WI.‘

vv'vvv' -w

F WM".

3 O‘: “
Q 2 "
n I .“

'I
o

5
°. 9?"
t

9

D;$
n
I
' 9
h

"s
2K
‘13:;

I
'
-
.s

.. .
1" -
b I

(inf In. a."
E...‘ '

.2
91

3““!
C '-
0 ‘g
'se-

-.s 9.5:.
fizz ~:

5
o
,0 ‘
96"

' at
U
0

‘Mrr
‘0 v
Q”.
ho.

rx

.1

"3
3A..
, \-
icky“

0 t

I
0‘: -
O
O

.Q'
S.
.

héain'

.fi
... a u
3,... 5
mm... ..
-. *-
" 2“ .V‘. n
.15.... ..
‘ .5
Mr-.. ~54...
.o
1.1. «Lu
' .fi'v' .o
O o... A..V
0 .u o.” a a.
3...... ....n
Law «1. a
. 7‘
“APO‘
T .
'5'-
. A” Q
«in

F
i
..
3“
o

r".
"u...

.fi. . o.¢oi “1.0
‘ . , .9 . a...
O. a... .. cIJ.
huh. I.» . n ..

a : .
It o no f“ n .OJ

q 1.\

y Or

m3:

._".a
‘1

d
n
. .
3
.fl-
, '0!-
-055 t In a

v.9
T.
b
.
0

fl
0
9'
or
I;
.
'7
.
I.

A

a.
l‘ u
an”

‘-“
z

3
A
I
3
.3.-

«.....U .

«mud
c...»

 

1‘}‘

"o':

ABSTRACT

CHARACTERIZATION OF THE GRANDIFLORA-MULTIFLORA
GENOTYPES IN PETUNIA HYBRIDA VILM. BY SEED
GERMINATION, GROWTH STUDIES AND
BIOCHEMICAL CONSTITUENTS

 

BY

Linda Lee Knowlton

Nine lines of B. hybrida Vilm., divided into three
sets, each including a homozygous multiflora gg, homozygous
grandiflora fig and hybrid gg were used for this study. The
relationship of seed germination, seedling growth rate,
leaf elemental composition, total chlorophyll, sugar and
starch content as well as quantitative sugar differences in
vegetative leaf tissue to genotype was investigated. Seed
germination was consistantly high for the hybrid gg (92%),
intermediate for gg_(77%), and low for gg (45%). The fresh
and dry weight of 28 day old seedlings was inconsistant
but, after 49 days the gg hybrid was the most vigorous
followed by the gg and gg genotypes. No differences were
observed in N, P, K, Na, Mn, Fe, Cu, Zn, or Al in alternate
leaf tissue of the three genotypes. Differences in Ca, Mg,
and B occurred but they were not uniform with respect to

genotype or to genotypes within a set. Calcium and Mg were

Gaga—8”,?—

Linda Lee Knowlton

generally highest in gg and lowest in gg. Boron in one of
two experiments showed the same pattern. These differences
suggested that there might also be a difference in chloro-
phyll content since each of these elements is involved in
the production of chlorophyll.

The amount of total chlorophyll was determined in
mature leaf tissue and in all three experiments it was not
specifically correlated to the grandiflora-multiflora
genotype. Similarly, sugar and starch content was examined
but again there were no significant differences in the
three genotypes. Qualitative differences of extractable
sugar compounds were observed using thin-layer chromato-
graphy. A band, Rf = 0.52-0.56, was resolved in extracts
obtained from vegetable tissue of fig and gg genotypic plants.
The extracts obtained from fig plants did not have this band.
This band did not co-chromatograph with any of the simple
sugar standards used including mannose, arabinose, glucose,
galactose, sucrose, and maltose. Gas-liquid chromatography
was attempted to further characterize the unknown compound

but the samples were insufficient for definitive results.

CHARACTERIZATION OF THE GRANDIFLORA-MULTIFLORA

GENOTYPES IN PETUNIA HYBRIDA VILM. BY SEED

 

GERMINATION, GROWTH STUDIES AND

BIOCHEMICAL CONSTITUENTS

BY

Linda Lee Knowlton

A THESIS

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

MASTER OF SCIENCE

Department of Horticulture

1976

ACKNOWLEDGMENTS

I wish to thank Dr. Ken Sink for the guidance and
encouragement he has given me over the years. I also
appreciate the support given to me by other graduate
students and friends, especially Nicolas Natarella,
Maarten Benschop, and James Liesman.

Special thanks are also due to Dr. Homma, LuAnn
Kent, and the other members of my committee.

And to my father, I dedicate this and all my other
accomplishments past and future. He and my family have
given me the moral support I have needed to continue and

finally finish my education.

ii

TABLE OF CONTENTS

Page
LIST OF TMLES . . . . . . . . . . . . . iv
LIST OF FIGURES. . . . . . . . . . . . . vi
LI TEMTURE REVIEW . . . . . . . . . . O O 1
STATEMENT OF PROBLEM . . . . . . . . . . . 5

INFLUENCE OF THE GRANDIFLORA-MULTIFLORA GENOTYPES
OF PETUNIA ON SEED GERMINATION, SEEDLING
GROWTH AND ELEMENTAL FOLIAR COMPOSITION
INTRODUCTION. . . . . . . . . . .‘ . . . 7
MATERIALS AND METHODS. . . . . . . . . . . 10
RESULTS AND CONCLUSIONS . . . . . . . . . . l4
INFLUENCE OF THE GRANDIFLORA-MULTIFLORA GENOTYPES
OF PETUNIA 0N TOTAL CHLOROPHYLL, STARCH, AND
SUGAR CONTENT AND ON THE QUALITATIVE SUGAR
CONTENT IN VEGETATIVE TISSUE

INTRODUCTION . . . . . . . . O . . . O . 30
MATERIALS AND METHODS. . . . . . . . . . . 33
ESULTS . . . . . O . . . . . C C O . 4 0
DISCUSSION . . ‘ . . . . . O . . . O O O 51
CONCLUSIONS AND RECOMMENDATIONS . . . . . . . 55

LITERATURE CITED 0 O O O O O O O O O O O 57

iii

10.

LIST OF TABLES

Petunia hybrida Vilm. breeding and hybrid
lines 0 O O O O O O O O O O O O

 

Seedling fresh and dry wts at 28 and 49 days
and plant ht at 49 days for grandiflora and
multiflora genotypes in experiment 1. . .

Seedling fresh and dry wts at 28 days and 49
days and plant ht at 49 days for grandiflora
and multiflora genotypes in experiment 2 .

Elemental analysis of 49 day old seedlings of
grandiflora and multiflora genotypes in
experimentl...........

Elemental analysis of 49 day old seedlings of
grandiflora and multiflora genotypes in
experiment 2. . . . . . . . . . .

Petunia hybrida lines utilized in this study.

 

Total chlorophyll, starch, and sugar content
of vegetative leaf tissue of plants in
eXPerj-ment 1 O O O O . 0 O O O O 0 0

Total chlorophyll, starch, and sugar content
of leaf tissue of plants in experiment 2
on a fresh-weight basis . . . . . . .

Total chlorophyll, starch, and sugar content
of leaf tissue of plants in experiment 2
on a leaf area basis . . . . . . . .

Total chlorophyll, starch, and sugar content
of leaf tissue in experiment 3. . . . .

iv

Page

11

17

18

21

24
34

41

42

43

44

Table

11. Total chlorophyll, starch, and sugar content
of leaf tissue in experiment 3 on a leaf
area basis . . . . . . . . . . .

12. R values and color reactions of standard
sugars and extracts from grandiflora-
multiflora petunia vegetative tissue . .

