Glabrous-leaf Cantaloupe: A Source of Host Plant
Resistance to Whiteflies

David Riley

University of Georgia, Coastal Plain Experiment Station, P.O. Box 748, Tifton, GA 31793

David Wolff

Sakata Seed America, P.O. Box 1118, Lehigh, FL 33970-1118

Dean Batal

University of Georgia, Coastal Plain Experiment Station, P.O. Box 748, Tifton, GA 31793

Additional index words. melon, muskmelon, SR-91, Cucumis melo, Bemisia tabaci, Bemisia argentifolii

Abstract. In 1995, 15 melon cultivars were compared over planting dates and insecticide treatments in a split-split plot experiment for susceptibility to whitefly damage. In this test, glabrous melons were found to reduce numbers of all stages of whiteflies on the crop. Genetic crosses were made to introduce the glabrous leaf trait into a commercial-type cantaloupe cultivar and evaluate the line in the field for response to whitefly and melon yield. In 1996 and 1997, F2 glabrous and pubescent selections were compared to 'Explorer' and 'Cruiser' in imidacloprid-treated and nontreated main plots in a split-plot design. The F2 progeny was from a TAM Sun (selfed selection out of 'Sunshine', an F1 hybrid from Ferry Morse) x 'SR-91' glabrous phenotype (single gene recessive trait, Foster, 1962) cross, backcrossed to TAM Sun. Whitefly counts were made through the season and yield data were collected to measure plant response. The glabrous-leaf melons consistently resulted in lower whitefly adult and nymph population densities compared to commercial pubescent-leafed cultivars. Also, the glabrous F2 line was not significantly different in yields compared with two commercial cultivars. Glabrous-leaf melons had significantly shorter vine length and percent sugars in Spring 1997, but not in Fall 1996, possibly due to seasonal differences. In 1998, the F3 of the same backcross was evaluated on plastic mulch beds under field conditions in Georgia and the F3 progeny was described.

The impact of silverleaf whitefly, Bemisia
argentifolii Bellows & Perring [previ-
ously b-strain sweetpotato whitefly, Bemisia tabaci (Gennadius)] on muskmelons can be severe with economic loss occurring with just 3 adults per leaf (Riley and Palumbo, 1995a, 1995b). Losses can be even greater when whiteflies are combined with whitefly-transmitted viruses (Brown et al., 1995). Host plant resistance to whiteflies has been documented in cucurbits (Kishaba et al., 1992) and is a desirable tactic for the control of this pest. Riley (1995) indicated that fewer leaf trichomes correlated with lower whitefly density on the leaf. A glabrous phenotype in the cultivar SR-91 (Foster, 1962) provided the opportunity to

pursue this trait as a possible source of host plant resistance to whiteflies in muskmelon.

The goals of these tests were to A) evaluate a selected glabrous-leaf melon cultivar for its potential as a source of host plant resistance to whitefly and B) develop a more commercial-type glabrous leaf cantaloupe. Two specific objectives under goal A were 1) to measure whitefly populations by monitoring adult and nymph populations throughout the season on glabrous-leaf melon compared to other melon cultivars and 2) to determine plant response of each melon cultivar under low and high whitefly infestation pressure. Two specific objectives under goal B were to 1) evaluate the F2 selections from a glabrous x commercial cantaloupe cross ('SR-91' x TAM Sun) as described under goal A and 2) describe the F3 progeny of this cross under field conditions in Georgia.

We thank the Texas Agricultual Experiment Station, Weslaco, Texas, the Coastal Plain Experiment Station, Tifton, Ga., and Sakata Seed America for lending the time and support to make this publication possible.

Cucurbitaceae '98


fly numbers. Yield was evaluated in terms of total harvested fruit by size class to estimate number of boxes. The number of harvested melons by boxes was reported per 100-ft2 plot. Analysis of variance were made using the GLM procedure (SAS Institute, 1988).

