CYTOGENETIC EVALUATION OF SECALE GERMPLASM
FOR FORAGE IMPROVEMENT

Thesis for the Degree of Ph. D.
MICHIGAN STATE UNIVERSITY

Mansour Nickneiad
I960

 

 

 

0-169

This Is to certify that the

thesis entitled

"“J'H - ‘ ‘ ‘ -' I" (\ 1:“! Y I "I ‘.‘ l '
V... i‘\/ Igug L -- U .'._l‘d -‘LA_.'V I“; - .4 K, in
‘4‘ IV ‘V Y ‘1 I . "
\,'- - _ - A.) ‘ Ii-l ‘J : LA‘aLJA .

" ,. '1' "34‘ ”’3”.-' ‘ 3:7
LK/sl. - \,;r,fir%$efiteflib‘jm-u-.T

has been accepted towards fulfillment
of the requirements for

_Eb_D__degree inm' s

ZZOLQ 2502.421:

‘7 Major professor

Date ll/VVI‘OO

 

LIBRARY

Michigan State
University

 

CYTOGENETIC EVALUATION OF SECALE GERMPLASM

FOR FORAGE IMPROVEMENT

BY

Mansour Nicknejad

AN ABSTRACT

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

DOCTOR OF PHILOSOPHY

Department of Farm Crops

1960

Approved

 

ABSTRACT

CYTOGENETIC EVALUATION OF SECALE GERMPLASM

FOR FORAGE IMPROVEMENT

by Mansour Nicknejad

The genus Secale has a great natural variability
which might be utilized in different breeding programs. At
present, the only species of this genus which is being
utilized is g, cereale or cultivated rye.

Eight species of Secale, each represented by one or
many samples, were assembled and grown in the greenhouse or
in the field. This collection also contained 92 samples of
weedy ryes from Iran. Many inter- and intraspecific crosses
were made, and cytologically analyzed. Meiosis was quite
regular in weedy ryes from Iran, and seven bivalents were

also generally observed in the F of weedy ryes from

18

separate areas. Cytological analysis of F 's between weedy

l
ryes and cultivated ryes revealed a similar pattern of pairing.
weedy ryes and cultivated ryes behaved similarly when crossed
with Secale montanum. In both cases a maximum association of
3 closed bivalents and a chain of 8 chromosomes were observed

at metaphase I. It is concluded that the two species differ

in three reciprocal translocations.

M. Nicknejad

Cytological analysis of F populations of interspecific

2

crosses between g, cereale x g, montanum revealed only two
kinds of plants, those with 7 bivalents at metaphase I and
those which were heterozygous for 3 translocations. Morpho—
logically F2 populations segregated into 1:2:1 ratio, g,

cereale, F and g. montanum respectively. Recombinants for

1!
major species characters were not found.
The chelating agent EDTA was used in an attempt to

break down species barrier at the diploid level. Inter-

specific F plants were treated during meiotic processes, but

1

no new recombinants were found in F2 populations grown from

seeds harvested from treated Fl plants. Frequency of interstitial

chiasma increased as a result of treatment with EDTA.
Interspecific barriers at the diploid level presented

a formidable handicap to the utilization of Secale germplasm in

desirable recombinations. To circumvent these barriers,

attempts were made to double the chromosome number in inter-

specific F hybrids, with colchicine or nitrous oxide. Nitrous

l
oxide treatments did not yield any tetraploids. With colchi—

cine treatments tetraploids were induced. Doubled Fl plants

were 50%,fertile in the greenhouse when crossed. One doubled

F2 population was cytologically and morphologically analyzed.

At metaphase I they showed an average of 10.72 bivalents per

M. Nicknejad

cell. Multivalents higher than chains of four were also
observed. Morphologically, the tetraploid F2 population was

quite uniform.

CYTOGENETIC EVALUATION OF SECALE GERMPLASM

FOR FORAGE IMPROVEMENT

BY

Mansour Nicknejad

A THESIS

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

DOCTOR OF PHILOSOPHY

Department of Farm Crops

1960

@1518!

ii

ACKNOWLE DGEME NT

The author wishes to express his sincere gratitude to
Doctor Fred C. Elliot for his guidance throughout the course
of this study and for his helpful advice in the preparation of
the manuscript.

The author is also deeply grateful to Dr. Everett H.
Everson for furnishing space in his experimental plots for the
author's thesis material, and help in planting them, and to
Dr. Carter M. Harrison for his valuable advice and assistance
during the preparation of the manuscript.

Finally the author is grateful to the Government of

Iran for financial support during the course of this study.

iii

TABLE OF CONTENTS

Page
LIST OF TABLES . . . . . . . . . . . . . . . . . . . . iv
LIST OF FIGURES . . . . . . . . . . . . . . . . . . . V
INTRODUCTION . . . . . . . . . . . . . . . . . . . . . 1
PART I--A STUDY OF WEEDY RYE SAMPLES FROM IRAN . . . . 3
1. Review of Literature 3
2. Materials and Methods 4
3. Results 6
4. Discussion 7
PART II--INTERSPECIFIC HYBRIDIZATION IN SECALE AND
SPECIES RELATIONSHIPS . . . . . . . . . . . . . . . 11
1. Review of Literature 11
2. Materials and Methods 12
3. Results 13
4. Discussion 21
PART III--USE OF CHELATING AGENTS TO FACILITATE
RECWBINATION O O O O O O O O O O O O O O O O O O O 37
1. Review of Literature 37
2. Materials and Methods 40
3. Results 42
4. Discussion , 47
PART IV--ALLOPOLYPLOIDY INDUCTION . . . . . . . . . . 55
1. Review of Literature 55
2. Materials and Methods 58
3. Results 59
4. Discussion 67

LITERATURE CITED ' 77

Table

1.

LIST OF TABLES

Types of pairingi11F1_interspecific hybrids.

Multivalent orientation at Metaphase I in
F1 hYbr ids O O O O O I O O O O O O O O 0

Analysis of anaphase I in F1 hybrids . . .

Per cent of tetrads with different numbers of
micronuclei in F1 hybrids . . . . . . . . .

A comparison between amounts of Ca, Mg, and P
in control and treated plants as measured
by spectrographic method . . . . . . . . . .

A comparison of anaphase I distribution in
control and treated plants of F1 inter-
specific hybrids . . . . . . . . . . . . .

Metaphase I analysis of doubled F hybrids .

2
Chromosome distribution at anaphase I of
doubled F2 hybrid rye . . . . . . . . . . .

Comparison of means and coefficients of
variability for several characteristics
in diploid and tetraploid Fz's of
interspecific crosses . . . . . . . . . . .

iv

Page

16

17

17

43

46

61

62

64

10.

ll.

12.

13.

14.

LIST OF FIGURES

Map showing locations of weedy rye samples
from Iran . . . . . . . . . . . . . . . . . . .

Map showing distribution of some Secale species
(adapted after vavilov, 1925) . . . . . . . . .

Metaphase I in F1 hybrids of S. cereale x.S.
montanum. 5 II (3 closed) + l I + 1 III. 951x .

Metaphase I in F1 hybrids of S, cereale x S,
montanum a chain of 8 plus 3 closed bivalents.
897x . . . . . . . . . . . . . . . . . . . . . .

Metaphase I of F1 hybrids of S, cereale X.§-
montanum. A chain of 6 showing alternate
orientation plus 4 bivalents. 938x . . . . . .

Anaphase I in F1 hybrids of S, cereale x S,
montanum, showing unequal distribution. 831x . .

Anaphase I in F1 hybrids of S, cereale x S,
montanum, lagging univalents start to divide.
970x . . . . . . . . . . . . . . . . . . . . . .

Anaphase I in F1 hybrids of S, cereale x S,
montanum, lagging univalents have divided. 861x.

Anaphase II in F1 hybrids of S, cereale x S,
montanum, unequal distribution at anaphase I
and unsynchronized division at anaphase II.
473x . . . . . . . . . . . . . . . . . . . . . .

Micronucleus in tetrads of F1 interspecific

hybrids of S. cereale x_S. montanum. 82X . . .
A tetrad with two micronuclei. 89X . . . . . .

Microspore with two micronuclei. 78X . . . .
Microspore with two nuclei. 89X . . . . . . . .

Seventeen univalents in metaphase I of meiosis
in asynaptic rye. 104OX . . . . . . . . . . . .

Page
10
28

29

29

3O

3O

31

31

32

32
33
33

34

34

Figure

15.

l6.

17.

18.

19.

20.

21.

22.

23.

24.
25.

26.

27.

28.

29.

Nine "pseudobivalents" at metaphase I in
asynaptic rye. 815x . . . . . . . . . . . . . .

Five bivalents and four univalents at meta-
phase I in asynaptic rye. 856x . . . .

Spontaneous chromosome breakage in asynaptic
rye with other disturbances. 846x . . . . . . .

"Pseudobivalents," univalents, and chromosome
fragments in asynaptic rye. 854x . . . . . . .

Diagrams of 4 chromosome associations in a
translocation heterozygote . . . . . . . . . . .

Diagrams of 4 chromosome associations in a
translocation heterozygote . . . . . . . . . . .

Diakinesis in F1 hybrids of S, cereale x S,
montanum. 849x . . . . . . . . . . . . . . . .

Metaphase I in F1 hybrids of S, cereale x.S.
montanum. Four closed bivalents and an inter-
stitial chiasma in chain of six. 1263X . . . .
Metaphase I in F1 hybrids. Four bivalents plus
a chain of six with an interstitial chiasma. A
later stage than Fig. 22. 1202X . . . . . . . .
Interstitial chiasma in a chain of four . . . .

Interstitial chiasma in a chain of six. 754x .

Two interstitial chiasmatajnchains of III and
IV. 844x 0 O I O O O O O O O O O O 0 O O O 0

Four closed bivalents plus a "frying pan"
configuration made up of six chromosomes. 1254X.

Qpen "frying pan" configuration. 824x . . . . .

