
... .... ....
.... .,.. ...
,J(‘ («(¢ .¢«,
. . , .
”.1. $3.35 ;.;.;.;
.... .... ...
.. _ .., .... ....
.... .... ..
..., .... ....
..,
:‘:’:l: :’:’:’:‘ ’\l\'\’
d) .. .
\ ~ \'\ \’\’\’\l ’\’\’\l
_ .... ..,. ....
.... .... ..,. ...
.... ..., ..., ....
.... .... ..., .,.
.... .,. ,.,. ..,.
.... .... ,... ...
,,.. ... .... ....
,.,. .... .... ...
,,.. .... .... ....
.,.. .... .... ...
- ,,.. ... .... ....
. .. .... .... ..,
,,., .,.. ,... ....
. ,, .... .... ...
..,, ... .... ....
.... .... .... ..,
,‘.‘,‘.‘ ‘.\,,.‘, ‘.‘.‘.‘. .‘.‘.‘.
.... .... .... ....
.... .... ..., ...
_. («a (3‘... «we
.... ... ,... .,..
.... .... .... ..
.... ... .... ...
.... .... .... ,
.... ... .,.. ....
.... .... .... ..
.... .... .... ...
.... .... t... ...
_ .... .... .... ....
.... .... .... ...
..., .... ..., ....
.... .... ..,. ...
,... ... ..,. .... . .
.... .... .... .t.
.... .... .... ....
..., .... .... ..,
,.., ..., .... ,...
I‘I‘I‘l‘ ‘I‘l‘l" ‘I‘I‘l‘l I‘I‘I‘t
- .... .... .... ,.. ,... . .
.... .... ..,. .,.. ..,.
.,., .... .... .., ....
..,. ... .,,. .... ,...
..,. t... .t.. .,. .... .
.‘.‘.‘,‘ ~.‘.‘,‘. ‘.\.\,‘. .‘,\.\, .‘,\.\.‘
I I I I I I I I I I I I I r t I '
. .. . .
-4

 

6701 1849 611 89-4-263 894-369 89-4-
Lines

Figure 11: Bar graph showing survival and ermination percentages of all the
lines used in the basal rot screemgng study using Allium interspecific
derivatives and normal A. cepa lines.

54

 

 

 

 

Screening Results
Pinkroot

 

"”"”""”"”I —

ion
1

ermina
urvrva

S

t

G
Adj.
rvival

   

Su

 

2.8.55

 

11 89-4-239 89-4-3EB 89-4-469
Lines

6

8155

420

es of all the

on percentag
tudy using Allium interspecific

ti

rmina

survival and ge
ng s

pinkroot screem

derivatives and normal A. cepa lines.

Bar graph sho
lines used in the

Figure 12

55

Screening Results

Control

 

-
"”””’l _

 

 

 

 

(

o»

(
r’””""””””"”"’l’ll'l4

 

 

 

~
~

\ a

\ \

4

 

 

“acumen

8155 89

Lines

e

rmrnatron percentages of all the
g study using Allium interspecific

Bar graph showmg survrval and g
lines used in the control screenin
derivatives and normalA cepa lines

Figure 13:

56

Table 13: Overall germination of selected A. cepa and interspecific lines used in
the Fusarium, pinkroot and control screening experiments!

 

 

 

Line # Rows Germination%?
420 5 90 a
611 9 94 a
1849 5 80 b
6701 5 84 ab
8155 5 40 c
89-4-2 e 5 80 b
89-4-3 0 4 90 a
8944 o 2 45 c
z The calculations were done on arcsine transformed data and pairwise
comparison was done using specific lsd for the pair.
y Means followed by the same letter are not significantly different at the 0.05
probability level.

Genetic Model: The screening data also provided an opportunity to speculate on
genotypes of the parent plants of the interspecific derivatives used in the study. All
the derivatives used were selfs and only two phenotypes could be considered for the
resistance character - alive or dead. Genes providing the resistance are likely to
have introgressed from the A. firmlosum, and the mode of genetic action of these
genes may not be the same as found for the resistance genes in the A. cepa lines. For
Fusarium resistance in A. cepa a two-gene incomplete dominant system has been
suggested (Bacher 1989). Jones and Perry (1956) have reported a single recessive

gene for resistance to pinkroot in the A. cepa system.