Texas test 1995. Selected melon cultivars were planted in 1 m beds by 9 m length field plots in early- (3 February) and late- (14 March) planted plots in Spring 1995. Treatments were arranged in a split-split plot design with two whole plots, early- and late-planted melons, two subplots, untreated and insecticide treated, and cultivars as the subsubplots. In the treated plots, the insecticide applied was imidacloprid 0.25 lb ai/a in furrow at planting and at one foliar application at 0.05 lb ai/a during the growing season. Foliar applications of insecticide were made with one overhead and two lateral hollow-cone spray tips (TX SS18, Spraying Systems Co.) at 3.5 kg·cm­2 CO2 pressure resulting in 45 l/ha spray volume. In-furrow applications were made with a single spray nozzle to the seed furrow. The melon cultigens planted into the subplots were 1) Honeybrew, 2) Morning Ice, 3) Hy. FMX 165, 4) 'Primo', 5) 'Hymark', 6) 'Cruiser', 7) 'Explorer', 8) 'Perlita', 9) PI 126125, 10) HMX 1601, 11) HMX 9584, 12) PI 125951, 13) PI SR91 pubescent, 14) 'Sunshine', and 15) PI SR91 glabrous. Due to low seed availability, cultivars in subplots 9 through 14 had fewer plants per plot, so yield data (based on a single harvest from a 9-m plot) were not reported for these subplots. Analysis of variance and correlations were made using SAS GLM procedures for a split-split-plot design (SAS Institute, 1988).

Texas tests 1996 and 1997. Selected melon cultivars were planted in 2 m beds by 15 m length field plots in Fall 1996 and Spring 1997. Treatments were arranged in a split plot design with two whole plots, untreated and insecticide treated, and cultivars as the subplots. In the treated plots the insecticide applied was imidacloprid 0.25 lb ai/a in furrow at planting and at least two foliar applications at 0.05 lb ai/a during the growing season, as in the 1995 test. The melon cultigens planted into the subplots were 1) 'Explorer', 2) 'Cruiser', 3) F2 glabrous, and 4) F2 pubescent. Whitefly counts were made throughout the sea

Materials and methods

The silverleaf whitefly (SW) was evaluated on selected melon cultivars, Cucumis melo L., in insecticide-treated and untreated field plots in 1995, 1996, 1997 at Weslaco, Texas, and in an untreated plot in Georgia in 1998. A glabrous-leaf cantaloupe, SR-91 (single gene recessive trait, Foster 1962), was evaluated against 14 other melon cultivars in 1995. The glabrous line was then crossed with a desirable commercial type cantaloupe, TAM Sun (selfed selection out of 'Sunshine', an F1 hybrid from Ferry Morse Seed Company) to produce the F2 selections used in the 1996 and 1997 tests. An F3 of the same cross was evaluated in the 1998 test.

Host plant resistance in terms of nonpreference, antibiosis, and tolerance (Horber, 1980) was assessed in these tests. Indications for nonpreference (the insect's ovipositional behavior) and antibiosis (survival of eggs to pupal stage) were calculated from season-long averages of egg/adult ratios and pupa/egg ratios, respectively. Whitefly egg, nymph, and adult stages were monitored weekly throughout the season. SW adults were counted per whole leaf at the 3rd node from the growing point by carefully turning the leaf. Immature SW were sampled on two leaf disks per leaf (total 7.6 cm2 of leaf area) sampled at the 6th node from the growing point (Riley and Palumbo, 1995a). Sample size was five leaves per plot for adults and ten leaves per plot for immature stages. Whitefly numbers were analyzed by development stage and ratios of the various whitefly life stages were compared using an analysis of variance (GLM) and lsd tests for mean separation (SAS Institute, 1988).

Plant tolerance in terms of yield response to direct whitefly damage was evaluated by using treated and nontreated plots in 1995, 1996, and 1997. One indicator of plant tolerance was assumed to be a lack of yield response to whitefly control using imidacloprid (Admire 2F, Bayer Corp., Kansas City, Mo.). This product was shown to provide excellent whitefly control and melon yield response in south Texas (Riley, 1994). This was evaluated by measuring number of fruit by grade, measuring insecticide treatment effects on yield, and correlating yield to season-long, white

Cucurbitaceae '98


Figure 1. Number of whitefly adults and total nymphs over time by week after plant emergence in early- and late-planted melons at Weslaco, Texas, Spring 1995.