Diagram comparing.means and variances for three
quantitative characters . . . . . . . . . . . .

vi

Page

35

35

36

36

49

50

51

51

52
52

53

53

_54

54

66

vii
Figure Page

30. Twenty-eight mitotic chromosomes of allotetra-
ploid rye at metaphase. 936x . . . . . . . . . 71

31. Mitotic anaphase in allotetraploid rye. 776x . . 71
32. Prophase in pollen mother cells of allotetra-
ploid rye. One nucleolus present. 977x . . . . 72

33. Prophase in PMC's of allotetraploid rye,
"bouquet" stage. 735x . . . . . . . . . . . . . 72

34. Metaphase I in PMC's of allotetraploid rye.
An aneuploid plant with 11 II and a chain of 7
chromosomes. 817X . . . . . . . . . . . . . . . 73

35. Metaphase I in PMC's of allotetraploid rye.
Eleven II plus a chain of 6. 873x . . . . . . . 73

36. Metaphase I in PMC's of allotetraploid rye, 8
II, 3 I, plus a chain of V and a chain of IV.
849x . . . . . . . . . . . . . . . . . . . . . . 74

37. Anaphase I in PMC's of allotetraploid rye. 14-
14 distribution. 977X . . . . . . . . . . . . . 74

38. Anaphase I in PMC's of allotetraploid rye.
Aneuploid plant with 12-15 distribution. 933x . 75

39. Lagging univalents dividing at anaphase I.
917x . . . . . . . . . . . . . . . . . . . . . . 75

40. Stainable and non-stainable pollen grains from
allotetraploid rye. 85X . . . . . . . . . . . . 76

I NTRODUC TION

Cultivated rye, Secale cereale L, is thought to be
the least ancient of the cereals. It was unknown to the
ancient Egyptians and Greeks. Rye has a wider potential
geographic range than wheat, both as to latitude and altitude.

Although other species of Secale are of potential
importance, especially for forage breeding purposes, utiliza—
tion of this genus has been almost completely confined to
S, cereale S. or cultivated rye.

After the pioneering work of Vavilov (56) on the
origin of cultivated plants, three broad classes of the genus
were recognized; wild, weedy, and cultivated rye. He recognized
four species in this genus, whereas Roschevitz (47) recognized
14. There is, however, a great store of variability found in
Secale germ plasm for both intra and interspecific character-
istics. For different purposes it is desirable to utilize
the total natural variability present in this genus through

recombinations. For example, wild rye, S. montanum Guss.

 

has desirable features such as leafiness, disease resistance.
and a perennial habit which may be recombined with the seed
characters of good cereale types. Such combinations would be
important for forage breeding purposes. Similarly, the
variability found in weedy ryes may be utilized in different
rye breeding programs designed for grain or forage improvement.
The greatest concentration of Secale germ plasm occurs

in southwest Asia. Although different species are somewhat

2
restricted to special areas, they overlap in their range of
distribution. In spite of sympatric distribution in some
areas, introgression of species delineating characteristics
has not been observed. Therefore, there must be some efficient
mechanisms which protect species identity. A study of such
mechanisms seems necessary because of their evolutionary
importance and to eliminate or circumvent them in successful
crosses.

In this study attempts will be made to survey the
variability in this genus and to find out the mean or means

by Which the desirable combinations might be obtained.

PART I

A STUDY OF WEEDY RYE SAMPLES FROM IRAN

1. Review 2: Literature

 

Although cereal rye is not cultivated in Iran, weedy
representatives are widely distributed in fields of wheat
and barley. On hilly sites where the soils are poor these
weedy ryes are more competitive than wheat and lead to the
popular notion that wheat actually is converted into rye.

There is little or no evidence suggesting that rye
has been cultivated in this area in recent or ancient times.
There is no special name for rye in the Sanscrit or Semitic
languages; it is merely called the "grain in wheat" or the
"grain in barley." Thus, it is clear that in southwest Asia.
rye was known first as a weed among fields of barley and
wheat and not as a distinct crop.

Collections of weedy rye from Iran were first made
by Vavilov in 1916. He classified these samples into 23
varieties according to spike color, toughness of rachis.
investment of grain, and glume characteristics. In later
expeditions he increased the number of varieties to 27.. Due
to the great variability present in weedy rye from Iran,
Vavilov considered Iran as a center of origin for rye. He.
also suggested that some weedy ryes of barley and wheat
fields gave rise to cultivated rye by means of domestication

and selection.

4

Another collection of weedy rye was made by Kuckuck
during 1952-54 and this collection was studied by Kranz (27).
In his study special reference was made to morphological
characteristics of the population in relation to their
geographical distribution. When these samples were classified
according to Vavilov's method, individual varieties did not
have a special distribution, but occurred at random through-
out Iran. When frequency of a trait was studied, significant
differences were found between populations from different
areas and the differences increased with geographic distance.
Evolutionary factors, such as isolation, selection, and
hybridization were considered to be responsible for this
differentiation.

Mflntzing (36) studied samples of weedy ryes from Iran
and found that two-thirds of the populations contained
accessory chromosomes. The frequency of plants with
accessory chromosomes, however, was quite low.

Cytogenetic analyses of crosses between and within
Secale species on a population level, with reference to
geographical distribution, have not been reported in the

literature.

2. Materials and Methods

 

Ninety-two samples of weedy ryes from 50 different
geographical locations in Iran were assembled (Fig. 1).

These samples were grown in the greenhouse or in the field,

and their morphological characteristics were assessed.
Individual spikes were isolated by means of glassine bags a
few days before anthesis and crosses were made by Changing
bags on two spikes. Vernalization of winter types was
accomplished by germinating seeds in petri dishes and leaving
them in the refrigerator between 340-400F for about six weeks.

In studying the cytogenetic relationships of this
collection, cytological analyses were carried out on the
following materials:

1. Plants grown from original seed samples.

2. Fl plants of crosses between parents from different
geographical areas.

3. Fl plants of crosses between the weedy ryes and
cultivated rye.

4. F1 plants of crosses between the weedy ryes and
perennial rye.

For cytological observations, whole spikes still
completely in the boot were fixed in Newcomer's solution
made up of the following:

6 parts isopropyl alcohol

3 parts propionic acid

1 part acetone

1 part petroleum ether

1 part dioxane

The fixed materials were kept in the refrigerator and

used at various intervals up to seven months.

6

Most observations were made from temporary propionic—

carmine smears. Some permanent slides were made by floating
the cover slip in 95 per cent alcohol for a few hours and

remounting them with diaphane.

3. Results

Individual plants showed different degrees of self-
fertility when bagged under greenhouse conditions, ranging
from 0.0 to 27 per cent. This might be an adaptive feature
of weedy rye isolated after continuous cultivation in wheat
fields for many generations.

Most samples of weedy ryes were susceptible to rust
and mildew, and only two showed complete resistance to both
diseases under greenhouse conditions. In the field, however,
these two samples also showed susceptibility.

When samples were grown in the greenhouse, no spike—
lets were detected with three florets, whereas the same
samples showed a high number of such plants when grown in
the field.

Meiosis was quite regular in weedy ryes from various
areas and the diploid number 2n = 14 was observed in all
cases.

Seven bivalents were also generally observed in the
Fl's of weedy ryes from separate areas. Several cases of
asynapsis of chromosomes were detected. Sometimes three

pairs of chromosomes did not pair and resulted in six univalents

and four bivalents. Such a case was found in crosses of
two neighboring samples.

Cytological analyses of Fl's between weedy ryes and
cultivated ryes also revealed a similar pattern of pairing
and seven bivalents were formed. Two accessory chromosomes
were detected in crosses which had in common a cultivated
rye from Denmark. No accessory chromosomes were found in
the original samples or in crosses between samples from Iran.

Hybrids between different samples of weedy rye and
mountain rye (S. montanum) showed the same cytological
properties as hybrids of cultivated rye and mountain rye,

the results of which will be discussed later.

4. Discussion

In spite of the great variability exhibited by weedy
rye when compared to cultivated rye, the karyotype is
identical through the range of distribution in Iran. Not
only are weedy ryes similar to each other but they are also
similar to cultivated rye as far as their gross karyotypic
structure is concerned. Cultivated ryes from different
sources were also similar in this respect, and there is no
report in the available literature that varieties of cultivated
rye differ in their gross chromosomal structure.

All the materials studied were diploid which is
especially important.when we consider that in other genera

of the tribe Hordeae, such as Triticum, evolution has been

achieved through polyploidy, and structural modifications
such as translocations differentiate varieties.

Uniformity of karyotype in weedy rye throughout Iran
and the gross structural similarity to cultivated rye emphasizes
an adaptive value related to cultivation. Whenever rye comes
under cultivation there is only one karyotype which can
survive, and that exists in all latitudes and altitudes where
rye is grown intentionally or unintentionally.

Cultivation as an important factor in the evolution
of crop plants was emphasized by Vavilov (56) and its
constituent factors outlined as follows by Darlington (10):

(a) Tillage conditions which lead to the selection of
larger forms which are better competitors at higher fertility
levels.

(b) Sowing conditions which lead to the selection of forms
with even and rapid germination and also to selection of forms
lacking the special mechanism for self-propagation.

(c) Harvesting conditions which lead to the selection of
forms with tough rachises.

Under cultivation special forms of rye of one'karyotype
were developed. In southwest Asia rye met such conditions in
fields of wheat and barley as a weed. Receiving similar
management as wheat and barley, rye became more similar to
the wheat and barley plants, and the grain became more
difficult to separate. When wheat was carried to higher.

altitudes and wider latitudes, rye, being more winter hardy

and better adapted to sandy soils, predominated in certain
areas and gradually became an independent crop.

Despite the fact that rye has been an important crop
in areas other than southwest Asia, it still has the most
diversity in this area. In southwest Asia rye meets the
greatest climatic diversity and a multitude of ecological
variations. Here, too, natural selection has produced a
great diversity of forms. Such a diversity is unknown to

cultivated rye elsewhere.

CV8

3r:
0
0°
0
O
0

Fig. 1.

10

U :23 S P.
CASI’IAN SL4
o
o o
o
O 00 o
o o o 0
0 O
o 0 °
0 O
0
'V‘ IV‘O’.‘
O O I Ed": 513151
e
o 0 °
0
0 Z
<
O f“-
o 22
o
o if:
0 o in
.L.
(t)
M.
T 'Q
l R A as!
40
'9
6‘
/
‘7
f‘.
O
(x
(A
V

 

Map showing locations of weedy rye samples from Iran.

11

PART II

INTERSPECIFIC HYBRIDIZATION IN SECALE

AND SPECIES RELATIONSHIPS

1. Review_2£ Literature

 

Longley and Sando (30) made interspecific crosses
between S. cereale and S, montanum and found that the
behavior of chromosomes in such hybrids was generally normal
with bivalent formation but that univalents occasionally
occurred.

Kostoff (25) found 75 per cent abortive pollen in
such hybrids and Duka (13) reported pollen viability ranges
between 5.0-38.0 per cent. Kostoff and Duka also reported
only bivalent formations with some univalents in hybrids.

