A CYTOGENETIC INVESTIGATION
OF SOME SPECIES OF HORDEUM (GRAMINEAE) '-

Dissertation for the Degree of“ Ph. D.
MICHIGAN STATE UNIVERSITY
GILBERT D. STARKS
1976

 

gim- m I lil’l . .

:33 I,
* LII..- FEARY

3 twins.- ‘biin 5%
it? 0337mm”?

 

3f\

      
   

   

This is to certify that the

thesis entitled
A CYTOGENETIC INVESTIGATION
OF SOME SPECIES OF HORDEUM (GRAMINEAE)

presented by

Gilbert D. Starks

has been accepted towards fulfillment
of the requirements for

Ph.D. Botany

degree in

 

 

    

Major professor

Date H’ ‘7'/77‘{

 

0-7639

I 4‘. A.“’ _J-A‘—...

-Ԥ-c-

ABSTRACT

A CYTOGENETIC STUDY OF

SOME SPECIES OF HORDEUM (GRAMINEAE)

BY

Gilbert D. Starks

0n the basis of the chromosome behavior of a backcross between
a colchicine-induced amphiploid (6X) from a cross between Hordeum
iubatum L. and fl;_compressum Griseb, and fl; iubatum (hX), the follow-
ing genome relationship between fl; iubatum, fl; ompressum and closely

related species is suggested: fl; iubatum, AAA'A', fl; brachyantherum,

 

AAA'A', fl; californicum, AA, fl; compressum A'A'. Also, an argument

 

is proposed for genetic control of chromosome pairing.

A cross between the amphiploid and autotetraploid fl;_vulgare
resulted in polyhaploids with varying chromosome numbers. Out of
five progeny, three were triploid, one was tetraploid, and one was
pentaploid. The tetraploid and the pentaploid had varying chromo-
some numbers while the triploids were stable.

Diploid fl; ulgare was crossed with the amphiploid in an
attempt to incorporate favorable wild genes into the genome of the
cultivated Species. However, low fertility and varying chromosome
numbers made the progeny from this cross, a selfing, and a backcross

economically unpromising.

A CYTOGENETIC INVESTIGATION

OF SOME SPECIES OF HORDEUM (GRAMINEAE)

BY

Gilbert D. Starks

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements

for the degree of

DOCTOR OF PHILOSOPHY

Department of Botany and Plant Pathology

I976

DEDICATION

To William G. Fields

Whose excellent teaching influenced me and many others

ACKNOWLEDGEMENTS

My appreciation is expressed to Dr. William Tai for being my
major professor; to Dr. John E. Grafius for sharing his research
material and giving me incouragement; to Dr. Leo W. Mericle for
philosophical insights into self discipline and high standards.
Thanks to Dr. Stephen N. Stephenson and Dr. Edward I. Klos for
their friendship and willingness to read my dissertation.

To my laboratory colleagues, Lynn Murry, Michael Christianson,
and Joanne Whallon, goes my thanks for friendship and support; and
Doug Stevens, for introducing me to cytology. Thanks to my crop
science friends, Bob Steidl and Rye Ho Huang, for suggestions and
instruction in the culturing and breeding of cereals.

My wife Kay and children Greg, Jeff, and Melanie put up with
an absentee husband and father above and beyond any reasonable
limit of expectation. My affection for them‘for this sacrifice
cannot be adequately expressed.

Financial support for part of my work came from a Donald F.
Jones Predoctoral Scholarship granted by the Research Corporation.

This support is very gratefully acknowledged.

TABLE OF CONTENTS

LIST OF TABLES . . . . . . . . . . . . . . . .
LIST OF FIGURES . . . . . . . . o . . o . . . .
INTRODUCTION- . . . . . . . . . . . . . . . . .

SECTION I Genome Analysis of Hordeum jubatum and
fl;_compressum . . . . . . . . . .

 

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

SECTION II Ponhaploids From a Cross Between
Autotetraploid fl; vulgare and C-l

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

SECTION III A Cytogenetic Study Involving the Genomes
of the Amphiploid (C-l), and Diploid

fl..- WIgare I O O O O O O O O I O

IntrOdUCtion O O O O O I O O O O O O O O O
Resu' ts O O O O O O O O O O O O O O O O 0
Discussion . . . . . . . . . . . . . . . .
SUMMARY 0 0 O O O O O O 0 O O O O O O O O O O 0
LI TE RATU RE CITED 0 O O O O O 0' O O O O O O O 0

APPENDIX 0 O O O O O O O O O O O O O O O O O O

Page

vi

l0
l2
13
I9

27
28
30

3l
#5

52
53
54
69
78
79

85

Table

LIST OF TABLES

Summary of Chromosome Association in the
F1, Amphiploid, C-1, and of Backcross Progeny
I”.fl; iubatum x E; compressum . . . . . . . .

Possible Genome Combinations and Chromosome
Pairing Without Gene Influence . . . . . . .

Possible Genome Combinations and Chromosome
Pairing Under Gene Control . . . . . . . . .

Summary of MI Chromosome Association in PHA
I to 3 (2N = 2]) O I O O O O O C O O O O O 0

Summary of MI Chromosome Association in PHA-h
(”TetraPIOEd“)o o o o o o o o o o o o o o o 0

Summary of M1 Chromosome Association in PHA-S
(I'PentaPIOid.‘)O O I I O O O O I O O C I O O 0

Summary of M1 Chromosome Association in I2-I
(2N = 28) O I O C O O I O O O O O I O O O O 0

Summary of MI and Diakensis Association in

12-1-8 o o o o o o o o o o o o o o o o o o 0

Summary of MT Association in l2-I-l . . . . .

Page

I3

2I

2k

3k

no

42

58

63
67

I2.

I3.
Th.

IS.

LIST OF FIGURES

Diagram of Barley Breeding Program . . . .

Diagram of a Cross Between 5; iubatum and
.fl; compressum . . . . . . . . . . . . . . .

 

Diagram of a Cross Between fl; iubatum and

C'looooocoooooooooooooo

Spike and Spikelet Morphology of u; iubatum,
fl; compressum, C-1, and CBJ l . . . . . . .

 

Stages of Microsporogenesis in u; iubatum,
C-1, and CBJ] o o o o o o o o o o o o o 0

Summary of Genome Formulae . . . . . . . . .

Diagram of a Cross Between C-1 and
Autotetraploid fl; vngare . . . . . . . . .

Spike and Spikelet Morphology of PHA

I to S O O O O O O O O O O O O O O O O O O O

Microsporogenesis in the PHA's . . . . . . .
PHA-S Microsporogenesis . . . . . . . . . .

A Diagram of a Cross Between C-1 and

fl; vulgare (2X) . . . . . . . . . . . . . .

Spike and Spikelet Morphology of E; vulgare
(2X), C-l, and I2-l-B Spikelet Awns are

InVESible O I O C O I O O O O I O O O O O O
Microsporogenesis in Plant l2-1 . . . . . .

Microsporogenesis in l2-l-B . . . . . . . .

Microsporogenesis in lZ-l-l . . . . . . . .

Page

l0

Il

IN

16
26

30

32
35
#3

5h

55
59
65

72

INTRODUCTION

Personnel in the laboratory of Dr. John Grafius, Cr0p and
Soil Science Department, Michigan State University have been
attempting to incorporate genes for winter hardiness and salt
tolerance into cultivated barley, Hordeum vulgare. With the
addition of such genes, it was forseen that the range for
.fl; vulgare could be tremendously extended, and hence a cor-
responding increase in crop production could be achieved. How-
ever, in trying to get the desirable genes into the germplasm of
_fl; vulgare, several problems were encountered. An interdepart-
ment c00peration developed to look at these problems, and with it,
this study had its beginning.

It was believed that a cytogenetic investigation would
demonstrate chromosome relationships between the involved species;
provide direct cytological evidence of ploidy level, abnormal
meiotic divisions, and source(s) of sterility. Hordeum vulgare
was known to cause cytogenetic instability (Kasha, l97h) and its
influence on the interspecific crosses would be particularly noted.
An explanation for the lack of genetic exchange between species
was anticipated from this work. Figure 1 explains that part of

the breeding program with which this study was concerned.

xm mo mmouxo c. mucosa:
oEOmoEOLLU Lu.3 >com0cau m <I¢ ucm : <Im

in _-~_ ume_mm u _._.N_ mn_o_am;>_oa mac“ n m o» _ <Ia
ocmmfly> .1 Cu commoLoxomn _uN_ u mu_uN_ Enumn:_ .I OH xomnmmoLo u nmu
mmoLo ocmm_:>.dfl x Fuu oz» mo >com0ca u _|~_ vmo_a_coEm com—om “n Flu

.EmcmoLa mc_noocn >o_cmn ozu c. uo>_o>c. muOumm
imamo>c_ .mucoEucmaopcouc_ ogu >3 cog: pomcmm ocoz co_umpcomm_u m_£u c. tom: mco_umcm_mop echo.” .

Aomm_v .._m um .cw_oocum .m .< >3 meow xco3e

 

 

 

 

Axmv m <xa
Axav : <12
AXNV F-_-N_ AXNQ m-_-~. Axmv m o“ _ <za
IIAxav _-~_niigx~v mum _:> an Axav mumm_a>. :_ Axmv N_ 0u _ smu
.2 Axmv mcmm_a> .1 0V _ u Axav Enumn:_ 4m

a..Ax
Rpm
a

Axov a_o_am£ae<

g

ucmEumoLh m:_owco_0u

a

%Axmv

Axwv EsmmoLQEoo 4m AXJV Enumnz.

Emcmocm mc_vmocm >m_cmm mo Emcmmmo ._ mc:m_u

_m_om

 

 

3

The first published meiotic chromosome counts of Hordeum
iubatum (2N = 28) were made by Aase and Powers (I926). Boyle
and Holmgren (I955) made the first genome analysis of H; iubatum
and concluded that it was an allotetraploid. They suggested AACC
as a genome formula for the species. A study of Hordeum species,
initiated by D. G. Hamilton, was carried out by Morrison (I959)
who suggested that mitotic chromosome morphology indicated a re-
mote evolutionary relationship between H; iubatum and H; vulgare.
Morrison, g£_gl. (I959) and Rajhathy and Morrison (I959) reported
that H; iubatum was crossed with H; vulgare, and that sterile
hybrids were obtained through embryo culture. Sterility and the
lack of chromosome pairing during meiosis demonstrated that there
is no close relationship between H; iubatum and H; vulgare. In

another cross between H; iubatum and fl; brachyantherum, Rajhathy

 

and Morrison (I959) showed there existed a close relationship
between these species. They concluded that 5; iubatum and fl;

brachyantherum belong to the same Medelian pOpulation. In a

 

three-way cross, very little autosyndetic pairing of chromosomes

occurred between the two species, fl; iubatum and fl; brachyantherum,

 

when crossed with H; vulgare. This was taken as evidence that
there was no homology existing between the genomes of H; iubatum.