Georgia test 1998. The F3 progeny of the glabrous SR-91 x TAM Sun cross was described under field conditions in Georgia. Plants were direct seeded into 5 ¥ 5 ¥ 7.6-cm cells in trays in a greenhouse using standard production methods on 5 May. Seedlings were exposed to a whitefly infestation of one to five adults per leaf immediately after plant emergence. After three weeks, 50 seedlings were transplanted into a 91 cm ¥ 91 m plastic mulch covered bed with 1.8 m between plants. The resulting glabrous- and hairy-leaf cantaloupe lines or segregates were maintained with stan

son and yield data were collected to measure plant response. Analysis of variance and correlations were made using SAS GLM procedures for a split-plot design (SAS Institute, 1988).

Table 1. Analysis of variance (P > F values) for the 1995 whitefly data using a split-split-plot design followed (below dotted line) by seasonal means of whitefly adults per leaf and immature stages per 7.6 cm2 of leaf area at the mainplot (df = 1), subplot (df = 1) and subsubplot (df = 14) levels.

Level Adults Eggs Small nymphsz Large nymphsy

Model 0.00 0.00 0.00 0.00

Mainplot (date) 0.55 0.00 0.00 0.00

Subplot (insecticide) 0.01 0.00 0.00 0.00

Mainplot ¥ subplot 0.94 0.06 0.03 0.00

Subsubplot (cultivar) 0.13 0.03 0.05 0.07

Mainplot ¥ subsubplot 0.80 0.42 0.65 0.50

Subplot ¥ subsubplot 0.27 0.82 0.84 0.43

Main

Early planted 13.4 81 18.0 27.4

Late planted 15.6 147 41.5 61.1

Subplot

Treated 6.5 44 17.8 12.6

Untreated 22.4 178 37.3 71.6

Subsubplot

Sunshine 14.4 232 ax 73.4 a 80.6 a

HMX 1601 14.9 120 bc 25.9 bcd 54.0 b

Morning Ice 18.1 122 bc 31.3 bc 50.5 bc

Honeybrew 17.6 139 a 41.3 b 47.8 bcd

Cruiser 14.1 117 bc 30.1 bcd 44.4 bcd

HMX 9584 12.9 103 bcd 26.7 bcd 42.8 bcd

Explorer 14.5 89 bcd 21.8 bcd 39.2 bcd

Hy. FMX 165 12.1 104 bcd 26.7 bcd 38.4 bcd

PI 125951 10.5 81 cd 19.8 cd 37.5 bcd

Perlita 14.3 113 bc 24.5 bcd 36.9 bcd

Hymark 16.7 102 bcd 27.3 bcd 36.5 bcd

Primo 15.4 85 bcd 21.4 bcd 33.9 bcde

PI 126125 10.2 88 bcd 13.7 cd 27.5 cde

PI SR-91 13.1 77 cd 21.2 bcd 23.9 de

PI SR-91 glabrous 10.2 56 d 10.4 d 11.6 e

zFirst and second instars.

yThird and fourth instars.

xMeans followed by the same letter not significantly different (P > 0.05, lsd).

Cucurbitaceae '98


Table 2. Analysis of variance (P > F values) for the 1995 melon plant data using a split-split plot design followed (below dotted line) by seasonal means of trichomes per 0.9 cm2 of leaf area, average percent soluble sugar, and boxes of melons per plot (9 m2) at the mainplot (df = 1), subplot (df = 1) and sub-subplot (df = 14) levels.