Table 14:

Possible model of a single dominant gene (P) for pinkroot resistance
in the Allium interspecific hybrid development study.

 

 

 

 

 

Observed
Postulated WW FXPected 2
Line Genotype Alive De Ratio X
89-4-2 Pp 12 3:1 2.40
89-4-3 PP 18 1:0 -
89-4-4 pp 3 0: 1 -
Table 15: Possible model of a single dominant gene (F) for Fusan'um resistance
in the Allium interspecific hybrid development study.
Observed
Postulated Expected 2
Line Genotype Alive Ratio X
89-4-2 Ff 21 3:1 0.40
89-4-3 FF 8 1:0 -
89—4-4 ff 1 0:1 -

 

58

Table 16: Possible model of a single dominant gene (R) for overall resistance in
the Allium interspecific hybrid development study.

 

 

Observed
Postulated f f Expected 2
Line Genotype Alive ead Ratio X
89-4-2 Rr 33 17 3:1 2.16
89-4-3 RR 26 4 1:0 -
89-4-4 rr 4 16 0:1 -

 

Tables 14 and 15 present parental lines of the interspecific selfs used in the disease
screening and the observed number of selfed seedlings which were alive (and also
which were dead) 21 days after sowing. The tables also present the assumed
genotypes of the parental lines along with the calculated X2 values for the expected
ratio of the alive and dead phenotypes in the selfed progenies of the heterozygous
parents. Table 14 presents the data for the pinkroot and table 15 the same for the
Fusan'um screening experiment respectively. A simple Mendelian model with a
dominant resistance gene is assumed in both the cases. The tables also show data
about assumed genotypes, dead and alive seedling at 21 days, and the expected ratio
according to the model and the calculated X2 values. The chi-square value for selfed
homozygous parents could not be calculated for the reason that a ratio of 1:0
requires division by zero. The number in the zero class, however, is low enough to
be attributed to chance alone. The calculated X2 values in all the cases are not
significant at one degree of freedom. The genotype for 89-4-2 is assumed to be
heterozygous, 89-4-3 to be homozygous dominant and 894-4 to homozygous

recessive. The resistance is assumed to be a dominant gene.

59

In nature, economic use of available resources is necessary in order to survive the
natural competition. The survival of fittest theory implies that the one with the most
efficient utilization system of resources will survive and produce progeny. With this
axiom, it is possible that both the P and F genes are the same R gene and thus are
pleiotropic. Table 16 shows the combined data for both the diseases. Again
assumptions about the model, as well as, the individual genotypes are the same as
above. The data fits well with the model, and the X2 values are not significant. It is
possible that the same gene is offering the resistance to both the diseases though the

approach of imparting resistance in each case may not be the same.

The empirical experience and knowledge about the fungal disease infection
physiology indicate that Fusarium and pinkroot diseases are multistep complex
processes. A lot of information about the host pathogen interaction is still not clear.
Nevertheless, current studies indicate that fungi, specially Fusarium, accumulate
cutinase on their outer cell surface when attached to a host. This cutinase plays the
key role in the penetration of the fungal haustoria (Kolattukudy 1985, Woloshuk et
al., 1986). Plants respond to such attack by accumulating hydrolyzing enzyme
chitinase in the surface cells (Mauch et al., 1989, Benhamou et al., 1990). Chitinase
kills the fungus by hydrolyzing its cell wall. In the particular case of this experiment,
the young seedling probably can not withstand the pathogen once it is penetrated.
The only way for the seedling to survive would be avoid the pathogen by stopping it
from penetration. The resistant / tolerant seedlings may have survived by either
hydrolyzing the fungal cell wall and / or by producing one or more of known
antifungal secondary metabolites like alkaloids, flavonoids or saponines (Wink
1988), which, when absorbed by the pathogen, kills it. Both these approaches are
general plant defense mechanisms. Since A. firtulosum is known to be resistant /

tolerant to a wide range of pathogens, it is not unreasonable to assume that it may

60

possess a few general resistance genes rather than many specific resistance genes.
Since the same interspecific line shows resistance / tolerance to both Fusarium and
pinkroot, 89-4-2 and 894-3 selfs; and the same line, 89-4-4 self, show susceptibility
to both the diseases, it is hypothesized that a single gene could be pleiotropically

imparting the resistance in both the cases.