In a cross between H; iubatgm and Secale cereale, Wagenaar

 

(I959, I960) was able to observe autosyndetic pairing between the
chromosomes of H; iubatum. From this evidence he concluded that
H; iubatum was a segmental allotetraploid. With additional infor-

mation, Rajhathy et al. (l96h) acknowledged that E; iubatum does

4
”pair autosyndetically in haploids over the presence of non-
homologous genome” and therefore is a segmental rather than a
typical allotetraploid. Redmann and Borgaonkar (I966) agreed
with these findings in their study of the Hordeae of the Dakotas.

Bowden (I962) included 5; brachyantherum as one of three

 

subspecies of H; iubatum, and gave it the name of_fli iubatum ssp.

breviaristatum Bowden. Previous works (Nevski, I934; Covas, l9h9;

 

 

and Hitchcock and Chase, I950) considered fl; brachyantherum a
separate species. However, Anderson (l9hh) and Covas (l9h9)
realized that the two taxa were closely related and probably could
hybridize in nature. Bowden (I962) stated that ”there is good

evidence that H; iubatum subsp X intermedium is of hybrid origin

 

from the crossing of subsp. iubatum with subsp. breviaristatum.”

 

This cross was artifically produced by Rajhathy and Morrison
(I959).

Mitchell and Wilton (l96h), from a relatively small sample
of plants at Palmer, Alaska, found no hybrids occurring between

H; iubatum and H; brachyantherum and concluded that they were

 

distinct Species. Rajhathy (l966) pointed out that experimental
hybrids between the two taxa had been achieved and the fertility
of the hybrids indicated an even closer relationship than would

crossability alone. Not only were the F hybrids of the two taxa

I
fertile, but partial fertility was maintained through successive
generations. Further, Rajhathy pointed out that both ”species”

are inbreeders, and it would not be unusual to find a lack of

5
hybridization in some sympatric populations. The_fl; iubatum -

brachyantherum group does not demonstrate inhibited embryo develop-

 

ment, a common phenomenon in interSpecific hybrids of Hordeum. The
lack of this isolation mechanism probably is significant in this
case.

Meiotic chromosomes of Hordeum compressum (2N = IA) were first

 

counted by Perak (l9hl). Covas and Schnack (l95l) produced a ster-

ile hybrid between H; compressum and fl; caIifornicum which closely

 

 

resembled fl; brachyantherum in its floral character. They felt

 

that the ancestors of H; brachyantherum were H; caIifornicum (or a

 

 

closely related Species), and a diploid Species from the group

including fl; stenostachys, fl; compressum, and fl: pusillum.

 

 

Schooler, et al. (I966) concluded that E; iubatum had a genome in

common with H; compressum.

 

Chromosomes were first counted in H; vulgare (2N = IN) by
Nakao (I9Il). No natural hybrids are known between the Species
in the section Cerealia and those in other sections of the genus
Hordeum. Although artificial hybrids have been made between
“H; vulgare and several other species of Hordeum, no fertile hybrids
have been produced (Price, I968). The heterospiculate nature of
the spikelets is characteristic of the genus, but Since H; vulgare
is isopiculate, many taxonomists have put it in a section (Cerealia)
by itself. The question of whether H; vulgare derived from a mono-
phertic or polyphyletic origin is unresolved. The degree of
relationship between Cerealia and the other sections of Hordeum is

unknown but it appears to be Slight.

6
In spite of the lack of close relationships between H; vulgare
and other intergeneric-intrageneric Species of the tribe Triticeae,
numerous crosses between H; vulgare and these Species have been
attempted. As early as I876, A. Wilson reported the successful
production of wheat-rye hybrid. Later workers have continued
making crosses with the goal of combining the best qualities of

two or more cereals into a new plant type. Secale cereale (rye)

 

has been crossed with H. vulgare and other species of Hordeum,
Thompson, I939, I940; Quincke, I940; Brink, COOper, and Ausherman,
I944; Brink and Cooper, I944, I947; Thompson and Johnson, I945;
MorriSon, gt_gl., I959; Wagenaar, I959), but in all cases the
hybrids were sterile.

Crossing difficulties have arisen not only because of hybrid
sterility, but also because of the lack of embryo and endosperm
deveIOpment. Embryo culture techniques have been developed to a
point where they greatly facilitate the making of crosses. Embryo
excision and in vitro culture techniques have been advanced by
Knudson, (I922); Cooper and Brink, (I944); Ziebur, §£_§l., (I950);
Konzak, §t_§l., (I95I); Morrison, §£_al., (I959); Davies, (I960);
Norstog, (I972, I973a, I973b); Jensen, (I974); Kasha, (I974).

Schooler (I960a, l960b), using embryo culture techniques,
took an indirect approach to crosses between wild Hordeum species
and H;. ulgare. This approach involves producing fertile amphi-

ploids between Hordeum Species and crossing the amphiploid with

it; llilfleiéo

7
The amphiploids can be used as an initial source for gene transfer
in a well designed backcross program. In essence this type of
program amounts to H; vulgare becoming enriched to a small degree
with genes coming from the amphiploid, thus broadening the base of
variability of H; vulgare and increasing the variety of recombina-
tion products which may be secured from it. (Allard, I960).
The objectives of this investigation were:
I. To determine the genome relationship between Hordeum

iubatum and Hordeum compressum.

 

2. To assess the influence of Hordeum vulgare genome

 

on the production of polyhaploids.
3. To observe meiotic behavior of hybrids, amphiploids,
and their back cross progeny.
4. To determine the relationship between meiotic behavior
of the hybrids and fertility.
5. To use genome relationships to determine the phylogenetic
relationships among the species studied.
6. To incorporate the information obtained through cytogenetic
studies into the barley breeding program.
The dissertation has been partitioned into three sections.

An overview of each follows:

Section I: Genome Analysis of Hordeum Jubatum and fl; compressum

 

 

It was necessary at the outset of this study to ascertain

the genome formulae for H; iubatum and fl; compressum. Without

8

this information conclusions about the phylogenetic relation-
ship between the two species would have been impossible to
make. The formulae were forthcoming after the amphiploid was
backcrossed to H; iubatum resulting in the I2 pentaploid
progeny (CBJ I to I2). The CBJ progeny were important not
only in providing data for genome formula determination but
also for concluding that chromosome pairing was under genetic

control.

Section II: Polyhaploids From a Cross Between Autotetraploid
.fl; vulgare and C-1

In an attempt to produce offspring with reasonably good
fertility, autotetraploid H; vulgare was crossed with the
amphiploid (C-I). The effect of the added H; vulgare genome
was chromosomal instability which resulted in offspring with
the varying ploidy levels: triploid (3 plants), tetraploid
(I plant with a varying chromosome number), and pentaploid
(I plant with a varying chromosome number). The resultant
triploids were comparable to the hybirds Schooler g£_gl.

(I966) obtained crossing fl; iubatum and fl; compressum. These

 

polyhaploids made it possible to confinn his work.

Section III: A CytOgenetic Study Involving the Genome of the
Amphiploid (C-I), and Diploid HE vulgare

The final part of this study deals with the cytology

of the crosses involving diploid fl; vulgar_. The amphiploid

9
(C-1) crossed with_fl;.!glgg£§ resulted in the l2-I progeny.
in these hybrids, chromosomes from the wild Species and
from H; vulgare contributed to the tetraploid I2-l. The I2-T
progeny were selfed in order to improve fertility and also
obtain favorable segregants. However, the offspring were
not economically promising, fertility was not increased,
and variation in chromosome numbers and configurations has

created a whole new set of questions.

SECTION I

Genome Analysis of Hordeum Jubatum
and fl; Compressum

 

I0

INTRODUCTION

The initial part of this study originated with Schooler's
work. The cross he made and the subsequent doubling of that

hybrid's chromosomes is diagrammed in Figure 2.

Figure 2. Diagram of a Cross between H; iubatum and fl; compressum
(Schooler et al., I966)

 

H. jubatum (4X) X H. compressum (2X)

 

FI Hybrid (3x)
Colchicine Treatment
(Doubling)

I

Amphiploid (6X)

I

Selfing

I

C-I

After the author examined the cytology of H; iubatum, H; 229‘
pressum, and C-1; the cytology of CBJ I to IZ was analyzed. The
following diagram (Fig. 3) illustrates the derivation of CBJ'S.

All twelve individuals resulting from this cross were pentaploid

II
and morphologically identical.
Figure 3. Diagram of a cross between_fl: jubatum and C-l.
lb_iubatum (4X) X C-l (6X)
CBJ'S (5X)

Cytotaxonomic studies by Bowden (I962) and Rajhathy and

Morrison (I959, I961) indicate that Hordeum jubatum serves as an

 

important link between some of the Old and New World species of
the genus. However, the true nature of its genomes remains
debatable. Following the study of Spontaneous hybrids between

_H; iubatum and Agrepyron trachycaulum, Boyle and Holmgren (I954)

 

suggested AACC as a genome formula for H; igbgtgm. After studying
hybrids between H; iubatum and other species of the genus, Rajhathy
and Morrison (I96I) assigned the genome formula AABB to H; iubatum
which clearly suggested that the Species was an allotetraploid.
Based upon a Size difference between the chromosomes of H; iubatum

and those of Secale cereale in a hybrid between the two Species,

 

Wagenaar (I959, I960) concluded that autosyndetic pairing occurred
and that H; iubatum was a segmental allotetraploid. Rajhathy g£_gl.,
(I964) later agreed with Wagenaar (I959, I960) that the two genomes
0f.fl; iubatum may be partially homologous, but the concept has not
been completely accepted because, as stated earlier by Boyle and
Holmgren (I954), there is a total lack of multivalent association

of chromosomes in E; iubatum.

l2
The purpose of this experiment is to ascertain the genome
formula for H; iubatum, and the genome relationships between

.5; iubatum, fl; compressum, and two related species, fl; brachyanth-

 

 

erum and fl; caIifornicum.

 

Materials and Methods

 

The F1 hybrid between H; iubatum and fl; compressum and the

 

colchicine-induced amphiploid of this hybrid were obtained by
Dr. A. B. Schooler (I966) of North Dakota State University. Seeds
(C-l) obtained from selfing the amphiploid were sent to Dr. John
Grafius, Cr0p and Soil Science Department, Michigan State University;
and our material was obtained from his laboratory. Florets of C-I
were emasculated and crossed with H; iubatum. The embryo culture
technique of Norstog (I973) was used to bring about the develop-
ment of the backcross prOgeny.

