Scott, J. W. and J. P. Jones. 1989. Monogenic resistance in tomato to

Fusarium oxysporum F. sp. lycopersici race 3. Euphytica 40:49-53.

Scott, J. W., G. Cameron Somodi, and J. B.Jones. 1993. Testing tomato

genotypes and breeding for resistance to bacterial wilt in Florida. Proc.

International Bacterial Wilt Symposium, Kaoshiung, Taiwan, ROC.

(in press).

Sonoda, R. M. and J. J. Augustine. 1978. Reaction of bacterial wilt-resis

tant tomato lines to Pseudomonas solanacearum in Florida. Plant Disease

Reptr. 62:464-466.

Young, P. A. 1963. Two-way varieties for hot or cold climes. Amer. Veg

etable Grower 11:13.

Proc. Fla. State Hort. Soc. 106:155-157. 1993.

IDENTIFYING POLYPLOIDS OF VARIOUS

CUCURBITS BY STOMATAL GUARD CELL CHLOROPLAST NUMBER

Fred McCuistion and Gary W. Elmstrom

University of Florida, IF AS

Central Florida Research and Education Center

5336 University Avenue

Leesburg, FL 34748-8232

Additional index words. Citrullus lanatus, Cucumis melo, Luff a

sp., stomate.

Abstract. Polyploids are useful in developing seedless fruits.

Tetraploid plants pollinated with diploid pollen produce tri-

ploid seeds. Plants from these triploid seeds produce seedless

fruits, which are preferred over seeded versions. For efficient

use of time and space, it is beneficial to identify polyploids

induced by colchicine treatment or by in vitro culture as rapidly

as possible. The ploidy level of plants may be determined by

counting the number of chloroplasts in stomatal guard cells.

This is accomplished by peeling the lower epidermis and

examining it under a light microscope. Tetraploids of water

melon [Citrullus lanatus) (Thunb.) Matsum & Nakai], musk-

melon (Cucumis melo L), and Luffa acutangula Roxb. have

been identified using this method, which can be used at an

early stage of plant development. One problem inherent in

this method is the inability to distinguish periclinal chimeras

in the colchicine-treated plants. However, progeny from peri

clinal chimeric plants will be diploid and can be eliminated

in the second generation.

In addition to seedlessness, polyploidy may result in

improved fruit characteristics such as flesh firmness, thick

ness, and possibly shelf life. Triploid seedless watermelon

sales in the U.S. have increased significantly in the past 5

years (Maxwell, 1992). Triploid watermelon seed is pro

duced by pollinating a tetraploid with a diploid (Kihara,

1951). The cellulose skeleton of the Luffa sp. fruit has sev

eral household and industrial uses, all of which require

removal of the seeds for commercial use (Iqbal and Zafar,

1993). Triploid seedless luffa would lower the cost of pro

cessing the fruits.

A major obstacle to improvement of triploid watermelon

varieties has been a lack of diversity in tetraploid breeding

lines. Tetraploids are produced through the application of

colchicine or occur spontaneously during in vitro culture of

diploids (Adelberg et al., 1990). These methods produce

tetraploids at a low frequency. Tetraploids can be detected

Florida Agricultural Experiment Station Journal Series No. N-00818.

by comparing stem, leaf, flower, and pollen size or fruit

and seed shape (Kihara, 1951; Roy and Ghosh, 1971).

However, these comparisons can be made only on mature

plants and growing large numbers of plants to maturity is

an inefficient use of resources.

It would be beneficial to screen a large number of plants

for ploidy level at an early stage of development. It has

been shown that the number of chloroplasts in stomatal

guard cells is directly proportional to the ploidy level of

the plant (Jacobs and Yoder, 1989; Singsit and Veilleux,

1991) such that the stomatal chloroplast number of tetra

ploid plants is twice that of diploid plants. Since seedlings

can be used for stomatal chloroplast number determination,

screening for putative tetraploids can be accomplished be

fore transplanting to the field. Therefore, it would seem

that stomatal chloroplast number determination is a more

reliable and earlier indication of ploidy level than compari

son of size and shape of morphological features. The

chloroplast number screening procedure is considerably

easier than making chromosome counts.