Page

45

47

LIST OF FIGURES

Figure Page

1. Percent seed germination in relationship to
the grandiflora and multiflora genotypes in
petunia O O O O O O O O O O O O 0 l6

2. Relationship of Boron, Magnesium and Calcium
content in vegetative tissue and grandiflora
and multiflora genotypes of petunia in
experiment 1 . . . . . . . . . . . 23

3. Relationship of Boron, Magnesium and Calcium
composition in vegetative tissue and grandi-
flora and multiflora genotypes in petunia in
experiment 2 . . . . . . . . . . . 26

4. Extraction technique for leaf tissue for
quantitative and qualitative analysis of
chlorOphyll, sugar and starch. . . . . . 36

5. Composite thin-layer chromatogram of sugar

standards and extracts from grandiflora and
multiflora petunia vegetative tissue . . . 49

vi

L ITERATURE REV IEW

Petunia hybrida Vilm. is one of the most popular

 

hybrid bedding plants in the United States and the present
varieties are the product of a long process of careful
breeding and selection from an initial cross between P.

axillaris which was introduced in Europe in 1793 by Lamarch

 

and P. violaceae which was also introduced about the same

 

time (Ferguson and Ottley, 1932). Since the original
hybridization of the two species, P. hybrida has emerged
as an extremely polymorphic species. Mutations have
occurred such as the double flower (Scott, 1937), grandi-
flora (Reimann-Philipp, 1968; Bianchi, 1959) and the
apetalous characteristic (Sink, 1973). Although the
inheritance of these traits has been well established,
little research has been conducted on the biochemical and
physiological effect of the grandiflora-multiflora gene
locus on petunia growth and development.

Multiflora and grandiflora petunias have been
differentiated by their growth habits and flowering
characteristics (Ewart, 1963). Multiflora petunias

generally have numerous, small flowers having small

calyxes with long, narrow sepals and slender, delicate
filaments, whereas grandiflora petunias have a small number
of large flowers with large calyxes, short, broad sepals
and short, thick anther filaments. Also, grandiflora
plants usually have a lighter green foliage than multiflora
ones.

The inheritance of the grandiflora—multiflora trait
has been investigated by several researchers. Bianchi
(1959) concluded from his work with the variety "weiBe
Wolke" that grandiflora and multiflora characteristics were
determined by alleles of a single gene, g and g respectively.
He also found that grandiflora homozygotes exhibited
various degrees of sub-lethality caused perhaps by the
apparently reduced chlorophyll content. However, this sub—
lethality could not explain deviations in the expected
monohybrid segregation. Thus from further back-cross
studies, he concluded that this pronounced certation effect
arose from a linkage of self-sterility alleles with the
alleles determining flower size.

Reimann-Philipp (1962) confirmed Bianchi's findings
but found no linkage between the self-sterility alleles
and those for flower size. He abscribed the lower number
of seeds to the action of a zygotic lethal factor (1).

Many individuals of the variety "WeiBe Wolke" showed a
linkage between E and l.
Seidel (1962) demonstrated that the g locus in

superbissima petunias (tetraploids) and in diploid

grandifloras are the same. Ewart (1963) confirmed the
single gene mode of inheritance for the grandiflora-
multiflora characteristic and also concluded that lethal
and sub-lethal alleles may be closely linked with the fi
allele resulting in a weak class of homozygoous dominant
petunias.

Chlebowski (1967a) found that the gene determining

flower size in P. hybrida Vilm. and in g. axillaris, one

 

of its ancestors (Ferguson and Ottley, 1932 and Natarella,
1974), were in fact the same. Later, Chlebowski (1967b)
demonstrated that fimbriate borders which is dominant over
smooth edges and green petal margins were both weakly
linked to the grandiflora trait. However, these linkages
like those involving lethality are not universal to the
species and are found only in certain lines. Kline (1972)
stated that the linkage between g and the lethal gene(s)
has been broken in inbred breeding lines used for hybrid
seed production.

Because of the inherent weakness of the homozygous
dominant (fig) plants, they can be inbred only to a limited
degree and fig and fig genotypes are difficult to distinguish
phenotypically. Thus, a method for determining between the
homozygous (fig) and heterozygous (fig) grandiflora plants
would facilitate breeding procedures. Testcrossing of F2
progeny is time consuming and costly; therefore, if a
method of determining the genotype could be developed,

breeding time could be shortened considerably.

Determining the effect of the grandiflora-multiflora gene

on petunia plant growth and development would lead to a
better understanding of the physiological action of this
gene locus. Using several of the methods used by Natharella
and Sink (1973) in determining the physiological effect of

a monogenic trait (double-single), an attempt was made to
determine such differences due to the grandiflora-
multiflora gene and their possible relationship to the

phenotypic characteristics of these genotypes.

STATEMENT OF PROBLEM

In the past few years there has been an increased
emphasis on genetic research of higher plants aimed at
elucidating the manner in which a specific gene mediates a
phenotypic trait through changes in biochemical processes.

Studies have been made on the inheritance and
phenotypic anatomical differences of the multiflora-

grandiflora character in Petunia hybrida Vilm. However

 

the relationship of genotype and plant morphology,

nutritional levels and physiology needs to be determined.

THE INFLUENCE OF THE GRANDIFLORA-MULTIFLORA
GENOTYPES OF PETUNIA ON SEED GERMINATION,
SEEDLING GROWTH, AND ELEMENTAL FOLIAR

COMPOSITION

INTRODUCTION

Petunia hybrida Vilm. varieties may be divided into

 

grandifloras and multifloras by their growth habits and
flowering characteristics. Multiflora plants generally have
dark green foliage and a large number of small flowers with
small calyxes and long, narrow sepals and slender filaments;
the grandifloras generally have lighter foliage, smaller
numbers of flowers with calyxes with short, broad sepals
and short, thick anther filaments (Ewart, 1963).

Several studies (Bianchi, 1959; Chlebowski, 1967a;
Ewart, 1963; Ferguson and Ottley, 1932) indicated that
gradiflora and multiflora characteristics were determined
by a single gene, g and g alleles respectively, and that
homozygous fig showed degrees of sub-lethality perhaps due
to lower chlorophyll content. Bianchi (1959) observed a
certation effect in addition to the sub-lethality. He
concluded certation arose by linkage of self-sterility
alleles with alleles determining flower size.

Reimann-Philipp (1962) found no linkage between the
self—sterility alleles and flower size and abscribed the

reduced number of seeds to a zygotic letha factor 1 (normal

allele g) which often reduced fertilization by pollen

tubes which carried it, so that its function could also

be explained as certation. He concluded that the low number
of grandiflora homozygotes was due to the sub-lethality of
the genotype fig caused by the chlorophyll defect and the
linked zygotic lethal factor.

Ewart (1963) also concluded that lethal and sub-
lethal alleles may be closely linked with the dominant fi
resulting in a class of weak homozygous dominant petunias
and further suggested that alleles of gene(s) controlling
vigor may interact with the large flower-viability gene
linkage.

Seidel (1962) showed that the g locus determining
large floweredness in superbissima petunias (tetraploids)
and in diploid grandifloras was the same. The genes
determining flower size in g. hybrida grandifloras and in

P. axillaris were found to be the same locus by Chlebowski

 

(1967a).

Green margins of petals in g. hybrida grandiflora
and E. hybrida vulgaris (multiflora) also appeared to be
linked with the grandiflora character (Chlebowski, 1967b).
This linkage, like that involving lethality and fimbriate
borders (Chlebowski, 1967b), is not universal to the
species but is found only in certain genetic lines. Kline
(1972) stated that the linkage between g and the lethal

gene(s) has been broken in breeding lines.

Hence, the grandiflora gene is a monogenic
inherited characteristic and it controls some as yet
unknown physiological action, the type of growth and
flowering of petunia plants. The research reported herein
was conducted in an effort to better understand the influ-
ence of this gene locus on the physiological aspects of
seed germination, seedling growth, and elemental content

of the vegetative leaves.

MATERIALS AND METHODS

Three sets of g. hybrida Vilm. each composed of a
homozygous multiflora (gg), homozygous grandiflora (fig),
and their hybrid (fig), were obtained from Harris Seed Co.
(Table 1). Seed germination was determined by placing 50
seeds on Whatmann No. 1 filter paper moistened with 5 ml of
deionized-distilled water in a 10 cm petri dish. The
dishes were placed in a chamber at 16 hrs light: 8 hrs
dark at 26°C and 22°C respectively. The number of germinated
seedlings was recorded daily beginning the second day, con-
tinued for 8 days and finally at 14 days. In this study,
the standard deviation for each line was computed and the
means of the three lines of each genotype were compared
using the Studentized t test (Steele, and Torrie, 1960).

In the remaining experiments, the nine lines were
considered as treatments and then partitioned into three
subclasses: genotypes, sets (each comprised of the three
genotypes) and the set by genotype interaction. In cases
where the genotypes were also significant in the AOV, the

mean separation for the entire genotype (the three lines)

10

11

Table l.--Petunia hybrida Vilm. breeding and hybrid lines.