Twelve backcross progeny (CBJ'S) were produced and grown in
the greenhouse. Spikes for cytological studies were fixed in"
Newcomer's (I953) solution and stored under refrigeration. Pol-

len mother cells of H; iubatum, and EL compressum, C-l, and the

 

CBJ'S were examined cytologically with the acetocarmine smear
technique. All plants used in this study were maintained in the
greenhouse and field nurseries at Michigan State University. A
minimum of five Spikes was observed for each plant, and a minimum

of 25 cells was examined for every cytological interpretation.

l3

Results

The spike and Spikelet morphology of the plants involved
in this section are shown in Figure 4.

Normal chromosome behavior was observed in H; iubatum and
fl; compressum as previously reported by others (Covas and Schnack,
I95l; Wagenaar, I959; Rajhathy and Morrison, l96l; Schooler g£_al.,
I966). In H; iubatum, I4 bivalents consistently occurred at MI

in the I00 cells observed (Fig. 5.A). In H; compressum, seven

 

bivalents were always present at MI in the 25 cells studied.
Chromosomes segregated normally at Al in both Species.

Since plants of C-l and the CBJ'S used in this crossing
program were direct descendants from Dr. Schooler's amphiploid,
his cytological data on the F1 hybrid and amphiploid are used
for comparison and are summarized in Table I.

TABLE I
Summary of Chromosome associations in the F], amphiploid C-I, and

of backcross progeny in H; lubgtum x H; gomEressum

 

 

Chromosome association at Metaphase I

 

 

 

 

 

 

I II III IV
No. of Range 2 Range 2 Range 2 Range ;
Cells
FI 25 2-I3 7.2 0-I2 6.36 O-I 0.I6 0-I 0.04
Amphi-
ploid 25 0-8 2.36 l4-2I I9.24 0-4 0-44 0-2 0.28.
C-I 6l 0-4 0.70 I3-2I l7.49 0-2 0.I0 0-4 I.44
CBJ II2 O-IO 3.68 7-I6 I0.89 l-6 3.I8 -——- -——-

l-I2

 

I4

Figure 4. Spike and Spikelet Morphology of H; IUbaEflfl:
‘fli compressum, C-1, and CBJ l.

 

A. Spikes of H; compressum, C-I, and fl; iubatum. (0.5x)

B. Spikelets of H; conpressum, C-l, and fl; igbatum. (I.Ix)

 

C. Spikes of C-I, CBJ l, and fl; iubatum. (0.5x)

D. Spikelets of C-l, CBJ I, and fl; ]ubatum. (I.Ix)

 

Figure 4

I6

“re 5. Stages of Microsporogenesis in H; iubatun, C-I, and

CBJ I.
Metaphase I in H, iubatufl_with I4 ll. (945x)

Metashas

m

I in C-I with I3 II + 1+ IV (solid arrows). (llZOx)

Quartet with 4 micronuclei in C-l. (l000x)

I

'5 a
a

t phase I in CBJ l with 3 I (open arrows) + I0 II

+ 4 III (solid arrows). (945x)

base I in CBJ I with 2 I (open arrows) + 9 II
III (solid arrows). (890x)

Anaphase l in CBJ I Showing a I5-20 disjunction of
chrowosomes. (900x)

l7

I8

The number of bivalents in C-I is similar to that of the
amphiploid. A significant difference is that the number of uni-
valents and trivalents decreased in C-I and the number of quadri-
valents increased (Fig. 5.8) from an average of 0.28 IV per cell
in the amphiploid to I.44 IV per cell in C-I which represents an
increase of almost five-fold. Normal segregation (2I-2l) of
chromosomes at Al was observed in 65% of the cells. The rest of
the cells had unequal disjunction or lagging chromosomes. Lag-
gards appeared in approximately 50% of the cells at All, approxi-
mately 45% of the quartets of C-I contained one to four micronuclei
(Fig. 5.C).

In the C-l selfed progeny (C-2) a simular number of quadri-
valents (I.I6 IV per cell) was observed. Anaphase I was character-
ized by lagging chromosomes in about 35% of the cells. Chromatin
bridges and fragments were occasionally observed. The frequency
of Iaggards at All was similar to that of C-I. Micronuclei,
ranging from I to 4 per quartet, occurred in 60% of the quartets.

AII twelve backcross progeny (CBJ'S) were pentaploid as
expected. At metaphase I, bivalents were the predominant associa-
tion with an average of II bivalents per cell. Univalents and
trivalents occurred on an average of 3.68 and 3.I8 per cell,
respectively (Fig. 5.0,E). No quadrivalent or pentavalent associa-
tions were observed. A (I7-I8) segregation occurred at Al in only
20 of 80 cells observed. Other segregate distributions were

found in I8 cells (Fig. 5.F), and 42 cells had I to 6 Iaggards

l9

an average of 2.4 Iaggards per cell. Laggards were observed in
less than I0% of the cells at A II. Micronuclei, which ranged

from I to 5 per quartet, occurred in 70% of the quartets.

Discussion

 

Rajhathy and Morrison (I959) demonstrated that H; brachy-
therum is conspecific with H; iubatum. Earlier, Covas (I949) made

the assumption that H; brachyantherum was an allotetraploid which

 

derived one of its genomes from 5; caIifornicum and another from

 

an undetermined diploid Species. Covas and Schnack (l95l) stated
that the undetermined species was probably one of the South American
diploid Species EL stenostachys, H; pusillum, or H, compressum.
Following a cross between the two diploid (2n = l4) species,

.fl; compressum and H; caIifornicum, Covas and Schnack (l95l)

 

 

suggested some homology existed between these genomes. A common

genome in_fl, caIifornicum and_fl; iubatum was postulated by Rajhathy

 

and Morrison (I96I); and Schooler et al., (I966) concluded that

H. jubatum and H. compressum also have one genome in common.

 

Therefore, from the published literature the possibility emerges
that HE jubatum is a segmental allotetraploid which has had its
chromosomes derived from two closely related diploid species

H. compressum and H. caIifornicum. Cytological data from the

 

 

present work has led to a Similar conclusion.

20

Although the l2 backcross progeny Showed an averaged chromo—
some association of 3.68 I + l0.89 II + 3.I8 III per cell, (Fig.
5. D,E) there were 4 cells which Showed a combination of II + 8 III
+ 6 III. Forty-three cells (38%) had at least 4 trivalents, and
78 cells (69%) had a minimum of 3 trivalents. It seems quite clear
that the maximum configuration of CBJ'S is 7 II + 7 III.

Until cytologically proved otherwise, the genomes from
_fl. iubatum are tentatively subscribed with a ”j” and those from
H; compressum with a ”c”. Theoretically, H. iubatum can be an
autotetraploid (A.A.A.A.), and allotetraploid (A.A.B.B.), or a

J J J J J J J J

segmental allotetraploid (AjAjA'jA'j). The genomic formula for

.5; compressum may be either ACAc (or A'CA'C) if this species has

 

a genome in common with‘fl. iubatum, or CCCc if a genome is not in
common with those of'fl. iubatum. Based on genome homology alone,
possible genomic formulae and chromosome associations are
summarized in Table 2.

It is quite obvious that H. iubatum is not an autotetraploid
due to the lack of quadrivalent associations. If H, iubatum were

an allotetraploid, the F hybrid between H. iubatum and fl. compres-

I
.§gm would have the possible genome combinations AijAc or AijCc.

That fl. compressum does not share a genome with H. iubatum is ruled
out due to the trivalent association which occurs in the FI hybrid
as observed by Schooler §§_§fl,, (I966). Wagenaar (I959) concluded

that‘fl. iubatum is a segmental allotetraploid, and Schooler et al.,

(I966) suggested that preferential pairing of homonIOgous

2I

TABLE 2

Possible Genome Combinations and Chromosome
Pairing Without Gene Influence

 

 

 

 

 

 

 

 

Parents F1 Hybrid
Expected
Maximum
fl. iubatum fl. compressum Formula Pairing
I I I I
Ac Ac AjAj Ac 7 III
A.A.A.'A.'
J J J J
l
CCCc AjAj Cc 7 I + 7 II
Amphiploid CBJ'S
Expected Expected
Maximum Maximum
Formula Pairing Formula Pairing
A.A.A.'A.'A 'A ' 7 VI A.A.A.'A.'A '
J J J J C C . J J J J C 7 V
A.A.A.'A.'C C A.A.A.'A.'C I + V
J J J J C C 7 II + J J J J C 7 7 I
7 IV

 

 

 

22

chromosomes in the absence of homologous partners explained the
distribution of trivalents and quadrivalents that he observed in
the F1 and induced amphiploid of H; iubatum crossed with-fl;,ggm-
pressum. We agree that H; iubatum is a segmental allotetraploid
with partial homology between the two genomes.

We view the chromosome behavior, as reported by Schooler §t_gl.,
(I966) and as observed in this laboratory, as evidence for genetic
control of chromosome pairing. Genetic influence on chromosome
pairing has been reported by Riley and Chapman (I958), Rajhathy
e£_§l., (I964), Feldman g£_al., (I966), and Gottschalk (I973),
among others.

If chromosome pairing is due only to homology between genomes;
and H; iubatum and fl; compressum have a genome in common; and fl;
iubatum is a segmental allotetraploid; then the III's observed
in the F1 hybridshould form IV's in the amphiploid. However, Vl's
have never been observed. Therefore, the following model was
proposed: Chromosome pairing in the plants under consideration
is genetically controlled. Possibly the gene(s) controlling
pairing is located in the Aj genome of H; iubatum which is the

genome not shared by H; compressum. When a single dose of this

 

gene (or a group of genes) is present, homoeologous pairing may
prevail, e.g. in the FI hybrid and in the autosyndetic pairing

found in the hybrid between H; iubatum and Secale cereale

 

(Wagenaar, I959).

'23

Accordingly, a maximum of 7 lI's and 7 IV'S is expected
to occur in the colchicine-induced amphihexaploid (Table 3).
An average of 0.28 IV'S observed falls short of this expectation.
By assuming genetic control of chromosome pairing, the five-fold
increase of quadrivalents in the C-l may be explained in terms
of gene segregation. The chromosome association in the backcross
pregeny further strengthens our hypothesis. Possibly the gene(s)
controlling pairing is located in the Aj genome of H; iubatum
which is the genome not shared with H; compressum. When a single
dose of this gene (or group of genes) is present, homoeologous
pairing may prevail. When a double dose is present, only home-
logous chromosomes may pair. Tentatively, we designate the
following genome formulae to the species related to this work:

Eh iubatum A A A'A', fl; brachyantherum, A A A'A',.fl; caIifornicum,

 

A A, and fl. compressum, A'A'.

 

24

TABLE 3

Possible Genome Combinations and Chromosome
Pairing Under Gene Control

 

 

 

 

 

 

 

Genome Formula F, Hybrid
Expected
Maximum
fl. iubatum H. compressum Formula Pairing
A' A' A.A' A' 7 l + 7 II
c c J j c
A.A.A' A'. . ,
J J j J CCCC AjA jcc 7 I + 7 II
Amphiploid CBJ I-I2
Expected Expected
Maximum Maximum
Formula Pairing Formula Pairing
A.A.A', |,A' AI A.A.A', ', ' II + III
JJJAJcc7||+7Vl JJJAJAC 7 7

A.A.A' ' 2l Il A.A.A'.A'.C + I4
J J jA jcccc J J J J C 7 I II

 

 

25

In summary the following points can be made:

l.