 

SUMMARY AND CONCLUSIONS

The sterility of the interspecific F1s between A. cepa and A. firtulosum has stopped
many projects trying to introduce desirable characters from A. fistulosum into
A. cepa. The current study is a part of a long term hybridization program which
appears to have been successful in introgressing some part of the A. fistrdosum
genome into the A. cepa background. The initial A. firmlosum genome source in the
diploid interspecific hybrid in this study, was the amphidiploid (2n = 32), Beltsville
Bunching. It seems that the polyploid environment of Beltsville Bunching provided a
better environment (probably by providing a buffered environment) for genetic
exchange between A. cepa and A. fistulasum genomes than one provided by the
direct cross of A. cepa and A fistulosum. Cytogenetic studies show aberrations such
as sticky chromosomes, heteromorphic chromosomes, fragments and chromatin
bridges, all of which are gradually decreasing in subsequent backcross and selfed
generations. These cytological formations which were important to verify the genetic
introgression, adversely affect the fertility of the developed lines. Many developed
interspecific lines show a high level of fertility indicating that the meiotic
chromosomal aberrations have significantly decreased probably due to
recombination by selfing and backcrossing with A. cepa. Some lines, however, do
indicate strong chromosomal mutations by expressing reduced fertility when used as
a female parent. The interspecifically developed plants morphologically resemble
A. cepa. By the 4th generation, they could not be distinguished externally from
A. cepa except for hybrid vigor. This study demonstrated that they are carrying
A. fistulasum genetic materials and, in many cases, with significantly improved

fertility.

61

62

The following conclusions are drawn from the study:

a.

Interspecific derivatives between A. cepa and A. fistulosum have been
developed which morphologically resemble A. cepa.

Many derivative lines show enhanced fertility and seem to approach

the fertility of A. cepa.

Meiotic irregularities are decreasing with the increase in the
generation level. Progress towards a regular meiosis is reflected by the

increased seed set by many developed interspecific lines.

Some of the derivatives carry a high degree of resistance / tolerance

to Fusarium and pinkroot.

The limited data indicates the mode of genetic action of the resistance
genes to be dominant for both the diseases. It is possible that a single

gene is pleiotropically imparting resistance in both the cases.

The derivatives are in an advanced stage ready to be directly used in a

breeding program to develop inbreds.

The derived plants could possibly be used to transfer additional
characters from A. firmlosum, as they should be easier to cross with
both A. cepa and A. fistulosum because of the genetic integration of

the two species.

63

The following further studies are suggested:

ii.

iii.

The seeds of the interspecific derived materials should be
increased and a comprehensive screening study should be done
for all the diseases A. fistulasum is known to show resistance.
This will identify the potentials of different derivative lines
which are already developed. Such a study would also find if
further crossing with A. fistulosum is needed to transfer
characters which were lost due to genetic drift / attrition in the

interspecific recombination.

With the results from section i, promising lines should be used
to introduce their genes to the elite inbreds of onion. The
crossing should be made in an open pollination block. Plants
grown from the seeds harvested from the A. cepa lines could be
screened with RAPD (using RW2 primer) for presence of
A. firmlosum DNA.

New A. cepa x A. fistulasum integration attempts using other
selected Beltsville Bunching lines could be initiated. An open
pollination crossing block procedure is recommended for
possible maximum seed set. Then RAPD (with RW2 primer)
could be used for screening of the selected plants from the
crosses. With this procedure, accelerated development of
interspecific hybrids could be achieved. Cytogenetic
observations are not necessary, but they would provide useful

information about the meiotic behavior of the cross.