Leaf Soluble Boxes/

Level trichomes sugars (%) plot

Model 0.00 0.04 0.03

Mainplot (date) 0.00 0.03 0.09

Subplot (insecticide) 0.00 0.18 0.00

Mainplot ¥ subplot 0.35 0.00 0.06

Subsubplot (cultivar) 0.25 0.00 0.00

Mainplot ¥ subsubplot 0.60 0.00 0.05

Subplot ¥ subsubplot 0.93 0.00 0.00

Main plot

Early planted 447 8.1 0.72

Late planted 613 6.2 0.64

Subplot

Treated 501 7.4 0.89

Untreated 545 6.9 0.47

Subsubplot

Hymark 534 bcdez 8.0 b 1.39 a

Sunshine 745 a 5.7 g ---

Explorer 579 bcd 7.9 b 1.26 a

Hy. FMX 165 573 bcd 7.3 cd 1.11 ab

PI SR-91 606 bc 7.0 de ---

HMX 1601 638 ab 6.6 ef ---

HMX 9584 587 bcd 7.1 cd ---

Primo 489 cde 7.4 c 0.94 bc

Perlita 484 be 7.1 cd 0.87 bcd

Honeybrew 483 de 9.1 a 0.79 cde

Cruiser 580 bcd 7.5 c 0.72 cde

Morning Ice 423 e 7.3 cd 0.61 de

PI 125951 598 bcd 6.4 f ---

PI 126125 544 bcd 6.5 f ---

PI SR-91 glabrous 0 f 4.9 h 0.12 f

zMeans followed by the same letter not significantly different (P > 0.05, lsd).

There was a significant planting date effect on whitefly eggs and nymphs with the later planting date resulting in greater whitefly populations, as expected from overall whitefly population trends (Figure 1). Also, the insecticide treatment of imidacloprid was effective in significantly reducing all stages of whiteflies significantly across all cultivars and planting dates without any significant interactions (Table 1). 'Sunshine' had high whitefly numbers, but this was partly an experimental artifact from this cultivar not being included in the early planting. Even so, 'Sunshine' and 'HMX 9584' had the highest nymph counts in the late-planted melons. 'SR-91' glabrous had the lowest numbers of whiteflies, regardless of planting date (Table 1); however, yield per plot was low for this cultivar (Table 2).

In general, the highest yielding cultivars had moderate to high trichome numbers on the leaves, but there was no clear trend that might suggest that leaf pubescence is correlated with the quality of fruit production (Table 2). It was evident from this test that the glabrous trait was effective in reducing whitefly numbers on the crop; however, the glabrous 'SR-91' line was not competitive in terms of yield and fruit quality (e.g., percent soluble sugars) with most commercial melons.

Texas tests 1996 and 1997. Moderate to high numbers of whiteflies occurred in these tests with greater numbers occurring early in the season in

Figure 2. Number of whitefly adults and total nymphs over time by week after plant emergence in melons at Weslaco, Texas, Spring 1996 and Fall 1997.

dard cultural practices and evaluated for commercial fruit quality traits, such as netting, percent soluble sugars, size of fruit, shape of fruit, etc. at crop maturity. The yield characteristics of the glabrous F3 plants were also described.

Results and discussion

Texas test 1995. High numbers of whiteflies occurred in this test, with greatest numbers in the late planted plots (Figure 1). Significant differences were found between melon cultivars in terms of whitefly adults and nymphs (Table 1) and yield (Table 2).

Cucurbitaceae '98


Table 3. Analysis of variance (P > F values) for 1996 whitefly data using a split-plot design followed (below dotted line) by seasonal means of adult, egg, and nymph numbers at the main plot (df = 1) and subplot (df = 3) levels.

Adults/ Whiteflies

Level leaf Eggs Nymphs

Model 0.00 0.01 0.00

Mainplot (insecticide) 0.00 0.01 0.00

Subplot (cultivar) 0.00 0.01 0.00

Mainplot ¥ subplot 0.00 0.03 0.00

Main

Treated 11.1 7.5 1.3

Untreated 64 .2 26.2 51.8

Subplot

F2 pubescent 51.0 az 26.1 a 42.7 a

Cruiser 44.6 ab 18.6 ab 30.5 b

Explorer 42.5 b 12.0 bc 24.4 b

F2 glabrous 12.5 c 10.6 c 8.7 c

zMeans followed by the same letter not significantly different (P > 0.05, lsd).