 

 

 

MSU Line Genotype and Set Description
750 99'1* Salmon Perfection
751 GG—l FII for Salmon Perfection,

very weak

752 Gg-l Hybrid, Salmon Perfection,
fringed, glowing salmon

753 99-2 Ace of Hearts

754 GG-2 FII for Ace of Hearts weak

755 G9'2 Hybrid, Ace of Hearts,
bright red

756 GG-3 FII for Roulette, weak

757 99-3 Roulette

758 Gg-3 Hybrid, Roulette, red and

white bicolor

 

*gg--multiflora

G---grandiflora

12

were compared. Tukey's hsd test at the 0.01 level was
used for mean separation (Steele and Torrie, 1960).

Plants for experimental purposes were obtained by
sowing the seeds on dampened peat-lite mixed with terra-lite.
After 2-3 true leaves appeared, 15-21 days from sowing, the
seedlings were transplanted into 2.5 X 2.5 cm plastic pots
filled with dampened peat-lite mixed with terra-lite.

These flats (32 pots/flat) were watered on alternate days
with 10 ml per pot of half-strengh Hoagland's nutrient
solution (Hoagland and Arnon, 1950) and water. The plants
were grown in a greenhouse at 22-27°C and with supple-
mental light when necessary to extend the day length to

16 hrs to induce flowering.

For growth data, seed was sown at two different
times; experiment 1 on 12 April 1972 and experiment 2 on
3 May 1972. The fresh and dry wts and no. of leaves were
determined at 28 days from sowing and at 49 days the same
parameters and the plant ht were recorded. In experiment 1,
4 plants were sampled at 28 days and 2 at 49 days for each
line studied; whereas, in experiment 2, 8 plants were
sampled at 28 days and 5 at 49 days. Each plant was con-
sidered a replicate.

Elemental analysis was conducted for the vegetative
foliage of mature plants according to the procedure of
Kenworthy (1960). Petunia seedlings grow in two stages,
the vegetative leaves are alternate and when flowering

begins, opposite leaves are produced. The alternate

13

leaves are usually larger than the opposite leaves and at
the stage sampled, they are also more abundant; therefore,
these were used for analysis. Three replicates when

obtainable were analysed for 10 major and minor elements.

RESULTS AND DISCUSS ION

The seed germination study comparing the multiflora
and grandiflora genotypes appears in Fig. 1. These results
are representative of 5 sets, each consisting of the three
genotypes, which were observed in a preliminary study. At
three days after sowing, the percent of seeds germinated in
lines having the same genotype were not statistically
different from one another and were thus combined. However,
the percent germination in the three genotypes were signifi-
cantly different from each other; the fig genotypes had
higher rates of germination than either fifi or gg. Maximum
germination occurred eight days after sowing and the gg
genotype was 78.4% of the hybrid fig while fig was only
46.0% of the hybrid (Fig. 1). These data indicated that
inbred multiflora and grandiflora breeding lines lacked
the vigor of their respective hybrids. Seed germination of
the hybrid appeared to be primarily due to heterosis in
contrast to the inherent weakness of the inbred lines.

In both experiments on seedling growth (Tables 2
and 3) lines or treatments were significantly different but

not genotypes for 28 days. This variance was due to set

14

15

.MHCSUTQ ca mommuocmm MHOHMHDHUE
can muoHMHccmum on» on mflnmcoflumamn CH coflumcflaumm Doom usmoumm

.H musmflm

4..

 

OO-

16

+9
1..

 

 

 

l l

O O O
O 0 V’

80338 OBIVNIWHBS :IO °/o

 

20"

 

2345678

l4

DAYS AFTER SOWING

l7

.omm m.>mxssuam>oa

Ho.o may pm OGONOMMHU MHUGMUAMHcmHm no: mum mnmupma mfimm may mcw>mn msmm2«««

.musmam N mo msmmz««
.mucmHm v we mammz«

 

 

 

mm.a ommm.o maooo.o ¢m¢.v Ho.o
mooe.o omm
o bbbhhw o ovom.na
o ov.m mmmm.e on eeeo.o mvmm.na on Hmme.o mumo
o mm.HH mmam.e on omeo.o omem.me Ommeme.o mumo
c om.HH mmem.e on meeo.o omsm.me on veem.o Hume
ntbbmmhb Olmwmwwm
o mm.e oqmm.o unwomoo.o mmm.s Ommmme.o mumm
Damm.m mvmm.o on mveo.o ome~.m on mnma.o Numm
c oo.ee mqmm.o o mwao.o mmsm.m o mamm.o H-mm
moomm.o momm¢.v
Dame.m mmme.o meeoo.o mmmv.m Haemo.o muww
mm~.m mmmm.o Ommeoo.o ommm.k OmHmOH.o Nuoo
mmo.s ooee.o ammo.o oemn.~ ...mmvmo.o Hugo
mEO CH Ramwmw av thmw mm NRm>MU av ¥m>MU mm
.«msmo me Om Dmm cum
#3 pecan us mun p3 ammum mmwuocmo
mGHHUTUM\mEmHm

 

.H pamEHummxm ca mmmhuosmm MHOHMAUHOE can muonacsmum
MOM Mame mv pm we Osman can mwmc me can mm um muz wuc can amoum mafiapmmmlu.m manna

l8

.omm m.>mxsauam>ma
Ho.o mcu um DCOHOMMHC mausmoHMHsmHm uoc mum mumupma mEmm Gnu mcw>ms mamm2«««

.musmam m mo mcmmz««

.mucmam m we mammz«

 

 

 

mm.e Amqm.o ammo.o omm.e nvm~.o Ho comm
e em.ma we moem.o O mnmo.o on mma.~e o mmmm.o mumo
o vm.oa we moen.o O mmmo.o on mme.me on Hmmv.o mnmu
c em.me moo evm>.o Ommmmo.o DO «mm.ea onmvmov.o Huwo
o vq.m mmeom.o osmao.o Ommve.m Dammam.o mumm
O mm.m Deon memo.o mnmao.o on mmH.NH onmhmmm.o mume
w mm.NH m memm.o ammomo.o o omm.me on emmv.o H.@@
n mm.m onmmomm.o moeeo.o mums.m Ommmom.o muoo
mmm.m con oqmm.o Emmeo.o Omemq.m Oomoam.o Nuoo
mmm.m woveq.o mvao.o ammo.m *«Rmheva.o Huoo

mac CA .Rmsmc av Amman mm .Rmmmc m4 «mama mm
*«mwmc as am umm can
u: unmam #3 who DB cmmum mmmuocmw
mcflaommm\mscum

 

.N usmEHuomxm ca mommuosmm MHOHMHDHDE can muoHMHccmnm
How m>MU mv um u: Osman cam m>MO mv Dam mwmo mm um mp3 amp Dam cmmnm mafiapmmmll.m Dance

19

and/or set by genotype interactions, independent of the
multiflora-grandiflora genotype influence. At 49 days, the
fresh and dry wts for each genotype were significantly
different. The fifi genotype was lowest in both parameters
followed by the gg genotype which was intermediate and the
hybrid (Gg) had the highest fresh and dry wts (Table 2).

In experiment 2, lines were significantly different
from one another but this was not due to significance of
the genotypes. Thus, the means of the three lines of the
same genotype were not combined as in experiment 1
(Table 3). However, in considering each set independently,
gg and fig genotypes were higher than the fig lines for fresh
and dry wts at 49 days. Plant ht followed the same trends
as in experiment 1, fig being the tallest followed by gg
and lastly fifi lines (Table 3).

The differences in vigor as indicated by this

growth data are related to genotype with F hybrids

l
accumulating up to twice as much fresh and dry wt as either
of the parents at 49 days. Since there was no interaction
between genotype and set, whether this increase is the
result of a more efficient genotype related metabolism or
other factors is unknown. The grandiflora inbreds, geno-
type fifi, were the slowest growing as indicated by their
reduced rate of increase in fresh and.dry wts, due perhaps

to the weakness of their genotype by lowered chlorophyll

and/or elemental content.