APPENDICES

 

Appendix A: Picture of bulb types of A. cepa, A. fistulosum and the F1. A. fistulas-um
does not produce bulbs, A. cepa is shown with regular onion bulbs and
the F1 has very slight bulbin . L to R : A. tulosum, without bulb
formation; two interspecific ksamples wit only very sli t tissue
swelling Sllmlbing); A. cepa, wit normal bulb formation. ( ourtesy -
Dr. Lowe C. Ewart).

64

65

Appendix B: Pedigree of the interspecific germplasm used in this study. (Source:

(3).
(b)-
(C)-
(d)-
(6)-
(0-

Ewart 1986).

A. cepa x Beltsville Bunching - > 73898 (F1, triploid)
A. cepa x 73898 -> 79353 (BC1, occasional trivalents)
79353 xA. cepa -> 82148 (BCz, diploid)

82148 xA. cepa -> 84462 (BC3, diploid)

A. cepa x 84462-2 - > 1253 (24 seeds obtained)
84462-2 x A. cepa - > 1255 (6 seeds obtained)

66

Appendix C: Arcsine transformed means, error and F-values for the treatments
used in Fusan'um, pinkroot screening and overall germination
(Fusarium, pinkroot and control combined) in the disease screening of

 

 

 

 

 

 

 

 

 

 

 

the Allium germplasm.
Fusarium:
Line # Obs Germ. Surv. Adj_Serv.
6701 4 68.785 45.000 53.060
1849 4 62.145 4.610 5.555
611 4 77.087 63.807 73.885
89-4-2e 3 66.147 56.790 66.830
Error 155.743 39.203 1 1 1.137
F-val 1.012 69.589 33.128
Probability 0.424 0.000 0.000
pinkroot:
Line # Obs Germ. Surv. Adj_Serv.
420 4 78.750 78.750 90.000
8155 4 37.442 31.390 64.742
611 4 76.172 67.867 78.362
89-4-29 2 60.115 51.335 69.540
89-4-3 o 2 90.000 71.560 71.560
Error 96.985 100.666 214.566
F-val 14.270 13.274 1.673
Probability 0.000 0.000 0.226
Overall Germination:
Line # Rows Germination
420 5 75.688
611 9 78.116
1849 5 64.030
6701 5 69.340
8155 5 38.954
89-4-2 e 5 63.734
89-4-3 0 4 80.139
89-4-4 e 2 42.115
Error 139.549
F -val 7.655
Probability 0.000

 

67

Appendix D: Pairwise calculated least significant differences (lsd) to compare
paired means in various Allium germplasm disease screening

 

 

 

 

 

experiments.
A. Fusarium :
Paired Obs Germination% Survival% Adj. Surv.%
3, 4 - 10.525 17.722
4, 4 - 9.745 16.407
B. pinkroot :
Paired Obs Germination% Survival% Adj. Surv.%
2, 2 21.676 22.083 -
2, 4 18.772 19.125 -
4, 4 15.327 15.615 -

 

C. Overall Germination

 

 

Paired Obs Germination%
2, 4 20.890
2, 5 20.182
2, 9 18.857
4, 5 16.182
4, 9 14.496
5, 5 15.256
5, 9 13.455

 

BIBLIOGRAPHY

BIBLIOGRAPHY

Abawi, GS. and Lorbeer, J .W. 1971a. Reaction of selected onion cultivars to
inaction by Fusarium oxysporum f. sp. cepae. Plant. Dis. Rep. 55(11):1000-
1 .

Abawi, GS. and Lorbeer, J .W. 1971b. Pathological histology of four onion cultivars
infected by Fusarium oxyspomm f. sp. cepae. Phytopathol. 61:1164-1169.

Abawi, GS. and Lorbeer, J .W. 1972. Several aspects of the ecology and pathology of
Fusarium oxysporum f. sp. cepae. Phytopathol. 62:870-876.