H, iubatum and fl, brachyantherum, if not conspecific,
are very closely related (Rajhathy, I966) and may

share the same genomes.

u. iubatum is a segmental allotetraploid (Wagenaar,
1959) .

The genome formula for fl,_igb§£um and fl;_brachyantherum
is AAA'A'.

H, compressum is one parent of fl;_iubatum (fl; brachyan-
therum) and has the genome formula of A'A' (Covas and
Schnack, l9Sl).

fl; caIifornicum is the other parent and has the genome
formula AA (Covas and Schnack, I95l).

According to the above and the model preposed in this

dissertation, the F between fl,_]ubatum and fl; com-

I
pressum (Schooler et al. I966) has the genome for-

mula of AA'A'.

The amphiploid has the genome formula AAA'A'A'A'.

CBJ'S have the genome formula AAA'A'A'.

26

Figure 6. Summary of Genome Formulae

H. compressum x H. caIifornicum

 

 

 

 

"(A'A') l '- (AA)
F1
(AA')
Doubling
1_/
fl; brachyantherum = fl; iubatum x H; compressum
(AAA'A') (AAA'A') 1 (A'A')
F1 ;/
(AA'A')
Doubling
_1_/
_fl, iubatum x Amphiploid
(A A'A') (AAA'A'A'A')
1/
CBJ'S
AAA'A'A'

1/ A double dose of the gene(s) in A genome controls
pairing among homologous chromosomes.

2] A single dose of the gene(s) in the A genome allows
pairing among homeologous chromosomes.

SECTION II

Polyhaploids From A Cross Between Autotetraploid
H; vulgare and C-l

27

28

INTRODUCTION

Polyhaploids, as defined by Rieger, g£_gl., (I968) are
individuals which arise from polyploid Species and have a
reduced ploidy level. Dewey (l96l) observed bivalent associa-
tions in polyhaploids which had arisen from tetraploid crested
wheatgrass and concluded that the tetraploids were aut0ploid in
origin. Therefore, studies of the chromosomal behavior in
polyhaploids may indicate the genome constitution of both the
polyploid and the derived polyhaploid. It Should be noted that
conclusions about phylogenetic relationships between Species fran
meiotic pairing and their hybrids may not be a valid assessment if
genetic control of chromosome pairing is not taken into considera-
tion (Starks and Tai, I974).

Riley and Chapman (I958) have shown that chromosome pairing
in hexaploid wheat is controlled by the gene SBL. Driscoll (I972)
reported on several minor genes affecting homeologous pairing in
hybrids between wheat and related genera.

In a review of haploid angiosperms, Kimber and Riley, (I963)
cited examples of 7l SporOphytic haploids representing 39 genera

in l6 families. In seven polyhaploids, they observed complete

29
synapsis of chromosomes and three of these plants were fertile.

Raven and Thompson (I964) in discussing the evolutionary
significance of haploids suggested that polyhaploidy is an evolu-
tionary mechanism in some polyploid groups. The shift between
the diploid and tetraploid states was thought to serve as a
selection mechanism adjusting the balance of fitness and flexi-
bility for some organisms.

Tai (I970) described the process of multipolar meiosis in
diploid crested wheatgrass and suggested that multipolar cell
division was a mechanism for reducing ploidy levels and forming
polyhaploids. After observing multipolar divisions in various
plant Species, Tai (I970) suggested that polyhaploidy has played
an important role in the evolution of chromosome numbers in
angiosperms.

Techniques for the induction of haploidy include x-ray
treatments (Goodspeed and Avery, I929; Katayama, I934), inter-
specific hybridization (Satina, g£_gl., I937; Kihara and
Tsunewaki, I962; and SYmko, I969), temperature shock (Mfintzing,
I937) and chemical treatments such as colchicine (Levan, I945)
and gibberellic acid (Subrahmanyam and Kasha, l97l). The theore-
tical and practical use of haploids is still increasing in both
cytogenetics and crop breeding (Kasha, I974).

The purpose of this article is to describe the cytology of

polyhaploids obtained through the reduction of ploidy level from

30

a synthetic amphiploid, and the comparative studies between these

polyhaploids and the original hybrid.

Materials and Methods

 

A hybrid between Hordeum jubatum and fl, compressum was

 

obtained by Schooler g£_§l., (I966). Its colchicine doubled
amphiploid was sent to Dr. John Grafius of the Department of Crop
and Soil Science, Michigan State University.

Pollen from autotetraploid H; vulgare was shed onto the
stigmas of C-I, a selfed, hexaploid progeny from the amphiploid
(Fig. 7). The embryo culture technique of NorStOg (I973) was used

to bring about the deveIOpment of the embryos from this pollination.

Figure 7. Diagram of a Cross Between C-I and Autotetraploid
.fl; vulgare.

C-I (6X) x H; vulgare (4X)
AAA'A'A' VVV

PHA l 3 (3x) PHA 4 ( 4x) PHA 5 ( 5x)
AA'A' AA'A' + vx AA'A'V + vx

3l

Five embryos developed into mature plants which were
identified as PHA l to 5. The cytology, fertility, and morph-
ology of the PHA'S were studied and compared with those charac-
ters of the parents. Spikes were prepared for cytological stud-
ies by fixation in Newcomer's (I953) solution and stored under
refrigeration. The pollen mother cells were examined with the
acetocarmine smear technique and phase microscopy. A minimum of
five spikes was observed for each plant, and a minimum of 30 cells
was examined for every cytological interpretation. Mature pollen
was checked for stainability with lz-KI, with pollen grains which
were fully expanded and darkly stained being considered viable.
All plants used in this study are maintained in the greenhouse
and field nurseries at Michigan State University.

The Spike and Spikelet morphology of the plants involved in

this section are shown in Fig. 8.

Results

The cytological results of PHA I to 3 are summarized in
Table 4. When a hexaploid plant is crossed with a tetraploid,
the progeny are expected to be pentaploids. However, the above
plants had the triploid, 2n = 2l, number of chromosomes. Meiotic
behavior at metaphase I (Fig. 9A,B) shows that univalents and

bivalents occurred with about equal frequencies with averages of

32

Figure 8. Spike and Spikelet Morphology of PHA l to 5.
A. Spike of PHA I. (0.75x)

B. Spikelet of PHA l. (l.5x)

C. Spike of PHA 4. (0.75x)

D. Spikelet pf PHA 4. (l.Sx)

E. Spike of PHA 5. (0.75x)

F. Spikelet of PHA 5. (I.5x)

33

 

Figure 8

34

TABLE 4

Summary of MI Chromosome Association in PHA I to 3 (2n = 2I)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Chromosome Association 3 Number

I II III Of “”5 7°

3 9 I I.25

b, 7 l I I.25

s 8 6 7.50

.5 5 2 5 6.25

6 6 I 40 50.00

7 7 23 28.75

J 4 2 I I.25
5 l 3 3.75

lfiverage 6.18 6.36 0.70 Total 80

 

 

 

 

35

Figure 9. Microsporogenesis in the PHA'S

Diakinesis with 6 I, I3 II, and III (arrows). (ll30x)
Metaphase I with 5 I, 5 II, and 2 III. (950x)
Anaphase I with a 9 + 8 distribution and 4 Iaggards. (900x)

Anaphase I with an 8 + 8 distribution and 5 lagging
chromosomes Showing precocious division. (900x)

 

 

. —I'._ ._‘r

_ —-—- —— .___

Figure 9

37

Figure 9. Continued
E. Anaphase II with a 9 + l0 distribution and 3 Iaggards. (950x)

F. Four quartets with micronuclei. (900x)

38

9

39

6.I8 and 6.36 per cell, respectively. Trivalents occurred in
50 of the 80 cells counted, or 62.5% averaging 0.70 trivalents
per cell (Table 4).

A (IO-ll) disjunction occurred at anaphase I in only five
of 36 cells observed. The remaining 3i cells showed a more un-
equal segregation and lagging chromosomes (Fig. 9C,D). TWO cells
had a (9-l2) disjunction, and the other 29 cells had one to six
Iaggards, an average of 3.7 Iaggards per cell. The average
number of Iaggards per cell at anaphase I seems to be consider-
ably lower than the average number of univalents observed at
metaphase I. This indicates that some of the univalents migrated
toward the poles and become included in the daughter nuclei. Lag-
gards were observed in all of the cells at anaphase II (Fig. 9E).
Micronuclei, which ranged from one to seven per quartet, occurred
in 90% of the quartets (Fig. 9F). Fertility of all three plants,
as tested by the staining reaction with Iz-KI seems negligible.
Only one pollen grain out of l800 was stainable, and this solit-
ary grain may have been a contaminant.

The plant PHA-4 had variable chromosome numbers in its PMC'S.
The chromosome number ranged from 22 to 28 per cell as summarized
in Table 5. Only univalents and bivalents were observed; unival-
ents averaged l6.6 per cell, and bivalents, 5.l per cell. Cells

having 28 chromosomes were the most prevalent accounting for 48%

q. _.

>>o.<..<.< um_:Ecom meccooee

.muco_m>_n me guess: umo;m_; mc_zo;me

 

JN
NN
cu
m.
m—
J.
N_ m: :N wN

wo.m om.m_

—

C‘ I

l‘

 

mm.m mm.m_

N_ 0 RN

 

40

J'—

‘a
fls

:_ hw_ m wu

 

mm.m mm.m_

.‘

__ :_ n mu

.‘ ‘

 

a

oo.m oo.:_ N :N

 

oo.: oo.m_ _ (MN

\ p
.“

 

F—u—r—v—l—MNu—NN ’0‘
umdudxorxwoLn:#\0Ln;rcflr\«oLn4rafla>h~u>usdtonce
m

‘5

E2
NNFT

oo.mi oo.N_ _ NN

 

 

 

__ _ m__ou ._ _ x m__ou ._ou com

 

 

omwce>< mo .02 co_um_00mm< OEOmOEOLLU mo .02 .EOuzu .oz

 

 

 

 

 

 

 

.<A:u_o_amcuou:v:u<:m c_ :o_um_00mm< oEOmQEOLzu —2 mo >cm553m

m u._m<._.

4I

of the cells sampled. Out of 50 cells counted, only four cells
had less than 25 chromosomes. Thus, the plant was considered to
be a tetraploid. In those cells with 28 chromosomes, the maxi-
mum configuration was 8 bivalents and l2 univalents, and the mean
chromosome association was approximately 5 bivalents and IS uni-
valents. Data were difficult to obtain due to considerable over-
lapping of chromosomes at metaphase I.

