Characterization of Desmodium heterocarpon (L) DC. and Desmodium ovalifolium wall

MISSING IMAGE

Material Information

Title:
Characterization of Desmodium heterocarpon (L) DC. and Desmodium ovalifolium wall
Physical Description:
v, 91 leaves : ill. ; 28 cm.
Language:
English
Creator:
McKellar, Martin Albert, 1950-
Publication Date:

Subjects

Subjects / Keywords:
Legumes   ( lcsh )
Legumes -- Germplasm resources   ( lcsh )
Agronomy thesis Ph. D
Dissertations, Academic -- Agronomy -- UF
Genre:
bibliography   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1988.
Bibliography:
Includes bibliographical references (leaves 86-90).
Statement of Responsibility:
by Martin Albert McKellar.
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001477670
oclc - 20959266
notis - AGY9594
System ID:
AA00011768:00001


This item is only available as the following downloads:


Full Text











CHARACTERIZATION OF DESMODIUM HETEROCARPON (L) DC.
AND DESMODIUM OVALIFOLIUM WALL.















By


MARTIN ALBERT MCKELLAR


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



'UNIVERSITY OF FLORIDA


1988














ACKNOWLEDGMENTS


The author thanks Dr. Kenneth Quesenberry for

guidance throughout this program of study and research.

Committee members Dr. David Knauft, Dr. Gloria Moore, Dr.

Carrol Chambliss, and Dr. Paul Lyrene offered sensible

advice for bridging murky matters and avoiding deadends.

Dr. Rex Smith made much of the isozyme work possible

by allowing the author generous use of his lab and his

knowledge of the subject. Dr. Albert Kretschmer provided

insight into the genus Desmodium and triggered new thoughts

on the subject with his very specific questions.

The author thanks Caryl McKellar for knowing when

patience and support were beneficial, and when lack of

patience was better.

David Moon deserves credit for making much of the

day-to-day research run smoothly and pleasantly.

Special thanks to Dr. Barbara Probert and the Math

Confidence Workshop, without which none of this would have

been possible.

















TABLE OF CONTENTS


Page
ACKNOWLEDGMENTS. . . .. .. ii

ABSTRACT . . ... iv

CHAPTERS

I INTRODUCTION. . 1

II MORPHOLOGY OF DESMODIUM HETEROCARPON AND
DESMODIUM OVALIFOLIUM . .. .13

Materials and Methods . .. .16
Results . . .. .22
Discussion. . .. .35

III DESMODIUM HYBRID IDENTIFICATION USING
ISOZYME ELECTROPHORESIS . 42

Materials and Methods . .. .44
Results. . ... 49
Discussion. . ... 59

IV CYTOLOGY OF DESMODIUM HETEROCARPON AND
DESMODIUM OVALIFOLIUM . .. .64

Materials and Methods . .. 65
Results . . 69
Discussion. . .. 74

V CONCLUSIONS .. . 83

LITERATURE CITED . ... 86

BIOGRAPHICAL SKETCH. . . 91








iii














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

CHARACTERIZATION OF DESMODIUM HETEROCARPON (L) DC.
AND DESMODIUM OVALIFOLIUM WALL.

By

Martin Albert McKellar

December 1988

Chairman: K. H. Quesenberry
Major Department: Agronomy


Characterization of Desmodium heterocarpon (Dh) and

D. ovalifolium (Do) germplasm was undertaken as the first

step of a breeding program. The objectives were to

characterize Dh, Do, and five interspecific hybrids for

morphological traits, isozyme markers, cytological behavior

during meiosis, pollen viability, and reproductive biology.

Fifty-seven lines of Dh and six lines of Do were grown

in two replications in the field in Gainesville, Florida, in

1985, or in two replications in the field in 1986 in

Dschang, Cameroon, or in both locations. Plant height,

plant width, leaf length, leaf width, flowering date, and

presence of marked leaves were measured. Plant height

ranged from 9 to 125 cm. Plant width varied from 13 to 78

cm. Leaf length varied from 2.9 to 6.5 cm. Leaf width

varied from 1.4 to 3.4 cm. Two lines of D. heterocarpon

subspecies angustifolium had leaf length/width ratios > 3.0

(a major taxonomic descriptor delineating this subspecies),










whereas a third line was < 1.5 in Gainesville and Cameroon.

Caged Do set little or no seed; caged Dh, and uncaged Do and

Dh, produced seed. Pollen stainability for parental and

hybrid lines was high.

Two Do lines, three Dh lines, and five Do x Dh

hybrids were grown in 10 cm pots both in a shade house and

in a greenhouse in 1987-1988. Meiotic analysis showed

univalents at metaphase I and laggards at first division

segregation in two of the five hybrids. Starch gel

electrophoretic analysis of isozyme markers in root and leaf

material showed staining differences for peroxidase between

parental lines and hybrids. Root material showed

differences for phosphohexose isomerase between parental

lines and hybrids.

Segregation for marked leaf in a Do x Dh F2 progeny

was 9:7, marked leaf:unmarked leaf. Modal F2 flowering

date was midway between parental flowering dates, but

individual plant flowering dates ranged over one month.

Segregation for flower color in a Dh x Dh cross was 3:1

purple:white.

These findings suggest that although Do and Dh are

separated by morphological differences, crossing barriers

between the two species are low. Morphological and isozyme

markers identified in this research will aid in recognizing

successful hybrids.















CHAPTER I
INTRODUCTION


Tropical forage legumes research in Florida

accelerated in the early 1970s as a result of rising prices

for crude oil which meant rising costs for nitrogen

fertilizer derived from oil and natural gas. This increased

cost made research into alternative sources of nitrogen

fertilizer more attractive, particularly research into

tropical forage legumes that fix nitrogen symbiotically.

Choice of a forage legume is complicated by disparate

seasonal temperature variation in Florida. Tropical summer

temperatures and cold winter temperatures make it unlikely

that one species can provide high quality forage throughout

the year. However, a perennial tropical forage legume could

improve forage crude protein during mid and late summer when

traditional forage grasses are of low nutritional content

(Sollenberger et al., 1988). The genus Desmodium contains

several tropical forage legumes with major agronomic

potential: D. canum (Gmel.) Schin. and Thell., D. intortum

(Mill.) Urb., D. ovalifolium Wall., D. sandwicense E. Mey.,

and D. uncinatum (Jacq.) DC. (Imrie et al., 1983).

Desmodium is a large genus containing an undermined

number of species variously listed as 200 (Takahashi, 1952),

350 (Bryan, 1966), 350-450 (Imrie et al., 1983), and 500










(Younge et al., 1964). Difficulty enumerating the species

number stems from the extreme variation in morphological

characteristics within the genus that extends across species

delineations (Imrie et al. 1983). Contributing to the

problem is the existence of two widely separated major

centers of origin; the southeast Pacific and central

America. No complete monograph is available on Desmodium.

The Asian species have been classified extensively by Ohashi

(1973) and the American species have been studied primarily

by Shubert (1963).

Desmodium heterocarpon, which has received some

research attention in Florida, is a case where questions

remain as to the exact delineation of species. Ohashi

(1973) described two subspecies of D. heterocarpon, one with

four varieties. The variety D. heterocarpon subspecies

heterocarpon variety heterocarpon includes both lines

typified by 'Florida" carpon desmodium [D. heterocarpon (L.)

DC. variety heterocarpon] as well as the species described

as D. ovalifolium Wall. The distinguishing characteristics

in Ohashi's key that grouped these two species together in

one subspecies were pods distinctly hooked-hairy,

inflorescence-rachides covered with spreading hooked hairs,

inflorescences densely flowered and pedicles 2-5(-8) mm long

in fruit. This classification reduced D. ovalifolium Wall.

to variety status, although it was listed as a separate

species by Schindler in 1928 (Ohashi, 1973) and continues to

be referred to as a separate species. Simultaneously, a











second D. ovalifolium, this one with authority Guill and

Perr., has been referenced (Hussain et al., 1987). Schubert

(1971) listed D. ovalifolium Guill and Perr. (described in

1832) as synomymous with D. adscendens (Sw.) DC. described

in 1825. Characteristics of Shubert;s D. adscendens were

herbaceous growth, creeping or forming tussocks, and leaves

that are pilose to glabrous on top and densely pilose on the

bottom. Shubert's description of D. heterocarpon was

shrubby or with procumbent habit with leaves glabrous to

moderately pilose above and sparsely to abundantly pilose

below. No mention was made of marked leaves in either

species by Shubert or Ohashi. Ohashi (1973) did not list D.

ovalifolium Guill and Perr. among the D. ovalifolium

synonymous with D. heterocarpon ssp. heterocarpon var.

heterocarpon, therefore the exact relationship between the

D. ovalifolium of Wall. and Guill and Perr. remains unknown.

The major distinctions between the two species are at

the gross morphological level. Imrie used D. ovalifolium in

his discussion of the genus because the agronomic type is

known by that name in Southeast Asia (Imrie et al., 1983).

Characteristics that distinguish D. ovalifolium from D.

heterocarpon are reported as prostrate versus erect growth

habit; leaves unifoliate when young versus trifoliate when

young, leaves glabrous on top versus opaque on top, leaves

unmarked versus marked with light green or white markings;

inflorescenses compact versus elongated; pods densely

pubescent versus glabrous to slightly pubescent











(Schultze-Kraft and Benavides, personal communication).

These wide morphological differences between D. ovalifolium

and D. heterocarpon have resulted in most researchers

continuing to refer to the agronomic types as distance

species (Schultze-Kraft and Benavides, personal

communication). In this dissertation the specific rank of

D. ovalifolium will be maintained.

The genus has an even distribution across latitudes

with 160 species in equatorial regions and 110 species in

temperate regions according to Index Kewensis (Imrie et al.,

1983). The center of origin of D. heterocarpon is reported

as Asia and the Pacific Islands (Kretschmer et al., 1979)

although it is now found in Africa, Australia, the Pacific

Islands, and Asia (Imrie et al., 1983). Recent collections

of D. ovalifolium have been made at its presumed center of

origin in Thailand and Malaysia (Schultze-Kraft and

Pattanavibul, 1985).

Desmodium heterocarpon originated in tropical

latitudes although high altitudes within these latitudes may

expose plants to cool weather and light frosts. Desmodium

heterocarpon has survived frosts and air temperatures down

to -6.60C (Kretschmer et al., 1979). Similar climate in

Florida and Australia at equivalent latitudes suggests that

tropical legume data from Australia should be applicable to

Florida (Kretschmer, 1970). Based on this, research was

started on D. sandwicense and D. heterocarpon at Fort

Pierce, Florida.










Kretschmer (1970) found D. intortum to be higher

yielding in association with Pangola grass (13,750 kg/ha)

than D. heterocarpon which produced 9,990 kg/ha. Desmodium

heterocarpon was compared again with D. intortum 'Greenleaf'

in association with three grasses and did not produce as

well as D. intortum (Kretschmer et al., 1973; Kretschmer et

al., 1976). Desmodium hetercarpon grown with 'Pangola'

digitgrass (Digitaria decumbens Stent.) or 'Pensacola'

bahiagrass (Paspalum notatum Flugge.) had total annual dry

matter yields of 9680 abd 9020 kg ha-1, respectively

(Kretschmer et al., 1979). A comparison of seven tropical

legumes with mixtures of six tropical grasses demonstrated

that D. heterocarpon outyielded many of the other legumes in

the summer, but when seed formation began in the fall, yield

often dropped below that of the other legumes tested

(Kretshmer et al., 1982).

Desmodium ovalifolium is reported to have drought

tolerance, shade tolerance, adaption to acid soils, and

persistence under grazing (Imrie et al., 1983). It has

produced over 7,000 kg/ha in eight months in Colombia (Imrie

et al., 1983). Contributing to high yield and persistence

under grazing is plant morphology. Bryan (1966) reported

that the large, axillary buds in D. uncinatum and D.

intortum provided rapid regrowth when the stem tip and

leaves were removed under light grazing but that removal of

much of the stem as happens under heavy grazing greatly

retards regrowth of both species. Therefore difference in










stocking intensity may influence legume persistence.

Kretschmer (1979) reported that D. heterocarpon persisted in

fields planted with 'Coastcross-1' bermudagrass (Cynodon

dactylon (L.) Pers.) for three reasons: the cattle preferred

'Coastcross-1' to D. heterocarpon, the growth form of D.

heterocarpon was prostrate under grazing, and the crowns

from which the plant regrew were situated low to the ground.

Crude protein in Desmodium heterocarpon was

significantly lower than in D. intortum or 'Siratro'

[Macroptilium atropurpureum (DC) URB] when separately grown

with each of three torpical grasses (Kretschmer et al.,

1973). Kretschmer et al. (1976) looked at seven legumes in

association with five grasses and determined that D.

heterocarpon produced the lowest amount of crude protein and

had the lowest in vitro organic matter digestion (IVOMD) of

the seven legumes tested, although all combinations were

above the seven percent crude protein suggested by Minson

adn Milford (1967) as necessary to sustain normal intake of

tropcial grasses. In the same experiment feeding pellets of

D. heterocarpon to cattle resulted in low digestibility,

which could possibly have been an artifact of the handling

and processing. The IVOMD ranged from 46.7 to 48.2 percent.

Toxicity is another aspect of Desmodium forage

quality that has received attention. Reports on Desmodium

species indicate toxicity levels are low. Zoebischh et al.