Bacher, J .W. 1989. Inheritance of resistance of Fusarium oxysporum F. sp cepae in
cultivated onions. M.S. Thesis, Michigan State University.

Benhamou, N. M. H. A. Joosten, and J. G. M. deWit. 1990. Subcellular localization
of chitinase and of its potential substrate in tomato root tissue infected by
Fusarium oxysporum )5 sp. radicis-lycopersici. Plant Physiol. 92:1 108-1 120.

Bino, R. J., J .M. van Vries and IN. de Vries. 1988. Breakin the interspecific
crossing barriers with the help of floral biology tec 'ques. Prophyta
42(6):147-148.

Brewster, J .L. 1977a. The physiology of the onion. Part 1. Hort. Abstracts.
47(1):17-23.

Brewster, J .L. 1977b. The Physiology of the onion. Part 2. Hort. Abstracts.
47(2):104-112.

Brewster, J .L. 1983. Effects of photoperiod, nitrogen nutrition and temperature on
inflorescence initiation and development in onion (Allium cepa). Ann. Bot.
5 1:429-440.

Brewster J .L. 1985a. The influence of seedling size and carbohydrate status and of

photon flux density during vernalization on inflorescence initiation in onion
(Allium cepa). Ann. Bot. 55:403-14.

Brewster, J .L. 1985b. The influence of seedling size and carbohydrate status and of

photon flux density during vernalization on inflorescence initiation in onion
(Allium cepa). Ann. Bot. 55:403-414.

Burr, B., S. V. Evola, F. Burr and J. S. Beckmann. 1983. The application of
restriction fragment length polymorphism to plant breeding. In Genetic
Engineering, eds J. Setlow and A. Ho aender, Plenum, N.Y.

Chang, W.N. and BE. Struckrneyer. 1976a. The influence of temperature on seed
development of Allium cepa. J. Amer. Soc. Hort. Sci. 101:296-298.

Chang, W.N. and BE. Strukmeyer. 1976b. Influence of temperature, time of da ,
and flower age on pollen germination, stigma receptivity, llen tube growth,
and fruit set of Allium cepa. J. Amer. Soc. Hort. Scr. 101(1 :81-83.

 

68

69

Clarke, AE. and H.H. Macka . 1946. A cytological study of some triploid onion
plants. J. Hered. 37(5):1 1-136.

Cochran, F. 1950. A study of species hybrid Allium ascalonicum x Allium fistidosum
and its backcrossed progenies. J. Amer. Soc. Hort. Sci. 55:293-296.

Corgan, J .N. 1985. Research on vfene in ogression into Allium ce a from A.
fistulosum. National Onion orkshop, Dec 5-6, 1985, Phoenix, .

Cronquist, A. 1968. The evolution and classification of Flowering Plants. Houghton
Mifflin, Boston. Chapter 5.

Cryder, C. M. 1988. A study of associations of heritable traits in rogeny from the
interspecific backcross (Allium fistulosum L. x Allium cepa ) x Allium cepa
L. Ph. D. Thesis, New Mexico State University, Las Cruces, USA.

Cryder, C.M., J .N. Corgan, N.S. Urquhart and D. Clason. 1989. A study of Iprogeny
from the interspecific backcross (A. gistulosum x A. cepa) x A. cepa. ational
Onion Workshop, Dec 10-11, 1989, oise, ID.

Currah, L. and Ockendon, DJ. 1978. Protandry and sequence of flower opening in
the onion. New Phytol. 81:419-428.

Davis, E.W. 1955. Nature of sterility in the amphidiploid Allium cepa- osum.
Amer. J. Botany 42:41-48.

Davis, G.N. and WJ. Henderson. 1937. The interrelation of the pathogenicity of a
phoma and a Fusan'um on onions. Phytopathol. 27:763-72.

de Vries, J. N. 1990. Onion chromosome nomenclature and homeology relationships
- workshop report. Euphytica 49:1-3.