20

Although significant differences were observed
between various lines used in the study for the other
elements, only those elements which had significant
differences at the 0.01 level for genotypes as well as
lines were investigated in further detail. Multiflora and
grandiflora genotypes indicated significant differences in
the % Mg and Ca in vegetative leaves of mature plants
(Table 4). In all sets, fifi was lower in Ca than gg and
in all but the second set from fig lines. The % Mg in sets
2 and 3 for the fig genotypes was intermediate between the
fig and gg lines but not statistically different. The fig
and gg lines in sets 2 and 3 were different with gg having
the highest Mg content. In set 1, the 3 genotypes did not
differ in % Mg (Table 4 and Fig. 2).

When the elemental analysis was repeated with
plants grown at a later date, significant differences were
observed for the elements Ca and Mg and also for B. In
sets 1 and 3, the fig lines had a greater % Ca than the fig
lines; the gg lines had the highest % Ca of the three
genotypes and was greater than fig lines for sets 1 and 2.
For % Mg, the gg lines had a higher quantity than fig lines
in sets 2 and 3 and the fig line in set 3. In set 1, there
were no differences in % Mg. In sets 1 and 3, B (Ppm) was
higher in gg than in fig and fig of set 1 was greater than
any fig genotype. There were no differences in set 2 in B

content (Table 5 and Fig. 3).

21

.OCDAOMMHC >Hucmoflmecmfim uos mud umuuma DEMm we»

mcfl>mc mcmmz«

 

 

 

 

Ho.o

owoc Omoc Omog Omoc Omoc Omoc Omoc MNOO ”MOO Omoc Omoc Owl: 0mm
moe mm 0.0m o.me om em mmem unmoa.e we em.m meq.o mm.o oo.m m-mo
no om H.mm H.mH mm om meme Omvo.H on nv.m moq.o o~.n om.m m-mo
em me H.qo m.vH mm mm Hemm mmm.o mco ow.m omm.o Sn.m on.m H-mo
BAH mm v.4v o.o~ mm mm ommv O ev.a m Sq.m moo.o em.n no.m m-am
ooe me «.mv H.mm OOH «ma Heme co cm.a m vm.m ome.o oo.e ev.q mnmm
mm om m.oo ~.nH Hm wee meme monmom.e um ne.m amm.o am.v Hm.m H-mm
AHA we m.mv m.me om oHH mvmq non mm.e numm.m «Av.o eo.m mo.v m-oo
med me m.mv n.9H mo wee mmmo nonmo~.H non qm.~ meq.o mm.m mo.m N-oo
mm he p.mv o.m~ we om mmmm Omoe.a «moo.~ mvm.o Hm.o oe.q ence
Ha cm m so we :2 «2 as no a x z umm new

5mm mammn #3 app :0 ucmEmHo mo unmoumm wmmuocmo

 

.H quEHuomxm
Ca momxuocom MMOHMHUHDE cam MHOHMHUGMHO mo wmcwacmmm UHO zap av mo mfimxamsm Hmucmamamun.v manna

Figure 2.

22

Relationship of Boron, Magnesium and Calcium
content in vegetative tissue and grandiflora
and multiflora genotypes of petunia in
experiment 1.

 

 

23

 

 

 

N.S.
had 00!
4.0 -
3.0 -
Ca % 20 __ Ihsd o-ou
I.O -

 

 

 

GG-I 66-2 66-3 99-l 99-2 99-3 69-I 69-2 69-3

24

.omm m.>mxdenao>ma Ho.o on» um ucmummmwp mauemoflwflcmflm no: mum umuumfl

Dawn OOH msH>MO mammz«

 

 

 

 

Ho.o

.m.: .m.a e.ma .m.: .m.: .m.: .m.a hm.o mo.o .m.c .m.: .m.c Omm
mom Mme Omm.vv m.m on Noe omem o ev.a co mo.m mm~.o m~.o vm.m mnmo
«OH Ame DOmo.ov o.m mo an mmmv OmmO.H O ne.m mmm.o mo.~a mm.m mumo
Ame NNH cu m.mm m.v mm mm mmmm mHo.H oO oo.m nem.o mo.m sm.m Hume
ONO mqe 0O m.mm o.H nm em moom o mn.O o mm.m mqm.o «m.m Oo.~ mumm
OMO OMO coO m.em o.vO mo woe movm o nw.H no mm.m mm~.o «m.m sm.m «loo
Ome om u m.mw o.m mo om «new DOmvm.O co em.m mne.o OH.S mv.m Hume
«we woe Hm.nm m.m am we mmmm o ov.O OEmm.N Omm.o «m.m ov.~ muoo
ONO «HO oOmm.n¢ o.n no me emom DO mm.H O mm.m mom.o mm.m oa.¢ muoo
owe mo Om~.vv o.m on me oven OmmH.O «moo.m vmv.o NA.OH mm.m aloe
Ha mm m so we Oz 62 a: no A M 2 Omm can

wmwuocmo

Ema mammn #3 map so quEmHo mo unmouom

 

.N ucwfiflummxo

cw mwmxuocmm UHOHMHHHSE cam MHOHMHpcmum mo mmcflapmmm 6H0 wmc mv mo mammamsm Hmucmfiwamnl.m magma

25

Figure 3. Relationship of Boron, Magnesium and Calcium
composition in vegetative tissue and grandi-
flora and multiflora genotypes in petunia in
experiment 2.

B-ppm 50

M9 “/0

C0 70

7O

60

4.0

3.0

2.0

LC

26

   
  

In“ con

 

 

 

In“ O-OI

1

 

 

GG-I 66-2 66-3 99-l 99-2 99-3 69-I 69-2 69-3

27

Ca was generally highest in lines with the gg
genotype, lowest in fig lines while Mg also followed this
pattern. Ca is involved in the maintenance of cell membranes
and functions in ion transport especially in chloroplasts
and mitochondria. In fact, these organelles are sites of
Ca accumulation along with Mg (Epstein, 1972; Larkum, 1968;
Noble, 1967; Stocking and Ongun, 1962). Magnesium is a
primary constituent of chlorophyll as well as being active
as a co—factor in numerous biochemical pathways, especially
those involving phosphate transfer. Unlike Ca, this is a
moble ion, readily translocated from older tissues to young
meristematic regions (Epstein, 1972). Boron is involved in
the translocation of sugars from leaves and especially in
switching the degradation of glucose to either glucolysis
or the pentose shunt (Epstein, 1972). Whether or not any
of these differences in ion concn is of physiological
importance cannot be determined from the above data alone
since in most cases, these elements are ordinarily present
in super-optimal concn in the plant.

Thus, differences in Ca, Mg, and B may be related to
chlorophyll content and/or metabolic differences. Since in
most cases, Mg is eSpecially high in the gg lines, this
may indicate that these plants do in fact have a higher
concn of chlorophyll than do the fig lines. Whether these
differences are the result of the grandiflora-multiflora
locus and a direct effect of these alleles, cannot be

definitely determined by this study. Studies dealing with

28

the actual chlorophyll content of these genotypes may
partially answer this question and will be pursued in
future studies. However, it must also be recognized that
these differences may be the result of other genes common
to the three sets and which may maSk or act synergisti-
cally with this particular gene locus especially since
these differences in elemental content were not consistent
in all three sets. The elemental values obtained for the
hybrids do closely resemble those reported by Carlson and
Carpenter (1972) for the grandiflora hybrid "Pink Magic"
as optimum, and thus the fact that similarities did appear
within all of the gg, fig, and fig lines indicates that
regardless of their genetic differences, the basic
physiology was very similar.

Although multiflora petunias appear to have greater
vigor than grandiflora inbreds as shown, both the germina-
tion and growth studies, the hybrids surpassed these in
terms of seed germination and early seedling growth.
Except for this difference in viability and vigor however,
all of the three genotypes appeared to have generally the
same basic physiology as indicated by the levels of

elements.

INFLUENCE OF THE GRANDIFLORA-MULTIFLORA GENOTYPES
OF PETUNIA ON TOTAL CHLOROPHYLL, STARCH, AND
SUGAR CONTENT AND ON THE QUALITATIVE SUGAR

CONTENT IN VEGETATIVE TISSUE

29

INTRODUCTION

In the past few years there has been an increased
emphasis on genetic research of higher plants aimed at
elucidating the manner in which a specific gene mediates a
phenotypic trait through changes in biochemical processes.
In these studies, there are two aspects to be considered:
(1) determination of the inheritance of a particular
phenotypic character to be investigated and (2) determine
physiological or biochemical processes which correlate
with the character.