Schiemann and Nurnberg-Kruger (48) confirmed the low
fertility of interspecific crosses and ascribed it to multi-
valent formation. They found a maximum association of four
bivalents and a translocation chain of six chromosomes.

Jain (20) also observed a maximum association of a
chain of six chromosomes plus four bivalents in the same
interspecific hybrids.

Riley (46) observed cells with chains of 3, 4, 5.
and 6 in similar hybrids. From this he concluded that S,
cereale differs from S. montanum by two large translocations
involving three pairs of chromosomes. Thus, he confirmed

the observations of Schiemann SE 21- On the evidence of a

12
single cell with chains of 3 and 5 and another cell with a
single chain of 8, he proposed a third translocation difference
between the two species.

Price (41) studied the same hybrids and found a maximum
association of 6 chromosomes in the first metaphase, as a
result of two tranlocations. The translocation configurations
were usually open chains and no closed multivalents were found.
Inversion bridges were reported in hybrids as well as in their
parents.

Price also irradiated pollen from S. montanum with
which he pollinated S, cereale. In hybrids produced in this
way he was able to detect a new translocation. This trans-
location involved chromosomes of the original chain of six
and another pair of chromosomes.

Stutz (54) found a maximum association of four ring
bivalents plus a translocation configuration made up of six
chromosomes. He also reported some cells Which had a complete
ring of six chromosomes.

Ossent (38) found that Fl's of the same crosses were
perennial with brittle rachises and the F2's contained 75 per
cent perennials with brittle rachises and one per cent

perennials with tough rachises.

2. Materials and Methods

In order to study interspecific relationships in this

genus, three different sources of S. montanum, two sources of

13
“S. segetale, one source of S. ancestrale, one source of S,
silvestre and many sources of both weedy rye and cultivated
rye were used. Crosses were made both in the greenhouse and
field. In self-sterile species it was possible to Obtain
hybrids without emasculation of the female parent and
morphological recognition of hybrids made this procedure
relatively simple to apply. Spikes were fixed in Newcomer
solution and most observations were made from temporary

propionic-carmine smears.

3. Results

In crosses between S, montanum and S. cereale, F1
hybrids were obtained only when S. cereale was used as the
female parent. In reciprocal crosses the seeds did not
germinate or germinated very poorly and the seedlings died
at early stages of development. Fl plants of interspecific
crosses were quite vigorous perennials and similar to their
wild parent S. montanum.with rachises brittle from the top
to the center of the spike. They were later flowering than
cultivated rye and slightly earlier than S. montanum. These
hybrids had 57.71 per cent stainable pollen and their thousand
kernel weight was about 23.6 grams.

Results of cytogenetic analysis of one thousand cells
at diakiesis or metaphase I in S. cereale x.S. montanum

hybrids are shown in Table l.

TABLE 1. Types of pairing in F

l

14

interspecific hybrids.

 

 

 

Type of pairing % P.M.C.
7 II 4.5
6 II + 2 I 1.7
5 II + 1 I + III 12.3
5 II + 1 IV 19.8
4 II + 3 I + III 0.7
4 II + 2 I + IV 0.8
4 II + l I + V 13.6
4 II + 2 III 6.4
4 II + 1 VI 37.5
3 II + 2 I + VI 0.7
3 II + l I + VII 0.3
3 II + 1 III 1 V 0.6
3 II + 1 VIII 1.0

 

More than 93 per cent of the cells analyzed at
metaphase or diakinesis showed multivalent formation.

About 40 per cent of bivalents observed at metaphase

I were open and 60 per cent closed. In most cases four or

less closed bivalents were observed and only in four per

cent of the cases as high as five ring bivalents were

detected. Six or seven closed bivalents were not observed'

in any cell.

This points out that one of the translocations

is quite large and prevents formation of closed bivalents.

15
The configuration which was observed most frequently was
four bivalents, three of Which were closed and a chain of
six chromosomes (Fig. 5). In such a configuration the
chromosomes associated with the nucleolus appear as a
bivalent and not associated in the chain of six. Five
bivalents, three of which were closed and a chain of four.
was next in order of frequency. Other configurations which
are the result of modification of original associations were
found with frequencies listed in Table 1.

There was no ring formation in multivalents observed
as reported previously (Stutz). About two per cent of the
cells showed configurations such as chains of VII, III + V
and a chain of VIII, all of which indicate existence of a
third translocation difference between the two species. The
other pair of chromosomes which joins the chain of six in
order to make a maximum association of three ring bivalents
and a chain of eight is the pair of chromosomes which is
associated with the nucleolus (Fig. 4). This explains Why
with such a high frequency of chains of six chromosomes there
was no ring of six.

Since only gametes which contain alternate chromosomes
of a translocation multivalent may survive, it is desirable
to know how these multivalents are oriented at metaphase I.
Results of such study in 273 multivalents are listed in I

Table 2.

16

TABLE 2. Multivalent orientation at Metaphase I in F

 

 

 

hybrids. 1
Chain IV V VI
Alternate
Segregation 70% 82% 49%

Non-Alternate
Segregation 30% 18% 51%

 

The differences between orientation of multivalents
may be due to the morphology of the chromosomes involved in
multivalent chains. In a chain of VI, orientation seems to
be quite random. In chains of IV it seems that more alternate
disjunction takes place and for chains of V it is definitely
nonrandom (x2 = 40.9 > 7.88). Anaphase was studied in about
600 cells, the results being listed in Table 3.

As the table shows, more than 78 per cent of gametes
received equal numbers of chromosomes.

Univalents which were present in more than 30 per cent
of the cells (Table 1) had a different behavior at anaphase.
In some cases the univalent or univalents passed undivided
to one pole in anaphase I (Fig. 6) or they did not go to any
pole and lagged behind to form a micronucleus. In other
cases univalent(s) divided mitotically (Fig. 7). Then each
sister-chromatid moved to opposite poles (Fig. 8). At second
anaphase, the behavior of univalents depended upon their

behavior at first anaphase. If they did not.divide at

17

 

 

o.m m.hH m.oa on mbmuumu X
.z.z .z.z .z.z .z.2 HoHUSGOHUHE
m N H oz mo Honesz

 

.mpHnnms Hm CH HoaodcouoHE mo muonfids psoHoMMHU SuHB mpmuuou no name Hem .¢ mqmfie

 

 

 

 

NH m.o m.m Rm v.2 suds
m
m u m H H H u .m. u H u H .. H H I H H u H
Ha. fie e/vam\ e s {m o s H e m e a. a
H

.mpHHQ>£ m CH H ommnmmcm mo mflmhamcfi .m mafiée

18
anaphase I then they divided at anaphase II. Otherwise, they
wereunable to divide again Whereupon they usually lagged on the
plate and were often lost in the cytoplasm to form micronuclei
(Figs. 10, ll, 12).

As the number of micronuclei could be used as an
indication of irregularity of meiosis, more than 900 tetrads
were analyzed for this purpose and the results are listed in

Table 4.

Analysis of F2 Population

 

F2 populations were obtained by growing seeds harvested
from F1 plants Whose cytological properties had been analyzed.

A population of 38 F plants was chosen for morphological

2
and cytological analysis.

Cytological analysis of the population revealed only
two kinds of plants. Of 38 plants analyzed, 17 were
structurally homozygous and formed only bivalents with
occasional univalents, and 21 plants were structurally
heterozygous. In heterozygous plants, cytological analysis
revealed exactly the same associations as reported for F1°s
(Table 1). There was no single heterozygous plant which
showed only a tetravalent association with no chains of six
and occasional chains of eight. This shows that these four
chromosome pairs segregate in parental combinations. I i

The homozygotes were of two kinds: ten of Which were

early flowering annuals with tough rachises, the other seven

19
were free tillering, late flowering plants with brittle
rachises.

Morphologically, the heterozygous plants were
essentially similar to Fl's. They were perennial, with
brittle rachises, and displayed low fertility. Mean fertility
for this group was I = 35.7 per cent whereas fertility for
homozygotes was E = 53.2 per cent.

An asynaptic plant was found in the F2 population
with a high number of univalents in metaphase I. The average
number of univalents was 4.81, for pseudobivalents 1.85 and
for true bivalents 2.20 per cell. A few multivalents were
also observed. There was a great disturbance in the karyotype
of this plant as reflected in the chromosome number (Figs. 14,
15), fragmentation (Figs. 17, 18), and distribution at anaphase.

In several cases of asynaptic rye reported in the
literature (40) none has been accompanied by chromosome
fragmentation. In Scilla sibirica, however, Rees (45) found
a similar situation in which asynapsis was in conjunction
with spontaneous chromosome breakage. But meiosis was normal
in cells which did not have fragments and frequency of cells
having fragments was about three per cent.

In the rye plant studied, meiotic disturbance was not
always accompanied by fragmentation. When anthers from
different florets of the same spike were examined, differences
were found. In some florets only asynaptic chromosomes were

found, either as univalents or pseudobivalents. In a few

20
florets true bivalents as well as asynaptic univalents were
found (Fig. 16).

A high percentage of tetrads showed micronuclei, and
fertility was as low as 5.9 per cent with caryopses being

shrunken.

Cytogenetic Analysis 2; Other Species Intercrosses

Other species of rye used in this study were S. segetale,
.§° ancestrale, and S, silvestre.

When S. seqetale and S, ancestrale were crossed with
.S. cereale they formed seven bivalents and in crosses with
‘S. montanum they behaved exactly as S. cereale.

In the case of S, silvestre the situation was
different. This is a self-fertile species which is quite
short with narrow leaves and extremely brittle rachises.
Attempts to cross this species with S, cereale were unsuccess-
ful in both directions. There is no report in the literature
of such a cross being made. When S. silvestre was used as
female and S. cereale,variety Gator as male, many embryos
were initiated but failed to reach maturity. Many of them
aborted prior to the dough stage. When hybrid embryos were
dissected and cultured on artificial media some of them
developed and produced vigorous plants. In reciprocal crosses
with S. cereale as the female, the embryos failed to show signs

of development. Incompatibility between the embryo and

surrounding tissue appeared to be the cause of failure in

21
crossing the two species. According to Jain (20) this species
crosses easily with S, africanum and gives a 4 II and VI
association in the meiotic metaphase of F1 plants. This

indicates that the karyotype of S. silvestre is similar to

 

that of S, cereale.

4. Discussion

 

In F1 hybrids between S. montanum and S. africanum
seven bivalents with occasional univalents were Observed.
This shows that the two species are similar in gross chromosomal
structure.