Dowker, B. D. and G. H. Gordon. 1983. Heterosis and h brid cultivars in onions. In
Monogr. on Theor. and Appl. Genet., Vol 6, rankel, R., Ed., Springer
Verlag, Heidelberg, 1983.

Dowker, B. D. 1990. Onion Breeding. In: H. D. Robiniowitch & J. L. Brewster
grids), Onions and allied crops, volume 11, pp 103-154. CRC Press, Inc., Boca
aton, Florida.

Emsweller, S.L. and HA. Jones. 1935. An interspecific hybrid in Allium. Hilgardia
9(5):265-294.

Emsweller, S.L. and HA Jones. 1938. Crossing-over, fragientation, and formation
of new chromosomes in an Allium species hybrid. t. Gaz. 99:729-772.

Emsweller, S.L. and HA Jones. 1945. Further studies on the chiasmata of the
gélzgfiscgga x A. firtulosum hybrid and its derivatives. Amer. J. Botany

Ewart, LC. 1986. A cytogenetic investigation of three families of the interspecific
cross between Allium cepa and A. firtulosum. IVT Rapport 226, Wagenmgen.

 

70

FAQ, Quarterly Bulletin of Statistics, Food and Agriculture Organization, Rome,
1988.

Felix, E. L. 1933. Disease resistance inA. fistularum I... Phytopath. 23:109-110.

Fiskesjo, G. 1975. Chromosomal relationshi between three species of Allium as
revealed by C-banding. Hereditas 81( 975 ):23-32.

Gabelman, W.H. 1988. Breedin for disease resistance in onion. In the 4th
EUCARPIA symposium, ellsboume, UK. pp 1 1-20.

Gorenz, AM. 1949. Factors affecting pathogenicity of pink root fungus of onions. J.
Agric. res. 78:1-18.

Hanelt, P. 1986. Pathways of domestication with regards to cm types (grain
legumes, ve etables), in The Origin and Domestrcation of Cu tivated P ants,
Barrigozzi, . Ed. Elsevier, Amsterdam, 1986, 179.

Hanelt, P. 1990. Taxonomy, Evolution and History. In: H. D. Robiniowitch & J. L.
Brewster (Eds), Onions and allied cr0ps, volume 11, pp 103-154. CRC Press,
Inc., Boca Raton, Florida.

Henning, E. and G. Mildenberger. 1988. Results of testing species and interspecific
hybrids of Allium for resistance to downy mildew, Peronospora destructor
Berk (Casp.). Archiv fur Zuchtungsforschung 18(6):407-416.

Jones, 1614;. and S.L. Emsweller 1933. Methods of breeding onion. Hilgardia 7:625-

J ones, H. A. and G. A. Perry. 1956. Inheritance of resistance to pinkroot in onion. J.
Hered. 47:33-34.

Jones, H. A., S.F. Bailey and S.L. Emsweller 1934. Thrips resistance in the onion.
Hilgardia 8(7):213-232.

Jones, HA. and AE. Clarke. 1942. A natural amphidiploid from an onion species
hybrid. Journal of Heredity. 33(1):25-32.

Jones, H. A. and L. K. Mann. 1963. Onion and Their Allies. Interscience Publishers
Inc. New York.

Jones, R. N. and H. Rees. 1968. Nuclear DNA variation inAllium. Heredity 23:591.

Jones, S.T. and AB. Kehr. 1957. The cytology and plant characteristics of an
amphidiploids derived from Allium ascalonicum x A. fistulosum. Amer. J.
Botany. 44:523-529.

Jones, R. N. 1990. Cytogenetics. In: H. D. Robiniowitch & J. I... Brewster (Eds),
aniogs and alhed crops, volume II, pp 103-154. CRC Press, Inc., Boca Raton,
on a.

Kalkrnan, E. R. 1984. Analysis of the C-banded karyo e of Allium cepa L.
Standard system of nomenclature and polymorphism, enetica, 65, 141.