In Petunia hybrida Vilm., multiflora and grandi-

 

flora cultivars, can be differentiated by their growth
habit and flowering characteristics. Multiflora petunias
generally have numerous, small flowers and calyxes, with
long, narrow sepals and slender, delicate anther filaments;
whereas, grandiflora petunias generally have a small number
of large flowers with large calyxes and short broad sepals
and short, thick anther filaments. Also, the grandiflora
genotypes generally have lighter green foliage than multi-
floras.

Inheritance of the grandiflora trait has been

investigated by several researchers. Bianchi (1959)

30

31

concluded that grandiflora and multiflora plant character-
istics were determined by alleles of a single gene, g and
g respectively. He also found that grandiflora homozygotes
show varying degrees of sub-lethality caused perhaps by the
apparently reduced chlorOphyll content they exhibited.

This sub-lethality could not explain deviations in the
expected monohybrid segregation thus, from further back-
crossing studies, he concluded that this pronounced certa-
tion effect arose by linkage of self-sterility alleles with
alleles determining flower size. Reimann-Philipp (1962)
confirmed Bianchi's findings but found no linkage between
self-sterility alleles and flower size and abscribed the
lower number of seeds produced in grandiflora homozygotes
to the action of a zygotic lethal factor (1). Ewart (1963)
confirmed the mode of inheritance and also concluded that
lethal and sub-lethal alleles may be closely linked with
the fi allele resulting in weak homozygous dominant grandi-
flora petunias.

Chlebowski (1967) demonstrated that fimbriate
borders and green petal margins were both weakly linked to
the grandiflora trait. However, these linkages that
involve lethality were not universal to the Species and
were found only in certain lines. Kline (1972) stated that
the linkage between g and the lethal gene(s) has been
broken in breeding lines. Thus, the mode of inheritance
of the grandiflora condition has been well established by

these investigations. In a previous study, Knowlton and

32

Sink (1975) found that the basic physiology of grandiflora
and multiflora genetypes was similar based on the levels

of elements in vegetative tissue and seed germination
studies. This paper will present the results of biochemical
studies conducted to determine total chlorOphyll composition
and total free sugar and starch as they relate to the
grandiflora-multiflora genotypes and qualitative evaluation

of soluble sugars by thin-layer chromatographic analysis.

MATERIALS AND METHODS

Nine lines of E. hybrida Vilm., divided into three
sets, each including a homozygous multiflora gg, homozygous
grandiflora fifi and hybrid fig were used for this study
(Table 6). The seeds were sown on the surface of moistened
peat-lite mixed with terra-lite (Jiffy Mix). After 2-3
leaves appeared (15—21 days from sowing), 16-32 seedlings
per line were transplanted into 2.5 X 2.5 cm plastic pots
filled with peat-lite mixed with terra-lite. The plants
were watered with half-strength Hoagland's nutrient
solution (Hoagland and Arnon, 1950) and at all other times
with tap water. The plants were grown in the greenhouse
with supplemental light used, when necessary, to provide a
16 hour photoperiod to promote flowering. Three groups of
seed were sown at different times of the year; experiment 1
on January 23, experiment 2 on May 4, and experiment 3 on
September 15. Vegetative leaf tissue was taken after
flowering had begun (75-100 days from sowing).

The chlorophyll, sugar and starch extraction
technique was modified from that of McCready, et al.

(1956). Alternate vegetative leaves were selected at

33

34

Table 6.--Petunia hybrida lines utilized in this study.

 

 

 

MSU Line Genotype and Set Description
750 gg—l* Salmon Perfection inbred
751 GG-l Inbred Salmon Perfection,

very weak

752 Gg-l Hybrid, Salmon Perfection,
Fringed, glowing salmon

753 gg-2 Ace of Hearts inbred

754 GG-2 Inbred Ace of Hearts, weak

755 Gg-2 Hybrid, Ace of Hearts,
Bright red

756 GG—3 Inbred Roulette, weak

757 gg-3 Roulette inbred

758 Gg—3 Hybrid, Roulette, red and

white bicolor

 

*gg--multiflora

G--grandiflora

35

random from plants of each line and placed on ice. In
experiment 1, approximately 1 gram of tissue was used
whereas in experiments 2 and 3, 20-l.2 cm. diameter discs
were punched out of the leaves and weighed. Three samples
were analyzed (Fig. 4) in each experiment for each of nine
lines. The tissue was placed in centrifuge tubes containing
10 ml of warm (SO-55°C) 80% ethanol. Each sample was

ground in a Tembroek homogenizer for 3-5 minutes and returned
to the centrifuge tube, the mortar and pestle rinsed into
the tube, 2X with 3-5 ml aliquots of 80% ethanol. The

tubes were placed in a 50-55°C water bath for 5-10 minutes,
centrifuged for 10 minutes at 10,000 X g and the supernatant
decanted into a 50 m1 test tube. Ten m1 of hot 80% ethanol
was added and the pellet, suspended and centrifuged as
above. The ethanol washing was repeated and the combined
supernatants stored under N2 gas at 2-4°C until the
following day.

Five m1 of distilled water and 6.5 m1 of 52%
perchloric acid were added to the pellet in the tube,
stirred occasionally for 20 minutes, then 10 m1 H20 was
added and centrifugation done at 10,000 X g for 10 minutes.
The supernatant was decanted and the residue treated again.
After 20 minutes, the residue and solution was washed into
the 50 ml tube and the combined sample passed through
Whatmann #1 filter paper on a Buchner funnel. The liquid
was brought to 50 ml volume, stoppered and stored in the

cold (2-4°C) until analyzed for total starch.

36

 

.MUHHO>£ mHaduwm

mo OSMMHH w>Humummm> OH mummsm Dam .OUHMUM .HHhOmonoHOo mo MHmmHmsm O>HumuHusmsv pew
.UHB .coHuommu wacuspcm How muomuuxo chuno 0» com: mouspmooum COHuomuuxm cam msHHmEmm .v musmHm

OOHuomou ocounuqfl
mHmmHmsm m>HumuHucmso

omm _
OOHz HE om cu Ho> mcHHO

HOOHHM new OGHOEOU

 

OOHummDMHHucoo

uHEo use

was“ umwu newEumouu

HE om oucH usom m>onm ummmmu
ucmumcuomsa4 umHHmm

 

m x ooo.OH Om :Hs 0H

How omDMHHucoo 0cm ON: HE OH

pom can» .OHE om Mom mHHMOOHmmooo
mean can meow oOOoHOoOmm «mm as m.m

.on HE m pea .Ooumum cam MHHOOO
mo MHsz HHOU mchucoo umHHmm

ADEHu pcoowm m

ummmouao>onm mm mmSMHuucwo

cam moum wow OoO as mHuOH nnmv
umHHom

 

UHF

 

 

as m.mvow com um wmmu
moum wow OOH; Ha ooa op Oo> HOHAO 6:6
HS om cu Ho> mcHHn vOmNmz Omsounu who

 

wmmOm Meson mmmnm Home:

 

OHS om Ucmum

OOH .umzum Edeouumm

HE ov nqu coHuHuumm
mHHmsmouoHOo
+ mummSM msHmucoo
Oomuuxm moum

AmCOHuoUum
ucmumcummsm m manEoov

ucmumcuwmsm

 

m x ooo.OH um cHE OH omOMHuucmo .CHE mum
umuuos xmoanma 6 OH cceuo .moum wow Roommuomv
uon HE 0H cum .mnsu coauHuucwo HE mm H OH momHm

 

 

AmUMHp ED ~.H|om no Emum HV
mamMHB MOTH O>Hp6pmmm>

37

To remove the chlorophylls, the 80% ethanol extract
was partitioned against 25 ml of petroleum ether. After
20-40 minutes, the lower phase was removed and washed
again with 25 ml of petroleum ether. The ether phases

were combined, filtered through dry Na $04, brought to

2
50 m1, and read on the spectrophotometer at 660 and 642.5
nm to determine the amount of total chlorophyll using a
standard formula (Association of Official Agricultural
Chemistry, 1965).

The ethanol sugar fraction was brought to 50 ml
with 80% ethanol and 10 ml removed for thin-layer chromato-
graphy (TLC). The remaining 40 ml were evaporated, brought
to 50 ml with distilled water and placed under N2 in the
cold (2-4°C) until assayed.