Cytological analysis of the hybrids between S. cereale
and S. africanum revealed the same pattern of pairing as
S. cereale x S. montanum hybrids. In both cases there were
three translocation differences between the two species. Two
of these translocations were large and caused multivalent
formation in many cases, whereas the third one was small and
only in about two per cent of cells formed chains of eight.

These translocations do not occur independently; that
is, they have some pairs of chromosomes in common. If they
did not involve any common pairs of chromosomes they would
have produced only three independent tetravalents. This.
of course, is not the case and translocations have brought
together four pairs of chromosomes in order to form a chain
of eight. This must have a selective advantage over three

independent translocations. Analysis of F2 populations

22
showed that only two kinds of zygotes can form,those which
are homozygous and are of S, cereale or S, montanum type and
those which are heterozygous and exactly similar to F1 plants.

The ratio was close to 1:2:1 S, cereale, F and S. montanum,

1
respectively. This suggests that in gametes produced by F1
plants, only parental combinations, as far as these four

pairs of chromosomes are concerned, can survive. Other
gametes being deficient or duplicated for segments of
chromosomes are discarded. Thus in rye, the four pairs of
chromosomes involved in translocations are transmitted as a
single Mendelian factor and segregate to maintain species
identity. In a translocation configuration, the homologous
parts of chromosomes which are not between a kinetochore and
translocated segment can recombine easily. In interstitial
segments crossing-over causes duplications and deficiencies

in most cases, especially.when multivalents consist of more
than four chromosomes, and interstitial chiasma are not
frequent. Therefore, we can assume that the genes determining
specific characteristics are located in interstitial segments.
and are transmitted as a linkage group and thus gene exchange,
as far as these characters are concerned, is suppressed.
Structural heterozygosity is not the only mechanism.which
keeps the two species separate. In the course of evolution
other barriers have also evolved. As was previously mentioned,
the cross between the two species is successful only when.

cereal rye is used as the female. In reciprocal crosses seeds

23
do not develop or seedlings die in early stages of growth.

Therefore, in nature. hybrids of S, cereale and S, montanum

 

can only occur in fields of wheat and barley where cereal

rye grows as a weed. The hybrid plants are perennial with
brittle rachises. Thus, in major species characteristics.

on which ecological adaptation is based, they resemble mountain
rye, whereas they occur in habitats suitable for cereal rye.
Under these conditions they are vigorously selected against.
The F1 hybrids are selected against, not only because they
cannot survive under cultivation, but also because they bloom
later than the mass of cereal rye. Being self-sterile and
only 28 per cent fertile in cross pollination, back crosses
are not very frequent in nature.

How different species of Secale are related is not
quite clear. .S. africanum seems to be derived from S. montanum.
Both are perennial species with similar morphological and
cytological characteristics and they intercross easily. The
only main difference between the two species is that S.
africanum is self-fertile and occupies only one mountain
range in South Africa, and is cut off from the main Secale
germ plasm which is found in southwest Asia and in the
Mediterranean region (Fig. 2).

It is not quite clear how this species occurs a long
distance from the main germ plasm of the genus. Anderson (1)
suggested that in the geological past the whole continent of

Africa was probably exposed to Secale and later unfavorable

24
conditions eliminated this genus in most parts. Only in South
Africa has this species persisted. Such a premise intimates
that at least in some other places in Africa with conditions
similar to the present habitat of S. africanum and probably
identical past geological changes, one should find this or a
similar species. There is no such report in the available
literature. Therefore, we may assume that long distance
dispersal of S. montanum by different means has given rise to

S, africanum.

 

.S. cereale is certainly younger than S, montanum.
Cereal rye exists only under cultivation as a result of man's
activity, a relatively recent event. Cultivation of wheat
and barley was practiced by man in the early days in southwest
Asia, where S. montanum is endemic. This practice created
a special habitat which did not exist previously. Many new
variants from S. montanum reached these areas as before, but
now some of them found suitable conditions. Probably at first
these new forms were only the result of gene mutations. But
still S, montanum occupied the surrounding areas, and the new
forms which had established themselves in cultivated lands
had to be protected from being lost as a result of inter-
crosses with mountain rye. Thus selection favored establishment
of translocations as a protective device and at the same time
as a linkage measure for species characteristics. A
With the rather wide distribution that weedy rye has

today, it is difficult to see how these genic and structural

25
changes have been fixed in the whole area. Actually cereale
rye started as a weed in fields of wheat and barley. The
areas in which these crops were grown at first were not large
and consequently the weedy rye populations were small. Natural
selection and a small population size could have easily fixed
genic and structural changes in the early weedy rye populations.

Wheat, being a valuable crop, was carried by man in
different directions to be grown in fertile valleys. Rye
grains mixed with the wheat were likewise transferred to
different regions. Thus, we may postulate that an originally
small weedy rye population, in which some characteristics had
been fixed, spread throughout the range of distribution in
which we find weedy rye today.

If the above assumption is true, then we might expect
to find a great uniformity between weedy ryes from different
regions. As far as gross chromosomal structure is concerned
we have shown that weedy ryes are similar. There are signifi-
cant differences in morphological characteristics between
populations of weedy rye from different regions. Such a
great variability of weedy rye might have been obtained in
two ways; first by direct action of natural selection on weedy
rye populations in each region; second by introgression of
nonspecific characteristics from S. montanum, an endemic in
those areas into which weedy rye was introduced. In this Way
weedy rye could have obtained a great deal of variability

from mountain rye Which was also adapted to its new habitat.

26

For S. silvestre the situation is different. This
species occupies the uncultivated sandy soils of the Sarmatian
basin north of the Black Sea and Caucasia (Fig. 2). This is
a completely self-fertile species with flowers Which have a
great tendency to cleistogamy. It is an annual with the.
karyotype of S. cereale.

Stutz (54) suggested that S. cereale might have been
derived from the products of interspecific hybridization
between S. silvestre and S, montanum. Such an assumption
seems to be at variance with the known facts about the three

species involved. First of all, self-fertility and cleistogamy

 

in S. silvestre prevent hybridization between this species

and S. montanum, and if such hybrids are occasionally produced.
translocation differences between the two species provide a
mechanism Which prevents recombination of certain characteristics.
Secondly, assuming that successful hybridizations occurred.

'S. silvestre has little to contribute to the Characteristics
found in S. cereale. The only major characteristics, which

.S. silvestre might have contributed to such a hybrid, is an

 

annual habit and it seems more reasonable to assume that such
mutants have been selected directly from S. montanum under
cultivation.

As an alternative assumption we may propose that S,
cereale in its upward movement, whether in wheat or as a
distinct crop, was brought in contact with sandy soils which.

were not cultivated. Under these conditions the tough radhis

27
of cereale types was a great disadvantage and, therefore.
mutants with a brittle rachis had a high selective value.
Drastic and unfavorable conditions in the sandy desert area
promoted self-fertility and a complete reproductive barrier
developed protecting this species from crossing with cereal
rye from surrounding areas. Morphological changes such as
short, thin, and spreading stems with narrow leaves provided

the species with better adaptive features in this habitat.

 

.,2‘.-‘~‘,“"-

\ I! :Vg} WWII J

 

LEGEND
vv secale eereoie

0 on .. montanum
0 u afrIccmurn
a?» « silvestre

 

 

 

 

Fig. 2. Map showing distribution of some Secale species
(adapted after vavilov, 1925).

 

‘1

Fig. 3. Metaphase I in F1 hybrids of S. cereale x S.
montanum. 5 II (3 closed) +1 I + 1 III. 951x.

l,
l
i
I

 

L_ ___,_.__
Fig. 4. Metaphase I in F hybrids of S. cereale x S. montanum
a chain of 8 plui 3 closed bivalents. 897x.

. e . I ‘

, it

 

 

.Fig. 5. Metaphase I of F hybrids in S. cereale x S. montanum.
A chain of 6 shoWing alternate orientation plus 4
bivalents. 938x

‘ I

 

 

Fig. 6. Anaphase I in F hybrids of S. cereale x S. montanum,
showing unequal distribution. 83lX.

91v

2; .
. "’ I.
. is.

l >-
I

Fig. 7. Anaphase I in F hybrids of S. cereale x S. montanum.
lagging univalents start to divide. 970x.

I

4':

3“.

Fig. 8. Anaphase I iniF hybrids of S. cerealeix S. montanum,
lagging univalehts have divided. 861X.

‘ I"? 32

 

\

 

Ll,
Fig. 9. Anaphase II in F1 hybrids of S. cereale x S. montanum.
unequal distribution at anaphase I and unsynchronized
division at anaphase II. 473x.

 

IV

 

 

Fig. 10. Micronucleus in tetrads of F1 interspecific hybrids
.of.S. cereale x S, montanum. 82X.

 

I

I, _l_r- 1__"____.____._*,_'

Fig. 11. A tetrad with two micronuclei. 89X.

 

:w" I w 7 ‘w‘A ' i” ‘— #'i 7 W— 7

Fig. 12. Microspore with two micronuclei. 78X.

33

34

 

L i
,—,"—\ ‘v E— A.— " I
Fig. 13. Microspore with two nuclei. 89X.
. -A_ 1k . -_ c.
l I
I o 0 '
D o . . ‘
e O . .

’7’ v—‘v fi—‘,

Fig. 14. Seventeen univalents in metaphase I of meiosis in
asynaptic rye. 1040X.

35

 

 

0
Fig. 15. Nine “pseudobivalents" at metaphase I in asynaptic
rye. 815x. v

I‘-

 

I
l
i
l
i
i
I
I

Ll l

w—-—.- ,_ I
..

 

Fig. 16. Five bivalents and four univalents at metaphase I
in asynaptic rye. 856x.

 

 

36

LA“ ,_,,_. ,1 1 1

Fig. 17. Spontaneous chromosome breakage in asynaptic rye
with other disturbances. 846x.

‘6
'0
e‘al"

'\
l"
l x r 'r

 

 

Fig. 18. "Pseudobivalents," univalents. and chromosome
fragments in asyhaptic rye. 854x.

37

PART III
USE OF CHELATING AGENTS TO FACILITATE RECOMBINATION

1. Review 2; Literature

It has been suggested that calcium and magnesium ions
play an important role in maintaining chromosome structure.

Mazia (31) reported dispersion of salivary gland
chromosomes of S. melanogaster by treatment with chelating
agents, such as EDTA, at low ionic concentrations. On the
basis of such experiments.a particulate structure for
chromosomes was suggested, the particles being held together
lengthwise by divalent cation bridges. He Showed that two
conditions should be met if the chromosomes are going to be
dispersed: first, treatment with an agent capable of binding
Ca++ or Mg++ ions and, second, provision of a medium of
sufficiently low ionic strength. He predicted that chromosome
breakage and crossing-over would be sensitive to the ionic
environment of the chromosome.