Materials and Methods

Watermelon plants, Citrullus lanatus 2n = 2x = 22, of

the diploid 'Small-Seeded Dixielee' and a tetraploid breed

ing line with the Chinese male sterile gene (Zhang and

Wang, 1990), cms, were grown from seed under greenhouse

conditions. The fifth true leaf was removed and a small

piece of the lower epidermis was peeled off using jewelers

forceps. Each leaf peel was placed in a drop of water on a

glass slide, covered with a cover slip, and examined at 400

magnification using a light microscope. Stomatal chloro

plasts in 30 individual guard cells were counted. The same

procedure was used to count the stomatal chloroplasts of

diploid muskmelon, Cucumis melo 2n = 2x = 24, 'Tenkei'

and 'Mission* and tetraploid muskmelon line 67-m6-100

(Nugent and Ray, 1992) and diploid accessions of Luffa

acutangula 2n = 2x = 26.

A drop of 0.2% aqueous colchicine was applied twice

daily for 3 to 4 days to the growing point of seedlings of

watermelon cultigens 'Picnic', 'Calsweet', 'Jubilee II', and

cms. Seedlings of Luffa acutangula and C. melo 'Mission'

were also treated. Leaf peels were taken from plants with

at least 5 true leaves. The stomatal chloroplast number for

each plant was determined and putative tetraploids were

transplanted to the field and self-pollinated. The stomatal

chloroplast number was determined for progeny of all self-

pollinated putative tetraploids at the 5th true leaf stage.

Proc. Fla. State Hort. Soc. 106: 1993. 155

Fig. 1. Leaf peels of diploid (top) and tetraploid (bottom) watermelons

showing stomatal chloroplasts.

Results and Discussion

The average stomatal chloroplast number of the diploid

'Small-Seeded Dixielee' was 11.4 whereas the average chloro

plast number of the tetraploid breeding line, cms, was 19.7

(Fig. 1). The triploid hybrid resulting from the cross of

these two cultigens had an average stomatal chloroplast

number of 14.0 (Table 1). The average stomatal chloroplast

number of the diploid muskmelon 'Tenkei' and 'Mission'

Table 1. The stomatal chloroplast numbers of various cucurbit diploid,

triploid, and tetraploid plants.

Crop

Watermelon

Muskmelon

Luff a acutangula

Variety

Small-Seeded Dixielee

cms1

cm x Small-Seeded Dixielee

Mission

Tenkei

67-m6-1000

Ploidy

2x

4x

3x

2x

2x

4x

2x

CNy

11.4

19.7

14.0

8.5

8.5

18.6

9.0

was 8.5 while the tetraploid line, 67-m6-100, had an average

of 18.6 (Fig. 2). No tetraploid luffa accessions were tested;

however, a diploid L. acutangula had a stomatal chloroplast

number of 9.0 (Fig. 3, Table 1). Stomatal chloroplast number

was a good indicator of ploidy level for all three species

examined in this study.

The results of the watermelon progeny test show that

approximately 60% of the plants that were identified as

putative tetraploids produced tetraploid offspring. This

number might have been higher except that the number

of seed obtained from several of the putative tetraploids

was small and many seeds did not germinate. Additionally,

self pollinations were not obtained from all putative tetra

ploids (Table 2). Data are not presented for progeny tests

of muskmelon as the material was still in the first generation

of colchicine treatment. The progeny test of L. acutangula

showed that 73% of the plants that were identified as puta

tive tetraploids produced tetraploid offspring.

The production of diploid progeny by plants that had

a tetraploid chloroplast number indicates that some were

probably periclinal chimeras. Therefore, this method of

early screening of colchicine treated material did not detect

"false positives" when selecting for 4x plants but will help

to eliminate all diploid material.

zcms = Chinese male sterile.

yCN = Number of chloroplasts in stomatal guard cells. Avg. no. in 20 cells.

Fig. 2. Leaf peels of diploid (top) and tetraploid (bottom) melons

showing stomatal chloroplasts.

156 Proc. Fla. State Hort. Soc. 106: 1993.

Table 2. Stomatal chloroplast numbers of colchicine treated plants and

their progeny.

Progeny

Fig. 3. Leaf peel of diploid Luff a acutangula showing the stomatal

chloroplasts.

Usually 150 seeds of a cultigen were planted; poor ger

mination and high mortality from colchicine treatment

resulted in about 100 plants left for stomatal chloroplast

number determination. Of these remaining plants, about

15% were putative tetraploids. An additional 15% were

identified as sectoral chimeras and 70% were discarded

based on a diploid chloroplast number. Since there is a

very low recovery rate of 4x plants from sectoral chimeras,

only plants that screen as 4x should be transplanted to the

field or greenhouse.