The anthrone reaction was used to determine sugar
and starch content in the ethanol and perchloric acid
fractions, reSpectively (McCready et al., 1950 and Clegg,
1956). In both cases, 1 gram anthrone was dissolved in
500 ml concentrated H2504. Duplicate aliquots (0.2-0.5 ml)
were used for each sample and 5 ml of the anthrone-sulfuric
acid reagent was added to each sample aliquot, stirred and
placed immediately in ice to stop the reaction. The
samples and standard aliquots were heated in a boiling
water bath for 7.5 minutes, cooled in ice and the %
transmission read at 630 nm on a Bausch-Lomb colorimeter.

The standard curve was computed from glucose standards

using linear regression equation and least squares

38

techniques after converting % transmission to O.D. using
a conversion table solution of this equation for the X

variable by the method of linear calibration where
X = Ug glucose equivalents/aliquot and y = O.D.

of the sample or standard was computed (Snedecor and Cochran,
1967).

Thin-layer plates covered with 250 nm of silica gel
G (Uniplate) were divided into 12 1.2 cm sections. The
sugar standards were prepared as 1 ug/ul concentrations
in 95% EtOH. The extract samples were reduced to dryness
and brought up to 5 ml with 95% ethanol. Fifty ul of each
reduced extract of each line was streaked onto the plates
as well as 25-50 ul of the standards. The plates were run
ascending in a solvent system of Acetone: Water: Chloroform:
Methanol (75: 5: 10: 10) to about 15 cm (Lewis and Smith,
1969). After air drying, the plates were sprayed with a
phenol-sulfuric acid reagent (Lewis and Smith, 1969) and
heated 10-15 minutes at 100°C. The chromatograms were also
observed under long and short UV light and the color and
position of each spot recorded.

A difference in TLC patterns for the gg and fig
lines versus fig in the center (Rf .49-0.53) of the plates
was identified; new plates were run in each experiment and
the middle third (Rf 0.33-0.67) of these plates scraped
off and eluted several times with 80% MeOH. These samples

were then dried and the residue treated with TMS

39

(trimethylsilyl) to prepare ethers for gas-liquid chromato-
graphy (GLC). In all cases, duplicate GLC runs were made
(Farshtchi and Moss, 1969). The column used was U-shaped
glass 6 ft X 15 mm in diameter, packed with 2% OV-l on
80-120 mesh Chromosorb. Temperatures were: Detector 225°C,
injector 225°C, and column 160°C. Nitrogen inlet pressure
was adjusted to 40 ml/min and was used as the carrier gas.
Hydrogen was adjusted to 40 ml/min and compressed air (02)
to 400 ml/min.

The compound that appeared on the TLC plates in
the gg and fig genotypes was also delineated in the GLC data
but the amount of compound in the tissue extracts was not

sufficient to quantitize the unknown compound.

RESULTS

In experiment 1, total chlorophyll content was
significantly higher in leaf tissue in the fig genotype
lines than in the gg or fifi genotype lines (1.20 mg/g
fresh weight vs. 0.98 and 0.77 mg/g fresh weight),
respectively (Table 7). However, in the remaining experi-
ments, no significant differences in chlorophyll content
were found (Tables 8-11) as determined on either a fresh
weight or leaf area basis.

Total free sugar on a leaf area basis was not
significantly different among the three genotypes in any
experiments. There were, however, significant differences
in sugar content on a per fresh weight basis in experiments
2 and 3. In experiment 2, the sugar content in lines with
the fig genotype (9.22 mg/g fresh weight) was significantly
higher than that in lines with the gg genotypes (6.38 mg.g
fresh weight) while that of the fig genotype was not signifi-
cantly different from either of the other two genotypes
(Table 8). In contrast, in experiment 3, fig lines had the
highest sugar content of the three genotypes and was

greater than that of the gg lines (10.95 mg/g fresh weight

40

41

.DDGDHDMMHU OOODHMHOOHM o: n .m.c

.Omm m.>mxsa HD>DH AH DOD
Om OODODMMHU uHucmDHchmHm uoc mum AN .h .xv HDOODH Damn DO» >O @D30HHOM mcmme OEOHOUO

.Omm m.>ste HD>DH we
TOO Om OODODMMHO wHucmDHchmHm Ooc mum Amnmv HDOODH Dawn TOO mO OD3OHHom mcmme quHoum

 

rmcmmac Ho.o

 

 

 

.m.c mn.o mH.o 0mm
Ho.o
vm.m mm.H vm.o 0mm
mv.m .N No.mH .N om.H
OD ov.h DO oo.mH D om.H mmn mumo
D oo.OH DO mm.mH DO mm.H mme Nico
DOD om.h O mH.HH DDO mmo.H mmn Hume
No.5 .x mv.m .x mh.o
OD mm.h m om.> m mo.o om» mloo
DO hm.m D mm.m DOM om.o «mp mime
m hm.m m mm.m m v5.0 Hmn Hnoo
mm.m .s A~.OO O.» mm.o
DO e~.m O Hm.oO OD mm.o Ame mumm
DOD NH.m O «O.DH DD OO.H mmn Numm
OD Dv.e DO DH.MO mDDOD Ho.O omh O-DD
OOmHmz Omoum m\DmoosHm OOmHmz Omwum m\DmoosHm uOchs Omwum c\ms DOHO Dom
muOm>HHHscm ma mucm>HHHsvm we HH>OmouoHOU Dam Dehuocwo
ummsm Ooumum

 

.H OODEHODQND
OH DOODHQ mo DDDDHO wmmH D>Humummm> mo ucmucoo Hamsm Dam .Ooumum .HHMOmouoHOD Hmuoelu.n DHOMB

42

.TDOTMTMMHO ucmDHMHOmHm on n .m.c

.omm m.»mxse OD>DO DO DOD

um OOTHTMMHO mHOOTDHMchHm uoc Tum A.>.xv OTOOTH Team TOO mO OTBOHHOM mOTTE OEOHOUO

.Omm m.>ste OD>DO OH
TOO Om OOTHTMMHO mHucmDHMHcmHm O0: Tum ATIDV OTOOTH Team TOO hO OT3oHHom TOOTH OESHOUT

 

 

 

 

.Ho.o
vo.N .m.c .m.c 0mm
Ho.o
vm.m .m.: .m.: 0mm
.A.x ov.m om.n no.0
.Ommwwmum. DD.OH AD.o mme mumo
ODOD mv.h mm.b mo.H mmb mnmo
DD mm.0H oo.m mH.H mmn Hume
.m mm.m om.m mm.o
DDO om.m 55.0 m>.o emu mnoo
OT mv.o NT.HH mm.o «me «:00
D o>.HH ~H.b Hm.o Hmn Hume
O.x mm.o hm.m m>.o
D ~>.v mv.q vv.o Ame mnom
D mm.m Hm.m qn.o mme Numm
DDDO mo.m he.m DN.H ome O-DD
OOmHTz OmTum m\TmoosHm OOmHTS OTTum m\TmoosHm OOmHTB OmTum m\mE TOHH uTm
TOOT>HHH5Um m8 TOOT>HHHsvm ms HH>OmouoHOU COT meuocTo

ummsm ODHTum

 

.mHmmO OOmHTznOmTum m
:0 N OOTEHHmeT OH TOOTHQ mo TOOTH» MOTH mo OCTOOOD Hmmsm DOT .Ooumum .HHwOmouoHOD Hmuoanl.m THOTB

.TDOTOTMMHO OOODHMHOmHm 0O u .m.O

.Omm m.mmxse HT>TH OH
TOO OO OOTOTMMHO mHOOODHMHOmHm OOO THO ATIOV OTOOTH TEOm TOO wO OT3oHHOm TOOTH OEOHOUO

 

43

 

.H0.0
.m.O .m.O .m.O emm
H0.0
.m.O .m.O mH.0 0mm
OH.m nm.mH mm.0
N0.0 mm.mH O mv.0 wmn mnmo
OH.O H0.0H DO 00.0 mmn Nico
n>.m 00.9 D «0.0 mmn H100
mm.m n~.vH mv.0
0n.m mO.HH O 00.0 Omn mnoo
mv.v 0m.OH D no.0 emu muoo
Om.m m~.vH O 00.0 Hmh H100
mH.O mm.m 0m.0
ov.m Hm.m .D mm.o ems muom
vm.m OD.MH Om ov.o Mme «-mo
om.m ma.m DD mn.o omn Humo
Omm¢\TmoosH0 «mmO\TmoosHm Omm¢\mfi TOHH OTm
mOOO>HHHOUm ma mOOO>HHHOvm ma HHwOQOOOHOU OOO TthOOTo
OOmOm ODOOOm

 

.mHmOO OTHO MOTH O co
m OOTEHOTQNT OH mOOOHm mo TammHO MOTH mo OOTOOOD OOmsm OOO .ODOOOO .HHmOQOOOHOD HOOoanu.m THOOB

44

Table 10.--Total chlorophyll, starch, and sugar content of
leaf tissue in experiment 3.