71

Kehr, A. E., M. J. O’Brian and E. W. Davis. 1962. Pathogenicity of Fusarium
axysporum . sp cepae and its interaction with Pyrenochaeta terrestris on
onion. Eup 'ca. 11:197-208.

Kokoreva, V.A. and NA Yureva. 1987. Reciprocal crosses of Allium fistulosum L.
and Allium altaicum Pall. In Progres. priemy v selectii plodoovoshch. Kulkur
Moscow, USSR (1987) pp 112-119.

Kolattukudy, P. E. 1985. Enzymatic penetration of the plant cuticle by fungal
pathogens. Ann. Rev. Phytopathol. 23:223-250.

Krueger, S. K. 1986. Resistance to Fusarium basal rot in onions (Allium cepa 1...).
MS. Thesis, University of Wisconsin. pp 46.

Levan, A. 1933. ological studies in Allium. IV. Allium fistulosum, Sven Bot
Tidsker., 27: 11.

Levan, A. 1941. the cytology of the species hybrid A. cepa x A. fistulosum and its
polyploid derivatives. Hereditas. 27:253-272.

Lu, C.C., L. Currah and EB. Peffley. 1989. Somatic embryoienesis and lant
regeneration in diploid Allium, firmlosum x A. cepa F1 hy rid onions. lant
Cell Reports 7(8):696-700.

Maeda, T. 1937. Chiasma studies in Allium firtulosum, Allium cepa and their F1, F2
and backcross hybrids. Jpn. J. Genet. 13:146-159.

Masuda, H. 1956. Studies on seed production in onion. Sci. Rep. Fac. Agric. Okla.
Univ. 8. 1956. 45-54.

Mauch, F. and A. Staehelin. 1989. Functional im lications of the subcellular
localization of ethylene-induced chitinase and 1,3-glucanse in bean leaves.
The Plant Cell 1:447-457.

Maude, R. B. 1990a. Leaf diseases of onions. In: H. D. Robiniowitch & J. L.
Brewster (Eds), Onions and allied crops, volume II, pp 173-190. CRC Press,
Inc., Boca Raton, Florida.

Maude, R. B. 1990b. Storage diseases of onions. In: H. D. Robiniowitch & J. L.
Brewster (Eds), Onions and allied crops, volume 11, pp 273-196. CRC Press,
Inc., Boca Raton, Florida.

McCollum, GD. 1971. Sterility of some interspecific Allium hybrids. J. Amer. Soc.
Hort. Sci. 96(3):359-362.

McCollum, GD. 1974. Chromosome behavior and sterility of hybrids between the
common onion, Allium cepa and the related wild A. oschaninii. Euphytica
23(1974): 699-709.

McCollum, GD. 1982. erimental hybrids between Allium fistulosum and A.
roylei. Bot. Gaz. 143 2):238-242.

McCollum, GD. 1976. Onion and allies. In N. W. Simmonds ed. Evolution of Crop
Plants. Longman, London.

72

Meer, Q.P. van der, and J .N . de Vries. 1990. An interspecific cross between Allium

roylei Steam and Allium cepa L., and its backcross to A. cepa. Euphytica 47
(1990) 29-3 1.

Michelmore, R. W., I. Paran and R. V. Kesseli. 1991. Identification of markers
linked to disease resistance genes by bulk segregant analysis: a rapid method
to detect markers in specific fienomic regions by using segregating
populations. Proc. Natl. Acad. Sci. SA 88:9828-9832.

Moll, RH. 1954. Receptivity of the individual onion flower and duration. Proc.
Amer. Soc. Hort. Sci. 64:399.

Peffley, EB and PD. Magnum. 1989. Cytogenetic and Electrophoretic evidence for
rlecorérbgnfition blelgween Allium genomes. National Onion Workshop, Dec 10-
1, 1 8 , oise, .

Peffley, EB. 1986. Evidence for chromosomal differentiation of Allium fistulasum
and A. cepa. J. Amer. Soc. Hort. Sci. 111(1):126-129.

Peffley, E.B., J .N. Corgan and SD. Tanksley. 1985. Electrophoretic analysis of
Allium addition hnes. Theor. Appl. Genet. 71:176-184.