(1952) fed 25 species of Desmodium (species unspecified) to

chicks and reported that none of the 25 species reduced










chick growth, although some intoxication was evident with

eight of the species. Bryan (1966) reported low alkaloid

determinations on D. uncinatum, small amounts of oestrogens

in D. uncinatum and D. intortum, and no reports of adverse

effects form feeding Desmodium in Australia. Tannin levels

in D. uncinatum and D. intortum border on levels that

according to Rotar are high enough to interfere with rumen

cellulase activity but according to Hutton and Coote are not

high enough to interfere with rumen cellulase activity

(Bryan, 1969). Recent work with D. heterocarpon and D.

ovalifolium indicated that tannin levels varied by 100% in

42 lines of D. heterocarpon tested, and by 50% in 20 lines

of D. ovalifolium lines tested (Albrecht and Quesenberry,

1987).

Researchers have also looked at factors that may

reduce yield and stand persistence such as pest and disease

susceptibility. Some causes of lack of persistence in

Desmodium in general are soil moisture stress during the

growing season, insect attack, poor seedling vigor, frost,

and competition from grasses (Imrie et al., 1983).

Web-worms and nematodes are the major pest problem

of D. heterocarpon in Florida. Desmodium heterocarpon

suffers from web-worm attacks on leaves and flowers,

especially if large amounts of leaf material accumulate

(Kretschmer et al., 1976).

Root-knot nematodes, Meloidogyne sp., are prevalent

in Florida, particularly in areas that have been used for











vegetable production before conversion to pastures.

Resistance to root-knot nematode varies among and within

Desmodium species. Desmodium heterocarpon showed mixed

resistance to root-knot nematodes in Florida (Kretschmer et

al., 1980). This test of seven accessions of D.

heterocarpon showed that some lines were tolerant to

root-knot nematode whereas 'Florida' is high susceptible to

root-knot nematode. The nematode tolerant D. heterocarpon

accessions yielded less than 'Florida' when all were grown

without nematodes, but their yield was equal to or higher

than 'Florida' when grown in the presence of root-knot

nematodes. Desmodium intortum in this test showed resistant

to root-knot neamtode. Baltensperger et al. (1983) reported

that one line of D. uncinatum was highly resistance to M.

incognita, M. arenaria, and M. javanica. In the same

experiment, four D. heterocarpon lines [including 'Florida'

and a line identified by Kretschmer et al. (1980) as

susceptible] were high susceptible to all three root-knot

nematodes, D. ovalifolium (U.F. 28) was slightly resistant

to M. incognita, and D. ovalifolium IRFL 1699 had some

resistance to all three lines. Desmodium uncinatum

'Silverleaf' was susceptible to M. javanica and M. hapla (no

authority given), highly susceptible to M. arenaria, and

resistant to M. incognita incognita and M. incognita acrita

(no authority given) (Minton et al., 1967). Tests in

Colombia indicated that D. heterocarpon and D. ovalifolium

were susceptible to Meloidogyne javanica (Treub) Chitwood










(Lenne, 1981). Recent testing of D. heterocarpon and D.

ovalifolium response to Meloidogyne arenaria race 1 (Neal)

Chitwood, Meloidogyne incognita race 3 (Kofoid and White)

Chitwood and M. javanica showed that both types varied among

lines for number of galls and number of egg masses

(Queseberry and Dunn, 1987). In this experiment, the

highest level of resistance in D. heterocarpon and D.

ovalifolium to all three root-knot species was identified in

D. ovalifolium.

Successful nitrogen fixation by legumes depends on

an association with the appropriate Rhizobium spp.

Introduction of D. heterocarpon into Florida must proceed

with the knowledge that a compatible strain of Rhizobium

spp. already exists in the soil or can be added with the D.

heterocarpon. The situation regarding Rhizobium spp.

nodulation of Desmodium in general is somewhat ambiguous.

Both D. uncinatum and D. intortum are reported to have

rhizobium requirements that are usually met by native soil

populations (Imrie et al., 1983). However in some cases

results were poor when D. intortum was planted without

inoculation. In other cases inoculation improved nodulation

but not yield. Additional work has shown that inoculation

with a specific rhizobium will generate only 5% of the total

nodules if a compatible native strain is present, but will

generate 100% of the nodules if no native compatible strain

is present. Kretschmer (1970) suggested that Florida soils

generally have sufficient native and introduced legumes for










which, except for virgin soils, inoculation should not be

necessary. Where necessary, a cowpeaa" type should be used

at twice the recommended rate.

Desmodium heterocarpon is slow to establish even

under ideal conditions (Kretschmer et al., 1979; Kretschmer

et al., 1982). Poor establishment of D. heterocarpon was

also encountered in experiments with legume-bahia grass

pastures at Ona, Florida (Pitman, 1983). Although the

plants produced a good seed set and survived well under

grazing, the initial difficulty with establishment has given

D. heterocarpon mixed acceptance among growers in Florida

(Pitman, 1986).

Traits of interest in a genetic improvement program

in D. heterocarpon are flowering time, chromosome number,

reproductive compatibility between types and species, and

inheritance of important traits. The flowering time for D.

heterocarpon appears to be determined by photoperiod.

Desmodium heterocarpon is triggered to flower by shortening

days. The chromosome number of all Desmodium species is

2n=22 (Rotar and Urata, 1967). Schifino (1983) investigated

two morphological forms of D. uncinatum that had the same 2n

number. The presence of similar numbers of chromosomes in

species and lines suggests the possibility of interspecific

crosses between Desmodium species. Chromosome pairing

failure has been reported in crosses between D. intortum and

D. sandwicense (Hacker, 1968). No evidence of

self-incompatibility has been reported (Imrie et al., 1983).










Most Desmodium spp. are thought to be primarily

self-pollinated and that, along with the minute size of the

flowers, makes artificial hybridization difficult.

Cytoplasmic male sterility has been reported in D.

sandwicense with restorer genes in D. intortum (McWhirter,

1969).

Flower color, leaf markings, and stem color are

controlled by a single gene pair whereas seed size,

internode length, raceme length, and nodulating ability are

thought to be polygenically controlled in D. intortum and D.

uncinatum (Rotar and Chow, 1971).

Interspecific crosses are often made to introduce

desirable genes from one species into another. In

Desmodium, crossing occurred naturally between D. intortum

and D. sandwicense. Desmodium intortum 'Greenleaf' is a

combination of three lines, one of which is now known to

have contained some genes introgressed from D. sandwicense.

The first indication of the presence of these genes was a

shift in date of flowering and yield in the population as

these introgressed genes were selected against (Imrie and

Blogg, 1982).

Desmodium heterocarpon offers promise for growers in

Florida, more so if problems involving establishment and

nematode susceptibility can be worked out. The objectives

of this research were to characterize D. heterocarpon and D.

ovalifolium as the basis for initiating a breeding program

to combine desired traits of both species. Specific





12




objectives were: (1)to determine range in morphological

variation by quantifying leaf size, plant size, flowering

date, stem color, and presence of marked leaves; (2)to

improve the ability to make crosses by determining

percentage outcrossing, pollen viability, and chromosome

behavior during meiosis; and (3)to develop a method for

verifying hybrids by using isozyme markers.















CHAPTER II
MORPHOLOGY OF DESMODIUM HETEROCARPON
AND DESMODIUM OVALIFOLIUM


The International Board for Plant Genetic Resources

(IBPGR) has published a list of descriptors for forage

legumes (Andersen and Davies 1984). These descriptors are

divided into four categories for accession data, collection

data, characterization data, and preliminary evaluation

data. These descriptors will, in part, alert researchers

involved with germplasm to the range of information needed

about each accession, as well as standardize the responses

so that results can be universally understood. Some

examples are given for white clover (Trifolium repens L.),

red clover (Trifolium pratense L.), alsike clover (Trifolium

hybridum L.), lucerne (Medicago sativa L.), and birdsfoot-

trefoil (Lotus corniculatus L.). Although these descriptors

were not written with Desmodium in mind, the

characterization descriptors indicate aspects of forage

legumes that are thought to be important. These include

sowing and planting date; evaluation planting design used;

length, width, and shape of leaf; presence of marked leaves;

growth habit; and date and color of flowers.

Current collections of Desmodium include over 11,000

samples in Australia and over 12,000 in Colombia, followed

by 830 in Brazil and 230 in the U.S.A. (Davies, 1984).











Desmodium uncinatum (Jacq.) DC., D. intortum (Mill.) Urb.,

and D. ovalifolium Wall. are among legumes listed as first

priority for further collection and evaluation in tropical

and subtropical areas of South America, Mexico and Central

America, Asia, and Australia (Davies, 1984). Attempts to

collect D. ovalifolium and D. heterocarpon (L) DC. germplasm

from Southeast Asia have resulted from cooperative efforts

by several research organizations including the Thailand

Institute of Scientific and Technological Research,

Livestock Department (Bangkok), the Centro Internacional de

Agriculture Tropical (CIAT), the Malaysian Agricultural

Research and Development Institute, and IBPGR. Most

recently, Schultze-Kraft and Pattanavibul (1985) collected

six samples of D. heterocarpon and 12 samples of D.

ovalifolium in eastern Thailand (Schultze-Kraft and

Pattanavibul, 1985). Desmodium germplasm was evaluated at

CIAT by Sobrinho (1982). He evaluated 18 accessions of D.

ovalifolium and one accession of D. heterocarpon and found

highly significant differences among accessions for all

parameters analyzed including dry matter yield and crude

protein. Sobrinho also reported in this study that the one

D. heterocarpon accession appeared morphologically and

biochemically different from the 18 D. ovalifolium

accessions studied. Schultze-Kraft (personal communication)

evaluated 83 accessions of D. ovalifolium for characters

such as establishment rate, date of flowering, dry matter

yields, plant height and lateral growth, and chemical








analyses. Schultze-Kraft noted continuous morphological

variation between D. ovalifolium and D. heterocarpon but

attributed this to natural hybrids between the two species.

Hybridization between Desmodium species has been

reported by several researchers (Chow and Crowder, 1972;

Chow and Crowder, 1973; Hutton and Gray, 1967; Park and

Rotar, 1968; Rotar and Chow, 1971). The marked leaf in D.

heterocarpon is a possible morphological marker for crosses

with plants with unmarked leaves. Silver leaf-mark of D.

uncinatum was dominant in crosses with D. intortum and D.

sandwicense (Chow and Crowder, 1973). Brown fleck leaf-mark

of D. intortum was dominant in crosses with D. sandwicense

E. Mey. and D. uncinatum (Chow and Crowder, 1973). Hutton

and Gray (1967) reported similar 3:1 segregation ratios for

silver and brown fleck leaf-marks in F2 generations

originating from D. uncinatum and D. intortum crosses. Park

and Rotar (1968) divided flower color in D. sandwicense into

a white group and a purple group. F2 segregation ratios

were 3:1 for purple over white.

The objectives of this research were threefold: (1)to

characterize the D. heterocarpon and D. ovalifolium

germplasm collection at the University of Florida,

Gainesville, for the morphological characters leaf width,

leaf length, leaf length/width ratio, plant height, plant

width, presence of marked leaves, and for reproductive

biology; (2)to determine segregation ratios for marked leaf

and flower color in the F2 population; and (3)to determine

date of flowering in an F2 population.












Materials and Methods



Three field experiments were conducted to

morphologically describe fifty-seven lines of D.

heterocarpon and six lines of D. ovalifolium. Seeds of D.

heterocarpon and D. ovalifolium lines were obtained from Dr.

A. E. Kretschmer, Jr., Agricultural Research Center, Ft.

Pierce, Florida; from the USDA Plant Introduction Service,

Experiment, Georgia; and from CIAT, Colombia. All line

identification numbers will be IRFL numbers unless specified

otherwise. Table 2.1. cross-references U.F., IRFL, and CIAT

identification numbers of material evaluated and location

where evaluated. F1 seed of crosses 191 (D. ovalifolium

U.F. 28 x D. heterocarpon 6105) and 18/185 (D. heterocarpon

purple flowered 6106 x D. heterocarpon white flowered 6150)

and F2 seed of 191 and 18/185 were produced by Dr. K.

Quesenberry, University of Florida, Gainesville, Florida.

Seeds were scarified and then germinated in petri dishes.

Individual plants of some lines were isolated in 0.5 x 0.5 x

0.5 m wood frame cages covered with window screening.

Experiment 1. Plants from seed germinated on May 5,

1985, in Gainesville, Florida, were moved to the greenhouse

one week later. Five seedlings per row of each accession

were planted in the field on June 6, 1985, in a randomized

complete block design of two replications. The plants were

spaced 0.5 meters in the row and 1.0 meters between rows.











Table 2.1. Desmodium heterocarpon and D. ovalifolium
introduction numbers and location where
evaluated.


Introduction Numbersa


UF IRFL CIAT Speciesb Locationb


20
137
143
144
145
146
28










136
449

444
445

447
-


588,6102
854,6104
1695,6105
1696,6106
1697,6117
1699
1946
3279
3280
3321
3322
3324
3325
3661
3663
3664
3776
6103
6107
6108
6109
6110
6111
6112
6113
6114
6115
6118
6119
6121
6122
6123
6124
6125
6135
6141
6143
6144
6146
6147
6148
6149


365
3653
3669
3670
3671
3673
3674


3781
3784
3787
3788
3755
3785
3793

3116
3672
3675
3678
3680
3687
3688
3700
3735
3749
3786
3810
3843
3984
13142
13146
13148
13162
13170
13173
13175
13177
13179
13180
13182


H
H/h/h
H
H/h/str
H/h/h
O
O
H
H
0
O
H
O
H/h/str
H
O
H
H
H/h/h
H/h/h
H/h/str
H
H
H
H/h/h
H
H/h/h
H
H
H
H
H/a
H
H/a
H/h/h
H
H/a
H/h/str
H
H/h/str
H/h/str
H/h/str


C,F
F
F
C,F
F
F
C,F
C
C
C
C
C
C
C
C
C
C
F
F
F
C,F
F
F
F
F
F
F
F
C,F
C,F
C,F
C,F
C,F
C,F
C,F
C,F
C,F
C,F
C,F
C,F
C,F
C









Table 2.1. Continued.