The plant PHA-5 approached a pentaploid constitution and
represents the expected chromosome number from such a cross.
Again the chromosome number of the PMC's varied, ranging from 25
to 34 per cell (See Table 6). The distribution of the number of
chromosomes per cell was rather evenly spread with no particular
number of chromosomes occurring in any large percentage of cells.
Relatively few trivalents, quadrivalents, and pentavalents were
seen, and they occurred on the average of 0.I5, 0.I4, and 0.03
per cell, respectively (Table 6). Occasionally, chromosome
fragments and what appeared to be uncondensed chromatic material
were observed at metaphase I. The uncondensed material was
stained the same as the chromosomes, but occurred in a long thin

thread.

42

._m_coume o_um50ccu nonconcouc: mc_>m: Amvm__ouee
.mucmEmmcm mc_>m; Amvm__oue

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

_ a. o
_ m_ m
oo.__ oo.N_ _ o. a. N.__ a :m
_ m cm
_ _ _ _ o. _
om.o om._ om.o om.m oo.: _ N m N Immm N mm
_ :_ :
om.m_ oo.m _ m. a mum N Nm
_ m. m.
om.o oo.m om... _ _ m o. m.m N _m
...c...~ N _ m
mm.m om.o_ «am o. o. m.n_ m on
N N m.
N o_ m
«N m. o.
mm.o m_.o mm... m_.o_ _ n a. 0.5. m wN
_ ~ _ m __
«a. o. A
oo.m oo.m _ m m m.m m RN
_ ml. __
_ m o.
a_ _ m A
om.o oo.o om.~_ _ m m_ A... : om
_ _ m A.
a. __ m
u... — w m
om.“ 0:.m N N __ N.:_ m mu
_ m m—
>, >_ ___ __ _ m__6u > >_ ___ __ _ x m__ou __6u toe
omoco>< we .02 co_um_o0mm< oEOmOEOLLQ mo .02 moEOmoeoLLU
we .02

 

 

 

 

 

 

.A:u_o_amucoazv mu<xm c_ co_um_u0mm< mEOmOEOLLU .2 we >LmFEJm

o m4m<H

J

43

Figure l0. PHA-5 Microsporogenesis.
A. Pachytene stage with 3 micronuclei. (840x)

B. Diakinesis showing I (arrow) + I3 II and I micronucleus

(arrow). (850x)
C. Anaphase II Showing 2 micronuclei. (ll00x)

D. Anaphase II with chromatin bridge and 3 micronuclei. (IIOOX)

 

Figure l0

45

Discussion

 

Hordeum jubatum has been considered to be an allotetraploid

 

(2N = 28) by Boyle and Holmgren (I954) and Rajhathy and Morrison
(I96l). Work by Wagenaar (I959, I960) and Rajhathy et al., (I964)
demonstrated thaf‘the species is a segmental allotetraploid. fl;

compressum, a diploid species, (2N = l4) was crossed with H; 'ub-

 

atum by Schooler et al., (I966) who synthesized the colchicine
induced amphiploid (2N = 42) of this hybird.

In the F between H; jubatum and fl; compressum the average

 

l

number of chromosome associations were 7.2 univalents, 6.36 bi-
valents, 0.I6 trivalents, and 0.04 quadrivalents. The averages
for the amphiploid were 2.36 univalents, I9.24 bivalents, 0.44 tri-
valents, and 0.28 quadrivalents (Schooler, g£_gl., I966). The
plant C-l, selfed progeney of the amphiploid, had averages of 0.70
univalents, l7.49 bivalents, 0.I0 trivalents, and I.44 quadrival-
ents (Starks and Tai, I974).

Because of the prevailing bivalent associations found in both

the F (where H; iubatum contributed the two genomes: AjA'j) and

I
C-I (having four genomes of H; jubatum: AjAjA'jA'j) it was suggest-
ed that when only one set of each genome Of.fl; jubatum chromoSomes

(AjA'j) were present, homeologous pairing would prevail (Wagenaar,

I959; Starks and Tai,l974). When both sets of each genome

46

(AjAjA'jA'j) are present, pairing is thought to occur only between
homologous chromosomes. In our previous work, we tentatively
suggested that H. jubatum and H; compressum shared a common

genome and assigned the following genome formulae to the species

and their amphiploid: H; jubatum (A A A'A'), H; compressum

 

(A'A'), amphiploid (A A A'A'A'A') and C-I (A A A'A'A'A').

The selfed seeds of the amphiploid grew into plants which we
coded C-l. These plants were used as the female parent in a cross
with tetraploid H; vulgare and the progeny produced was coded
PHA l to 5.

Modes of haploid formation have recently been treated by Kasha
(I974). In general, haploids are produced by androgenesis, parthen-
ogenesis, interspecific hybridization, chromosome elimination, and
tissue culture technique.

The origin of PHA l to 3 by androgenesis, the development of
a haploid embryo from a male nucleus (Rieger §t_al., I968) was dis-
regarded for two reasons. The chromosome number of the three plants
exceeds the number of chromosomes expected in the spenn nucleus of
the tetraploid male parent. Furthermore, the morphology of PHA
l to 3 had little characteristics of H; vulgare; in particular,
auricles were absent.

It is possible that the three plants resulted from the

deveIOpment of an unfertilized egg. The chromosome numbers are

47

the same as would be expected in the female gametes of C-I which
has a high percentage of regular anaphase segregation (2l-2l)
(Starks and Tai, I974). Another reason for suspecting that the
plants arose from an unfertilized egg is that the cytological
data of the original FI synthesized by Schooler (I966) is compar-
able to that of PHA l to 3. II PHA I to 3 arose from unfertilized
eggs, they should have the same chromosome constitution as the

original F1, with two genomes (AjA'j) from H; jubatum and one

 

(A'c) from H; compressum. The morphological description of the

I

seems to be in agreement with that of PHA l to 3.

F which is similar to C-I and lacking characteristics 0f.fl; vulgare

Another explanation of the origin of PHA I to 3 may be that
they resulted from fertilization and subsequent chromosome elimina-
tion by way of multipolar cell divisions (Tai, I970) or the selec-
tive elimination (Ho, I973) of individual chromosomes. There is
evidence to support this explanation. H; vulgare has often been
involved in chromosome elimination as demonstrated in crosses
between H; vulgare and H; bulbosum (Lange, l97l; Subrahmanyam and
Kasha, I973; and Ho, I973). But, depending upon the chromosome
dosage of H; vulgare, relative to that Cf.fl; bulbosum, genomes of
.5; bulbosum are eliminated. Only the combination in the ratio of
one H; vulgare genome to two genomes of H; bulbosum consistently
produced hybrids in which H; bulbosum chromosomes are not lost

(Kasha, I974). Whatever the case, there appears to be consider-

48

able incompatibility whenever the H; vulgare genome is involved
in an interspecific cross. A similar effect may be causing the
genomes of H; vulgare to be eliminated in this cross. Again,

morphological and cytological similarities between the original

F and PHA l to 3 support the view that H; vulgare chromosomes

I
are not present. We believe that fertilization did occur, be-

cause PHA 4 and 5 were obtained from the same procedure and under
the same conditions.

At this time we have not determined if fertilization occurred
in the production of PHA l to 3, and if so, by what mechanism the
chromosomes were lost. Work done by Kasha and co-workers (Kao and
Kasha, I969; Subrahmanyam and Kasha, I973; and Ho, I973) in crosses
between_fl: vulgare (2X) and_fl: bulbosum (2X) suggest that chromo-
somes are individually eliminated during embryonic deveIOpment. In
this process the chromosomes apparently decondense one after another
and disappear into the cytoplasm.

Huskins (I948) has shown that chromosomes can be eliminated
in a single step by a process he coined ”somatic meiosis." Tai
(I970, l97l) frequently observed the loss of whole genomes by
multipolar cell division during meiosis in various plant species.
Pera and Rainer (I973) working with tissue cultures of kidney
epithelium and fibroblasts have shown that chromosomes of multi-
polar mitosis are separated into complete genomes. Rizzoni, §£_§l.,

(I974) working with cultures of Rhesus kidney cells demonstrated

49

the frequent occurrence of multipolar mitosis by whole genomes in
polyploid cells. They Showed that segregating cells tend to give
rise to further segregation processes and concluded that multi-
polarity can be inherited. The heritability of multipolar cell
division was suggested by Tai (I970) and Ho (I973) and the phenomenon
has been observed in our laboratory (Chen, I975 unpublished Ph.D.
thesis). Camenzinal (I974) demonstrated chromosome elimination
in the third cleavage division of the gall midge. In this work,
55 out of 66 chromosomes were eliminated in one division. In the
next division a solitary chromosome was eliminated in a single
step very early in the development of embryos, thus giving rise to
haploids or polyhaploids, Work is currently underway in our labora-
tory to investigate whether chromosomes involved in this study were
eliminated gradually, in genomic sets, or through a combination of
both means during embryonic development.

Fertilization undoubtedly occurred in the development of
PHA-4. The chromosome number, although variable, exceeded the
gametic number of the normal egg cell of C-1. Further evidence for
H; vulgare contributing chromosomes to this plant came from general
morphological characteristics, particlarly the deveIOpment of
auricles. Neither C-I nor its parents have auricles; therefore,
the genes for auricles in PHA-4 had to be contributed by H; vulgare.
Only univalents and bivalents occurred in the “tetraploid”, and it

can therefore be assumed that there are two genomes present from

50

H;_]ubatum, one genome from H; compressum, and one genome from
H; vulgare (AjA'jA'c+Vx). Presumably the other genome from H;
vulgare was eliminated following fertilization, but it may haye
been eliminated prior to fertilization. It seems probable that
most of the chromosomes were eliminated by means of multipolar
cell divisions followed by the sequential elimination of a few
chromosomes in the manner observed by Camenzinal (I974).

If a single dose of the A genome is present, homeologous
pairing and the formation of multivalents are expected to occur.
This is the case in PHA I to 3 (AjA'jA'c) and PHA-5 (A'jA'jA'cV+VX).
Why it does not occur in PHA-4 is unexplained unless multivalent
formation is in some way inhibited by the partial genome 0f.fl;
vulgare or an error was made in interpretation.

All the PHA-5 have a considerable number of bivalents (Fig. lOB).
These may be explained by homologous pairing of H; vulgare chromo-
somes, by homeologous pairing between the genomes of H; jubatum,

and by homeologous pairing between the chromosomes of H; jubatum

and H. compressum.

 

The cytological evidence in this study lends support to the
view of taxonomists (Nevski, I934; Aberg, I940; Bowden, I959) that
_fl; vulgare is not closely related to the wild species of Hordeum.
The influence of the H, vulgare genome on chromosome elimination
is not only interesting but may be economically important. If

chromosome elimination can be controlled, desirable chromosomes

SI

may be left inside prot0plasts while chromosomes carrying genes for
undesirable traits may be expelled. The remaining genes would have
an Opportunity to pair in such a way with the genes of cultivated

barley that crossing over and recombination of genes would follow.