Peffley, EB. 1987. Interspecific Hybridization in Allium. In The breeding of
horticultural cro 5. Proceedings of an international symposium held at
National Chung ing University, December 1986, Taipei, Taiwan.

Porter, D. R. and H. A. Jones. 1933. Resistance of some of the cultivated species of
Allium to pink root Phoma terestrir. Phytopath. 23:290-298.

Ranjekar, P. K., D. Pallotta and J. G. Lafontaine. 1978. Analysis of plant genome. V.
Com arative study of molecular properties of DNAs of seven Allium species.
Bioc em. Genet. 16:957. '

Saini, SS. and G.N. Davis. 1967. Compatibility in some Allium species. J. Amer. Soc.
Hort. Sci. 91:401-409.

Saini, SS. and G.N. Davis. 1970. Karyotypic analysis of Allium species. J. Amer. Soc.
Hort. Sci. 95(1):]02-105.

Smith, I... 1947. The acetocarmine smear technic. Stain Tech. 22:17-31.

Soni, S. K. and P. P. Ellis. 1990. Insect pests. In: H. D. Robiniowitch & J. L.
Brewster (Eds), Onions and allied crops, volume 11, pp 213-272. CRC Press,
Inc, Boca Raton, Florida.

Steel, R. G. D. and J. H. Torrie. 1980. Principles and Procedures of Statistics. A
biometrical approach. 1980. Second Edition. McGraw-Hill Book Company.

Swanson, C. P., T. Merz and W. J. Young. 1981. Cytogenetics. The Chromosomes in
Division, Inheritance and Evolution. Second edition. Prentice-Hall Inc.
Eaglewood Cliffs, NJ 07632.

73

Tanksley, S. D. N. D. Young, A. H. Paterson and M. W. Bonierbale. 1989. RFLP
m gprgém plant breeding. new tools for an old science. Biotechnology

Tarasova, EM. and K. Pole. 1984. Karyolo 'cal results of the interspecific cross A.
cepa X A. oschaninii O. Fedtsch. In ogres. priemy v selectii plodoovoshch.
Kulkur Moscow, USSR pp 88-89.

Tashiro, Y. 1980. Cyt qgenetic studies on the origin of Allium wakegi Araki. Bull.
Fac. Agr., Saga mv 55. 125- 129.

Tashiro, Y. 1981. ogenetic studies on the origin of Allium wakegi Araki II. Agr.
Bul. of Saga mv. 50:85-95.

Tashiro, Y. 1984. oenetic studies on the origin of Allium wakegi Araki. Bull.
Fac. Agr., Saga niv. 56:1-63.

Walker, ’018: C. 1961. Onion Diseases and Their Control. U. S. Dep. Agric. Handbk.
2 27 pp.

Walker, J .C. and E. C. Tims. 1924. A Fusarium bulb rot on onion and the relation of
environment to this development. J. Agric. Res. 28(7):683-692.

Wietsma, W .A. and C. Kik. 1989. Resistance to downy mildew within reach.
Groenten en Fruit (1989) 45(10)71.

Wietsma, WA. and J .N. de Vries. 1988. Inters specific crosses provide onion with
much-needed genetic variation. Prophyta 2(5) 114-116.

Williams, J. G. K., A. R. Kubelik, K.J. Livak,J. A. Rafaiski and V. Tingy.1990.
DNA polymorphisms amplified b arbitrary primers are useful as genetic
markers. cle Acid Res. p:18(22) 531-6535.

Wink, M. 1988. Plant Breeding: importance of plant secondary metabolites for

protection against pathogens and herbivores. Theor. appl. Genet. (1988)
5 225-233.

Woloshuk, C. P. and P. E. Kolattukudy. 1986. Mechanism by which contact with
plant cuticle tr1 ers cutinase dgene expression in the spores of Fusarium
solani f. sp. pisi. oc. Natl. Aca Sci 83:1704-1708.

‘i1111111011“