Introduction Numbers


UF IRFL CIAT Species Location


- 6150 13183 H/h/h C,F
- 6151 13184 H/h/h C,F
- 6152 13185 H/h/h C,F
6153 13186 H/h/str C,F
- 6154 13187 H/h/str C
- 6155 13189 H/h/str C,F
- 6158 13119 H C,F


a Plant introduction numbers are numbers assigned by
Dr. K. H. Quesenberry, Agronomy Department, University
of Florida (UF), by Dr. A. E. Kretschmer, Jr., AREC, Ft.
Pierce, Florida (IRFL), and by the Tropical Pastures
Program, Columbia (CIAT).


b Species:


H = species heterocarpon
O = species ovalifolium
H/a = species heterocarpon subspecies
angustifolium
H/h/h = species heterocarpon subspecies
heterocarpon variety heterocarpon
H/h/str = species heterocarpon subspecies
heterocarpon variety strigosum


c Location evaluated: C = Cameroon, F = Florida.










The soil type was a Wachula sand (sandy, siliceous,

hyperthermic Ultic Haplaquad). Notes were taken on plant

height, plant width, leaf length, leaf width, flowering

date, and presence of marked leaf on August 15, 1985. Leaf

length to leaf width ratio was calculated for each leaf

measured. Flowering was noted on dates August 15, September

4, September 12, September 19, October 3, October 23.

Individual plants of line 6155 D. heterocarpon, line CIAT

13110 D. ovalifolium, and line CIAT 3670 D. heterocarpon

were caged from October 22, 1986, until November 5, 1986.

Fertilization was evaluated in terms of number of pods

formed, and presence of seeds in the pods.

Experiment II. Seed was germinated in Dschang,

Cameroon on Dec. 3, 1986, and transplanted to peat pots one

week later. Seedlings in peat pots were moved outside each

morning and returned to the lab each evening. Five

seedlings of each accession were planted per row in the

field on Jan. 20, 1986 in a completely randomized design

with two replications. The plants were spaced 0.5 meter

apart in the row and 1.0 meter between rows. Notes were

taken on plant height, plant width, leaf length, leaf width,

and marked leaf, on May 21. Flowering date was recorded

noted on May 21, July 21, October 23, 1986, and January 10,

1987. Individual plants of lines 3325 D. ovalifolium and

1699 D. ovalifolium were caged from July 5, 1986 until

August 4, 1986. Successful pollination was evaluated in

terms of number of pods formed, number of segments per pod,










and presence of seed.

Experiment III: Seed was germinated on May 15, 1987,

in Gainesville, Florida. Seedlings were moved to the

greenhouse, then to the field on June 20, 1987. Forty-nine

F2 plants of hybrid 191, 45 F2 plants of hybrid 18/185,

7 ramets of Fl hybrid 191, 9 plants of IRFL 6105, and 9

plants of 6102 were planted 0.5 m apart in the row and 1.0 m

between rows. Data on segregation for marked leaf among 191

F2's were taken on September 3, 1987. Data on flowering

date of F2191 were taken on September 13 and 19, October

3, 15, 20, and 28, and November 3 and 17, 1987. Data on

flower color of 18/185 F1 were taken as plants came into

flower, either in the field, or later, in the greenhouse.

Three plants of 6105 D. heterocarpon and three plants of

6102 D. heterocarpon were caged on September 1, 1987. Seed

weight produced per plant was used to evaluate mode of

pollination.

In addition to caging data, notes were taken on

greenhouse seed set for lines listed in Table 2.2.

Accessions in flower in the greenhouse between January 18,

1988, and September 20, 1988, were periodically observed for

percentage successful pod formation. Pod formation was

noted four days after flower opening. Lines 6150, U.F. 28,

U.F. 160, and hybrids 195, 191 which did not self-trip early

in the morning, were tripped and self-pollinated. Lines

6150. U.F. 28, U.F. 160, and hybrids 195, 191, 83-4-3,

83-4-2 were tripped and cross-pollinated with another plant




























-l






4.)
U)
















U4
0



-4









1-1










O
C























r
90
00






























0 0
C4J'


-4




o o










-i

00




(N


-1Q
G(0


(U)
-4



-4
0
c0







0
a


03



















(U

-)



4 4-
-4 -4














()

CU
-4


(0

E
enC


U)
Uo
V




0






U]
0


o





a0

04
U)










o




0

0
r-4








4-)
Q)
U]
V
















0
0U
a




















en
0











U)

0
I-4
Vl










U]

(H
U)
0,
-4


r- o o0 0
o (N













o0 LA --i (N




CO r-q
(N

(N Ln 0 *
S 00 A .
H H M *
~lD D


0 V
CN





o0 r-i






in
o LD O
tr in .-


N tf
'H(N


N 'H
c (N


-1 0

(N 00


o o0 "r
o






o '- H-








U]
00 00 -

I I
>4I I LA

'H
*riA C


co r Mr















Ln *'T -
rNi us


-I n 0
'-H rN yo


(NN







0 0
1 en '


I
I
03


I







0
I
I
I
I
I

I

I










of the same line. Successful pod formation was noted four

days later.



Results



Plant height, plant width, leaf length, leaf width,

leaf length/width ratio, flower date, and presence of marked

leaf for D. heterocarpon and D. ovalifolium grown in

Gainesville in 1985 are listed in Table 2.3. Although plant

height ranged from 9 cm to 39 cm when these measurements

were taken, some lines eventually grew to mean heights of 1

meter before frost. Plant width ranged from 13 cm to 84

cm. Leaf length varied from 2.8 to 6.5 cm. Leaf width

ranged from 2.7 to 6.7 cm. Three lines (6123, 6124, 6125)

had leaf length/width ratios of higher than 3.0 cm. Two

lines (6123, 6125) were D. heterocarpon ssp. angustifolium.

The third line was specified as D. heterocarpon without

subspecies or variety ranking. A third line (6143)

identified by CIAT as D. heterocarpon ssp. angustifolium had

a leaf length/width ratio of only 1.4. Leaf length/width

averages ranged from 1.0 to 3.7.

The majority of D. heterocarpon lines screened had

marked leaves. Most D. ovalifolium and D. heterocarpon ssp.

angustifolium had unmarked leaves. One line of D.

heterocarpon spp. heterocarpon var. heterocarpon (6108) and

one line of D. heterocarpon species and variety unspecified

(6158) had unmarked leaves.


























O O- 0-S ac a


I
r-
-S


4-I



-S ~. ^.
s


-S


--S-

--S.~ -


rn M e n
ON N N OCAMn C4CNC7a m rnN(N( (NC
q4 .-4 r-4 V '4 V'r 'r W v- -.-- i -4 -1-lV 'T --- 'V .
00\a CD0 P 0-0
I .**^*-. a m e m H >*>^.-.*>. .'*.* m a o o **>>.o


0



.-I

o-4
(





-4 i


0 r.





E


r- .4-
o





4-1 -4
-.4
E








04
4 -







U)


00


-41




U a
4-1
* (















.0



C) o
o
















enl


S-4
U W



















0
4
uc+
c~l
e (
3eIQ


Ol rmo m NOo r
* .** *


I

I
I 0r CN ^










000

D0D 0 D 0% 0 1?0
zMzAIH4M
000


in or-
000
-4 r--
%D k.0 V0


- in a n mm 0 o

m *r m N m


,-i N in o r

m**y*^** *


v-4 kD W M M -1 r- onoMoi- o Mn in 0o r



-CN CN -N lA -4 Nn"fmmmmn-mCN


ww omm~~~mewmw~


0 C r-4 --4I-4 -4 V-4
r.D-4 -0'S
\00O00 k 0o


o00mNoN0 (NmpoNM(NfO omiOMN

S. .* o *v % n tn m o m* n *v *o %o v. v *v m m


U
0)








30(
0>1




E-4


>-I



? 0)


S i-l


S-4



3



410)







4
C(
m-4





Ca

40





0

-r





















4J 4- 4J 4-J 4J 44



rnmra tmmmmmm
rn m ) m mj j m n to mW







mmmmm(wm0mm X t





4NNNNNNN"NNNN NN.









CQ r N ON o f


0)





>




041
0
'o








*t
Q)o
0 4-1




















a) i4
*Cu
4)-14J
f0 c0
(U *^


(N r- r-

4,


L OW0-I r-"N o0 0 O -4 %D r N 0n O
S O** ** ** 9
wwwramwmwmwwwmwma


00


Un
*
inimewmn boo.(Ni d o


(N 0 C 0 -


0)
-4


Cu


E








'-4


0

'4
<4







(Di
*1-
(


SI
44











$-4


*4
c


0)


Cu

Cu





* 2-
(01






0)
E
ll ll


-4
>




0 r.

w 04
> 0



0
l O)



4 0







4-J
LO
0) 1-


04 0a
u a

0
U *








0 U
a 0

u uQ
0 U
Cu 1-1



S fl

I0 O










5u o.S
*.- o'& -
m- u <


E

-r










0
4
O






0




0)
a




C





0
a









0



UQ
Vu
u Na

II 1
Oa


0


u

0

4-4
c)

0 U

(4 (0

0 U
O-

CU)
Cu
0) II





-a
0)

Se-4

3 U


In rr- r- rLn -

m"" CN'' m"^ m qw In W


mMmnrCr- MNN~N~
" 14-4MOfM OfN -lM -










Flowering started with line U.F. 145 on August 15.

Neither D. ovalifolium line (U.F.28 and U.F.146) flowered

before November 4, 1985. Other D. ovalifolium lines

observed in other years have shown flowering dates of Nov. 1

to Nov. 30. Ten lines flowered between mid-September and

mid-October. Twenty lines started flowering between

mid-October and mid-November.

Results of pod set as influenced by caging are

given in Table 2.4. Three plants each of D. heterocarpon

ssp. heterocarpon var. strigosum 6106 and 6155 were enclosed

in individual screened cages to exclude insect pollinators

while five plants of each line were left uncaged. Pod set

for 6155 was high for both caged and uncaged plants. Pod

set for caged (3,646) and uncaged (4,189) was not different

(P>0.05). Pod set for 6106 for caged (28) and uncaged

(1,142) was different (P<0.05). Three plants of six D.

ovalifolium (CIAT 13110) enclosed in individual screen cages

produced a mean of 34 legumes. Three uncaged plants

produced a mean of 1705 legumes per plant that was different

at P=0.10.

Plant height, plant width, leaf length, leaf width,

presence of marked leaves, leaf length/width average, and

flower date for D. ovalifolium and D. heterocarpon grown in

Cameroon in 1986 are shown in Table 2.5. Plant height

varied from 8 to 125 cm. Plant width varied from 52 to 240

cm. Leaf length ranged from 2.6 to 6.4 cm. Leaf width

varied from 1.2 to 3.4 cm. Leaf length/width average varied






26



Table 2.4. Desmodium heterocarpon and D. ovalifolium caged
and uncaged pod set in Gainesville in 1986.


Line
Number


Plant Caged


No. of
Pods


Seed
present


IRFL 6155 1 yes 1,056 yes
D. heterocarpon 2 yes 912 yes
3 yes 1,678 yes
4 no 700 yes
5 no 1,443 yes
6 no 279 yes
7 no 2,204 yes
8 no 293 yes


CIAT 13110 1 yes 46 yes
D. ovalifolium 2 yes 22 yes
3 yes 33 yes
4 no 117 yes
5 no 2,000+ yes
6 no 3,001 yes


IRFL 6106 1 yes 8 yes
D. heterocarpon 2 yes 2 no
3 yes 18 yes
4 no 23 yes
5 no 59 yes
6 no 16 yes
7 no 751 yes
8 no 293 yes


a mean number of legumes between caged and uncaged was
not different (P>0.05) for 6155, was different
(P=0.10) for 13110, and was different (P<0.05) for
6106.







27












04- 0
-W

oC

oos00 caa== coo=o=c0o0a0


I
Ilgn


M

CN

r-4
I '-41I


'-4
(N
- S


E









0


O
,-U
E
c
'U




Efl













0 0
r4







-4
S*4
00










QC)
U '-4


a C

SU






0 0
(a
V
-i4E















0 r

E a)








0 41
ErU
04
> G



SO
*
























1-
'C




',U
(0 -

U 4-
0 (0
-
(U 1
-p








*,
in
*)


(U


C.,

rl--SoV

Ln r- iI


U


M 4 "-4 CN .-N N -1 CN N r-4 CN .-4 r-4 -1-4 -4
SW 0 NM OmVv N lNC va











-I *wrc***-4r-en c- *n* *r-onuv*oo o-




I --4






l: rZtUfCDCOOOO N C0(NrM * S r-4 -4 -q (N CNn M M M W r-4 -4.-- r4 -4
IOmo^omrmmfmmmonommO~oW


U
a)








'(Q




04-'





'E
(U(
a; 4-

(0 (U


1



4-1


4
a) 1-C







C 4






40

C- 3s


r-4
CN

I n I

























N
4j
4-)
10


a a aa
-5~C23~


'~-4


ww-t
4) 4- 4
CD CD ,
S-li. +J
ui5 5n u 5.5 x
x~xx: ; x
aaa a a


m o


ON
IBM I-ir


a 0
-4 r qr-4

0 (N
VI V -4


N -i o -. oo. r, -- o n


4U 4J- 4J>
w in to
NNN









'-4
(-4
5-'c


NNNNCNNNNNNNNCN -


0

(Me i


a o m oo Ln


I .*. 9 9




U

I o r- r-n r-o rn-L anr- Ln 0mom
-4W MNr Ir-' 04 4 M-n4Mmo






v w v-4 ir o in r. r- N %D m r~- oLn ,o -W
i 14 -H 4


Ln n -4 m- 0 r-
N (Vl v 3' -t r


Oo0-INmw'L
rtn mn )o Ln u Ln LI


3NNO-NN(NNN NNN


(0
E





cJ



























M
Ci
44

k-4








z0
u











r -4




l-4
a
Cu




a 0







4CPN


()

0

E0


.a

44 -W
(U-1
0) .9


C C




4-1

c O








r-4
Sa
S3


>1
4-1
0)

-4
(c



0

E
I U


0 0



ON
44 ()
-4 4.)
4-1 W

01





u Q
CU



ul
OU O-









*0 U
(n




04 0
D &

Sm




e 0

01
E4.



0> u



(D u
-4 110)

0- 4-
0)0




4) II U

II I .C
.5-
II (U c
05 K 3S


>4
0)
-4

CU


0
a






0








.c



C
0






CL


1,4
0









*U
C4-

WX:
0

0)



4<
U





01
cD,


&5
.5-


-4
0)
04
C
Zn


.-4




.r4
r-4
0 II




.0
oN
e c










from 0.8 to 4.6. Desmodium heterocarpon ssp. heterocaron

var. angustifolium lines 6123 and 6125 had leaf length/width

averages over 4.0 but 6143 had a leaf length/width average

of less than 1.5. All three D. heterocarpon ssp.

heterocarpon var. angustifolium accessions had unmarked

leaves. Four D. ovalifolium (U. F. 28, 3321, 3322, and

3325) had unmarked leaves but two other lines (U. F. 146 and

3664) had some marked leaves. Sixteen of 37 lines flowered

during a year of observation in Cameroon. First flowering

occurred on July 21 (D. ovalifolium lines U. F. 146 and

3325) and last flower date was D. heterocarpon 6146 blooming

on December 10, 1986.