Kaufmann and McDonald (23), on the other hand,
attributed the effects of EDTA to alteration in general
metabolism of the cell rather than to direct breakage of
chromosomes through chelation of divalent cations serving
as bridges between macromolecular complexes of protein and.
nucleic acid. By comparison of the action of EDTA.with that
of ribonuclease in modifying crossing—over in Drosophila

they indicated that the latter agent may be more effective

38
than the former under specified conditions of treatment.

Steffensen (53) reported induction of chromosomal
alterations during meiosis in Tradescantia by raising plants
under magnesium deficient conditions. Using Solution I of
Hoagland and Arnon; he adjusted the concentration of calcium
to three different levels. The first solution contained 2.5
ppm of calcium. After three months clones grown in this
solution were transferred tova solution lacking Ca++ altogether.
The second solution contained 10 ppm of Ca++ for the first
three months and 2.5 ppm thereafter. The third solution
contained 250 ppm of calcium, an optimal supply.

Buds were collected during the sixth and seventh
months of incubation.

Steffensen found that plants grown in the first
solution contained only abortive cells, none of which.were
undergoing microspore division. Chromosomal alterations
could not be analyzed in these anthers. Plants grown in the
second solution, which was suboptimal, exhibited at least
17 times more chromosomal aberration and micronuclei than did
plants grown with optimum calcium. From these results, he
concluded that bivalent cations may bind together the
macromolecular constituents of the chromosome and thereby
provide structural stability.

Kirchner (24) removed calcium and magnesium from
bacterial cells either by treatment with Versene or by culture

in a minimal medium lacking these ions. He obtained an

39

inorease in the frequency of recombination.

Hyde (19) treated non—excised root tips of onion and
Vicia §§Q§_with .01 M solutions of EDTA. The first noticeable
effect within three hours of the treatment was suppression
of anaphase movement of chromosomes. Chromosomes swelled
in metaphase and late prophase. This swelling was not
accompanied by any marked loss of stainability. Therefore.
Hyde concluded that EDTA treatment breaks chromatin into
smaller units. When he treated root tips with a 1:1 molar

ratio of CaCl and EDTA the cells continued to divide

2
normally.
Wolff and Luippold (58) soaked seeds of y. £222 for
varying times in a .001 M solution of EDTA. They found
chromosome breakage as a result of the treatment and the
amount of chromosome breakage produced by combined calcium
deficiency and radiation was more than the additive amount
of both when applied alone.
Davidson (11) treated root tips of onion and flowering
shoots of Tradescantia with solutions of chelating agents
(EDTA and quinaldic acid) and CaCl

MgCl and NaCl. He

2’ 2'
found various abnormalities at meiosis and mitosis. Chromo-
some breakage occurred but only in shoots and after a long
delay. The Chelates did not appear to be as specific SgLyiyg

as they were lg giggg. Therefore, he concluded that the effects

found after treatment could not be attributed to the inactiva—

tion of particular cations.

40
Eversole and Tatum (15) found variation in the amount
of crossing-over after treating the cells of biochemical
mutant strains of Chlamydomonas reinhardi with a chelating

agent or with MnCl The frequency of crossing—over was

2.
increased by these treatments in two of three chromosome
intervals tested. The effect of both agents were reversed
by subsequent incubation of treated cells in high concentra-
tions of calcium and magnesium ions. It is suggested that

the action of MnCl may involve the replacement of Mg by

2
Mn. They concluded that the sensitivity of crossingvover

in S. reinhardi to ionic environment is consistent with the
theory that chromosomes are composed of macromolecular unit
particles linked together by divalent ions.

Caldwell and Burdich (5) reared Drosophila melanogaster
on media containing various amounts of EDTA. They studied
uptake, viability, and crossing-over resulting from these
treatments. They found that crossing-over was increased
at all levels of EDTA, the maximum effect being produced at

50 per cent of maximum uptake level where viability was still

equal to the control.

2. Materials and Methods

Four solutions of EDTA with concentrations of 0.1 M,
0.01 M, 0.001 M and 0.0001 M were prepared and the pH
adjusted to about 7.0.

Fl plants of crosses between S, cereale and S, montanum

41
were used which have a maximum association of a chain of
eight and three closed bivalents at metaphase I.

Spikes still in the boot stage were chosen for treat—
ment in which meiosis would continue to occur for a few days.

For each treatment, including the control, four plants
were allocated. The plan was to apply two different rates
of EDTA for each concentration. This was not quite possible.
because the culms did not have the same capacity even in the
same plant and equal amounts of EDTA did not produce similar
observable effects.

Concentrations of 0.1 M were too strong and in most
cases killed the spike in the boot, and 0.0001 M did not
produce any observable effects. Therefore, they were both
discontinued.

Hypodermic needles were used to inject the solution
into the plants. An attempt was made first to inject solutions
into the base of spikes, but this was found impossible to
achieve. Unless the amount was very small the solution was
pressed out by the boot. Internodes of the culm were found
to be a suitable place for injection, since they are hollow
and the nodes are solid.

In order to inject the solution in the culm, the
needle was first inserted at the top of the internode to allow
air to escape while solution was injected into the lower part
of that internode. Glue was applied to both holes and it
prevented leakage of the injected solution. The capacity of

internodes varied from less than 0.1 ml to more than 1.0 ml,

42
depending on the vigor of the plant. Therefore, the capacity
of the internode was taken as the approximate vigor of the
spike and in each case it was filled with solution. Cytological
observations were carried on, on the spikes, which showed
visible damage when they were fixed for observation.

All the plants used for treatment were grown in the
greenhouse. Spikes of treated culms were bagged together
for interpollination. Seeds from these Fl plants were grown

in the field in early spring. Many of these F plants stayed

2
vegetative and did not produce spikes, those which produced

spikes were cytologically analyzed.

3. Results

Treated culms Showed symptoms after 24-48 hours.
First an injured region appeared along the midrib of the
leaves above the treated internode: then the edges of the
leaves began to curve inward. Gradually injured spots appeared
over the whole surface of the leaf and chlorosis developed.
In most cases the first three leaves from the top died.
Chemical analyses were carried out by spectographic
methods in order to determine whether changes in chemical
composition had occurred in treated plants.' Three parts were
taken and analyzed separately; the first including the spike
plus two top leaves, the second the third leaf, and the third
the first internode. Results of the analysis are listed in
Table 5. Samples were taken when visible damage from treatment

was obvious.

43

TABLE 5. A comparison between amounts of Ca. Mg. and P in
control and treated plants as measured by
spectographic method.

 

 

 

% Ca % Mg % P

Spike plus lst and 2nd leaf:

treated 0.70 0.35 0.73

control 0.48 0.26 0.85
3rd leaf:

treated 0.75 0.42 0.73

control 0.57 0.36 0.91
lst internode:

treated 0.26 0.22 0.80

control 0.26 0.22 0.95

 

As shown in the table, the amount of calcium and
magnesium was greater in treated parts than in the control.
except for the first internode in which the amount remained
unchanged. In fact, the difference between control and
treated plants decreases as less active parts of the plant
are compared. The largest differences were found between
young leaves. No differences were observed between treated
and control internodes.

Cytological analysis. EDTA treatments produced
changes in the morphology of the chromosomes. These changes
ranged from complete agglutination to only very slight
changes. In many cases, metaphase chromosomes had swollen,
and fragmentation was not observed in any case. Analysis
of metaphase I in pollen mother cells revealed the same

configurations as in controls and the treatment did not seem

44
to have produced any gross structural changes such as breakage.
new translocations, and inversions. During the analysis.
however, it was found that the number of interstitial chiasmata
had increased. These are the chiasmata which form in a trans-
location configuration in that part of the chromosomes between
the point of interchange and the centromere. According to
Darlington (8), these interstitial chiasmata cannot terminalize;
therefore, they are arrested between the point of interchange
and the centromere. Several workers have given views in
support or rejection of this hypothesis, but none seem to be
conclusive.' Among thoSe who have rejected Darlington's
hypothesis is Price (42) who worked with an interspecific
hybrid of S, cereale L. x'S. montanum, Guss., the same
material used in this experiment. He found a special multi-
valent in the first metaphase which he calls "frying pan."
This association was found in an irradiated hybrid plant
obtained by pollinating S. cereale with x—rayed pollen of
S, montanum. This ”frying pan" configuration is exactly the
same as is shown in Fig. 27. He suggests that this configura-
tion is the result of terminalization of an interstitial
chiasma. Thus, he refutes Darlington's hypothesis that
terminalization of chiasmata is arrested at points where
homologies change. In a diagram as reproduced on page 49,
he shows how this terminalization takes place.- As far as.
terminalization of interstitial chiasma is concerned,

configurations observed in this study confirm Price's views.

45
But as to how this terminalization takes place, a somewhat
different point of view may be suggested. Fig. 28 and its
diagrammatic interpretation on page!i)suggest that there is
no three-chromosome terminal chiasma as diagrammed by Price.
Fig. 28 shows that a multivalent association opens When an
interstitial chiasma is terminalized. This breaking of
multivalents is the result of terminalization of regular
chiasmata formed between two homologous parts of non-homologous
chromosomes in a translocation configuration. Thus chromatids
involved in interstitial chiasma have to pass through two
chiasmata in order to terminalize and this is accompanied by
opening of the multivalent at least one point. Therefore, it
seems reasonable to assume that interstitial chiasmata
terminalize later than normal chiasmata. Such a delay makes
them more suitable for counting than regular chiasmata.

When the frequency of interstitial chiasmata in controls
and treated materials were compared, marked differences were
found. In the controls. they occurred in 4.8 per cent of
multivalents with a coefficient of variability of 32 per cent.
In treated materials corresponding values were 23.4 and 67.0
per cent, respectively. These chiasmata occurred in chains of
III, IV, V, and VI.

Anaphase I analyses revealed more unequal chromosome
distribution in the treated material than in the controls.

The results, however, were not very consistent and considerable
variation was found from plant to plant. Table 6 shows the

result of anaphase analyses.

46

TABLE 6. A comparison of anaphase I distribution in control
and treated plants of F1 interspecific hybrids.

 

 

 

7 I - 7 I 6 I - 82[ Others
Control 72.8% ' 13.4% 13.8%
Treated , 58.9 30.4 10.7

 

It seems reasonable to assume that higher chiasma
frequency in treated materials maintains chromosome
associations for a longer time and this may enhance chances
of unequal distribution especially in multivalents.