—_—. C. 1..--
.—

 

 

 

 

Genotype and Chlorophyll Sugar
Set mg/g Fresh Weight mg/Fresh Weight
gg-l 0.91 11.44 bca
2 0.82 5.44 a
3 0.76 5.34 a
0.83 7.40 x'b
GG-l 0.92 11.89 bc
2 0.99 8.20 ab
3 0.68 7.89 a
0.86 9.327 x'y'
Gg-l 0 83 14.24 c
2 0'93 12.11 c
° 6.50 a
0.76 10.95 y'
HSD0.01 n.s. 3.89
HSD n.s. 2.907
0.01

 

aColumn means followed by the same letter (a-c) are
not significantly different at the 1% level Tukey's HSD.

bColumn means followed by the same letter (x, y)
are not significantly different at the 1% level Tukey's
HSD.

n.s. = no significant difference.

45

Table 11.--Total chlorophyll, starch, and sugar content of
leaf tissue in experiment 3 on a leaf area

basis.

 

 

 

 

Genotype and Chlorophyll Starch Sugar
Set mg/AREA mg/AREA mg/AREA
gg-l 0.63 ab 12.46 7.97 bca
2 0.54 ab 3.15 3.62 a
3 0.62 ab 6.89 4.41 a

0.60 6.89 5.33
GG-l 0.51 ab 4.72 6.63 bc
2 0.74 b 6.43 6.06 ab
3 0.44 a 8.63 5.09 a
0.56 6.59 5.93
Gg-l 0.43 a 8.12 7.57 bc
2 0.50 ab 6.59 6.84 bc
3 0.39 a 3.98 3.34 a
0.44 6.23 5.92
HSD 0.29 n.s. 1.85
0.01
HSD n.s. n.s. n.s.
0.01

 

aColumn means followed by the same letter (a-c) are

not significantly different at the 1% level Tukey's HSD.

n.s. = no significant difference.

46

vs. 7.40 mg/g fresh weight) (Table 10). In this experiment,
the fifi lines were intermediate between the fig and gg lines
and were not significantly different from these two geno-
types (9.33 mg/g fresh weight) (Table 10).

Starch content also varied among genotypes in the
three experiments, in experiment 1, the total starch
content was greater in the gg and fig lines than in the fig
lines, while the fig lines were also significantly greater
than gg lines in starch content (Table 7). Another pattern
emerged in experiment 2, fig lines were again highest in
starch content followed by the fig lines and lastly by the
gg lines. The fig lines were higher in starch level than
the other two genotypes and the fig lines were higher in
starch level than the gg lines (Table 9). However, when
the starch content was placed on a fresh weight basis, no
significant differences were observed between the three
genotypes (Table 8). In experiment 3, there were no
differences in starch content among the three genotypes on
either a fresh weight or leaf area basis (Tables 10 and 11).

The Rf values for the sugar extract developed on
TLC plates showed that the gg and fig lines resolved an
additional band approximately Rf 0.52—0.56 and was absent
in the aliquots obtained from the fig lines (Table 12 and
Figure 5). The color of the spots and the standard sugars
are recorded on Table 12. The most intense Spot for each
line was at Rf 0.33-0.37 and corresponded to the glucose

standards sugar, Rf .34.

47

Table 12.--Rf values and color reactions of standard sugars
and extracts from grandiflora-multiflora petunia
vegetative tissue.

 

Compound R Spray

 

Number f UV (see text) Sourcea

l 0.01 green -- 99

2 0.16 green -- gg-l

3 0.36 -- yellow gg

4 0.50 -- green gg

5 0.54 -- green 99

6 0.20 green -- GG-l

7 0.35 -- yellow GG

8 0.50 -- green GG

9 0.01 green -' G9
10 0.18 green -- Gg-l
11 0.35 -- yellow Gg

12 0.50 —- green Gg
13 0.53 -— green Gg
14 0.60 grey -- Maltoseb
15 0.49 grey -- Arabinose
16 0.34 -- yellow Glucose
17 0.54 green -- Mannose
18 0.57 grey -- Sucrose
19 0.44 red pink Galactose

 

aAverage of three lines of each

duplicate plates.

b

genotype and

Standard sugar solutions 1 ug/ul ethanol.

Note: -- indicates no reaction under these
conditions.

48

.TsmmHO T>HOOOT0T> OHOOOTQ OOOHMHOHOE OOO OOOHMHOOOOm
ECHO mODOOOxT OOO OOOOOOOOO OOmOm mo EOODOOOEOOOD OTmOHuOHOO TOHmomEoo

.m THOmHm

HO!"

SOlVIN 1

49

e o
0,9 9

<——(O|80|8 53 Cl) IONVNIIHS HIOIOI01H38IIIVM8INOJIOV

 

 

fl

AIAIINOSI OWCOSI IAN NOS! SUCIOSI OALACTOSI

g! 1. ”no“

50

The GLC data indicated a peak in the fractions from
gg and fig lines but not in fig lines. This peak was found
in extracts from all three experiments and ranged in
retention times from 6.2 to 8.4 minutes depending on when
the samples were run. The samples were too small however

to characterize the unknown by running it with standard

sugar samples.

48

.TOmmHO T>HOOOT0T> OHOOOTQ OOOHMHOHOE OOO OOOHMHOOOOT
Scum mODOOOmT OOO OOOOOOOOO OOmOm mo EOOTOOOEOOOD OTmOHuOHOO TOHmomEoo

.m musmHm

DISCUSSION

In a previous study, it was observed that differ-
ences found in Ca, Mg, and B may be related to chlorophyll
content and/or metabolic differences (Knowlton and Sink,
1975). Also, apparent differences in the color of the
foliage tend to indicate that the fi locus in petunia may,
in fact, affect the metabolism of chlorophyll. Thus, it
appeared that determining the chlorophyll content of the
leaves would be a logical place to begin since the leaves
of the multiflora genotype petunias generally appear to be
much darker green in color than the homozygous or hetero-
zygous grandiflora lines. However, the results in all
three experiments (Tables 7-ll) did not support these
visual observations and inferences from previous research.
In experiment 1, the chlorophyll content was highest in
the hybrid lines on a fresh weight basis and not the
multiflora lines. This may be explained by heterosis of
the hybrid lines or interference by other compounds such as
carotenoids. However, in the other two experiments,
differences between genotypes either on a fresh weight or

leaf area basis were observed. This can be best explained

51

52

by consideration of the relationships of ChlorOphyll con-
centrations in the cells of the various genotypes. In
multiflora leaves, the chlorophyll concentration may be
high and thus the leaves appear deep green while in the
grandiflora genotype, the cells are larger and even though
there is a similar amount of chlorophyll, the leaves appear
lighter because the relative concentration is much lower
due to the larger volume of the individual cells. The
weight of these cells could very easily be similar in the
various genotypes if the change in volume were due not to

a larger number of oganelles but to a less dense cell sap.
The discrepancy observed in experiment 1, might be explained
by the more favorable environmental conditions under which
those plants were grown, especially favoring the hybrid
genotype.

Carbohydrate content was considered since it may
influence water retention of the cells and with the larger
leaves and flowers found in grandiflora lines, more energy
substrates might be required as increased levels of sugar
and starch. The starch content of the lines was different
in experiment 1 on a fresh weight basis and in experiment 2
on a leaf area basis but the results of these analysis were
inconclusive since the patterns were different in both
cases (Tables 7 and 9). The amount of starch accumulated
in cell walls is probably not necessarily correlated to
genotypes but more likely influenced by environmental

conditions and their complex interaction with the

53

grandiflora-multiflora genotypes. This integration is
regulated not by one gene but by many which may or may not
be linked with the grandiflora—multiflora locus.