Three of six plants of D. ovalifolium 3325 and

U. F. 146 were enclosed in individual screened cages (Table

2.6.), while three of each line were left uncaged. Caged

plants of both lines produced no pods, except for plant # 1

of line 3325. This one pod consisted of one segment that

did not contain a seed. The remaining four 3325 and three

146 plants were uncaged, and set pods with seeds, except for

the seventh plant of line 3325 and one of the uncaged U. F.

146 plants which were not flowering.

Marked leaf and flower color F2 segregation data

are given in Table 2.7. The observed segregation ratio for

presence or absence of marked leaves was approximately 9:7

in the F2 generation of 191 (U. F. 28 x 6105). The F1

had marked leaves. The observed segregation ratio for

flower color was 3:1 for presence of color versus absence of







































SI I >I >i>i i







O0 00 0








021 > > > >
WWWWO)WO






w w 0 0 0 0














E

LC r-4
NO
m -4
S-4


I- 0

I-rI


00000
01 cn
0)Q00






III I >1 >1
ooo o r-







02 0 2 0 0 0
> C >> >z



0203202
WW(U(U00 0


c
-4


Un

,a
0


a

CP
O






04


Oc



S0
U












. 0

0-4
3 P
4r-.









0 fa
0 u
00



E-
m U
-410



0 m



NI0
r-4


E
*-l
,-I
0
I-l
-4-4
(0
>
r 0

=3 Q|



















Ln
to
,-4





0
-4
0
U


a)

o
0















CN





i -4
0

















-4
--4




w r












0
S-4
o >





0
a)




U-,4
OC









S-4
*m




a 0
>C)






N






.a
-4


a0

O-

04






x



ro
a)

4- 0


(0
O i


*4


3.
0
4-1




0
z





4-
-4




z

4J
Z
















."4.

E-4


(U
.-4







.Q
>1
33


0I m














CN O-











mN
N CN


r-4
o 0






Ln
0



-4


Q1
Lin




x








0
co













a)


S(




E

-4




a



0
>4-



0)


Q-I

a,

e -



-1-4
0



-45-
a)1
-44(0
01 W


LA
-4


x
o









-4
4o
X











a4
3



O





0

to
u












0
a




a



Oa)
c











o 4
o







a,





0








S(U
oat

0 o


14-1
01*r
cn x








r-l (
'*>O1-
00(U


,-I

E
4-4












color in the F2 generation of hybrid 18/185 (6106 purple

flowered x 6150 white flowered).

Frequency distributions for flower dates for 6105,

191 F1, and 191 F2 are shown in Table 2.8. Female

parent, U. F. 28, was not planted in 1986 in Gainesville.

This line, however, did not flower by day 307 in 1985 in

Gainesville. 6105 mean flower date for nine plants was day

267. Three ramets of 191 F1 had a mean flower date of 298

days. Flowering dates for 33 191 F2 plants had a mean of

293 days, excluding 7 plants that had not flowered by day

321 when plants were taken to the greenhouse to avoid frost.

Results are given for caging D. heterocarpon 6105 and

6102 in Gainesville in 1987 in Table 2.9. Three of eight

plants of 6105 and three of six plants of 6102 were enclosed

in individual screened cages. All plants of both lines,

both caged and uncaged, produced seed. Total seed

production varied from 0.71 gms for an uncaged plant of line

6102 to 26.32 gms of seed produced by a caged plant of line

6105.

Seed set on Desmodium in the greenhouse is presented

in Table 2.2. 6150, U.F. 28, U.F. 160, hybrid 83-4-3, and

hybrid 83-4-2 set no seed on flowers that were not tripped

manually. Percentage pod set on unmanipulated flowers of

other lines ranged from 0% to 25%. On five lines which were

self-pollinated manually, pod set ranged from 11% to 73%

whereas seven lines which were cross-pollinated had pod set

of 08% to 57%.






























-N






-4
ru

4-14







0o



c 0







ON
or,
0





Ori
-4 -







4-r
4 M


0 T3

4-. -4
S4


o 0



r. r-



a0
.-I

->


cu




-) r4


o
aa-
0 to
o r-4
(M o


a)
4-)
(0



.1.





c




m
-4

'n












-4
r-
.-I
,











.z1
--4


cu


ra









>1
ca
-4

















U)
*41
o















il
0





0



O
r (U






4


CN Oc




0
- -4


9

ON


0
C(



c 0




U
0



0 r

4-4


a)



N


U)









ro
0 Q)












4 -
0





0 >







o 0
OO U





>
c 0


(n
S3












Ua0
MMC
cO *U



i>



C4 On
>me





> 4-
a)- a)
c-iM Q


>4



i


r-
(N
CM4




r-
0




.-4
0


O


o





(N




I-
NC
(N




(N




o
to
c-
(N
















Table 2.9.


Line
Number


Desmodium heterocarpon caged and uncaged seed
production in Gainesville, Florida in 1987.


Plant
Number


Caged


Seed
Weight


---gm---
U. F. 6105 1 yes 2.78
D. heterocarpon 2 yes 26.32
3 yes 19.95
4 no 8.00
5 no 1.35
6 no 1.69
7 no 1.38
8 no 5.46


IRFL 6102
D. heterocarpon


yes
yes
yes
no
no
no


7.9
10.05
7.80
.71
3.89
1.05


a Mean seed weight between caged and uncaged was
not different (P>0.05) for 6105 and 6102.











Discussion



Interest in percentage outcrossing versus selling in

D. heterocarpon and D. ovalifolium was generated by observed

differences in pod set in the greenhouse. Desmodium

heterocarpon appeared to set profuse amounts of pods with no

manipulation, whereas D. ovalifolium-type lines set few pods

if not manipulated. Caging experiments were conducted in

the field to determine whether either species needed flower

tripping by insects or cross-pollination to set pods. Both

species appear to be photoperiod sensitive and triggered to

flower by shortening daylength. This photoperiodism

complicated caging experiments by restricting the number of

lines available to be caged. The variation in daylength in

Cameroon (latitude 30N) is so slight that during the time

available for caging studies during the summer of 1986, only

two D. ovalifolium lines flowered. These two lines (3325

and U. F. 146) gave uniform results. Caged plants produced

no pods while uncaged plants produced pods containing seed.

It is interesting to note that flowering response was not

uniform within these two lines. Each line had one plant

that did not flower. Presence of immature seed in pods was

noted as an indication of successful pollination. Pods were

harvested before seeds were mature to allow analysis of

results during the time allowed in Cameroon. An additional

line of D. ovalifolium was caged in Gainesville, Florida.









This line (CIAT 13110) gave results similar to caging

results from Cameroon, although caged plants did produce a

small number of pods containing seed. Two uncaged plants of

this line produced large numbers of pods ( >2,000) per

plant. A T test for homogeneity does not show caged versus

uncaged results to be different (P,0.05), but this is most

likely an artifact due to high variability within the

uncaged plants where one plant produced only 117 pods. Two

D. heterocarpon that most resemble the agronomic description

of D. hetercarpon ssp. heterocarpon var. heterocarpon (6105

and 6102) were caged in Gainesville. The results were

measured in seed weight, because the high number of pods set

precluded counting individual pods. Caged and uncaged

plants produced large numbers of seed. Large amounts of

seed were produced by the caged plants. This may be due to

increased humidity within the cages, or lower temperatures.

Rotar and Chow (1971) have reported that high humidity and

low temperatures increased success for controlled crosses

between D. intortum, D. sandwicense, and D. uncinatum. Two

lines of D. heterocarpon ssp. heterocarpon var. strigosum

were caged and gave mixed results. 6155 set high numbers of

pods whether plants were caged or not. 6106 gave

significantly different results with caged plants producing

very few seed. These plants were caged late in the year

after they had already been flowering for some time. All

flowers and seeds were removed from the plants at the start

of this experiment, but plants tended to resume vegetative









regrowth rather than flowering. The microclimate produced

by the cages may have affected flowering.

Results from caging experiments indicated

that D. ovalifolium required some manipulation of flowers

denied it under caged conditions. Either the plant required

insect cross-pollination, or it required insects to trip the

flowers. Desmodium heterocarpon types appeared to set seed

regardless of whether the plants were caged or not,

indicating that flower manipulation and cross-pollination

were not necessary for seed set. These conclusions are

drawn from a small sample, but indicate that a distinction

between D. ovalifolium and D. heterocarpon may be the need

for flower manipulation or outcrossing in D. ovalifolium and

not in D. heterocarpon.

Casual observations of desmodium in the greenhouse

indicated that 6102 and 6105, both D. heterocarpon, set

large amounts of pods without manipulation of the flower.

Furthermore, the flowers on these two lines self-tripped

early in the morning. This was in contrast to the D.

ovalifolium whose flowers remained untripped until late in

the day. Neither D. ovalifolium line produced seeds from

tagged flowers that were not tripped, nor did two hybrids

(83-4-3 and 83-4-2). The remaining three hybrids each

produced one pod, indicating that seed production from

untripped flowers was extremely low. In an attempt to

determine if tripping and self-pollinating would increase

pod set, three lines were tripped and selfed at 10:00 A.M.









each morning. Here the distinction between D. ovalifolium

and D. heterocarpon breaks down. Accession 6150, although

morphologically neither a typical D. heterocarpon or D.

ovalifolium type, is classified as D. heterocarpon ssp.

heterocarpon var. heterocarpon. It has white flowers that

do not self-trip, small non-pubescent leaves, and a

shrublike growth habit. It required tripping in the

greenhouse for pod set. When tripped, pod set increased

11%. Tripping and self pollinating the D. ovalifolium lines

increased pod set to by 60 to 73%. The same three lines

were tripped and cross pollinated but this did not improve

pod set. These results indicate that the D. ovalifolium

lines are not obligate outcrossers, but that flowers of

these lines do require manipulation for pods set.

Marked leaf has been used by many researchers as an

indication of successful cross pollination in crosses that

involve maternal unmarked leaf and paternal marked leaf

parents. Segregation data for the F2 generation of such a

cross (U. F. 28 unmarked leaf D. ovalifolium x 6150 marked

leaf D. heterocarpon) gave a 9:7 ratio for marked leaves.

The marked leaf parent when self-pollinated did not appear

to segregate for marked leaf indicating that the accession

was homozygous for the trait. In several other crosses made

by Dr. K. H. Quesenberry, anytime the F1 appeared to be

morphologically intermediate between a unmarked-leaf D.

ovalifolium parent and a marked-leaf D. heterocarpon parent,

the F1 has always had marked leaves.










Another morphological marker used to indicate

a successful cross is flower color. The data fit the 3:1

purple:white flower ratio expected for a single gene

dominant trait. Usefulness of this marker is restricted by

the limited number of white flowered lines in the U. F.

germplasm collection although the white flower trait could

be incorporated into more lines.

Several accessions of Desmodium, particularly D.

ovalifolium types, flower too late in the year to produce

seed before frost. Wide variability exists within the

germplasm for date of flower. Flowering seems to be in

response to shortening photoperiod. Table 2.8. shows the

spread of flower date for the segregating F2 population of

a cross made between lines with widely divergent flower

dates (U. F. 28 x 6105). U. F. 28 flowered so late in the

year that the Fl cross was made in the protected

environment of a greenhouse during the winter. IRFL 6105,

on the other hand, flowered early in the year (day 262) and

produced a heavy seed set before frost. The F1 of cross

U.F. 28 x 6105 flowered on day 298 which was approximately

the median between the two parents. The bloom dates for the

F2 population were from day 276 to later than day 307.

The mode for the F2 bloom date was the same as the bloom

date for the F1 plants, day 298. This variation in bloom

date indicates that change in bloom date should be

responsive to manipulation through controlled crosses and

selection.










Ohashi (1973) has proposed a taxonomic key for the

species D. heterocarpon that divided the species into two

subspecies, heterocarpon and angustifolium. His most

readily apparent distinction was the leaf length to width

ratio for subspecies angustifolium which was given as 3 -

6.5 as opposed to subspecies heterocarpon which he described

as having leaf length to width ratios of less than 3. Three

lines were classified as D. heterocarpon subspecies

angustifolium among the lines tested. 6125 and 6123 had

leaf length to width averages that were greater than 3 when

grown both in Cameroon and in Florida. 6143 had leaf length

to width averages (0.8 in Cameroon and 1.4 in Florida)

outside of the specifications for subspecies angustifolium.