The amount of seed set in treated Fl plants was 15
per cent which is about half of the controls with 27.7 per
cent. This can be interpreted as the result of increased.
meiotic disturbances, in structurally heterozygous plants.
due to EDTA treatments. But when structurally homozygous
plants of cultivated rye, variety Gator, were treated with
0.010 M and 0.001 M EDTA the amounts of seed set was 17.8
and 38.7 per cent, respectively. Controls of the same
variety had a seed set of 51.8 per cent under similar
conditions. Cytological analyses of treated Gatorplants did
not reveal any visible changes. Therefore, reduction of seed
set in treated F1 plants is due not only to increased meiotic
irregularities, but may be attributed to a general metabolic
disturbance.

Cytological analysis of F2 plants obtained from seeds

47
harvested from treated F1 plants showed no significant
difference from those of the control F2 populations. Of 41
plants analyzed, 18 were structurally heterozygous and
morphologically similar to Fl's and 23 were structurally
homozygous. Weight of 1000 caryopses was calculated, the
mean being 19.2 grams and coefficient of variability 25.1
per cent. Therefore, there was no significant difference
between treated and control F2 populations with regard to

the seed weight or its variance.

4. Discussion

With the knowledge that three translocations are the
major factors which keep the two species separate, attempts
were made to induce new translocations or to reverse natural
ones.

It is suggested that Ca++ and Mg++ play an important
role in maintaining chromosome structure. Plants deficient
in these ions can be produced by growing them in an artificial
mediwm lacking these ions. But this is a long process and
causes an over-all disturbance in plant metabolism. Rapid
inactivation of metallic ions in solutions can be achieved
by chelating agents. These form chelate-metal complexes and
though soluble, are not ionized. EDTA shows a marked
affinity for calcium ions.

In this experiment EDTA did not produce any chromosome

breakage and two cells each with one additional translocation

48
cannot be indicative of such effects. As the treated cells
are the last or close to the last ones in their lineage to
form microspores, any effective treatment will produce very
limited cells with modified chromosomes. It has been shown
by several workers that chelating agents produce more re-

combination of genetic material in bacteria and Drosophila.

 

Information postulating chiasma formation as the cytological
mechanism for genetic recombination are not conclusive.
Formation of interstitial chiasma in structurally heterozygous
hybrids of S, cereale X.§- montanum and their increase by
'means of EDTA seemed at first to be a suitable case for such
investigation. If more recombinant types had been recovered
in F2 populations from treated Fl's than in F2 populations
from control Fl plants, then we could have established a
positive correlation between chiasma frequency and genetic
recombination. This was not the case. Recombinants did not
appear either from cytological analysis of treated F2's or
from their morphological characteristics. Systems of three
translocation differences makes it very improbable for
products of interstitial crossing-over to be recovered.

There should be other cross—overs in the same chain to compen-
sate for chromosomal unbalance produced by the first interstitial
cross-over. As the treated Fl plants had only 15 per cent
seed set or about half of that of the controls, those gametes

which survived may have been those with parental combinations.

49

 

 

 

 

 

 

 

 

 

 

 

0 1:37:12. ‘ ‘ ‘

Fig. 19. Diagrams of 4 chromosome associations in a
translocation heterozygote. A, pachytene
association with chiasmata located so that
a "frying pan" configuration would result
at first metaphase; B, first metaphase with
terminalization of proximal chiasma arrested
at the point of interchange; C, first meta-
phase with terminalization completed despite
changes of chromosome homology. (After
Price, 1959.)

50

 

Fig. 20. Diagrams of 4 chromosome associations in a trans-
location heterozygote. A, pachytene association,
B, metaphase I, and C, the same association at a
somewhat later stage. Complete terminalization
of interstitial chiasma is accompanied by opening
of the "frying pan" configuration.

 

I
I

0
--~__._ '— —._.___ _—_.._ 5
..

Fig. 21. Diakinesis in F1 hybrids of S. cereale x‘S. montanum.
849x. ~

\

 
 

£315.
9

 

I
J

g_h___-_. 11 if

.__—.‘ —.—‘

Fig. 22. Metaphase I in F1 hybrids of S. cereale x S. montanum.

Four closed bivalents and an interstitial Ehiasma
in chain of six. 1263X.

52

 

Fig. 23. Metaphase I in F1 hybrids. Four bivalents plus a
chain of six with an interstitial chiasma. A
later stage than Fig. 22. 1202X.

I ' [

 

 

 

LLB V _ 1
Fig. 24. Interstitial chiasma in a chain of four.

 

3

l
I ‘ ' I
. - ~——~— — , H *"-—-——1x‘

Fig. 25. Interstitial chiasma in a chain of six. 754x.

 

‘— _ _ --.

;. ..-

I

 

 

I T .
Fig. 26. Two interstitial chiasmata in chains of III'and
IV. 844K.

53

 

_--1_ i" r.‘7 i_11

Fig. 27. Four closed bivalents plus a "frying pan"
configuration made up of sixzhhromosomes. 1254X.-

. o
z 3 '
O

I
I

j
i
111 11 11 1 11 111 1 1111

Fig. 28. Open "frying pan" configuration. 824x.

55

PART IV
ALLOPOLYPLOIDY INDUCTION

1. Review 9: Literature

In the process of organic evolution, nature has long
utilized polyploidy in creating new species. Recently plant
breeders have used this method to produce more suitable plants.
In flowering plants, nearly half the species, and in the
Gramineae about 75 per cent of the species owe their origin
to polyploidy. In general, the origin of polyploids in the
Gramineae has been through hybridization and subsequent
chromosome doubling of the hybrids. With such a high occurrence
of polyploidy in the grass family, differentiation in Secale
has occurred entirely at the diploid level (2n - 14). Lack
of polyploid series and almost complete barriers to recombina-
tion of specific characteristics at the diploid level
encourages allopolyploidy induction in this genus.

In order to produce polyploids artificially different
methods have been utilized such as tissue wounding, grafting.
heat treatment, colchicine techniques and nitrous oxide.

Blakeslee and Avery (3) opened a new era in polyploidy
investigation when they announced in 1937 the discovery of
colchicine as a powerful chemical agent in inducing polyploidy
in plants. Since that time many polyploid plants have been 1
produced by this method. Colchicine techniques usually

produce plants which have polyploid chimeras. Attempts have

56

been made to double the embryonic cells as close to the first
mitotic division of the zygote as possible. This may result
in plants Whose chromosomes are doubled in all or most of the
tissue. For this purpose gases such as nitrous oxide which
have a good penetrating power have been utilized.

I Ostergren (39) in 1944 treated Crepis capillaris with
N20 under pressure cxf ten atmospheres for four-six hours at
the first or second zygotic division. By this method he
obtained a fairnumber of polyploids.

Nygren (37), also using nitrous oxide, was able to

induce polyploidy in Melandrium.

 

Cooper (7) used nitrous oxide at 50, 75 and 100 pounds
pressure in order to induce polyploidy in 21212- Plants
receiving 75 pounds pressure or more did not set any seeds.
No tetraploids were found in the samples of treated materials.

The first autotetraploid plants in rye were induced
by Dorsey (12) in 1936, using the heat shook method developed
by Randolph (43). After the discovery of the colchicine
technique many more autotetraploid ryes were produced.
especially in Sweden.

Chin (6) analyzed metaphase I of autotetraploid rye
and found that the maximum number of quadrivalents was four
in one cell. He also observed bridges at the first anaphase.
The material under study was directly obtainbd from the)
diploid parent by doubling the chromosome number and no

selection had been practiced at the tetraploid level.

57
Tetraploids had a seed set of about 23.8 per cent.

Mflntzing (35) studied rye autotetraploids for several
different characteristics and compared them to their diploid
parents. These tetraploids had been selected for seven
generations for desirable agronomic characteristics. At
metaphase I of this material, an average of four quadrivalents
were found per cell with many univalents. Tetraploid plants
were about 15 per cent taller than diploids and their degree
of tillering was reduced about 12 per cent. Tetraploids had
fewer flowers per spike than diploids but their caryopsis
weight exceeded those of the diploids by more than 50 per cent.
In the best tetraploid varieties the average seed set was
about 62 per cent, whereas the corresponding value for
diploids was 80 per cent.

Morrison (33) studied autotetraploid rye selected for
fertility over a 15 year period. He did not find any
significant difference between this rye and "Steel-rye"
produced by Mflntzing and selected for fertility for seven
generations with respect to meiotic behavior. Therefore, he
concluded that selection for increased fertility was not
accompanied by changes in pairing behavior. Hilpert (18),
on the other hand, believed that good fertility was correlated
with regular meiosis.

Although several workers have tried to combine
desirable characteristics of different rye species at the
diploid level, there is no report of utilizing allopolyploidy

for this purpose.

58

2. Methods and Materials

In order to induce polyploidy two methods were
employed:

1. Nitrous oxide under pressure.
2. Colchicine.

Nitrous oxide was applied under five different
pressures; namely, 75, 100, 125, 150, and 200 pounds per
_ square inch. Duration of treatments was 6, 12, 18, and 24
hours. Heads were bagged individually and pollinated when
completely receptive. Treatments started 12, 24, 48, and 72
hours after pollination. Plants in pots were treated by
putting them in an air-tight chamber. None of the plants
died as a result of treatment. No seeds were obtained from
plants receiving pressures of 125 pounds or more. In seeds
obtained from 75 and 100 pound treatments, no polyploids were
found.

The second method employed for polyploidy induction
was colchicine treatment.

Interspecific hybrid seeds were germinated in petri
dishes. When coleoptiles and roots were about 2-5 mm long.
they were immersed in a thin layer of 0.1 or 0.2 per cent
of aqueous colchicine for 3 or 1.5 hours, respectively.
Petri dishes were placed in an oven with a 27°C temperature
during treatment. After treatment, seeds were washed

thoroughly with water for about ten minutes and then spread

59
over'moist blotting paper in petri dishes at room temperature
for about four days and then transplanted to flats in the
greenhouse. Fifty-six plants out of 180 seeds treated
reached maturity in which seven tetraploid heads were detected.
Tetraploid heads were found as sectors and in no case was the
Whole plant doubled. MorpholOgically, tetraploid sectors
were quite distinct from diploid parts by having larger
leaves, thicker and longer stems.

Of the seven tetraploid heads produced, five were
crossed to Mota rye, an autotetraploid, because they bloomed
at different times and it was not possible to cross doubled
Fl's together. The other two bloomed simultaneously and
were intercrossed. Seeds obtained from these two heads were
*planted in an isolated area in the field.