Table 5. Analysis of variance (P > F values) for 1996 melon plant data using a split-plot design followed (below dotted line) by seasonal means of weekly vine length measurements (cm), percent soluble sugar, and boxes per plot at the main plot (df = 1) and subplot (df = 3) levels.

Vine Soluble Boxes/

Level length sugars (%) plot

Model 0.00 0.03 0.34

Mainplot (insecticide) 0.00 0.00 0.66

Subplot (cultivar) 0.50 0.27 0.10

Mainplot ¥ subplot 0.02 0.23 0.27

Main

Treated 234 13.2 2.5

Untreated 134 9.6 2.3

Subplot

Explorer 185 11.7 3.8

Cruiser 194 10.8 1.6

F2 pubescent 179 11.3 3.0

F2 glabrous 176 12.2 1.3

Table 4. Analysis of variance (P > F values) for 1997 whitefly data using a split-plot design followed (below dotted line) by seasonal means of adult, egg, and nymph numbers at the main plot (df = 1) and subplot (df = 3) levels.

Adults/ Whiteflies

Level leaf Eggs Nymphs

Model 0.00 0.22 0.08

Mainplot (insecticide) 0.00 0.00 0.00

Subplot (cultivar) 0.00 0.27 0.09

Mainplot ¥ subplot 0.01 0.28 0.11

Main

Treated 6.9 0.04 0.07

Untreated 27.2 0.72 2.89

Subplot

F2 pubescent 20.7 az 0.40 2.00

Cruiser 18.4 a 0.48 2.13

Explorer 17.4 a 0.53 1.61

F2 glabrous 11.5 b 0.11 0.19

zMeans followed by the same letter not significantly different (P > 0.05, lsd).

Table 6. Analysis of variance (P > F values) for 1997 melon plant data using a split-plot design followed (below dotted line) by seasonal means of weekly vine length measurements (cm), percent soluble sugar, and boxes per plot at the main plot (df = 1) and subplot (df = 3) levels.

Vine Soluble Boxes/

Level length sugars (%) plot

Model 0.00 0.00 0.08

Mainplot (insecticide) 0.66 0.53 0.86

Subplot (cultivar) 0.00 0.00 0.10

Mainplot ¥ subplot 0.28 0.10 0.66

Main

Treated 83 8.2 2.3

Untreated 83 7.9 2.2

Subplot

Explorer 83 10.2 az 2.4

Cruiser 85 7.7 b 2.7

F2 pubescent 86 7.1 b 1.8

F2 glabrous 76 7.1 b 2.1

zMeans followed by the same letter not significantly different (P > 0.05, lsd).

Fall 1996 than in Spring 1997 (Figure 2). Partly because of this early, high influx of whiteflies in the fall crop, the main and subplot treatment effects were significant for all whitefly stages. The F2 pubescent line had significantly greater whiteflies and the F2 glabrous had significantly fewer whitefly adults, eggs, and nymphs than at least one of the commercial melon cultivars (Table 3). In both 1996 and 1997, the glabrous-leaf F2 melons consistently had reduced whitefly adults and nymphs compared to the other cultivars.

Generally, greater yields were associated with

lower SW numbers, but the response varied with cultivar. The F2 glabrous line was not significantly different in yield compared with commercial standards, 'Explorer' and 'Cruiser' in 1996 and 1997, but yield tended to be low (Tables 5 and 6). Even so, this represented a substantial improvement over the 1995 field evaluation of the SR-91 glabrous melon. Glabrous-leaf F2 melons had significantly shorter vine length and lower percent sugars in Spring 1997 (Table 6), but not in Fall 1996 (Table 5), possibly due to seasonal differences in

Cucurbitaceae '98


Table 7. Analysis of variance (P > F values) for the 1996 and 1997 whitefly ratio data using a split plot design followed (below dotted line) by seasonal means of ratios of eggs to adults (eggs + 1/adults + 1) and nymphs to eggs (nymphs + 1/eggs + 1) at the mainplot (df = 1) and subplot (df = 3) levels.