Ohashi (1973) remarked that D. heterocarpon was one of the

most highly polymorphic species. This exception to leaf

length to width ratio was either an example of morphological

variation which in this case extended outside the range of

the subspecies or indicates this was an incorrectly

identified line.

No clear-cut differences existed between D. heter-

carpon and D. ovalifolium lines in plant height or plant

width, although D. ovalifolium lines tended to be grouped

among the shorter lines. Leaf mark also did not strictly

divide along lines of D. heterocarpon and D. ovalifolium.

Most D. ovalifolium lines had unmarked leaves, although 3664

had marked leaves. Most D. heterocarpon ssp. heterocarpon

var. heterocarpon accessions had marked leaves, although





41



some, as in line 6108, did not. Thus presence or absence of

marked leaves is not a steadfast indicator of whether a line

is D. ovalifolium or D. heterocarpon, if these lines are

correctly classified.















CHAPTER III
DESMODIUM HYBRID IDENTIFICATION
USING ISOZYME ELECTROPHORESIS


The genus Desmodium is pan-tropical with various

species found in sub-tropical and tropical areas of

Southeast Asia, Africa, and Central America (Imrie et al.,

1983). Interest in Desmodium heterocarpon (L) DC. as a

tropical forage legume for Florida results from the fact

that the species is able to fix nitrogen and that it is the

best adapted perennial pasture legume for summers in Florida

(Pitman and Kretschmer, 1984). Taxonomically D.

heterocarpon has been considered to include Desmodium

ovalifolium Wall. by some but not all researchers (Ohashi,

1973; Schultze-Kraft, personal communication). The cultivar

'Florida' carpon desmodium, Desmodium heterocarpon, was

selected from plant introduction evaluations and released in

1979 (Kretschmer et al., 1979). 'Florida' has shown several

desirable attributes including perenniality, good seed

production, and some tolerance to wet soils, but it has been

shown susceptible to root-knot nematode (Kretschmer et al.

1980). Initial screening for nematode resistance in lines

of Desmodium has shown that some lines of the D. ovalifolium

type have root-knot nematode resistance, a valuable asset in

Florida soils (Quesenberry and Dunn, 1987). Ideally,

breeding for nematode resistance in Desmodium could










combine the root-knot resistance of the D. ovalifolium types

with the agronomic characters of earlier flowering and cold

hardiness of the released cultivar Florida.

Most D. heterocarpon have marked leaves whereas most

D. ovalifolium types have unmarked leaves. The leaf-mark in

D. uncinatum and D. intortum have been shown to be

controlled by dominant genes (Chow and Crowder, 1973). Use

of this leaf mark to verify hybrids requires that crosses be

made between female D. ovalifolium parents and male D.

heterocarpon parents. At times it may be desirable to make

the reciprocal cross or to cross two parental lines both of

which have marked leaves. Thus identification of other

morphological or other type markers are needed. An

alternative method of identifying parents and hybrids would

be the use of isozyme markers to indicate successful hybrids

of D. heterocarpon x D. ovalifolium. Isozyme markers have

been used to identify hybrids in bananas (Musa spp.)

(Jarret, 1987), peach [Prunus persica L.(Batsch)] (Durham et

al, 1987), and papaya (Carica papaya L., Carica cauliflora

Jacq. and hybrids) (Moore and Litz, 1984). Isozyme markers

have also allowed researchers to distinguish between

morphologically similar lines within a germplasm collection

of bananas (Jarret, 1987), alfalfa (Medicago sativa L.)

(Quiros, 1980), Lolium perenne (L.) (Hayward and McAdam,

1977), and cassava (Manihot esculenta Crantz) (Ramirez et

al., 1987). Isozyme markers have been used to verify

hybrids of Desmodium intortum (Mill.) Urb. x Desmodium










uncinatum (Jacq.) DC., and Desmodium sandwicense E. May. x

D. intortum (Chow and Crowder, 1973), to detect the presence

of Desmodium sandwicense genes within D. intortum

'Greenleaf' (Imrie and Blogg, 1982) and, with Desmodium

ovalifolium, to distinguish lines of a germplasm collection

(Hussain et al, 1987). The primary objective of this

research was : (1) to identify an enzyme system which showed

isozyme variability between D. hetercarpon and D.

ovalifolium and, (2) to evaluate the use of such a system to

distinguish Fl hybrids from their parental types.



Materials and Methods



Experiment 1: Desmodium ovalifolium lines U. F. 28

and U. F. 160 and D. heterocarpon lines 6102 and 6105 were

screened for activity using 24 stains and 6 buffers (Table

3.1. and Table 3.2.). Leaf tissue came from mature,

greenhouse grown plants and was selected from the eight most

recent, fully expanded leaves on any branch. Leaf tissue

was prepared according to three methods. In the first

method, a leaf was placed on a glass plate, a 6 x 6 mm

section of Whatmann #1 filter paper (wick) placed over it,

and the head of a pestle rolled over the paper squeezing

cell sap from the leaf into the paper. In the second method

leaf tissue was ground in a mortar with a pestle in 1%

glutathion in 0.1 M Tris HC1 at pH 7. The third method was

identical to the second, except that 0.015 gm of polyvinyl










Table 3.1. Enzyme activity stains and buffer systems inves-
tigated for the detection of isozyme banding pat-
terns from leaf extracts of D. heterocarpon and
D. ovalifolium.


Enzyme system Reference Buffer system


Acid phosphatase (EC 3.1.3.2)
Alcohol dehydrogenase (EC 1.1.1.1)
Ascorbate oxidase(EC1.10.3.3)
Catalase (EC 1.11.1.6)
Diaphorase (EC 1.6.4.3)
Esterase (EC 3.1.1.2)
Formate dehydrogenase (EC 1.2.1.2)
beta-D-Galactosidase (EC 3.2.1.23)
Glucose-6-phosphate
dehydrogenase (EC 1.1.1.49)
Glutamate oxaloacetate
transaminase (EC 2.6.1.1)
Glutathione reductase (EC 1.6.4.2)
Hexokinase (EC 2.7.1.1)
Isocitrate
dehydrogenase (EC 1.1.1.42)
Leucine aminopeptidase (EC 3.4.11.1)
Malate dehydrogenase (EC 1.1.1.37)
Malic enzyme (EC 1.1.4.0)
Peroxidase (EC 1.11.1.7)

Phosphoglucomutase (EC 2.7.5.1)

6-Phosphogluconate
dehydrogenase (EC 1.1.1.44)
Phosphohexose isomerase (EC5.3.1.9)

Shikimate
dehydrogenase (EC 1.1.1.25)
Superoxide dismutase (EC 1.15.1.1)
Urease (EC 3.5.1.5)


TC
H 6.3a
TB,LBTCa,K
TB, LBTC, K
H
TB, LBTC, K
H
TB, LBTC, K
TB, LBTC


b TB, LBTC, K


TBa,K
TC
LBTC, TC


b H
b LBTC, H 6.3
b TC
b TB, LBTCa,
Ka, TCa
b H, LBTC,
H 6.3a
c H, LBTC,
H 6.3
b TBa,LBTC,
Ka, H 6.3a
c TB, LBTC, K


TBa, LBTC, K
TB, K


a Buffer system with visible results
b Durham (1986)
c Vallejos (1983)
d Wendel and Parks (1982)
e Shaw and Prasad (1970)










Table 3.2. Electrophoretic buffer systems used for analysis
of D. heterocapon and D. ovalifolium leaf and
root extracts.

Electrode Gel
Buffer system buffer bufferb


0.065 M histidine
0.02 M citric acid
pH 5.7 with
citric acid

0.065 M histidine
0.02 M citric acid
pH 6.3 with
citric acid

0.016 M LiOH
0.192 M boric acid
pH 7.2



0.038 M trizma base
0.002 M citric acid
pH 8.6

0.05 M trizma base
0.016 M citric acid
pH to 7.0 with
citric acid

0.18 M trizma base
0.1 M boric acid
0.004 M Na2EDTA
pH 8.6


0.009 M histidine
0.003M citric acid
pH 5.7 with
citric acid

0.009 M histidine
0.003 M citric acid
pH 6.3 with
citric acid

0.0016 M LiOH
0.019 M boric acid
0.007 M citric acid
0.046 M trizma base
pH 7.7

0.03 M boric acid
pH 8.5


0.017 M trizma base
0.005 M citric acid
pH to 7.0 with
citric acid

0.045 M trizma base
0.025 M boric acid
0.001 M Na2EDTA
pH 8.6


a H= Histidine citrate (pH 5.7), Histidine citrate (pH
6.3), LBTC=Lithium borate/tris citrate,
TB=Tris/borate, TC=Tris citrate (Durham 1986).

b 10 mg NADP was added to buffer in cathodal tank and gel
buffer of systems that included analysis for enzymes
that required NADP as a cofactor.


H 6.3


LBTC









polypyrrolidone (PVPP) was added.

Starch gels (11%) were prepared by heating 21 grams

Sigma and 10 grams of Connaught starch in 300 mis of the

appropriate buffer, degassed, and poured into 19 x 19 cm gel

molds to a thickness of 0.5 cm. NADP, when used as a

cofactor, was added during the heating process. Each gel

contained 12 wicks representing the 12 combinations of four

lines prepared each of 3 ways. As a result of this

screening, stains for peroxidase and phosphohexose isomerase

were chosen for more extensive trials.

Experiment 2: Leaf tissue from five parental lines

and five hybrids (Table 3.3.) were evaluated for response to

peroxidase and phosphohexose isomerase. Leaf tissue (2 cm x

1 cm) was taken from the most recent, fully expanded leaf.

Root tissue (5 cm long) was taken from approximately five

terminal root sections, although older roots were used when

new growth was not available. Five plants of each parental

line, and one of each F1 hybrid, were evaluated on each of

three different days on three different gels. Leaf tissue

was macerated in five drops of Tris HC1 (pH 7.8 0.1M). Root

tissue was macerated in five drops of Tris HC1 (pH 7.8 0.1M)

with the addition of 2mg PVPP. Crude extract was absorbed

onto 6 x 6 mm filter paper wicks (Whatmann #1) and inserted

into a slit in the gel.

Starch gels (8.3%) were prepared as above except that

only 25 grams Sigma starch and no Connaught starch was

used. Initially several concentrations of starch ranging






48







Table 3.3. Desmodium ovalifolium and D. heterocarpon
lines and hybrids examined for isozyme
markers.


Line #


Species


IRFL 6102

U.F. 28

IRFL 6105

U.F. 160


D. heterocarpon

D. ovalifolium

D. heterocarpon

D. ovalifolium


IRFL 6106 D. he

Hybrid
191 F1 (28 x 6105)

195 F1 (160 x 6102)

83-4-3 F1 (160 x 6106)

83-4-2 Fl (160 x 6106)

83-3-3 F1 (160 x 6106)


terocarpon










from 25 to 30 grams were used but 25 grams (8.3%) produced

gels that were easier to make and handle than 10% gels, and

gave equivalent results.

Leaf tissue was analyzed for anodal peroxidase using

Tris-citrate buffer. Root tissue was analyzed for cathodal

peroxidase and phosphohexose isomerase using K buffer (Table

3.2.). All gels were run at constant 250 volts, and

terminated when Bromo methyl blue marker indicated the

anodal front had moved 8 cm. Gels were sliced into three

horizontal sections, with the top section thrown away. The

middle and bottom section of the cathodal end of K-buffer

gels were stained for peroxidase activity (PER: E.C.

1.11.1.7) using the method of Durham et al. (1987). The

middle and bottom section of the anodal end was stained for

phosohexoseisomerase (PHI: Glucophosphate isomerse, E.C.

5.3.1.9) using the method of Durham et al. (1987). The

middle and bottom section of anodal ends of tris citrate

(TC, Table 3.2.) gels were stained for peoxidase.



Results



Results of an overall stain evaluation program are

given in Table 3.1. Many staining systems were tried only

once. Those that gave no visible results were not

repeated. All buffer systems marked with a superscript a

were systems that produced visible banding.










A zymogram of cathodal peroxidase patterns for D.

ovalifolium U.F. 28, D. heterocarpon 6105, and their

F1 191 is given in Fig. 3.1. 6105 had two bands of

activity: a slow band extending from Rf 0.10 to 0.13. and

a fast band extending from Rf 0.22 to 0.26. U.F. 28 also

had two bands of activity: a slow band extending from Rf

0.18 to 0.21 and a fast band from Rf 0.23 to 0.27. Hybrid

191 had three bands of activity: a slow one from Rf 0.11

to 0.15, another slow one from Rf 0.17 to 0.20, and a fast

one extending from Rf 0.23 to 0.28.

A zymogram of cathodal peroxidase patterns for D.

ovalifolium U.F. 160, D. heterocarpon 6102, and their F1

hybrid 195 is given in Fig. 3.2. U.F. 160 had two bands of

activity: a slow one extending from Rf 0.18 to 0.22 and a

fast one extending from Rf 0.26 to 0.30. 6102 had two

bands of activity: a slow one extending from Rf 0.12 to

0.18 and a fast one extending from Rf 0.27 to 0.32.

Similar to hybrid 191, hybrid 195 had three bands of

activity: a slow band extending from Rf 0.12 to 0.17,

another slow band extending from Rf 0.20 to 0.25, and a

fast band extending from Rf 0.28 to 0.33. A photograph of

cathodal peroxidase is shown in Fig. 3.3.