One person screened 100 or more treated plants per

day using this technique. More than 100 tetraploids from

25 different watermelon cultigens and 16 tetraploids of

one accession of L. acutangula have been generated with

this system in one year.

Crop Variety

Watermelon

cms2

Jubilee II

Calsweet

Picnic

Luff a acutangula

Putative 4xy

(no.)

28

12

12

11

26

2x

(no.)

6

6

3

3

6

CNX

11.1

11.2

11.1

10.0

9.0

4x

(no.)

14

3

5

7

16

CN

20.2

20.6

21.1

18.4

18.4

zcms = Chinese male sterile.

yScreened positive for tetraploid chloroplast number; however, some are

known to be periclinal chimeras.

XCN = Number of chloroplasts in stomatal guard cells. Avg. no. in 20

cells.

Literature Cited

Adelberg, Jeffrey, B. B. Rhodes, and Halina Skorupska. 1990. Generat

ing tetraploid melons from tissue culture. HortScience 25:73.

Iqbal, M. and S. I. Zafar. 1993. The use of fibrous network of mature dried

fruit of Luff a aegyptiaca as immobilizing agent. Biotech. Tech. 7:15-18.

Jacobs, J. P. and John I. Yoder. 1989. Ploidy levels in transgenic tomato

plants determined by choloroplast number. Plant Cell Repts. 7:662-664.

Kihara, H. 1951. Triploid watermelons. Proc. Amer. Soc. Hort. Sci. 58:

217-230.

Maxwell, K. 1992. Aiton: Seedless sales "exploding". The Produce News.

2 May 1992, p. 23.

Nugent, P. E. and D. T. Ray. 1992. Spontaneous tetraploid melons.

HortScience 27:47-50.

Roy, R. P. and J. Ghosh. 1971. Experimental polyploids oiLuffa echinata.

The Nucleus 14:111-115.

Singsit, C. and R. E. Veilleus. 1991. Chloroplast density in guard cells of

leaves of anther derived potato plants grown in vitro and in vivo.

HortScience 26:592-594.

Zhang Xing-ping and Ming Wang. 1990. A genetic male sterile (ms) wa

termelon from China. Cucurbit Gen. Coop. 13:45-46.

Proc. Fla. State Hort. Soc. 106:157-159. 1993.

PRELIMINARY DETERMINATION OF PARAMETERS TO DEVELOP

AN OBJECTIVE PROCEDURE FOR ASSESSING TIPBURN IN LETTUCE

M. L. Stratton and R. T. Nagata

Everglades Research and Education Center

IFAS, University of Florida

Belle Glade, Florida 33430-8003

Additional index words, calcium, fertility, temperature,

humidity, breeding, Lactuca sativa.

Abstract. Tipburn resistant varieties of lettuce, Lactuca sativa

L. provide an important measure of control of the disorder.

Florida Agricultural Experiment Station Journal Series No. N-00839.

The authors gratefully acknowledge the funding support of South

Bay Growers, Inc., South Bay, Florida, and the support of Conrad Fafard,

Inc., Springfield, Ma. who provided the peat-lite soilless mix used in this

study. The authors extend a special thank to Vivian Johnson for technical

support.

The lack of a rapid and reliable screening test for tipburn has

made selection for resistance time-consuming and difficult. A

screening procedure for tipburn was tested utilizing 'Dark

Green Boston', a tipburn susceptible line. Plant ages of 21,

38, and 56 days old at the start of testing were evaluated.

Prior to testing, plants were grown at 25C and 55% RH, with

regular fertilization. During testing, plants were grown in a

clear polyethylene tent with high temperature (40C), high

humidity (RH > 90%), and high fertility, conditions that pro

vided for rapid plant growth. Tipburn development was asses

sed daily upon initiation of testing. Tipburn symptoms began

appearing within 1 or 2 days after test initiation for 38 and

56 day old plants. Twenty-one day old plants developed tip

burn by the third day of testing. Thirty-eight day old plants

were most severely affected, having a tipburn index (number

of affected leaves/total number of leaves) of 0.58. Twenty-one

and 56 day-old lettuce had tipburn indexes of 0.29 and 0.19,

Proc. Fla. State Hort. Soc. 106: 1993. 157