SECTION III

A Cytogenetic Study Involving The Genome of The
Amphiploid (C-I), And Diploid H; vulgare

52

53

INTRODUCTION

As Stated previously, deficiencies such as limited pubescence
and intolerance to cold and salt might be overcome by the transfer
of genes for these characteristics either through crosses of
_H; vulgare with species in other sections of the genus or through
wider crosses within the tribe Triticeae. In I940, Quincke reported
a cross between H; jubatum and H; vulgare in which 70% of the florets
set seed and 9% of the seeds germinated. Kerber (personal comnunica-
tion) made similar crosses obtaining highly sterile hybrids in which
preliminary cytological observations indicated that chromosome
numbers varied from l7 to 22 with one preparation yielding cells
with a chromosome number up to 38. Kerber concluded that the chromo-
some behavior during meiosis was extremely abnormal and that the lack
of chromosome pairing during meiosis demonstrated little if any
homology between the parent plants. Others.who synthesized and
studied this interspecific hybrid were Morrison, S£_il-’ (I959),
Vinogradova (I946, cited in Smith, I95l; and Price, I968), and
Rajhathy and Morrison (I959). Recently, Murry (I975) analyzed micro-
sporogenesis in the reciprocal cross, H; vulgare x H;_jubatum, pro-
duced by Robert Steidl, Department of Crop and Soil Science, Michigan
State University. She also found complete sterility, variable chromo—
some numbers from l6 to 23, and abnormal meiosis. She concluded that

the parental species demonstrated either a complete lack of homology

54

or genetic incompatibility relative to chromosome pairing.
Because of the high sterility in the hybrids between

_H; jubatum and.fl;.¥21§i£§’ an indirect approach to obtaining

gene transfer was undertaken. The H; jubatum x_fl; compressum

amphiploid, C-l, was crossed with H; vulgare in an attempt to

transfer the desirable traits of H; jubatum into the genetic

constitution of diploid H; vulgare. The resulting cross was

called 12-1 (Fig. 11).

Figure II. A Diagram of a Cross Between C-I and H; vulgare (2X)

 

C-l (6X) x .H; vulgare (2X)
(A A A'A'A'A') 1 (vv)
H. vulgare (2X) x l2-I (4X) x l2-l (4X)
(vv) (A A'A'V) (A A'A'v)
12-1-3 12-1-1

The purpose of the research in this section was to ascertain the
meiotic chromosome behavior of the crosses presented in Figure II

and to determine the genome constitution of the progeny.

Beselts
The spike and spikelet morphology of the plants involved in

this section are shown in Figure l2.

55

Figure l2. Spike and Spikelet Morphology of H; vulgare (2X),
(C-I, l2-l, and 12-1-8. Spikelet Awns are
Invisible.

A. Spikes of C-1, l2-I, and H4_vulgare. (0.3x)

B. Spikelets of C-I, and H; vulgare. (I.Ix)

C. Spikes of l2-l, l2-I-B, and H; vulgare. (0.3x)

D. Spikelets of l2-l, l2-l-B, and H; vulgare. (I.Ix)

 

57

The metaphase I data of 12-1 are summarized in Table 7. The
chromosome number of lZ-I was, as expected, tetraploid (2N = 28).
Meiotic behavior at metaphase I (Table 7) showed an average l5.6
univalents and 4.6 bivalents per cell. No trivalents were
recorded, but quadrivalents occurred on the average of 0.7 per
cell. See Figure I3.

Plant culture 12-1 was backcrossed to H; vulgare (2X),

(Fig. II), in an attempt to improve the seed Size of I2-l. The
hybrid obtained, coded lZ-l-B, (Table 8), averaged 3.6 I, 8.6 II,
O.l4 III, 0.06 IV, l.00 fragment, and 0.49 ”uncondensed” chromatin
material at metaphase I (Fig. l4).

It was expected that l2-I-B would display a triploid constitu-
tion (3N = 2l), however, the chromosome numbers observed, not
including fragments or ”uncondensed” chromatin, ranged from l4 to
32. Only 6 cells approached the triploid number, and three of
these cells contained fragments (Fig. I4). The widely variable
chromosome numbers suggest that processes such as endomitosis,
chromosome elimination, and multipolar cell divisions might be
occurring.

Plant l2-I was allowed to self pollinate, and chromosome
behavior of the progeny, 12-1-1, was observed. The cytological
data were surprisingly variable and are summarized in Table 9.
Averages of chromosome configurations were meaningless as chromo-
some numbers ranged from 6 to 96. Over 27% of the cells recorded

had a ploidy level between 6X and 7X. Multivalents were observed,

58

TABLE 7

SUMMARY OF MI CHROMOSOME ASSOCIATIONS IN l2-l (2N=28)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Chromosome Association Number of Cells %

I ll IV
I0 9 2 5
IO 7 I I 2
l0 5 2 I 2
I2 6 l 2 5
I2 4 2 I 2
“I 1 2 2 5
IA 5 I h 9
IA 7 5 II
I6 6 I 2
I6 h I 8 I8
I6 2 2 3 7
I6 3 I 2
I8 5 3 7
I8 3 I h 49
20 h 5 II
22 3 l 2

Total hh

Avg.

Per -

Cell IS.6 h.6 0.7

 

 

 

 

 

 

 

59

Figure 13. Microsporogenesis in Plant I2-I

Interphase cell pinching into 3 cells. (ll00x)

Metaphase l with lh I and 7 II° (900x)

Telophase I. (800x)

Anaphase II showing evidence of multipolar cell division

and cells with small numbers of chromosomes resulting
from chromosome elimination. (680x

 

 

 

 

60

 

Figure l3

61

Figure l3 COTI't.
E. "Quartet" of 8 cells (2 cells out of focus). (750x)

F. A normal quartet. (750x)

 

Figure l3 (cont'd.)

‘ "I.

’5

I
. “i
"A .

F

63

TABLE 8

SUMMARY OF M1 AND DIAKENESIS CONFIGURATIONS FOR

l2-l-B

 

IV

Frag.

Uncondensed

2N

# of Cells

 

I6

 

l4

 

Ih

 

lh

 

l6

._s—A—LNN

 

.l32

 

22

 

23

 

23

 

26

 

26

 

C‘O‘J—‘kOCfiCD-‘N

21

 

I6

 

l7

 

l8

 

20

 

2l

 

O‘U‘IbNd

25

 

I8

 

2l

 

2l

 

2l

 

22

 

23

 

23

 

23

 

u: h) n: L: :-x» to

2A

 

26

 

27

dI—I

 

 

\Okomo‘u‘lmm-F‘MML’

 

to \0 Lo \0 KO \9 \o \D to \o \0 NO (n c» c» c» (n (n ~q ~q \4 \4 ‘4 \4 \4 \J-u ~q ~q (m

 

 

 

 

 

30

d

 

 

 

Total
Avg.
Per
Cell

61+

TABLE 8--Cont1nued

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

I II III IV Frag. Uncondensed 2N # of Cells
IO 9 1 3] 1
'0 20 1
I lO 2] 1
I l0 1 2] 2
2 l0 2 22 1
3 l0 2 23 1
h lO 2h 1
h l0 3 2h 1
6 10 26 2
6 10 1 26 1
6 ‘0 2 26 1
6 10 2 26 1
7 10 27 1
8 I0 2 28 1
IO l0 4 30 1
‘I 22 1
2 l‘ 7 2h 1
2 ‘I L 21+ 1
I ll L, 26 I
_§_ II I l 27 l
6 11 ._ 28 1
9 “ 31 1
*3 '2 2 27 1
h 12 2 28 1
‘3 2 26 1
2 L12 28 ‘ 1
“ L3 1 230 1.
2 ‘3 1 2 1
2 15L. 412 1
237 56 9 h 65, #32 6S

3.6h 8.55 .1h .06 1.00 .69

 

 

 

 

 

 

 

 

 

65

Figure lb. Microsporogenesis of l2-I-B.

A. Diakinesis with ID ll. (950x)

8. Diakinesis with 2 I, II II, and 7 fragments (Arrows). (950x)
C. Diakinesis with 3 I, 9 II, and h fragments (Arrows). (950x)

0. “Quartet” with 7 cells. (950x)

 

Figure I“

7;' \
Q’
.1_-
'(
{1‘7”
4 "
w ;.
'53.”
,1, ’1".
.1
‘ /

KU

67

TABLE 9

SUMMARY OF MI CONFIGURATIONS FOR l2-l-I

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Chromosome Associations Number of Number of
I II III IV V Chromosomes Cells Ploidy
6 6 I X
A 8 I

43 2 I IO I

43 I II I

8 2 I2 I

h h I2 I

I0 I I2 3

6 A IN 2 2X
8 3 IA 2 2X
4 5 IA I 2X
9 I I I4 I 2X
9 3 l5 2

IO 2 I6 I

7 5 I7 I

9 I8 I

8 8 22 I

7 h I I 22 I

I8 2 I 25 I

I h 25 I

II 6 l 26 I

I7 5 27 I

8 5 3 27 1

8 IO 28 I 11x
I4 8 30 2

I7 7 3] I

I I7 35 I 5X
13 9 2 37 I

I3 II I 39 I

I6 9 I I Al I

 

 

 

 

 

 

68

TABLE 9--Continued

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Chromosome Associations Number of Number of
I II III IV V Chromosomes Cells Ploidy
I2 l5 A2 3 6X
I2 2 L1 L12
1 5 12 1 113 1
I5 9 2 l A3 I
I2 I3 2 AA I
l2 9 3 I AA I
II II 2 l AA l
IA I5 AA I
I7 8 I 2 AA I
23 II A5 I
I6 IO 2 I A6 I
IO I3 2 I A6 I
I5 IA I A6 I
I2 I3 2 A6 I
IA I6 A6 I
I5 IO 2 2 A9 I 7X
IA 7 30 7A I
A2 8A I I2X
II 37 85 l
43 A2 87 l
1,1, 88 I
I} A0 93 I I
8 AA 96 I
A8 96 5
6 L13 1 96 1 IAx-z

 

 

 

 

 

 

69

but their frequency was relatively low compared to the frequency
of univalents and bivalents. Out of 66 cells counted, 23 cells
had multivalents. The highest number of multivalents (A IV's)
was observed in a single hexaploid cell. Neither fragments nor

“uncondensed” chromatin was observed.

Discussion

 

The overall morphology of l2-I was very similar to the amphi-
ploid, C-I, (Fig. 12); however, the expression of auricles in I2-I
indicated the presence of the EL vulgare genome. The lack of
fertility of l2-I was evidenced in the absence of stainable pollen
and seed set.