A zymogram of anodal PHI for root tissue for D.

ovalifolium U.F. 160, D. heterocarpon 6102, and 160 x 6102

hybrid 195 is given in Fig. 3.4. U.F. 160 has two bands of

activity of which the slow one, Rf 0.18 to 0.22, was

consistent. The fast band extended from Rf 0.37 to 0.46,























Origin




O 0 IRFL 6105
D. heterocarpon


0D 0 191 Fl
(28 x 6105)


S]0 U. F. 28
D. ovalifolium
0.7 0.6 0.5 0.4 0.3 0.2 0.1 0

Migration (Rf)


Figure 3.1. Illustrative drawing of cathodal
peroxidase isozyme banding patterns for
U. F. 28, 6105, and their hybrid.

























Origin





O I IRFL 6102
D. heterocarpon


D D 195 Fl
(160 x 6102)


SU. F. 160
D. ovalifolium

0.7 0.6 0.5 0.4 0.3 0.2 0.1 0

Migration (Rf)



Figure 3.2. Illustrative drawing of cathodal
peroxidase isozyme banding patterns for
6102, U. F. 160, and their hybrid.





53
























... ....... .. jiI










Figure 3.3. Cathodal peroxidase banding patterns
from left to right, U.F. 160 (five
plants in five lanes), hybrid 160 x
6102 (material from one plant in
each of three lanes, each prepared
with a different extraction
buffer),and 6102 (five plants in
five lanes).
























Origin

0 ^--- 0


DD 0 I -L IRFL 6102
D. heterocarpon


jr I 195 F1
(160 x 6102)

nU. F. 160
D. ovalifolium
0.7 0.6 0.5 0.4 0.3 0.2 0.1 0

Migration (Rf)


Figure 3.4. Illustrative drawing of anodal PHI
isozyme banding patterns for IRFL 6102,
U. F. 160, and their hybrid.










but was variable, as were other less distinct areas of

staining. 6102 has four bands of activity with the slow one

extending from Rf 0.22 to 0.28. The other three bands

extended from Rf 0.33 to 0.36, 0.41 to 0.45, and 0.46 to

0.48. The hybrid 195 has three bands of activity of which

the slow one extends from Rf 0.21 to 0.30 with two others

from Rf 0.32 to 0.35 and from 0.40 to 0.50.

A zymogram of anodal PHI for root tissue for D.

ovalifolium U.F. 160, D. heterocarpon 6106, and 160 x 6106

hybrid 83-4-2 is shown in Fig. 3.5. 6106 had four bands of

activity with three of those bands close together in the

slow region. These bands are at Rf 0.18 to 0.23, 0.25 to

0.27, and 0.30 to 0.33. A faster band consistently showed

up between Rf 0.42 and 0.48. U.F. 160 is the same as in

Fig. 2.4. Hybrid 83-4-2 has three bands of activity with

the slow one extending from Rf 0.18 to 0.27. Two other

bands extended from Rf 0.31 to 0.40, and from 0.45 to

0.50.

A composite photograph of anodal PHI is shown in

Fig. 3.6. with five lanes of U.F. 160, one lane of 195, five

lanes of 6102, one lane of 83-4-2, and five lanes of 6106.

Root tissue from a different plant was used to load each

lane.

A zymogram of anodal peroxidase for D. ovalifolium

U.F. 160, D. heterocarpon 6106, and 160 x 6106 hybrids

83-4-2, 83-4-3, and 83-3-3 is given in Fig. 3.7. U.F. 160

had two bands of activity: a slow one extending from Rf
























Origin




E 0 10 D IRFL 6106
D. heterocarpon

F- I I 83-4-2 Fl
(160 x 6106)

] U. F. 160
D. ovalifolium
0.7 0.6 0.5 0.4 0.3 0.2 0.1 0

Migration (Rf)


Figure 3.5. Illustrative drawing of anodal PHI
isozyme banding patterns for U. F. 160,
IRFL 6106, and their hybrid.












































Figure 3.6. Composite photograph of isozyme
banding patterns observed for
anodal PHI from left to right,
U.F. 160 (five plants in five
lanes), hybrid 195 (U.F.160 x
6102), 6102 (five plants in five
lanes), hybrid 83-4-2 (U.F.160 x
6106), 6106 (five plants in five
lanes).























Origin


m
E3
I I


0.7 0.6 0.5 0.4 0.3 0.2 0.1
0.7 0.6 0.5 0.4 0.3 0.2 0.1


IRFL 6106
D. heterocarpon

83-4-2 F1
(160 x 6106)

U. F. 160
D. ovalifolium

83-4-3 F1
(160 x 6106)

83-3-3 F1
(160 x 6106)


Migration (Rf)


Figure 3.7. Illustrative drawing of anodal
peroxidase isozyme banding patterns for
U. F. 160, IRFL 6106, and their hybrids.


_












0.02 to 0.06 and a fast one extending from Rf 0.35 to

0.40. 6106 had two bands of activity: a slow one extending

from Rf 0.02 to 0.05 and a fast one extending from 0.32 to

0.36. Hybrid 83-4-2 had two bands of activity: a slow one

extending from 0.02 to 0.06 and a fast one extending from

0.32 to 0.40. Hybrid 83-4-3 had two bands of activity: a

slow band extending from Rf 0.02 to 0.05 and a fast one

extending from 0.31 to 0.40. Hybrid 83-3-3 had two bands of

activity: a slow one extending from 0.02 to 0.05 and a fast

one extending from 0.31 to 0.40.

Figure 3.8. shows a representative photograph of

anodal peroxidase for U.F. 160 (five lanes), 160 x 6106

hybrid 83-4-2 (one lane), and 6106 (five lanes). Leaf

tissue from a different plant was used to load each lane.



Discussion



Banding patterns for D. heterocarpon 6105, D.

ovalifolium U.F. 28, and 6105 x 28 hybrid 191 were distinct

for each line. The F1 hybrid had three bands while

each parent had two bands. One of the two slower bands

migrated at the same speed as the slow band of U.F. 28. The

other slow band migrated at the same speed as the slow band

of 6105. The fast band in the F1 hybrid migrated at the

same speed as the fast band of each parent. This banding

pattern could be explained by gene action of two loci. The

locus for the fast band is homozygous and the same for each












































Figure 3.8. Banding patterns observed for anodal
peroxidase from left to right, five
lanes U.F. 160 (five different
plants), hybrid 83-4-2 (U.F. 160 x
6106), five lanes of 6106 (five
different plants).












for 6105 is probably homozygous as is the slow band Rf

0.18 to 0.21 for U.F. 28. The hybrid locus is heterozygous

for this trait with bands at Rf 0.11 to 0.15 and Rf 0.17

to 0.20, that is bands roughly equivalent to the two

parental slow bands.

Banding patterns for D. heterocarpon 6102, D.

ovalifolium U.F. 160, and 6102 x 160 hybrid 195 were

distinct for each line. It should be noted that cathodal

Rf values were calculated as if the cathodal front had

moved 8 cm while the anodal front was moving 8 cm. The F1

hybrid had three bands while each parent had two bands. The

three hybrid bands appeared to be two slow bands, each of

which was similar to one of the parent slow bands, and one

fast band identical to the fast band of each parent. The

most likely interpretation of this banding pattern is

similar to 191 above. The locus for the fast band is

homozygous and the same for each parent and the hybrid. The

locus for the slow band Rf 0.12 0.18 for 6102 is

probably homozygous as is the slow band Rf 0.18 to 0.22

for U.F. 28. The hybrid locus is heterozygous for this

trait with bands at Rf 0.12 to 0.17 and Rf 0.20 to 0.25,

that is bands roughly equivalent to the two parental slow

bands. Parental D. ovalifolium lines 160 and 28 are very

similar in morphology, as are parental D. heterocarpon lines

6102 and 6105, therefore it is not surprising that there

hybrids 191 and 195 are similar in morphology and in isozyme

marker patterns.










Desmodium ovalifolium U.F. 160, D. heterocarpon

6102, and their hybrid 195 each showed distinct banding for

root material stained for PHI isozymes. U.F. 160 showed two

bands of activity, 6102 four bands, and the hybrid three

bands. The bands of interest for distinguishing parental

lines and their hybrids from one another were the slow

bands. The slow band in U.F. 160 extended from Rf 0.18 to

0.22 while the slow band in 6102 extended from Rf 0.22 to

0.28. The hybrid slow band was larger and somewhat

intermediate, extending from Rf 0.21 to 0.30. The

uniqueness of the hybrid band makes it adequate for hybrid

verification without determining the number of bands that

actually compose it.

Desmodium ovalifolium U.F. 160, D. heterocarpon 6106,

and the hybrid 83-4-2 each showed distinct banding for root

material stained for PHI. U.F. 160 showed bands as

described above. The bands of interest for purposes of

distinguishing parental lines from each other and their

hybrid were the slow bands. The slow band in U.F. 160 was

as listed above while the slow banding area in 6106 was

composed of three bands that extended from Rf 0.18 to

0.23, 0.25 to 0.27, and 0.30 to 0.33. The hybrid slow band

was larger and somewhat intermediate, extending from Rf

0.18 to 0.27. This banding pattern was sufficient for

verification purposes to conclude that the parents and

hybrids were different.










The banding pattern for anodal peroxidase showed two

bands for D. ovalifolium U.F. 160, D. heterocarpon 6106, and

the hybrids 83-4-2, 83-4-3, 83-3-3 all of which were 160 x

6106 F1 hybrids. The slow band was the same for each

species and hybrid covering an area roughly Rf 0.02 to

0.06. The fast band in U.F. 160 migrated at Rf 0.35 to

0.40 and in 6106 migrated at Rf 0.32 to 0.36. In the

hybrids, the fast band covered the area of both parental

bands extending from roughly Rf 0.31 to 0.40 and probably

is two bands and could indicate heterozygosity for a monomer

at that locus. However, the gene action is speculative.

This staining system is capable of verifying hybrids.

Isozyme electrophoresis of Desmodium using starch

gels is a straight forward procedure that can be used to

verify a large number of hybrids in a short period of time.

Crosses with D. heterocarpon and D. ovalifolium could be

made in either direction without concern for the presence of

a dominant morphological marker in the male parent. This

system could also be used to delineate gene action by

looking at segregation ratios in the F2 populations.

Additional work needs to be done using a large population of

D. heterocarpon and D. ovalifolium plants .to verify whether

isozyme markers can distinguish among lines in each species.














CHAPTER IV
CYTOLOGY OF DESMODIUM HETEROCARPON
AND DESMODIUM OVALIFOLIUM


The haploid chromosome number of 11 was reported for

North American East Coast Desmodium (24 species) by Young

(1939-1940). Several workers have observed that difficulty

in obtaining differential staining between the cytoplasm and

the chromosomes was a serious impediment to cytological work

(Young, 1939-1940; Rotar and Urata, 1967; Schifino, 1983).

Schifino (1983) looked at four species of Desmodium and

reported that the 2n number was 22 for all four species.

She examined two morphologically different forms of D.

uncinatum (L.) DC. and found that their similar chromosome

number supported Oliveira's contention that they are one

species.

Failure of chromosome pairing was noted in Desmodium

intortum (Mill.) Urb. x Desmodium sandwicense E. Mey F2

hybrids (Hacker, 1968). Hacker reported that one of the

F2 plants had laggards in 25% of the cells studied and

showed 78% pollen stainability. However, Hacker's pollen

stainability varied within plant, from flower to flower, and

even from anther to anther. Hacker also referred to the

possible formation of multicellular tetrads in Desmodium

instead of micronuclei as a result of delayed cytokinesis.










'Florida' carpon [Desmodium heterocarpon (L) DC.])

was released for use in Florida in 1979 (Kretschmer et al.,

1979). 'Florida' carpon has perenniated in central Florida,

has good seed production, and showed tolerance to wet soils,

but was susceptible to root-knot nematode (Kretschmer et

al., 1979). Initial screening of Desmodium ssp. germplasm

indicated that D. heterocarpon (L) DC. was less resistant to

Meloidogyne arenaria race 1 (Neal) Chitwood, M. incognita

race 3 (Kofoid and White) Chitwood, and M. javanica (Treub)

Chitwood than was D. ovalifolium (Quesenberry et al.,

1986). Attempts to hybridize D. ovalifolium and D.

heterocarpon in the greenhouse resulted in few successful

crosses (Quesenberry and McKellar, unpublished results).

These two species are taxonomically considered one species

by Ohashi (1973.), therefore this difficulty in making

crosses was not anticipated. The objectives of this

research were twofold: (1) to evaluate D. heterocarpon and

D. ovalifolium pollen viability, and (2) to characterize

meiotic regularity in parent lines of D. heterocarpon and D.

ovalifolium and their hybrids.



Materials and Methods



Mature plants of D. heterocarpon, D. ovalifolium, and

their hybrids were maintained in the greenhouse at

University of Florida. Seed sources of parental lines were

the Centro Internacinal de Agriculture Tropical (CIAT)










and Dr. A. E. Kretschmer, Jr., Agricultural Research and

Extension Center, Fort Pierce, Florida (Table 4.1.).

Desmodium hybrids evaluated in this study are also described

in Table 4.1. Some, but not all, CIAT D. heterocarpon

accessions were taxonomically classified as to species,

subspecies, and variety. When available, this

classification was followed. Other D. heterocarpon lines

were taxonomically ranked by CIAT only to species.

Desmodium ovalifolium and D. heterocarpon not ranked by CIAT

were listed as D. ovalifolium or D. heterocarpon based on

gross morphological characters. All plant material was one

year old or older.

Bud material was collected in the form of entire

racemes before buds were visible between the bracts.

Racemes were placed immediately in 95% ethanol:chloroform:

glacial acetic acid (6:3:1, v:v:v) and refrigerated for 24

hours. Racemes were then placed in 70% ethanol and

refrigerated until use, usually within two weeks and always

within one month.