Lastly, the sugar content in the three genotypes
was on the whole not different. And, again, different
patterns arose in the three experiments (Tables 7-11).
This also indicates that environmental effects are probably
responsible for these differences and that there is an
unknown optimum carbohydrate level of available sugars.

Even though there appeared to be no quantitative
differences in total chlorophyll, starch, or sugar content
that could be attributed to the presence of the g or g
alleles, a qualitative difference was observed in the form
of a sugar compound found both by TLC and GLC techniques.
This extra band was evident whenever the recessive g
allele was present, i.e., in the multiflora gg and grandi-
flora heterozygous fig lines. Thus, it appears that the
presence this allele in some way modifies the presence of
endogenous substrates of the cells in the vegetative leaves
in these genotypes. The gene—enzyme-substrate mechanism
for gene action might also be the place of modification.

Further work on identifying and isolating the
unknown sugar should be carried out using GLC with larger
samples and standard sugars, mass spectroscopy would also
be of value as indicated by the reaction of the unknown to

the anthrone reagent. It is relatively certain from

54

these results that the unknown substance is a carbohydrate
of low molecular weight and may possible be used in
identifying heterozygous grandiflora plants for breeding

purposes.

CONCLUS IONS AND RECOMMENDATIONS

As shown by the results of elemental analysis and
determination of total sugar and starch content of
vegetative tissue, grandiflora-multiflora breeding lines
and hybrids of petunia are physiologically similar. While
it appears that multiflora plants have darker foliage than
grandiflora plants, the amount of chlorophyll extracted
from the various genotypes was not significantly different
possible indicating that the density of the chlorophyll
varies in the cells of the two phenotypes rather than the
total amount present or could be modified by the presence
of other pigments. The unknown compound found by TLC
separation seems to be present only in sugar extracts from
genotypes carrying the recessive g allele. This allele
may code for a different amino acid than the fi allele and
thus result in a different protein structure that in turn
contributes to the production of a Specific sugar. Since
the molecular weight was not determined nor a positive
identification made using standard sugars, the relationship
to the plant's metabolism is unknown. It is also possible

that this unknown is, in fact, present in the fig genotype

55

56

but in amounts undetectable by the techniques used in this
study and that this characteristic is quantitative rather
than qualitative. The unknown may be an indirect factor
affecting cell division and density discussed above and
thus in turn could regulate the growth rate and germination
and vigor of the various genotypes.

Further studies on characterizing the unknown could
be conducted using paper chromatography techniques, GLC
and mass spectroscopy to further determine the compound's
structure. The role of the unknown in the plant's
metabolisms should also be investigated by using other
biochemical tests, assays, and histochemical techniques.
Cytological studies on leaf and other tissues that appear
to be larger in the grandiflora phenotypes than in the
multifloras could also aide in determining whether the fi

locus does in fact affect cell division and density.

LITERATURE CITED

LITERATURE CITED

Association of Official Agricultural Chemists. 1965.
Methods of analysis. W. Howitz (ed.).

Bianchi, F. 1959. Onderzoek naar de Erfelijheid van
Bloemvorm bij Petunia. Acad. Proefschrift ter
Vertrijging van den Graad Doctor in de Wis-en
Natuurkunde aan de Universeit van Amsterdam.

Carlson, W., and W. J. Carpenter. 1972. Optimum soil
and plant nutrient levels for petunias. Mich.
Florist. 496:16-18.

Chlebowski, B. E. 1967a. Genetic studies on Petunia.
III. Inheritance of the grandiflora character
in a cross between E. hybrida randiflora and
g. axillaris. Genet. P01. 8: 7-73.

 

 

Chlebowski, B. E. 1967b. Genetic studies on Petunia.
IV. Inheritance of green margins of petaIs In
B. hybrida grandiflora and g. hybrida vulgaris.
Genet. Pol. 8:75-97.

 

 

Clegg, K. M. 1956. The application of the anthrone
reagent to the estimation of starch in cereals.
J. Sci. Food Agric. 7:40-44.

Epstein, E. 1972. Mineral Nutrition. Lg E. Epstein
(ed.) Mineral Nutrition of Plants: Principles
and Perspectives. John Wiley and Sons, Inc.,
New York. pp. 285-322.

Ewart, L. C. 1963. The inheritance of flower size
in grandiflora and multiflora petunias. Ph.D.
Thesis. PA. State Univ. 73 P-

Farshtchi, D., and C. W. Moss. 1969. Tetracyano-
ethylated pentaerythriol: an efficient polar
liquid phase for analysis of trimethylsilyl
derivatives of sugars and sugar alcohols. J.
Chromotogr. 42:108-111.

57

10.

ll.

12.

l3.

14.

15.

16.

17.

18.

19.

20.

21.

58

Ferguson, M. C., and A. M. Ottley. 1932. Studies in
Petunia. Amer. J. Bot. 19:385-405.

Hoagland, D. R., and D. I. Arnon. 1950. The water
culture method for growing plants without
soil. Calif. Agric. Expt. Sta. Circ. 347 p.

Kenworthy, A. L. 1960. Photoelectric spectrometer
analysis of plant materials. Report presented
at the annual meeting of ASHA. Stillwater,
Oklahoma.

Kline, J. 1972. Personal communication.

Knowlton, L. L., and K. C. Sink, Jr. 1975. Influence
of the multiflora-grandiflora genotypes of
petunia on seed germination, seedling growth,
and elemental foliar composition. J. Amer.
Soc. Hort. Sci. 100:509-512.

Larkum, A. W. D. 1968. Ionic relations of chloro-
plasts lg vivo. Nature. 218:447-449.

Lewis, B. A., and A. Smith. 1969. Sugars and
derivatives. lg E. Stahl (ed.). Thin-Layer
Chromatography--A Lab Handbook. Springer-
Verlag. New York, Inc. pp. 807-837.

McCready, R. M., J. Griggoly, V. Silvera, and S. H.
Owens. 1950. Determination of starch and
amylose in vegetables. Anal. Chem. 22:
1156-1158.

Natarella, N. J., and K. C. Sink, Jr. 1973. Thin-
layer chromatography of free amino acids,
indoles, and phenols of double and single
flowered petunias. J. Amer. Soc. Hort. Sci.
98:374-378.

Natarella, N. J., and K. C. Sink, Jr. 1974. A
chromatographic study of phenolics ancestral
to Petunia hybrida Vilm. J. Hered. 65:
85-90.

 

Nobel, P. S. 1967. Calcium uptake, ATPase and photo-
phosphorylation by chloroplasts i3 vitro.
Nature. 214:875-877.

Nobel, P. S. 1969. Light-induced changes in the
ionic content of chloroplasts in Pisum sativum.
Biochim. Biophys. Acta. 172:134-143.

 

22.

23.

24.

25.

26.

27.

28.

29.

59

Reimann-PhiliPP. R. 1962. Untersuchungen uber die
Verebung des grandiflora-Merkmals bei Petunia
X hybrida Vilm. Z. Pflanz. 48(2):l43-l76.

Seidel, J. 1962. Beitrage zur Genetik and Zuchtung
der tetraploid Superbissima-Petunien (Petunia
X hybrida Vilm Superbissima-Gruppe). Z.
Pflanz. 48(4):327-359.

Scott, G. W. 1937. A genetical and cytological study
of petunia with special reference to the
inheritance of doubleness. Ph.D. Thesis,
University of California. 42 p.

Sink, K. C. Jr. 1973. The inheritance of apetalous
flower type in Petunia hybrida Vilm. and
linkage tests with the genes for flower
doubleness and grandiflora characters and its
use in hybrid seed production. Euphytica.
22:520-526.

 

Snedecor, G. W., and W. C. Cochran. 1967. Sixth
Edition. Statistical Methods. The Iowa State
University Press. Ames, Iowa. 593 p.

Steele, R. G. D., and J. H. Torrie. 1960. Principles
and Procedures of Statistics. McGraw-Hill
Book, Co. New York. 481 p.

Stahl, E. (ed.). 1969. Thin-Layer Chromatography--A
Lab Handbook. Springer-Verlag. New York,
Inc. 1041 p.

Stocking, C. R., and A. Ongun. 1962. The intracellar
distribution of some metallic elements in
leaves. Amer. J. Bot. 49:284-289.

 

”'11113’1111111111 (111)11111111ES