It seems worthwhile to mention that the following year
exactly the same colchicine technique was employed and
despite a larger number of seeds treated, no tetraploid heads
were detected. The difference seems to be due to the cold
period which followed colchicine treatment in the second
year, but which was not applied during the first colchicine
treatment. The cold period employed for the purpose of
vernalization of winter types seems to retard development of

tetraploid sectors and results in their abortion.

3. Results

Detailed cytological analysis of doubled F1 plants

was not possible due to the limited amount of pollen mother

60
cells available. Eleven cells which were analyzed showed an
average of 1.1 univalents, 9.4 bivalents, 0.63 trivalents.
and 1.0 tetravalent. Analysis of 322 microspores showed that
23 per cent had one, 21 per cent had 2, 0.9 per cent had 3,
and 84 per cent had no micronuclei. When flowers of the two
doubled heads were crossed, the average seed set was about

51 per cent. Thus. compared to diploid F 's with an average

1
seed set of 27.7 per cent, tetraploids showed an improvement
of 80 per cent in the first generation.

In the F2 generation rather detailed cytogenetic
analysis of the doubled hybrids was carried out.

I At the prophase stage the tetraploids showed only one
nucleolus instead of two (Fig. 32) and no cells were found
which contained more than one nucleolus. Thus. doubling did
not seem to have affected number of nucleoli.

Analysis of all cells at metaphase I was not possible.
due to the large number of chromosomes and their overlapping.
The results of metaphase I analyses of 100 cells in which
the chromosomes were well spread and readily countable are
shown in Table 7. This table shows the frequency of
different multivalents found in 100 cells and not the
combination of different multivalents in each cell. The first
column in the table indicates the chromosome configurations,
the second frequency of their occurrence, the third their.

relative frequency, and the fourth the average number of each

multivalent per cell.

61

TABLE 7. Metaphase I analysis of doubled F2 hybrids.

 

 

 

 

Average
Configuration

Configurations Frequency Per Cent Per Cell
I 74 5.6 0.74
II 1072 81.9 10.72
III 70 5.3 0.70
IV 64 4.9 0.64
V 14 1.1 0.14
VI 10 0.76 0.10
VII 5 0.38 0.05
1309 99.99 28 I

or
14 II

 

On the average more than ten bivalents were found in
each cell at metaphase I. In many other cells in which
complete analysis was not possible, at least seven bivalents
could be detected. The average number of quadrivalents
per cell was 0.64 and in no case were seven quadrivalents
present. Quadrivalents were mostly oriented in zig—zag
chain configurations and rings of IV were not observed.
Multivalents higher than quadrivalents were also observed.

These were due, perhaps, to homologies caused by translocations.

One hundred sixty—three cells were analyzed at

anaphase I and the results are shown in Table 8. The first

62
column shows the kind of distribution, and the second column
their relative frequency. Behavior of univalents was the
same as described for diploid rye. Other irregularities were

detected which are listed in Table 8.

TABLE 8. Chromosome distribution at anaphase I of doubled F

 

 

 

hybrid rye. 2
Distribution Frequency
14 - 14 59.5%
13 - 15 17.8

12 - 16 3.1

13 - l - 14 3.1

13 - 2 - 13 2.4

13 2/2 - 13 2/2 1.8

13 1/2 - 14 1/2 * 8.0

13 — 14 2/2 1.8

12 2/2 - 14 2/2 1.2

12 — 2 - 14 1.2

 

Microspore analysis showed that 76 per cent had no
micronuclei, 21 per cent had one, and 3 per cent had two.
The total number of micrOSpores counted was 412.

Pollen grains were much larger in tetraploids than in
their diploid parents. Tetraploids had 87.21 per cent

stainable pollen grains whereas the corresponding value for

63
diploid Fl's was 45.5 per cent. Thus, tetraploidy increased
the amount of stainable pollen 92 per cent over their diploid
parents.

Seed fertility in the tetraploid F2's averaged 40.4
per cent compared to 27.7 per cent for diploid hybrids and
represents a 45 per cent improvement in the amount of seed
set. Average thousand kernel weight in the tetraploids was
31.1 grams, whereas that of diploid hybrids was 23.6 grams.
an increase of about 32 per cent in seed weight. Average
height of tetraploid plants was.63.6 inches, the corresponding
value for diploid Fl's was 48.2, an increase of 32 per cent
in plant height due to polyploidy. Average diameter of culms
measured in the middle of the second internode was 5.72 mm.
that of the parent diploid 3.20 mm or about 79 per cent
increase in culm diameter.

The above tetraploids were compared with their
related diploid F2's from the same original crosses with
respect to five agronomically important characteristics. The
results of such comparisons are listed in Table 9. Plant
height, culm diameter, and kernel weight were larger for the
tetraploids. Tillering and fertility were relatively
constant between the diploid and tetraploids. As far as
fertility was concerned, tetraploid plants were very uniform
in flowering and during this period unfavorable weather
prevailed. In a favorable condition, higher fertility would

be expected. It seems more reasonable to expect higher

64
higher fertility, especially when we consider that doubled
Fl's had about 51 per cent seed set when pollinated under
bags, in the greenhouse. In the fields some tetraploid plants
had fertility as high as 68 per cent and variability between

the heads of the same plants was very low in all cases

 

 

 

 

studied.

TABLE 9. Comparison of means and coefficients of variability
for several characteristics in diploid and
tetraploid F2's of interspecific crosses.

Diploid Tetraploid
Coefficient of Coefficient of
Mean Variability Mean Variability

Plant height (inJ 45.86 19.07 63.63 7.4

Tillering 40.35 53.5 41.63 43.5

Stem diameter (mm) 3.36 24.7 5.72 13.11

Thousand kernel

weight (grams) 21.9 27.2 31.1 12.67

Fertility (per 43.5 56.1 40.4 37.88

cent)

 

A comparison of coefficients of variability for
characteristics in the diploids and tetraploids reveals less
variability in all cases for each characteristic in the
tetraploids than in the diploids.

A comparison of three characteristics whose means
were increased in the tetraploids; namely, plant height, culm

diameter, and 1000 kernel weight reveals how means and

65
variances have changed in tetraploids when compared to
diploids (Fig. 29). In this figure means are plotted on the
vertical axis and variance on the horizontal axis for each
of the above characteristics. For all of these three cases.
tetraploid means have increased and variances have decreased.
Variance decreases also for tillering and fertility Whose means
remain unchanged in the diploid and tetraploid F2's.

It should be pointed out that chromosome doubling
in cultivated rye causes a considerable reduction in degree
of tillering. In this case allotetraploidy seems to compensate
for that disadvantageous effect.

Tetraploid F 's were partially perennial and continued

2
to send up new shoots and form heads which again produced
seeds. When we consider that diploid Fl's are strongly
perennial, then doubling the chromosome number has actually
changed this characteristic from dominant to an intermediate
condition.

In contrast to diploid hybrids Which have brittle or
semibrittle heads, which usually break before they are
harvested, the doubled F2 hybrids had persistent heads Which
remained intact during harvest and showed only slight brittle-
ness at the top after drying. For all practical purposes.
they may be considered non-brittle. A similar situation was
observed in doubled Fl's. Thus, for this character doubling

the chromosome number in hybrids has reversed dominance almost

completely.

MEAN

Fig.

29.

 

66

4N

 

2N

 

VARIANCE

Diagram comparing means and variances for three
quantitative characters, namely plant height,
culm diameter, and 1000 kernel weight, in diploid
and tetraploid F2 populations from interspecific
crosses. (Adapted after Stebbins, 1956.)

67

4. Discussion

If reproductive barriers at the diploid level present
a formidable handicap to recombination of desirable character-
istics in interspecific crosses of Secale, it is possible by
doubling the chromosome number in hybrids to overcome these
barriers. Nature has utilized this device extensively to
circumvent the reproductive barriers existing between species.
In fact, most naturally known amphidiploids have appeared
in hybrid cultures. In such cases, hybridization leads to the
sharing of gene combinations, and.polyploidy fixes these
combinations with their heterosis, and at the same time
insures the reproductive potential by restoring fertility.
In Secale, natural polyploids have not been reported. This.
of course, does not mean that in nature these polyploids
have never been produced or are produced in a very low
frequency. Rather, after occurrence they may not establish
themselves and are easily eliminated. For example. polyploids
have been very successful in wheat. The reason is certainly
in the different genetic systems Which these genera possess
and evolution in each of them has taken different pathways.
Rye is naturally a highly cross—pollinated genus and a high
potential for recombination can be maintained in this way,
whereas in wheat, because of self-pollination, larger numbers
of chromosomes are necessary to produce a high recombination
index. In rye occasional natural tetraploids will outcross

to diploids, and the triploid embryos abort, Whereas in wheat

68
self-pollination preserves these tetraploids.

The fact that three pairs of chromosomes in S. cereale
and S, montanum are identical as far as their gross morphological
structure is concerned, make their doubled hybrids a segmental
allopolyploid and not an amphidiploid. The other four pairs
of chromosomes which are not similar in the two species, have
translocation differences. This is why we find at metaphase I
of doubled hybrids neither 14 bivalents, as would be the case
for amphidiploids, nor many tetravalents as might be expected
in an autotetraploid. They show an average of 10.72 bivalents,
which includes the eight pairs of non-similar chromosomes
with some additional ones. By means of selection for more
regular meiosis and higher fertility, it might be possible
to increase the number of bivalents close to 14 in later
generations. Occasional chains of V, VI, and VIII were
observed in metaphase I which are the result of association
of similar segments in different pairs of translocated chromo-
somes. Multivalents higher than chains or rings of four
usually do not occur in autotetraploids of rye.

In spite of the strong self-incompatibility which
exists in diploid S. montanum, S. cereale, and their hybrids,
allotetraploids exhibited different degrees of self—compatibility
which in some cases reached 22 per cent when individual spikes
were isolated under glassine bags. Selection for this A
characteristic would certainly result in more self-compatible

races. Such self-compatible races would be more useful in

69
areas where weather conditions are not usually favorable for
cross-pollination. This is especially important for tetraploid
rye which is much more uniform in pollination time and a
short period of unfavorable weather during this critical
period may cause serious losses. Such self-compatible races
will not degenerate very fast as a result of inbreeding, as
diploids usually do; because they carry a greater load of
heterozygosity and have a higher recombination index. Changes
in genetic systems brought about by selection for self—
fertility would certainly be of practical and theoretical
importance.