The consistent count of 28 chromosomes along with the average
chromosome configurations of l5.6 univalents, A.6 bivalents, and
0.7 quadrivalent leaves no doubt that plant I2-I has a tetraploid
constitution. The average number of univalents is higher than the
maximum expected number of IA. However, this high number may be
accounted for by the early separation of bivalents. It would not
have been unexpected to find trivalents, but none were recorded.
Perhaps their absence may be explained by the presence of the
.fl; vulgare genome which may in some way inhibit multivalent forma-
tion.

The maximum number of associated chromosomes, II or II + IV,

observed might be explained by secondary associations (Sadasivaiah

70

and Kasha, (I971) between the chromosomes of u; vulgare and limited
homology between the chromosomes of u; jubatum and fl; compressum.

It was impossible to determine which chromosomes were pairing, but
it was assumed that the chromosomes from fl; jubatum pairing auto-
syndetically would account for seven bivalents. Work done by Kerber
(Appended), Murry (I975), Rajhathy and Morrison (I959), supports the
conclusion that no significant homology exists between E; iubatum
and L 29.1.2229.-

In plant I2-I-B, the average chromosome configurations at
metaphase I were 3.6 univalents, 8.6 bivalents, 0.IA trivalent,
0.06 quadrivalent, I.OO fragment, and 0.A9 ”uncondensed” chroma-
tin. If these averages are added, the plant appears to have a
chromosome constitution somewhat in excess of 2I chromosomes.
However, out of the total number of 65 cells observed, 60% of the
cells have chromosome numbers ranging from 2l to 28. Perhaps
this indicates the range of greatest chromosome stability in the
plant.

The egg of I2-I would have a A A'A'V genome constitution if
reductional division did not occur during xmegasporogenesis.

The addition of one fl;_vulgare genome from diploid fl; vulgare would
make the l2-I-B zygote a pentaploid with genome formula A A'A'VV.
This is the same genome formula as the “pentaploid” from the poly-
haploid investigations (see Section 2). The two plants have very
similar morphology (Steidl, personal communication) as would be

expected if the hypothesized genome formulae are correct.

7l

At this time there is no way of determining which of the
12-1-8 chromosomes are pairing. It is reasonable to assume that
the bivalents are autosyndetically formed by the H; iubatum genomes
and by the allosyndetic pairing of the genomes of H; vulgare.

Plant IZ-I-I is extremely interesting cytologically exhibiting
a wide range of chromosome numbers among its pollen mother cells.
It obviously undergoes multipolar cell divisions (Fig. ISL), and it
shows an unusually high frequency of ring bivalents in some cells
(Fig. 151:).

The plant has no pollen stainability or seed set and is
virtually sterile. It lacks auricles or other obvious characteris-
tics of H._ ulgare, yet, its spike does not resemble that of the
amphiploid.

The I2-I-I is thought to have originated from the selfing of
I2-I. However, it is difficult to explain the high chromosome
numbers observed. At this time, it is suggested that the plant
doubled its chromosome number, produced an unreduced egg and was

self pollinated.

72

Figure I5. MicrosporOgenesis in IZ-I-I.

A.

Metaphase I with 9 cells; 7 cells have different
chromosome numbers. (700x)

Metaphase I, one of the cells from above with

II I and 37 II. (890x)

Metaphase I with 6 cells showing different chromosome
numbers. (700x)

Metaphase I, one of the cells from above with A8 II.

(9A0x)

 

Figure l5

7A

Figure l5. continued.

E. Metaphase I with AA II. (9A0x)

F. Metaphase I, two cells with different chromosome numbers. (890X)
G. Metaphase II.(lA00x)

H. Metaphase II with a 2-18 distribution. (l320x)

75

 

 

 

Figure l5

76

Figure l5. continued.
I. Metaphase II with a A-IA distribution. (I320x)

J. Metaphase II showing 2 microceIIs with ca. 7 chromosomes
each. (I320x)

K. Anaphase II, tetrapolar separation. (I320x)

L. Late teIOphase II, tripolar. (I320x)

 

Figure l5 (cont'd.)

78

Genome formulae were established for I_-I_,_ caIifornigJfl, Ii; 39p;-
pressum, fl. brachyantherum, and H‘.iubatum. In the A genome genetic
control for homologous pairing of chromosomes was suggested.

Multipolar cell division seemed to occur whenever the fi;,vulgare
genome interacted with the genomes of C-I. This interaction, re-
sulting in chromosome loss may have future significance particularly
in the production of haploids and polyhaploids.

Guidelines have been established for future breeding programs
using the materials in this research. However, the need for further'

work was indicated by some unanswered questions (particularly in

Section 3).

LITERATURE CITED

Aase, H. C. and L. Powers. I926. Chromosome numbers in crOp
plants. Amer. J. Bot. I3:367-372.

C
Aberg, E. I9AO. The Taxonomy and phylogency of Hordeum L. Section
Cerealia ands. Symb. Bot. Upsal. Vol. A (2), I56 pp. iIIus.

Allard, R. W. I960. Principles of Plant Breeding. John Wiley
and Sons, Inc. A85 pp.

Anderson, E. I9A9. Introgressive Hybridization. Wiley and Sons.
Bowden, W. M. I962. Cytotaxonomy of the native and adventive

species of Hordeum, Eremopyrum, Secale, Sitanion, and
Triticum in Canada. Can. J. Bot. A0:I675-I7II.

 

 

Bowden, W. M. I959. Chromosome Numbers and Taxonomic Notes on
Northern Grasses. l Tribe Triticeae. Can. J. Botany 37:IIA3-IISI.

Boyle, W. S. and A. H. Holmgren. I95A. A cytogenetic study of
natural and controlled hybrids between Agropyron trachycaulum
and Hordeum iubatum. Genetics A02539-SA5.

 

 

Brink, R. A. and D. C. Cooper. I9AA. The antipodals in relation
to abnormal endosperm behavior in Hordeum lubatum X Secale
cereale hybrid seeds. Genetics 29:39I-AO6.

 

Brink, R. A. and D. C. C00per. l9A7. The endosperm in seed deveIOpment.
Bot. Rev. l3zA23-5AI.

Brink, R. A.; D. C. Cooper, and L. E. Ausherman. I9AA. A hybrid between
Hordeum jubatum and Secale cereale. J. Hered. 35:67-76.

 

 

Camenzinal, R. I97A. Chromosome elimination in Heteropeza pygmaea.
Chromosome (Berl.) A9:87-98.

 

Chen, Y. R. I975. Multipolar cell division in AgrOpyron cristatum
(L.) Gaertn. (Gramineae). PH.D. Dissertation, Michigan State
University 90 pp.

 

Cooper, 0. C. and R. A. Brink. I9AA. Collapse of the seed following
the mating of Hordeum jubatum X Secale cereale. Genetics

29:370-390.

 

 

Covas, G. I9A9. Taxonomic observations on the North American Species of
Hordeum. Madrono IO:I-2I.

Covas, G. and B. Schnack. I95]. Hibridacion interspecifica en Hordeum II.
fig compressum X H; caIifornicum. Rev. Argent, Agron. I8:I3-l7.

 

 

79

8O

Davies, D. R. I960. The embryo culture of interspecific hybrids
of Hordeum. New. Phytol. 59:9-IA.

Dewey, D. R. I96I. Polyhaploids of Crested Wheatgrass. Crop
SCE. 1:2h9-25h0

Driscoll, C. J. I972. Genetic suppression of homeologous chromo-
some pairing in hexaploid wheat. Can. J. Genet. Cytol.

IA:39-A2.

Feldman, M. I966. The effect of chromosomes SB, 50, and SA on
chromosome pairing in Triticum aestivum. Proc. Natl.
Acad. Sci. U.S. 55:19A7-I953.

Goodspeed, T. H., and P. Avery. I929. The occurence of a
Nicotiana glutinosa haplont. Proc. Nat. Acad. Sci. U.S.
I5:502-SOA.

 

Gottschalk, W. I973. The genetic control of meiosis. Genetics

7A:S99.

Hitchock, A. S. and A. Chase, I950. Manual of Grasses of the
United States. U.S. Dept. Agr. Misc. Publ. No. 200.

IOSI pp.

Ho, K. M. I973. Genetic control of chromosome elimination in
interspecific hybrids leading to haploid formation in
barley. (Hordeum vulgare L.) PH.D. Thesis, University
of Guelph, Guelph, Ontario.

Huskins, C. L. I9A8. Segregation and reduction in somatic tissues.
I. Initial observations on AlIium cepa. J. Hered.

39:31]“3250

Jensen, C. J. I97A. Production of monOploids in barley: A
progress report. pp. l67-l97. In: PonpIoidy and
Induced Mutations in Plant Breeding. IEAE Vienna.

Kao, K. N. and K. J. Kasha. I969. Haploidy from interspecific
crosses with tetraploid barley. pp. 82-88. In: Barlev
Genetics II. Ed. R. A. Nilan. Washington State University.

Kasha, K. J. I97A. Haploids from somatic cells. pp. 67-87. In:
Haploids in Higher Plants. Proceedings of the First Inter-
national Symposium. Ed. K. J. Kasha. The University of
Guelph. '

8i

Katayama, Y. I93A. Haploid formation by x-rays in Triticum
monococcum. Cytologia 5:23A-237.

 

Kerber, E. R. l97A. Personal communication.
Kerber, E. R. l97A. Unpublished data.

Kihara, H., and K. Tsuenewaki, I962. Use of alien cytoplasm as
a new method of producing haploids. Jap. Jour. Genet.

37:310-313.

Kimber, G. and R. Riley. I963. Haploid angiosperms. Bot. Rev.
29:A80-53I.

Konzak, C. F., L. F. Randolph, and N. F. Jenssen. I95I. Embryo
culture of barley species hybrids. J. Hered. A2:l25-I3A.

Kundson, L. I922. Nonsymbiotic germination of orchid seeds. Bot.
Gaz. 73:l-25.

Lange, W. I97I. Crosses between Hordeum vulgare L. and H; bulbosum L
II. Elimination of chromosomes in hybrid tissues. Euphytica

 

Levan, A. I9A5. A haploid sugar beet after colchicine treatment.
Hereditas 3I:399-AIO.

Mitchell, W. W. and A. C. Wilton. I96A. The Hordeum jubatum -
calspitosum - brachyantherum complex in Alaska. Madrono

 

 

Morrison, J. W., A. E. Hannah, R. Loiselle, and S. Symko. I959.
Cytogenetic studies in the genus Hordeum II. Interspecific
and intergeneric crosses. Can. J. Plant Sci. 39:375-383.

Morrison, J. W. I959. Cytogenetic Studies in the Genus Hordeum. I
Chromosome Morphology. Can. J. Botany 37:527-538.

Muntzing, A. I937. Note on a haploid rye plant. Hereditas 23:A0l.

Murry, L. E. I975. A Cytogenetic Investigation of X Agrohordeum
pilosilemma. PH.D. Dissertation, Michigan State University.