1996 1997

Level Eggs/adults Nymphs/eggs Eggs/adults Nymphs/eggs

Model 0.04 0.00 0.06 0.00

Mainplot (insecticide) 0.07 0.01 0.02 0.00

Subplot (cultivar) 0.01 0.08 0.00 0.01

Mainplot ¥ subplot 0.12 0.07 0.95 0.02

Main

Treated 0.70 0.29 0.46 1.05

Untreated 0.50 2.07 0.41 2.01

Subplot

Explorer 0.43 bz 1.50 a 0.39 b 1.51 ab

Cruiser 0.53 b 1.26 ab 0.42 b 1.63 a

F2 pubescent 0.63 ab 1.23 ab 0.42 b 1.86 a

F2 glabrous 0.83 a 0.73 b 0.50 a 1.11 b

zMeans followed by the same letter not significantly different (P > 0.05, lsd).

climate or in the lower whitefly pressure in the spring as compared to the fall. Percent soluble sugars for the F2 glabrous line was quite high during the fall, suggesting that an improvement in sugar content was achieved with the cross without the loss of resistance to whiteflies. These data suggest that progress toward a glabrous commercial cantaloupe cultivar was accomplished, but that much still needs to be done in the improvement of yield and quality potential.

Ratios of eggs to adults were significantly higher for the glabrous F2 than other cultivars in both the 1996 and 1997 tests (Table 7). Conversely, ratios of nymphs to eggs were lower for the glabrous line than the other cultivars tested. At the same time, adult numbers on the glabrous lines tended to be low (Tables 3 and 4). The interpretation we make of these data is that the glabrous trait is resulting in nonpreference for the adult whitefly as a landing site, but once on the plant more eggs are oviposited per adult compared with commercial melon cultivars. However, a greater majority of the eggs do not survive to the nymphal stages either due to antibiosis or greater predation, parasitism, or other factors. Since no obvious differences were noticed in percent parasitism, the authors feel that antibiosis (e.g., greater desiccation of nymphs) is a more likely explanation.

Georgia test 1998. The results of the 1998 test will be presented at the 1998 Cucurbitaceae Meeting at Pacific Grove, Calif., not in this paper. Instead, a general description of the SR-91 gla

brous melon is as follows. The fruit are round to oval, weighing 1.6 kg each. The net is medium, and the fruit have slight vein tracts. The vine is small to medium, and the leaves are smoother and have a thicker texture compared to normal pubescent leaves. Under heat stress there is a tendency of scalding of some leaf tissue. The average yield of fruit per plant is nine fruit weighing a total of 14 kg. Specific characteristics of the leaf tissue were previously described by Foster (1962).

Literature cited

Brown, J.K., D. Frolich, and R. Rosell. 1995. The sweetpotato or silverleaf whiteflies: Biotypes of Bemisia tabaci or a species complex? Annu. Rev. Entomol. 40:511­34.

Foster, R.E. 1962. Glaborous, a new seedling marker in muskmelon. J. Hered. 54:113­115.

Horber, E. 1980. Types and classification of resistance, p. 15­21. In F.G. Maxwell and P.R. Jennings (eds.). Breeding plants resistant to insects, Wiley, New York.

Kishaba, A.N., S. Castle, J.D. McCreight, and P.R. Desjardins. 1992. Resistance of white-flowered gourd to sweetpotato whitefly. HortScience 27:1217­1221.

Riley, D.G. 1994. Insecticide control of sweetpotato whitefly in south Texas. Subtrop. Plant Sci. 46:45­49.

Riley, D.G. 1995. Melon cultivar response to Bemisia. Subtrop. Plant Sci. 47:39­45.

Riley, D.G. and J.C. Palumbo. 1995a. Action thresholds for Bemisia argentifolii (Homoptera: Aleyrodidae) in cantaloupe. J. Econ. Entomol. 88:1733­1738.

Riley, D.G. and J.C. Palumbo. 1995b. Interaction of silverleaf whitefly (Homoptera: Aleyrodidae) with cantaloupe yield. J. Econ. Entomol. 88:1726­1732.

SAS Institute. 1988. Technical report P. 179. Release 6.03. SAS Inst. Inc., Cary, N.C.

Cucurbitaceae '98