Cytological examination of chromatin and pollen

tetrads was done with individual buds removed from the

raceme under a dissecting scope. Buds were placed in one

drop of propionic carmine and the anthers were removed. The

anthers were gently squashed with a flat spatula. All

visible plant tissue was removed from the slide, leaving the

propionic carmine containing suspended pollen mother cells

(PMC), meiotic material, tetrads, or pollen, depending













Table 4.1. Desmodium heterocarpon and D. ovalifolium lines
and hybrids examined for tetrad formation,
pollen stainability, and meiotic behavior.


Introduction Numbers a

UF IRFL CIAT Species Cross


Parental lines

20 588, 6102 365 H
28 1946 3674 O
143 1695, 6105 3669 H
144 1696, 6106 3670 H/h/str
160 O
- 6150 13183 H/h/h

Hybrid lines

83-3-3 160 x 6106
83-4-2 160 x 6106
83-4-3 160 x 6106
195 160 x 6102
191 28 x 6105



a Plant introduction numbers are numbers assigned by
Dr. K. H. Quesenberry, Agronomy Department, University
of Florida (UF), by Dr. A. E. Kretschmer, Jr., AREC, Ft.
Pierce, Florida (IRFL), and by the Tropical Pastures
Program, Columbia (CIAT).


b Species:


H = species heterocarpon
O = species ovalifolium
H/h/h = species heterocarpon subspecies
hetercarpon variety heterocarpon
H/h/str = species heterocarpon subspecies
heterocarpon variety strigosum










on the age of the anthers. Material was covered with a

cover slip and examined briefly under a 10X objective on a

Wild M20 microscope equipped with a quartz halogen light

source. Material that appeared promising was then

immediately destined by first heating the slide over a

flame, sealing two opposite sides of the cover slip with

sticky wax, flowing 45% acetic acid under the cover slip,

then sealing the remaining two sides with sticky wax. The

material, once covered with a cover slip, was not squashed

as this tended to disrupt cells and disorganize cellular

contents. The entire staining and destaining process was

limited to ten minutes as this was the time differential

between which chromatin took up stain and the surrounding

cytoplasm became stained.

Tetrads were counted while still enclosed within

the clear envelope within which they developed. Tetrads

were evaluated on whether they contained four equal sized,

stained cells (normal) or they contained three or fewer

normal sized, stained cells (abnormal).

Pollen was collected at 10:00 A. M. from greenhouse

grown plants. Untripped flowers were tripped in a manner

that caused pollen to be deposited on slides, which was then

covered with a cover slip. When flowers had self-tripped

before 10:00 A. M., pollen that remained on the anthers was

shaken onto the slides. Pollen was stained with propionic

carmine within four hours of collection, and scored

immediately after staining using a 10X objective on a light









microscope. Normal sized pollen that absorbed stain was

classified as stained, whereas abnormally small or shriveled

pollen that did not absorb stain was classified as

shriveled.



Results



Results of meiotic analysis for D. ovalifolium U.F.

160, D. heterocarpon 6106, and three F1 hybrids are shown

in Table 4.2. Both parental lines and their three hybrids

showed regular first division segregation. Hybrid 83-4-3

had one cell out of twenty that had two laggards at

metaphase I. All other cells showed regular pairing at

metaphase I (Figure 4.1.).

Hybrid 191 (U.F. 28 x 6105) showed some abnormal

pairing at metaphase I and some abnormal segregation during

first anaphase division (Table 4.2.). Parental line U. F.

160 had normal metaphase I pairing and first division

segregation. Parental line 6102, D. heterocarpon, had

normal metaphase I pairing (Figure 4.2.), but had one cell

out of twenty that had two laggards during first division

segregation. Five of twenty cells of F1 191 had two

univalents at metaphase I (Figure 4.3.). The same line had

four cells with one laggard, one cell with two laggards, and

one cell with five laggards at first division segregation

(Figure 4.3.).










































00 OC 0-





00 0 m m
CN (N r-


00
(C (4


000
('NC' ("


C
0

.,-4

>




-4


--4





O
a)






SC-






0,-


0
--4
CC-







E4
Oc










0
um
E-U
0 0)





E-l 0)




0 fl(
M-I
CE4
E)

-4









-(


o CNm
000
-4 -4-4
o t0 .


0 C4oo -4


U) -

O 4-
' 0


O(
'riZ







O
r-4


0



*z
4-O
-Z
0 a)
2 U


(N
0
-


X I I I
M0 1 1

-4 m0 oo
01-1 o o


CM
0
r-






,r.-I
x


,-.4-4


i-l

X


CN r-i


.4-i4 N


000 0
CN 04CN -





000 0 0
(N' (CN (N (






00-4 0 Un'











00o 0 o i
NC rC (-4 N -





000 0 0
(N CNN (N (N


,- -
.c 1
4.)
*.l4 (N

+



0I
- 4
(U 0


U -
04-




2 -
4.4


--4
oa)


00 OO C











00 00o *
NCN (N CN l





00 00CD
N( C4 NCM -































WaIw


*~ ~'CME ij
4.


Figure 4.1. First division segregation in 6106
with no laggards.



















*es 4 *1* ^.^f
rr *






V9* *^ :J. *-
rr






4t~4


S."


a. &


t


't #


S&


a


Figure 4.2. Metaphase I pairing.
A. Line 6102 showing 11 bivalents.
B. Hybrid 83-3-3 showing 11
bivalents.














































.ij


B. -



Figure 4.3. Meiosis in hybrid 191.
A. metaphase I pairing with
univalents (arrows).
B. First division segregation with
five laggards (arrows).










Parental line U.F. 28 (D. ovalifolium) was regular in

both metaphase one pairing (Figure 4.4.) and in first

division segregation. Parental line 6105 showed two cells

with univalent chromosomes at metaphase 1 (Figure 4.4.) and

one cell with six laggards at fist division segregation

(Figure 4.4.). Hybrid 195 had normal pairing during

metaphase I but had two of twenty cells that showed laggards

during first division segregation (Table 4.2.). One of these

cells had one laggard (Figure 4.5.) while the other had three

laggards (Figure 4.5.).

Tetrad formation and pollen stainability are shown in

Table 4.3. Lines 6102, U.F. 28, and hybrid 195 had 100%

normal tetrads of the samples counted. Pollen stainability

was high in all cases, ranging from 75% in hybrid 83-3-3 to

100% in accession 6106.



Discussion



The results for the two D. ovalifolium lines (U.F. 160

and U.F. 28) showed no problems with chromosome pairing

during metaphase I or first division segregation indicating a

complete complement of homologous chromosomes. Tetrad

formation was normal and pollen viability was high, as would

be expected. Desmodium heterocarpon lines 6102, 6105, and

6106 were only slightly less uniform in their meiotic

behavior. Line 6106 was taxonomically ranked as D.

heterocarpon ssp. heterocarpon var. strigosum by CIAT. The












































r


I


t


`C~
I
P ~+
"rt ,li
I d


4 1#


Figure 4.4. Meiotic chromosome configerations.
A. Metaphase I pairing in parental
line U.F. 28.
B. Diplotene pairing in parental line
6105


r
~


a 1 r
rr II
I; r;


tr


^ *






















































Figure 4.5. First division segregation in hybrid
195 showing one laggard (arrow).






























o (N O iln
m^ m m 0 m a m
-4








T-4 N 1
OC00o(Ln







0 v 0 0 0 0
0 n o ooo
CXdHQMnMH


0




-,-I
a)






S-H,
O o




0o




-4
GC C













C o

0l


-O


E >

44 -


Or




4-
O
*
II


O


CO



&4-1









,a
















o
0 m











z
cP


r nO OI0
CNlAm r io k

I r-- o or-
C U'n Ln I' u 1


(N Ln %


0o o
000
-" -4-I--


0Ln


m M
^oCM r-
i-l *


Q^*
M 3


en cr Ln Ln V





v:r Ln CM4
b*d' i-i


LnLn or-

NLO
NC0-I
C0


CM (N rC 00 00









-4









m r0 rLn N 0



rL ) Ln l
mo0e
ma0m







ma0m0



hom0





*. I I irl

4-4 00 00 00 r--i
-Q


C0) 00

r-i


o r-
0a~
I -


4-1
0
z



4-4
0 (
* -4
Oro

0 )
z





a)

*H(
;r


r0
a)
v-

>C




II II

13^
4-1


Ma a










remaining two lines were classified only to the species level

of D. heterocarpon. Line 6106 paralleled the D. ovalifolium

lines in that it had normal pairing during meiosis and normal

segregation at first division. Line 6105 had two out of

sixteen cells with univalents at metaphse I and one cell out

of twenty with six laggards at first division segregation.

Tetrad formation and pollen stainability were both 92% for

this line. Line 6102 had only one cell with laggards and no

univalents at metaphase I. Its tetrad formation was as

expected, 100% normal, and its pollen stainability was high

(90%).

Hybrids between D. ovalifolium and D. heterocarpon

have been somewhat difficult to obtain (Quesenberry and

McKellar, Unpublished results). The F2 generation of

hybrids 83-3-3, 83-4-2, 83-4-3 showed morphological

characteristics often associated with hybrid breakdown

including severe stunting, chlorosis, and inability to form

flowers. Hybrid breakdown often occurs in wide crosses

(Stebbins, 1958). Such breakdown is usually associated with

some chromosome pairing problems during meiosis. This was

not the case with hybrids 83-4-3, 83-4-2, and 83-3-3. The

metaphase configerations of these three hybrids showed only

one cell with univalents out of a total of sixty. All cells

examined were normal for first division segregation. Tetrad

formation was not entirely normal but still high. Pollen

stainability for these three hybrids was lower than the

parental lines, ranging down to 75%. Quesenberry and Taylor










(1976) reported a similar situation in a Trifolium alpestre

L. x T. heldreichianum Hausskn. cross. The F1 hybrid had a

low incidence of univalents at metaphase I and few

micronuclei at the quartet stage. Nonetheless it had a mean

pollen stainability of 35% less than the parent lines, and

low seed set and F2 seedling vigor.

Hybrids 191 and 195 showed more meiotic abnormalities

than hybrids 83-3-3, 83-4-2, and 83-4-3. These two F1

hybrids appear morphologically similar to the D. heterocarpon

agronomic type male parent with strongly marked, pubescent

leaves and upright growth habit. Hybrid 195 had two out of

twenty cells with laggards. Hybrid 191 had five out of

twenty cells with univalents at metaphase I and six out of

twenty cells with laggards at first or second division

segregation. Surprisingly, normal tetrad formation was high

for both lines with hybrid 195 having 100% normal tetrads.

Pollen stainability was also high for both hybrids. Pollen

production, as the end result of the meiotic process, would

be expected to be reduced by problems originating from lack

of chromosome pairing and segregation. The high pollen

stainability levels obtained for the five hybrids in this

experiment would argue against extreme evolutionary distance

between the two species D. ovalifolium and D. heterocarpon.

Conversely, Hacker (1968) noted highly variable pollen

stainability which he attributed to environmental factors.

It may be desirable to test more pollen over a wider range of

environmental conditions (including conditions representative










of the extremes of the greenhouse and field environment) to

determine what, if any, environmental factors might be

influencing pollen viabiliy. Hacker (1968), as previously

mentioned, also noted that 75% pollen stainability when

tetrad formation was 25% abnormal.

Initial examination of meiosis and tetrad formation

was complicated by nucleosome formation that mimicked

micronuclei (Figure 4.6.). Second division segregation would

occur, followed by chromatids condensing as small, individual

nucleosomes within one nuclear membrane. This is possibly an

artifact of the preparation technique but nevertheless does

not appear to influence eventual normal tetrad formation.

Chromatin material within each nuclear membrane of the

tetrads eventually merged into one nucleus. Different lines

gave different results for this trait, so that it seems

unlikely that this is an artifact of preparation.

An explanation for low success rates in crosses

between D. ovalifolium and D. heterocarpon is not obvious

from cytogenetic data or pollen stainability. In all cases,

pollen stainability was high. Thus parental lines should

have been adequate as male parents. Female parental lines,

when selfed, produced seed, thus these lines were adequate as

female parents. If the contributions of male and female

lines are functional, yet hybrids are difficult to obtain,

the problem maybe a lack of chromosome homology between D.

ovalifolium and D. heterocarpon. Slight irregularities as

might be caused by lack of chromosome homology, were observed































(n*


Figure 4.6. Multiple nucleosomes in 83-4-2.












in meiosis of some F1 hybrid lines, indicating possible

evolutionary distance between the two species.

The above results indicate that problems with

hybridization appear related to environmental factors, and to

crossing techniques and methods. There appears to be no

major barrier to transfer of genetic information between the

D. ovalifolium and D. heterocarpon types. Thus one should be

able to transfer resistance to the desired agronomic type.













CHAPTER V
CONCLUSIONS


Caged D. ovalifolium lines set few, if any, seed.

This could mean that flowers need to be tripped,

particularly because tripped D. ovalifolium lines in the

greenhouse set seed, or that D. ovalifolium lines need cross

pollination, which is less likely. Incidental cross

pollination, although not required for successful seed set,

might be expected if flowers need to be tripped, presumably

by insects. Differences in flowering time effectively

isolated D. heterocarpon from D. ovalifolium in

Gainesville. Although D. heterocarpon is capable of

producing flowers at the same time as D. ovalifolium, the D.

heterocarpon plants generally have stopped flowering earlier

after having produced large quantities of seed.