96 pp-

 

Nakao, M. I9II. Cytological studies on the nuclear division of the
pollen mother-cells of some cereals and their hybrids. Jour.
Coll. Agr. Tohoku Imp. Univ. A:l73-I90.

Nevski, S. A. I93A. Hordeae Benth. In: V. L. Komorov, Flora
USSR. Vol. II. Leningrad. pp. 590-728.

82

Newcomer, E. H. I953. A new cytological and histological fixing
fluid. Science II8:I6I.

Norstog, K. I972. Early development of the barley embryo: Fine
structure. Amer. Jour. Bot. 59:I23-I32.

Norstog, K. I973a. New synthetic medium for the culture of
premature barley. In Vitro 8:307-308.

Norstog, K. I973b. Personal communication.

Pera, F. and B. Rainer, I973. Studies of multipolar mitosis in
eupIoid tissue cultures. Chromosoma (Berl.) A2:7I-86.

Perak, J. T. I9Al. Numero do cromosomas de algunas especies de
Hordeum espontaneas en Argentina. An. Instit. Fitot.
Sta. Catalina. 3:7-II.

Price, P. 8. I968. Interspecific and intergeneric crosses of
barley. pp. 85-95. In: Barley: Orgin, Botany, Culture,
Winterhardiness, Genetics, Utilization, Pests. USDA Agr.
Handbook No. 338, Washington.

Quincke, F. L. I9AO. Interspecific and intergeneric crosses
with Hordeum. Can. J. Res. I8:372-273.

Rajhathy, T. I966. Notes on the Hordeum iubatum complex.
Madrono, S. Fransisco. I8:2A3—2AA.

Rajhathy, T., and J. W. Morrison, I959. Cytogenetic studies in
the genus Hordeum IV. Hybrids of H; jubatum, H; brachyanthe-
LEE1.fl; vulgare and a hexaploid Hordeum sp. Can. J. Genet.
Cytol. l:I2A-l32.

Rajhathy, T. and J. W. Morrison, l96l. Cytogenetic studies in
the genus Hordeum V., H. jgbatum and New World Species.
Can. J. Genet. Cytol. 3:378'390.

 

Rajhathy, T.; J. W. Morrison; and S. Symko. I96A. Interspecific
and intergeneric hybrids in Hordeum. pp. I95-2I2. In: Barley
Genetics I. Proceedings of the First International Barley
Genetics Symposium. Wageningen.

Raven, P. H., and H. J. Thompson. I96A. Polyhaploid and angio-
sperm evolution. Amer. Nat. 98:25I-252.

Redman, R. E. and D. S. Borgaonkar, I966. A cytological study
of the Hordeae in the Dakotas. Cytologia. 3I:2I9-23l.

83

Rieger, R., A. Michaelis and M. M. Green. I968. A Glossary of
Genetics and CytOQenetics. Springer-Verlag, New York, Inc.

Riley, R. and V. Champman. I958. Genetic control of the cytologically
diploid behavior of hexaploid wheat. Nature I82:7I3-7I5.

Rizzoni, M.; F. Palitti; and P. Perticone, l97A. EupIoid segregation
through multipolar mitosis in mammalian cell cultures. Chromosoma
(Berl.) 45:151-162.

Sadasvaiah, R. S. and K. J. Kasha. I97I. Meiosis in Haploid Barley--An
Interpretation of Non-Homologous Chromosome Associations. Chromosoma

(Berl.). 35:247-263.

Satina, 5.; A. F. Blakeslee; and A. G. Avery, I937. Balanced and unbalanced
haploids in Datura. Jour. Hered. 28:I93-I98.

Schooler, A. B. I9603. Wild Barley hybrids. I. Hordeum compressum X
Hordeum pusillum. J. Heredity 5l:l78-I8l.

 

Schooler, A. B. I960b. Wild barley hybrids. II. Hordeum marimum X
Hordeum compressum. J. Heredity 5I:2A3-2A6.

 

Schooler, A. 8., Jill Nelson, and Ann Arften. I966. Hordeum jubatum L.
crosses with Hordeum compressum Griseb. CrOp Sci. 6:l87-I90.

Smith, L. I95I. Cytology and genetics of Barley. Bot. Rev. I7:I-355.

Starks, G. D. and W. Tai. I97A. Genome analysis of Hordeum iubatum
and H; compressum. Can. J. Genet. Cytol. l6z663-668.

 

Subrahmanyam, N. C. and K. J. Kasha, l97I. Increased barley haploid
production following gibberellic acid treatment. Barley Genetics
Newsletter. I:A7-50.

Subrahmanyam, N. C. and K. J. Kasha, I973. Selective chromosomal
elimination during haploid formation in barley following inter-
specific hybridization. Chromosoma A2:III-I25.

Symko, S. I969. Haploid barley from crosses of Hordeum bulbosum (2X)
X H. vulgare (2X). Can J. Genet. Cytol. Il:602-608.

Tai, W. I970. Multipolar meiosis in diploid crested wheatgrass,
Agropyron cristatum. Amer. J. Bot. 57:ll60-II69.

 

Tai, W. I97I. Multipolar Meiosis and Spindle Organizers in Various
Plant Species. Genetics 68:566.

Thompson, W. P. I939. The frequency of fertilization and the nature
of embryo and endosperm development in intergeneric crosses
in cereals. VII Int. Congr. Genet., Proc. (Abs.).

8A

Thompson, W. P. I9AO. The causes of hybrid sterility and
Incompatability. Trans. Royal Soc. Canada, Ill, Sect. 5.
3A:I-3. 1

Thompson, W. P. and D. Johnson, I9A5. The cause of incompatibility
between barley and rye. Canad. Jour. Res. C. 23:l-l5.

Vinogradova, N. M. I9A6. (Hybridization between cultivated barley
and wild species of Hordeum). Trudy Zonal. Inst. Zernavovo
Khoziaistva Nechernozemnoi Polosy SSR I3:I3A-I37.

Wagenaar, E. B. I959. Intergeneric hybrids between Hordeum jubatum L.
and secale cereale L. J. Hered. 50:l9A-202.

 

Wagenaar, E. B. I960. The cytology of three hybrids involving
Hordeum jubatum L.: The chiasma distributions and occurrence
of pseudo ring-bivalents in genetically induced asynapsis,

Can. J. Bot. 38:69-85.

 

Wilson, A. 5. I876. Wheat and Rye hybrids. Bot. Soc. Edinb.
Trans. and Proc. I2:286-288.

Ziebur, N. K.; R. A. Brink; L. H. Graf; and M. A. Stahmann. I950.
The effect of casein hydralysate on the growth in vitro
of immature Hordeum embryos. Amer. Jour. Bot. 37:IAA-IA8.

APPENDIX

85

PRELIMINARY OBSERVATIONS ON THE CYTOLOGY OF THE
F‘ HYBRID BETWEEN HORDEUM JUBATUM
AND COMMON BARLEY

By E. R. Kerber”

Preliminary observations of meiosis have been made on the

F hybrid between Hordeum jubatum and common barley. The somatic

 

I
chromosome number of the iubatum parent is 28 and that of common

barley is IA. The expected number in the hybrid is 2l. However,
cytological examination of microsporocytes of the hybrid reveal
considerable variation of chromosome numbers. In one preparation
the numbers vary from l7 to 2], while cells of other slides pre-
pared from anthers of a single floret show from 20 to 22 chromo-
somes. The occasional anther appears to have all cells with 2]
chromosomes. One preparation is of particular interest in that
some of the cells have up to 38 chromosomes, including eight bi-
valents. This variability in chromosome number apparently is

not confined to pollen mother cells. Micronuclei are found
associated with tapetal and ordinary somatic cells of anther tis-
sue. Twenty metaphase chromosomes were counted in one cell
undergoing mitotic division, while in another approximately 75
chromosomes and a fragment were observed at mitotic anaphase.

The chromosome numbers in dividing root-tip cells of the hybrid

has not been determined.

86

87

Chromosome Behavior at Meiosis

Preliminary cytological observations are confined primarily
to first meta - anaphase. At early metaphase numerous univalents
and a few bivalents are found. The number of bivalents varies
from one to five, the most common being from two to three, and
these are usually of the open type. The bivalents are found on
the metaphase plate, while the univalents are grouped at either
pole. Together with these univalents are found double or triple
end-to-end associations of univalents. Side by side associations
of two or three univalents are also noted. These types of associ-
ations do not show chiasmata or diakinesis or at metaphase.
Occasional univalents are observed lying at the plate with the
bivalents. When no bivalents are present the univalents are more
or less scattered throughout the cell. At later metaphase the
bivalents separate and the daughter chromosomes move to opposite
poles. During this time some of the univalents move from the poles
to the plate where they appear longitudinally double. Other uni-
valents and the daughter halves of the bivalents also appear double
at this stage. None of these double appearing univalents has been
observed to resemble the characteristic, four-armed anaphase config-
uration of daughter halves in normal diploids. At the stage which
may be described as late anaphase the double appearing univalents
move to the poles, apparently without dividing. Lagging univalents
which do not reach the poles in time to be included in the telophase
nuclei form micronuclei. Most frequently the telophase nuclei are

of different sizes.

88

Second division of meiosis in the hybrid is very irregular
as indicated by bridge formation and lagging chromosomes at anaphase
and by numerous micronuclei. The number of daughter cells formed
varies greatly. At the 'tetrad' stage as many as eight spores may
be formed from one original pollen mother cell. The individual
spores usually have one large nucleus plus one or more micronuclei.

No well developed pollen grains are observed.

Discussion

 

The above preliminary observations on the cytology of the
hybrid fl. iubatum x common barley are of interest in several aspects.
Firstly, the behavior of the chromosomes at meiotic division is
extremely abnormal. Secondly, the lack of chromosome pairing and
complete sterility of the hybrid indicates there is little homology
between the chromosome sets of the two parental species. The pair-
ing that is observed may be the result of allo- or autosyndesis.
Thirdly, and perhaps the most striking observation is the variabil-
ity of chromosome numbers in both meiotic and mitotic cell divisions.

Further cytological studies on the FI hybrid are to be under-
taken with the object of determining the precise behavior of the
chromosomes throughout meiotic division, in particular during first
metaphase. Preliminary observations indicate the behavior of the
univalents and bivalents is similar to that found and described in

a haploid of Triticum vulgare by Mr. Person. It may be possible

 

to determine or substantiate the validity of his hypothesis regard-
ing univalent - bivalent behavior in relation to the action of the

spindle mechanism.

89

Another informative study can be made by investigating the
nature and causes of the variability of chromosome numbers observed
in meiotic and mitotic cell division. The preliminary observations
presented in this paper are from a clone of an F] plant. Clones
from six additional FI plants involving different parental varieties
or stocks are available for study and comparison with the present

observations.

”E. R. Kerber, Research Branch, Research Station, 25 Dafoe Road,
Winnipeg, Manitoba, R3T 2M9