Higher fertility of allotetraploids in comparison to
diploid hybrids confirms the fact that partial sterility of
the F1 plants is not genic, rather it is chromosomal. By
doubling the chromosome number of the F1 hybrids, each
chromosome finds an exact homologue, pairing and segregation
becomes more regular, and, as a consequence, higher fertility
is achieved.

A comparison of variances in diploid F 's and

2

tetraploid F2's showed that tetraploids were buffered more
in their reactionto environmental factors than diploids.
While diploids readily release their variability in the form
of new genotypes, tetraploids have their heterozygosity
distributed among all individuals. As a partial compensation

for lack of genotypic variations, the whole population is

similarly buffered against the environment. Such a developmental

70
flexibility permits the population to become adapted to its
environment without losing its genetic variability. Thus,
tetraploids and especially amphidiploids, obtain a great
selective advantage in areas where drastic changes of
environmental factors prevail. In such areas the intensity
and direction of natural selection changes periodically and
diploid species whose genetic variation is released in the
form of new individuals soon exhaust their store of variability.
Due to their buffering capacity, polyploids may be favored

in the survival of the species.

Fig. 30.

Fig. 31.

71

4....---

 

b—r —

 

Twenty-eight mitotic chromosomes of allotetraploid
rye at metaphase. 936x.

H—-.
1m

 

 

 

Mitotic anaphase of allotetraploid rye. 776x.

 

L7 7 7 7 7 1 fl 77:7?

Fig. 32. Prophase in pollen mother cells of allotetraploid
rye. One nucleolus present. 977x.

J

 

 

Fig. 33. Prophase in PMC's of allotetraploid rye. "bouquet"
stage. 735x.

i

73

 

Fig. 34. Metaphase I in PMC' s of allotetraploid rye. An

aneuploid plant with 11 II and a chain of 7
chromosomes. 817X.

 

 

Fig. 35. Metaphase I in PMC's of allotetraploid rye.
Eleven II plus a chain of 6. 873x.

74

   

1
I

Fig. 36. Metaphase I in PMC's of allotetraploid rye, 8 II,
3 I. plus a chain of V and a chain of IV. 849x.

. I ‘

 

 

 

Fig. 37. Anaphase I in PMC's of allotetraploid rye.
14-14 distribution. 977X. ‘

_ _ .1 1 --. ”Al-bul In.“ 75

 

 

Fig. 38. Anaphase I in PMC's of allotetraploid rye.
Aneuploid plant with 12—15 distribution. 933x.

I
l
I
1
i

I
I

I A I ,

Fig. 39. Lagging univalents dividing at anaphase I. 917X.

 

Fig. 40.

 

Stainable and non—stainable pollen
grains from allotetraploid rye. 85X.

76

10.

ll.

12.

13.

77

LITERATURE CITED

Anderson, E. Evolution After Darwin, Vol. III. Ed. by
S. Tax. University of Chicago Press (1960).

Antropov, V. and V. Rye in USSR and in the adjoining
countries. Supp. 36-th of the Bull. Appl. Bot.
(1929).

Blakeslee, A. F. and A. G. Avery. Methods of inducing
doubling of chromosomes in plants. J. Heredity,
28:393-411 (1937).

Bremer, G. and Bremer-Reinder, D. E. Breeding of
tetraploid rye in the Netherlands. I method and
cytological investigations. Euphytica, 3:49-63

Caldwell, R. R. and Burdick, A. B. Uptake and effect
on crossing-over of EDTA in Drosophila melanogaster
The Nucleus, Vol. 11:125-130 (1959).

Chin, T. C. Cytology of autotetraploid rye. Bot. Gaz.,
Vol. 104, pp. 627-632 (1943).

Cooper, R. L. "Hybridization in vetch," thesis for the
degree of M.S., M.S.U. (1958).

Darlington, C. D. Chromosome behavior and structural
hybridity in the Tradescantiae. J. Genetics,
21:207-286 (1929).

Darlington, C. D. The origin and behavior of Chiasmata
VIII S. cereale (n = 8). Cytologia, 4:444-452
(1933).

Darlington, C. D. Chromosome Botany. George Allen &
Unwin Ltd. (1956).

Davidson, D. The effect of chelating agents on cell
division. Exp. Cell Res., 14:329—332 (1958)..

Dorsey, E. Chromosome doubling in the cereals. J.
Heredity, 30:343-395 (1939).

Duka, S. K. Cytological research on the interspecific
hybrids S, cereale x S, montanum. Plant Breeding
Abstract, 6:869 (1935).

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

78

Elliot, F. C. Plant Breeding and Cytogenetics. McGraw-
Hill Book Company (1958).

Eversole, R. A. and Tatum, E. L. Chemical alteration
of crossing-over frequency in chlamydomonas.
Proc. Natl. Acad. Sci., 42:68-73 (1956).

 

Grant, V. Chromosome repatterning and adaptation. Adv.
in Genet., 8:89—107 (1956).

Grant, V. The regulation ofirecombination in plants.
Cold Spring Harbor Symp., Vol. 23:337-363 (1958).

Hilpert, G. Effect of selection for meiotic behaviour
in autotetraploid rye. Hereditas, 43:318—22
(1957).

Hyde, B. B. The effect of versene on the structure of
plant chromosomes. Biol. Bull., 109:347 (1955).

Jain, S. K. Cytogenetics of rye. Bibliographia
Genetica, Vol. 19:1-86 (1960).

Kaufmann, B.P., Gay, H. and.McElderry, M. J. Effect of
ribonuclease on crossing-over in Drosophila.
Proc. Natl. Acad. Sci., 43:255 (1957).

Kaufmann, B. P. and McDonald, M. R. Organization of the
chromosomes. Cold Spring Harbor Symp., 21:233
(1956).

Kaufmann, B. P. and McDonald, M. R. The nature of the
changes effected in chromosomal materials by the
chelating agent EDTA. Proc. Nat. Acad. Sci.,
43:262 (1957).

Kirchner, C. E. J. The influence of divalent cations
on genetic exchange in bacteria. Plant Breeding
Abstract, p. 216 (1960).

Kostoff, D. Pollen abortion in species hybrids.
Cytologia, 3:337-339 (1932).

Kostoff, D. Interspecific hybrids in Secale. [Summary
in P1. Breeding Abst., 7:392 (1937)].

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

79

Kranz, A. R. Populationsgenetische untersuchungen am
iranischen primfijveroggen. Zeit. Pflanzenz, 38:
101-146 (1957).

Levine, R. P. Chromosome structure and the mechanism
of crossing-over. Proc. Nat. Acad. Sci., 41: 727-
730 (1955).

Levine, R. P. Chromosome organization and gene recom-
bination. Cold Spring Harbor Symp., 21: 247 (1956),

Longley, A. E. and Sando, W. J. Nuclear divisions in
the PMC's of Triticum, Aegelops, and Secale and
their hybrids. Jour. Ag. Res., 40:683 (1930).

 

Mazia, D. The particulate organization of the chromo-
somes. Proc. Natl. Acad. Sci., 40:521-527 (1954).

McDonald, M. R. and B. P. Kaufmann. Production of
mitotic abnormalities by EDTA. Expt. Cell Res.,
12:415-417 (1957).

Morrison, J. W. Chromosome behavior and fertility of
tetrapetkus rye. Canad. Jour. Ag. Sci., 36:
157-165 (1956).

Mfintzing, A. The evolutionary significance of auto-
polyploidy. Hereditas, 21:263-378 (1936).

Mfintzing, A. Cytogenetic properties and practical value
of tetraploid rye. Hereditas, 37:17—84 (1951).

Mfintzing, A. Frequency of accessory chromosomes in rye
strains from Iran and Korea. Hereditas, 43:682
(1957).

Nygren, A. Polyploids in Melandrium produced by nitrous
oxide. Hereditas, 41:287-290 (1955).

 

Ossent, H. P. Pernnierender Kulturroggen. Zfichter,
2. 221 (1930).

Ostergren, G. Polyploids and aneuploids of Crepis
capillaris produced by treatment with nitrous
oxide. Genetica, 27:54-64 (1954).

 

40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

50.

51.

52.

80

Prakken, R. Studies of asynapsis in rye. Hereditas,
29:475-495 (1943).

Price, S. Irradiation and interspecific hybridization
in Secale. Genetics, 40:651-667 (1955).

Price, S. Chiasma terminalization in a structural
heterozygote in Secale. Genetics, 44:705-712
(1959).

Randolph, L. F. Some effects of high temperature on
polyploidy and other variations in maize. Proc.
Natl. Acad. Sci., 18:222-229 (1932).

Randolph, L. F. An evaluation of induced polyploidy
as a method of breeding crop plants. Am.
Naturalist, 75:347 (1941).

Rees, H. Asynapsis and spontaneous chromosome breakage
in Scilla. Heredity, 6:89-97 (1952).

Riley, R. The cytogenetics of the differences between
some Secale species. Journ. Agric. Sci., 46:
377-383 (1956).

Roshevitz, R. Y. A monograph of the wild, weedy, and
cultivated rye. In Russian. P1. Breeding Abstract
(1948).

Schiemann, E. and U. Nurnberg-Krfiger. Neue Untersuchungen
an Secale africanum Stapf. Die Naturwiss, 6:
136-137 (1952).

Stebbins, G. L. Variation and Evolution in Plants.
New York, U. P. Columbia (1950).

Stebbins, G. L. Artificial polyploids as a tool in
plant breeding. Brookhaven Sym. Biol. NO. 9:
37-52 (1956).

Stebbins, G. L. The inviability, weakness, and sterility
of interspecific hybrids. Adv. Genet. 9:147-215
(1958).

Steffensen, D. Induction of chromosome breakage at
meiosis by a magnesium deficiency in tradescantia.
Proc. Natl. Acad. Sci., 39:613-620 (1953).

53.

54.

55.

56.

57.

58.

81

Steffensen, D. Breakage of chromosomes in Tradescantia
with a calcium deficiency. P.N.A.S., 41:155-160
(1955).

Stutz, H. C. A cytogenetic analysis of the hybrid
Secale cereale S. x S, montanum Guss. and its
progeny. Genetics, 42:199-221 (1957).

Vavilov, N. I. On the origin of cultivated rye. Bull.
Appl. Bot., 10:561-590 (1917).

Vavilov, N. I. Studies on the origin of cultivated
plants. Bull. Appl. Bot., 16:1—248 (1925).

Whittaker, T. W. The Geobotanical significance of
polyploidy. Evolution, 5:417 (1951).

Wolff, S. and Luippold, H. E. The production of two
chemically different types of chromosome breaks
by ionizing radiation. P.N.A.S., 42:510-514
(1956).

,1} -Y