Desmodium heterocarpon and D. ovalifolium, although

separated by morphological and biochemical differences,

appear to have no major crossing barriers. In spite of the

opportunity for outcrossing, the D. ovalifolium lines show

little variation for morphological traits, particularly for

unmarked leaves. Desmodium heterocarpon types show the same

uniformity, in that, when selfed, D. heterocarpon types do

not segregate for marked leaves. A 9:7, marked:unmarked,

F2 hybrid ratio indicates that marked leaves are not a

single gene controlled dominant trait. A test combining










isozyme marker examination of potential hybrids could

confirm whether all hybrids in an unmarked leaf x marked

leaf cross have marked leaves. Another possibility would be

using flower color to verify the effectiveness of marked

leaves as an indicator of successful hybridization, although

the current D. ovalifolium germplasm does not include any

white-flowered lines.

Isozyme electrophoresis for phosphohexose isomerase

(PHI) distinguished between a D. ovalifolium line, a D.

heterocarpon subspecies heterocarpon var. strigosum, and a

D. heterocarpon subspecies not specified. The subspecies

unspecified line (6102, 'Florida' carpon desmodium) was

nonetheless morphologically very different than the

strigosum variety and similar to the heterocarpon agronomic

type in growth habit and in having clearly marked leaves.

Variation in staining for PHI for five D. ovalifolium plants

of line U.F. 160 indicated that this line is not

homogeneous. This lack of uniformity in an D. ovalifolium

line might also be expected based on its pollination

biology.

Cytological studies indicated that failure of chromo-

some pairing during meiosis did not reduce fertility in F1

hybrids. Apparent general chromosome homology indicated

that parental lines are similar, though not identical, in

chromosome compliment. Limiting factors for successful

crossing in the greenhouse may be environmental conditions

such as humidity or temperature. An experiment addressing





85


these environmental parameters using growth chambers might

resolve these questions.















LITERATURE CITED


Albrecht, K. A. and K. H. Quesenberry. 1987. Genetic,
environmental, and morphological effects on tannin
concentrations of tropical forage legumes. Agronomy
Abstracts 140.

Andersen, S. and W. E. Davies. 1984. Descriptor list for
forage legumes. IBPGR. Rome.

Baltensperger, D. D., M. Abd-Elgawad, R. A. Dunn, and K.
H. Quesenberry. 1983. Greenhouse nematode screening
test results for various forage legumes tested at
Gainesville, Florida. Agronomy Research Report
AY-84-2. University of Florida. Gainesville,
Florida.

Bryan, W. W. 1966. The pasture value of species of
Desmodium. Proc. of the 10th Int. Grass. Cong.
Helsinki 10:311-315.

Bryan, W. W. 1969. Desmodium intortum and Desmodium
uncinatum. Herbage Astracts 39(3):183-191.

Chow, K. H. and L. V. Crowder. 1972. Hybridization
of Desmodium canum (Gmel.) Schin. and Thell. and D.
uncinatum (Jacq.) DC. Crop Sci. 12:784-785.

Chow, K. H. and L. V. Crowder. 1973. Hybridization of
Desmodium species. Euphytica 22:339-404.

Davies, W. E. 1984. A plan of action for forage
genetic resources. IBPGR. Rome.

Durham, R. E., G. A. Moore, and W. B. Sherman. 1987.
Isozyme banding patterns and their usefulness as
genetic markers in peach. J. Amer. Soc. Hort. Sci.
112(6):1013-1018.

Hacker, J. B. 1968. Failure of chromosome pairing in a
probable Desmodium intortum x D. sandwicense hybrid.
Aust. J. Bot. 16:545-550.

Hayward, M. D. and N. J. McAdam. 1977. Isozyme
polymorphism as a measure of distinctiveness and
stability in cultivars of Lolium perenne. Z.
Pflanzenzuchtg 79:59-68.











Hussain, A., H. Ramirez, W. Bushuk, and W. Roca. 1987.
Identification of cultivars of forage legume
(Desmodium ovalifolium Guill et Perr.) by their
electrophoretic patterns. Can. J. Plant Sci.
67:713-717.

Hutton, E. M. and S. G. Gray. 1967. Hybridization
between the legumes Desmodium intortum, D. uncinatum,
and D. sandwicense. J. of the Aust. Inst. of Agric.
Sci. 33:122-123.

Imrie, B. C. and D. Blogg. 1982. Variability in isozyme
frequency in the tropical pasture legume, 'Greenleaf'
desmodium. Trop. Agric.(Trinidad) 60(3):193-196.

Imrie, B. C., R. M. Jones, and P. C. Kerridge. 1983.
Desmodium. In: The role of Centrosema, Desmodium,
and Stylosanthes in improving tropical pastures.
Ed.: R. L. Burt, P. P. Rotar, J. L. Walker, and M. W.
Silvey. Westview Tropical Agriculture Series No. 6.
Westview Press. Boulder, Colorado 97-140.

Jarret, R. L. 1987. Biochemical/Genetic Markers and
Their Uses in the Genus Musa. In:Banana and plantain
breeding strategies. Edited by: G. J. Persley and E.
A. De Langhe. ACIAR Proceedings No. 21. 182-185.

Kretschmer, A. E., Jr. 1970. Production of annual and
perennial tropical legumes in mixtures with
pangolagrass and other grasses in Florida. Proc. of
the llth Int. Grass. Cong. Surfer's Paradise,
Hawaii 144-153.

Krestschmer, A. E., Jr., J. B. Brolmann, G. H. Snyder,
and S. W. Coleman. 1976. 'Florida' Carpon
Desmodium, a perennial tropical legume for use in
south Florida. Soil and Crop Sci. Soc. of Fla.
Proc. 35:25-30.

Kretschmer, A. E., Jr., J. B. Brolmann, G. H. Synder, and
S. W. Coleman. 1979. 'Florida' Carpon Desmodium.
Agricultural Experiment Station Circular S-260.
University of Florida. Gainesville, Florida.

Kretschmer, A. E., Jr., J. B. Brolman, G. H. Snyder, and
G. J. Gascho. 1973. Production of six tropical
legumes each in combination with three tropical
grasses in Florida. Agronomy Journal 65:890-892.

Krestschmer, A. E., Jr., R. M. Sonoda, and G. H. Snyder.
1980. Resistance of Desmodium heterocarpon and other
tropical legumes to root-knot nematodes. Trop.
Grasslands 14(2):115-120.










Krestschmer, A. E., Jr. and G. H. Snyder. 1982.
Comparison of mixtures of seven tropical legumes and
six tropical grasses in South Florida. Soil and Crop
Sci. Soc. of Fla. Proc. 41:67-72.

Lenne, J. M. 1981. Reaction of Desmodium species and
other tropical pasture legumes to the root-knot
nematode Meloidogyne javanica. Trop. Grasslands
15(1):17-20.

McWhirter, K. S. 1969. Cytoplasmic male sterility in
Desmodium. Aust. J. Agric. Res. 20:227-241.

Minson, D. J., and R. Milford. 1967. The voluntary
intake and digestibility of diets containing
different proportions of legume and mature pangola
grass (Digitaria decumbens). Aust. J. Exp. Agric.
Anim. Husb. 7:546-550.

Minton, A. M., I. Forbes, and H. D. Wells. 1967.
Susceptibility of potential forage legumes to
Meloidogyne species. Plant Disease Reporter
51(12):1001-1004.

Moore, G. A. and R. E. Litz. 1984. Biochemical markers
for Carica papaya, C. cauliflora, and plants from
somatic embryos of their hybrid. J. Amer. Soc. Hort.
Sci. 109(2):213-218.

Norris, D. 0. 1967. The intelligent use of inoculants
and line pelleting for tropical legumes. Trop.
Grasslands 1(2):107-121.

Ohashi, H. 1973. The Asiatic species of Desmodium and
its allied genera (Leguminosae). Ginkgoana.
Contributions of the flora of Asia and the Pacific
region. Academia Sci. Book, Inc. Tokyo.

Park, J. S. and P. P. Rotar. 1968. Genetic studies
in Spanish clover, Desmodium sandwicense E. Mey. I.
Inheritance of flower color, stem color, and leaflet
markings. Crop Sci. 8:467-474.

Pitman, W. D. 1983. Initial comparisons of tropical
legume-bahiagrass pastures and nitrogen-fertilized
bahiagrass pastures in peninsular Florida. Soil and
Crop Sci. Soc. of Fla. Proc. 42:72-75.

Pitman, W. D. 1986. 'Florida' Carpon desmodium
establishment studies. Soil and Crop Sci. Soc. Fla.
Proc. 45: 161-163.





89





Pitman, W. D. and A. E. Kretschmer, Jr. 1984. Persist-
ence of selected tropical pasture legumes in
peninsular Florida. Agron. J. 76:993-996.

Quesenberry, K. H. and R. A. Dunn. 1987. Variability
among Desmodium species for response to root-knot
nematodes. Crop Sci. 27:1234-1237.

Quesenberry, K. H., M. A. McKellar, and R. A. Dunn. 1986.
Desmodium spp. response to root-knot nematodes.
Clovers and Special Purpose Legumes Research.
19:22-28.

Quesenberry, K. H. and N. L. Taylor. 1976. Interspecific
hybridization in Trifolium L., Sect. Trifolium Zoh.
I. Diploid hybrids among T. alpestre L., T. rubens
L., T. heldreichianum Hausskn., and T. noricum Wulf.
Crop Sci. 16: 382-386.

Quiros, C. F. 1980. Identification of alfalfa plants by
enzyme electrophoresis. Crop Sci. 20:262-264.

Ramirez, H., A. Hussain, W. Roca, and W. Bushuk. 1987.
Isozyme electrophoregrams of sixteen enzymes in five
tissues of cassava (Manihot esculenta Crantz)
varieties. Euphytica 36:39-48.

Rotar, P. P. and K. H. Chow. 1971. Morphological
variation and interspecific hybridization among
Desmodium intortum, Desmodium sandwicense, and
Desmodium uncinatum. Tech. Bulletin No. 82. Hawaii
Agricultural Experiment Station, Hawaii

Rotar, P. P. and U. Urata. 1967. Cytological studies in
the genus Desmodium species: some chromosome counts.
Amer. J. Bot. 54: 1-4.

Schifino, M. T. 1983. Chromosome numbers and meiotic
behavior in five native Brazilian forage legume
species. Brazil. J. Genetics. 6(2):357-362.

Schubert, B. G. 1963. Desmodium: Preliminary studies
IV. Journal of the Arnold Arboretum, 44:284-297.

Schubert, B. G. 1971. Desmodium In: Flora of tropical
East Africa. Edited by E. Milne-Redhead and R. M.
Polhill. Royal Botanical Gardens, Kew. 451-461.

Schultze-Kraft, R. and S. Pattanavibul. 1985.
Collecting native forage legumes in eastern
Thailand. IBPGR Newsletter. 9(1):4-5.









Shaw, C. R. and R. Prasad. 1970. Starch gel
electrophoresis--a compilation of recipes. Biochem.
Genet. 4:297-320.

Sobrinho, J. M. 1982. Yield performance and other
agronomic characters of eighteen accessions of
Desmodium ovalifolium Wall. and one accession of D.
heterocarpon. M.S. Thesis, New Mexico State
University, Las Cruces

Sollenberger, L. E., W. R. Ocumpaugh, V. P. B. Euclides,
J. E. Moore, K. H. Quesenberry, and C. S. Jones, Jr.
1988. Animal performance on continuously stocked
'Pensacola' bahiagrass and 'Floralta' limpograss
pastures. J. of Prod. Agric. 1(3):216-220.

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

Takahashi, M. 1952. Tropical forage legume and browse
plants research in Hawaii. Proc. 6th Int. Grass.
Cong. 1411-1416.

Vallejos, C. E. and S. D. Tanksley. 1983. Segregation
of isozyme markers and cold tolerance in an
interspecific backcross of tomato. Theor. Appl.
Genet. 66:241-247.

Wendel, J. F. and C. R. Parks. 1982. Genetic control of
isozyme variation in Camellia japonica L. J. Hered.
73:197-204.

Young, J. 0. 1939-1940. Cytological investigations in
Desmodium and Lespedeza. Botanical Gazette
101:839-850.

Younge, O. R., D. L. Plucknett, and P. P. Rotar. 1964.
Culture and yield performance of Desmodium intortum
and D. canum in Hawaii. Tech. Bulletin No. 59.
Hawaii Agriculural Experiment Station, Hawaii

Zoebisch, O. C., M. Rosenberg, and M. Takahashi. 1952.
Assaying toxicity in legumes by chick test. Proc. of
the 6th Int. Grassl. Cong. 2:1531-1531.














BIOGRAPHICAL SKETCH


Martin Albert McKellar was born July 22, 1950, in

Saginaw, Michigan.

Mr. McKellar graduated from the University of the

Virgin Islands in 1981 from which he received a Bachelor of

Arts in Biology degree. He received a Master of Science

from the fruit crops department at University of Florida in

1984.

Mr. McKellar worked in Cameroon, West Africa, for the

academic year 1985-1986 teaching at the University Center of

Dschang as a University of Florida Assistant in Agronomy.

Mr. McKellar is married to Caryl Wiesenfeld, and has

two sons.


















I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.





Kenneth H. Quesenberry, Chairman
Professor of Agronomy


I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.





Paul M. Lyrene/
Professor of Fruit Crops


I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.





Gloria A. Moore
Associate Professor of Fruit
Crops
















I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.




Carrol Chambliss
Associate Professor of Agronomy


I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.




David A. Knauft
Associate Professor of Agronomy


This dissertation was submitted to the Graduate Faculty of
the College of Agriculture and to the Graduate School and
was accepted as partial fulfillment of the requirements for
the degree of Doctor of Philosophy.


December 1988
n, Co lege o Agricu ture


Dean, Graduate School









































UNIVERSITY OF FLORIDA
IIDIIIIIIIII 1111 111111111 I111111ii 11II III
3 1262 08557 0074




Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EBGK7278L_77UOTF INGEST_TIME 2012-08-13T16:56:35Z PACKAGE AA00011768_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES