Sources of Whitefly-induced Silvering Resistance in Cucurbita

Linda Wessel-Beaver

Department of Agronomy and Soils, University of Puerto Rico, Mayagüez Campus,
P.O. Box 9030, Mayagüez, PR 00681-9030

Additional index words. Bemisia argentifolii, silverleaf, insect resistance, germplasm assessment

Abstract. Many Cucurbita accessions were screened for silverleaf and whitefly resistance from 1992 to 1996. Selected accessions were then evaluated and either discarded or selfed over four generations. Remaining after four generations were two C. pepo lines and 11 C. moschata lines: five derived from 'Butternut', five from PI 162889, and one from 'Waltham'. In 1997, these 13 lines plus 'Butternut', 'Waltham', PI 162889, silverleaf-susceptible checks 'Soler' and PRSYN3, and three silverleaf susceptible x resistant F1 crosses were field tested in order to verify the silverleaf resistance of the selected lines and to determine if they were also resistant to the silverleaf whitefly. Lines selected for silverleaf resistance had many fewer silvered plants than did the susceptible checks. The silverleaf resistance in lines derived from the Paraguayan land race PI 162889 appears to be distinct from that of 'Butternut' and 'Waltham'. In the tested lines silvering resistance was not associated with lower numbers of whiteflies.

Silvering of tropical pumpkin (Cucurbita moschata [Duch. ex Lam.] Duch. ex Poir.) was first observed in Puerto Rico in commercial plantings on the south coast of the island in 1987 (Segarra-Carmona et al., 1990). Silvering is a physiological plant response to feeding by immature whiteflies (Costa et al., 1993; Yokomi et al., 1990). Bellows et al. (1994) has recently proposed that the silverleaf whitefly is a new species, Bemisia argentifolii Bellows and Perring.

Foliage of all local tropical pumpkin cultivars including 'Boriquen' and 'Soler' becomes highly silvered even in the presence of low whitefly populations. Typically, fruits of tropical pumpkin are not silvered, thus the economic impact of silvering is not as great as in C. pepo L. and C. maxima Duch. ex Lam. However, if intense silvering occurs at an early stage, tropical pumpkin plants will often be slow-growing and unproductive.

Cultivars of C. pepo have been reported to show varying levels of resistance to silvering (McAuslane et al., 1996; Paris et al., 1993a, 1993b). Between 1992 and 1996, I field-screened a large number of Cucurbita accessions from the U.S. National Germplasm System including 420 C. moschata, 350 C. pepo, 294 C. maxima, 150 C.

argyrosperma Huber and a few wild Cucurbita in an attempt to identify sources of silverleaf and whitefly resistance (Wessel-Beaver, 1997). All accessions were rated for silvering and visually inspected for adult whiteflies on several occasions. No accessions were free of whiteflies. Large numbers of whiteflies were always found on highly silvered accessions. No accessions of C. maxima and C. argyrosperma were free of silvering. Some plants of some accessions of C. moschata and C. pepo had, however, little or no silvering and what appeared to be reduced numbers of whiteflies.

The objective of this research was to verify the silverleaf resistance of selected lines and determine to what degree these lines are resistant to the silverleaf whitefly.

Materials and methods

Potentially silverleaf-resistant accessions (identified during the screening process reported by Wessel-Beaver (1997) were evaluated in the field (unreplicated) over four generations. The local cultivar 'Soler' was grown to confirm the presence of silverleaf whiteflies. Trials alternated between stations at Isabela and Mayagüez, Puerto Rico. Only plants with very little or no silvering were advanced to the next generation. During this

Cucurbitaceae '98

phase whiteflies were not counted, but plants hosting many whiteflies were discarded. After 3 generations of selfing, 32 lines remained: 31 C. moschata and 2 C. pepo. Undoubtedly there are many more C. pepo lines that merit evaluation. Because this species is poorly adapted to tropical conditions, mature selfed seed was very difficult to obtain. The C. moschata lines were derived from five sources: 'Butternut', lines related to the original 'Butternut' (PI 550689, 550690, 550691, and 550694), PI 162889 (a Paraguayan land race), 'Waltham Butternut', and PI 419137.

In the fourth generation of selfing, PI 419137 exhibited some silvering and was discarded. Some plants of a C. pepo breeding line from Florida, ZUC78SLR, were slightly silvered. However, this line was advanced to include C. pepo in the final trial. Lines derived from PI 550689, 550690, 550691, and 550694 were selfed but not included in the replicated trial because they were closely related to, but less vigorous than, the 'Butternut'-derived lines. Several 'Butternut'-derived lines were crossed to local silverleaf-susceptible cultivar 'Soler' in order to include F1 plants in the trial reported here and to begin transferring silverleaf resistance to Soler.

In total, 20 cultigensfive 'Butternut'-derived lines, five PI 162889-derived lines, one 'Waltham Butternut'-derived line, two ZUC78SLR-derived lines, three 'Butternut'-derived x 'Soler' F1 hybrids, and susceptible checks 'Soler' and PRSYN4 (a local breeding line)were direct-seeded on 6 Feb. 1998 in Mayagüez, Puerto Rico, in a randomized complete block design with four replications. Plots consisted of a single row of five plants spaced 0.5 m apart. Rows were spaced 3 m apart. Plants were drip-irrigated. A highly susceptible C. maxima (Sunre 9901) was planted between each plot to attract whiteflies. In each plot the percentage of silvered plants was recorded at 4, 6, and 8 weeks after planting. At 8, 10 and 12 weeks the oldest healthy leaf was sampled from three plants per plot. Three 1-cm leaf disks were cut from one side (left or right) of each leaf and number of eggs and nymphs on the abaxial surface were recorded.

Several data transformations (log, square root, arcsine) were tested but none were found to adequately satisfy the assumptions of analysis of

variance. Therefore, a nonparametric analysis of variance was used where, for each variable, the data within a block was ranked from lowest to highest percentage silvering or number of insects (Conover and Iman, 1981; Eskridge, 1995). Both ranked and original data are shown in tables 1 and 2, however, all analyses and single degree of freedom contrasts were done with rank-transformed data.


Lines derived from 'Butternut' and 'Waltham Butternut', as well as certain lines derived from PI 162889, were completely free of silvering at 8 weeks after planting (Table 1) and until maturity (data not shown). In contrast, at least 75% of all plants of 'Soler', PRSYN4, and the F1 hybrids were silvered by 8 weeks after planting. There were no significant differences among lines for whitefly counts taken at 8 and 10 weeks after planting (data not shown). At 12 weeks after planting, the number of eggs found on selected (derived) C. moschata lines as well as on the F1 hybrids were not different from the susceptible checks (Table 2). 'Butternut', 'Waltham', and PI 162889 derived lines generally had higher numbers of whitefly nymphs and eggs plus nymphs combined than did the silverleaf-susceptible checks. The F1 hybrids were not different from the checks. The C. pepo lines also had higher egg and nymph counts then did the susceptible checks.


Whitefly-induced silvering is phenotypically distinct from that of leaf mottling. Leaf mottling occurs in patches in the vein axils and is genetically controlled (Coyne, 1970). Whitefly-induced silvering first parallels the leaf veins and later extends into the interveinal area of the leaf. Many, if not most, of the C. moschata accessions screened for silverleaf resistance had the mottled leaf phenotype. Most accessions were susceptible to silvering whether with mottled or green leaves. But it is interesting to note that most of the initial selections were green-leafed genotypes, and all selected lines in this study had green leaves. The silverleaf susceptible checks had mottled leaves.

The silverleaf resistance observed in 'Butter

Cucurbitaceae '98

nut' and 'Waltham' derived lines was distinct in appearance to that of the PI 162889-derived lines. While no typical silvering symptoms appeared in 'Butternut' and 'Waltham' lines, leaves of these genotypes had a somewhat yellow-green, not completely normal, appearance. Leaves of PI 162889 lines maintained a healthy dark green color throughout the growing season.

'Butternut' lines were much more uniformly resistant than PI 162889. The original seed lot of PI 162889 had seed of various size and color suggesting it was an open pollinated population. The five PI 162889 lines had from 0% to 92% silvered plants at 8 weeks after planting while all 'Butternut' and 'Waltham' lines remained free of silvering (Table 1). Plants of the PI 162889 line with 92% silvered plants

(9706-13-1) had only a faint trace of silvering and otherwise appeared healthy and vigorous.

Most lines derived from PI 162889 had brown or golden color seed. Brown seed is a trait thought to be found only in land races of South American origin (Laura Merrick, personal communication). Thus, seed color, and the fact that the passport data of this PI indicates a South American origin, is good evidence that PI 162889 lines are of a genetic background quite distinct from 'Butternut' and 'Waltham' which originate in the United States (Mutschler and Pearson, 1987). 'Butternut' is part of 'Waltham's' pedigree suggesting that silverleaf resistance in these lines may be genetically similar.

The F1 hybrids had mottled leaves like 'Soler'.

Table 1. Silvered plants (%z) and average silvering rank of lines selected for resistance to whitefly-induced silvering compared to related genotypes and susceptible checks.

6 weeks 8 weeks 10 weeks

Silvered Silvered Silvered

Genotype Descriptiony plants Rank plants Rank plants Rank

9706-3-1 Butternut-derived line 0 6.4 0 5.5 0 5.4

9706-4-5 Butternut-derived line 0 6.4 0 5.5 0 5.4

9706-5-1 Butternut-derived line 0 6.4 0 5.5 0 5.4

9706-24-1 Butternut-derived line 0 6.4 0 5.5 0 5.4

9706-24-3 Butternut-derived line 0 6.4 0 5.5 0 5.4

Butternut Butternut 0 6.4 3 6.9 3 6.8

9706-11-5 PI 162889-derived line 14 8.8 30 11.4 30 10.4

9706-12-4 PI 162889-derived line 19 11.4 44 13.1 59 13.6

9706-13-1 PI 162889-derived line 84 18.0 75 16.8 92 17.0

9706-14-4 PI 162889-derived line 0 6.3 0 5.5 0 5.4

9706-15-2 PI 162889-derived line 0 6.3 0 5.5 0 5.4

9706-21-1 Waltham-derived line 0 6.4 0 5.5 0 5.5

Waltham Waltham 0 6.5 0 5.5 0 5.5

9706-3-1 x Soler Resistant x susceptible F1 56 16.0 64 14.8 77 15.1

9706-4-5 x Soler Resistant x susceptible F1 40 14.5 66 15.4 78 15.0

9706-5-1 x Soler Resistant x susceptible F1 77 17.5 88 18.2 91 16.8

Soler Susceptible check 82 17.8 88 18.2 96 17.8

PRSYN4 Susceptible check 78 17.4 80 16.6 87 16.0

9706-22-1 C. pepo 10 9.9 48 13.4 45 11.9

9706-22-3 C. pepo 36 11.6 44 12.1 68 14.1

Probablility of >F value <0.01 <0.01 <0.01

Probability of contrasts

PI 162889-derived vs. susceptible checks <0.01 <0.01 <0.01

Butternut-derived vs. susceptible checks <0.01 <0.01 <0.01

Waltham-derived vs. susceptible checks <0.01 <0.01 <0.01

Butternut-derived vs. PI 162889-derived <0.01 <0.01 <0.01

Butternut parents vs. F1 hybrids <0.01 <0.01 <0.01

F1 hybrids vs. susceptible checks 0.17 0.20 0.22

Butternut-derived vs. Butternut 1.00 0.26 0.26

Waltham-derived vs. Waltham 0.98 0.91 1.00

C. pepo vs. susceptible checks <0.01 <0.01 <0.01

zCounts were ranked from lowest to highest in each block. Genotype means are the average of ranks in four blocks.

yAll genotypes are C. moschata unless otherwise indicated.

Cucurbitaceae '98

Coyne (1970) showed that this trait is dominant. Also like the susceptible checks, the F1 plants were generally silvered, suggesting that the silvering resistance in 'Butternut' is a recessive trait. Unfortunately, lines from PI 162889 have proven to be difficult to self and cross. For that reason no PI 162889 x Soler hybrids could be included in this study.

The silverleaf resistant lines apparently have little, if any, whitefly resistance (Table 2). The data are, however, not completely clear on this point. In general, there were few differences in whitefly counts between silverleaf resistant lines, or between those lines and the silverleaf susceptible checks. Where there were differences, the susceptible checks tended to have lower numbers of whitefly juveniles. The growth habit of the silver

ing susceptible checks and F1 hybrids was very different from that of 'Butternut', 'Waltham', and PI 162889 lines. The F1 crosses and susceptible lines were very large, viney plants. This may have an effect on whitefly behavior in the plant canopy. Sampling techniques also may need to be modified depending on plant type.

More work is needed to determine the inheritance and mechanism of silverleaf resistance found in this study. If the resistance is simply inherited it would be important to know if the resistance of 'Butternut' and 'Waltham' (which is assumed to be the same) is allelic to that of PI 162889. If the genes are not allelic, then how do the mechanisms of resistance differ? More conclusive testing is needed to determine what relationship, if any, there is between

Table 2. Number of whitefly eggs and nymphs 12 weeks after planting in lines selected for resistance to whitefly-induced silvering compared to related genotypes and susceptible checks.

Eggsz Nymphsz Eggs + nymphs

Genotype Descriptiony Count Rankx Count Rank Count Rank

9706-3-1 Butternut-derived line 2.0 13.6 1.0 13.1 3.0 13.4

9706-4-5 Butternut-derived line 0.0 6.1 1.0 11.6 1.0 8.5

9706-5-1 Butternut-derived line 1.0 10.3 0.3 8.5 1.3 9.7

9706-24-1 Butternut-derived line 2.0 11.5 1.8 11.6 3.8 13.1

9706-24-3 Butternut-derived line 1.2 11.4 0.0 5.4 1.2 9.4

Butternut Butternut 0.2 7.9 0.0 5.4 0.2 5.5

9706-11-5 PI 162889-derived line 1.2 8.8 1.2 12.0 2.5 10.8

9706-12-4 PI 162889-derived line 0.0 6.1 3.2 12.8 3.2 10.5

9706-13-1 PI 162889-derived line 0.8 11.4 2.0 12.5 2.8 11.9

9706-14-4 PI 162889-derived line 0.5 9.9 0.2 7.5 0.8 8.4

9706-15-2 PI 162889-derived line 0.0 6.1 0.5 9.3 0.5 6.8

9706-21-1 Waltham-derived line 1.0 12.0 1.8 11.9 2.8 11.9

Waltham Waltham 2.0 12.8 1.5 14.8 3.5 15.0

9706-3-1 x Soler Resistant x susceptible F1 1.0 10.6 0.8 11.4 1.8 10.4

9706-4-5 x Soler Resistant x susceptible F1 0.0 6.3 0.5 9.8 0.5 7.3

9706-5-1 x Soler Resistant x susceptible F1 0.0 5.7 0.0 5.3 0.0 3.7

Soler Susceptible check 1.0 10.3 0.0 5.4 1.0 7.5

PRSYN4 Susceptible check 0.0 6.1 0.0 5.4 0.0 4.0

9706-22-1 C. pepo 1.2 11.1 1.5 11.4 2.8 11.1

9706-22-3 C. pepo 2.8 16.4 2.8 11.6 5.5 16.6

Probablility of >F value 0.06 0.08 0.04

Probability of contrasts

PI 162889-derived vs. susceptible checks 0.89 0.01 0.06

Butternut-derived vs. susceptible checks 0.21 0.02 0.02

Waltham-derived vs. susceptible checks 0.10 <0.01 0.02

Butternut-derived vs. PI 162889-derived 0.14 0.59 0.47

Butternut parents vs. F1 hybrids 0.22 0.29 0.12

F1 hybrids vs. susceptible checks 0.23 0.14 0.60

Butternut-derived vs. Butternut 0.27 0.07 0.05

Waltham-derived vs. Waltham 0.87 0.49 0.46

C. pepo vs. susceptible checks 0.01 0.01 <0.01

zCounts were totaled from three 1-cm leaf disks from each of three plants. Means are an average of total counts in four blocks.

yAll genotypes are C. moschata unless otherwise indicated.

xCounts were ranked from lowest to highest in each block. Genotype means are an average of ranks in four blocks.

Cucurbitaceae '98

silverleaf and whitefly resistances. Although silvering resistance itself is of economic importance in Cucurbita, whitefly resistance is probably the more important breeding goal.

Literature cited

Bellows, Jr., T.S., T.M. Perring, R.J. Gill, and D.H. Headrick. 1994. Description of a species of Bemesia (Homoptera: Aleyrodidae). Ann. Entomol. Soc. Amer. 87:195­206.

Conover, W.J. and R.L. Iman. 1981. Rank transformation as a bridge between parametric and nonparametric statistics. Amer. Stat. 33(3):124­129.

Costa, H.S., D.E. Ullman, M.W. Johnson, and B.E. Tabashnik. 1993. Squash silverleaf symptoms induced by immature, but not adult, Bemisia tabaci. Phytopathology 83:763­766.

Coyne, D.P. 1970. Inheritance of mottled-leaf in Cucurbita moschata Poir. HortScience 5:226­227.

Eskridge, K.M. 1995. Statistical analysis of disease reaction data using nonparametric methods. HortScience 30:478­481.

McAuslane, H.J., S.E. Webb, and G.W. Elmstrom. 1996.

Resistance in germplasm of Cucurbita pepo to silverleaf, a disorder associated with Bemisia argentifolii (Homoptera: Aleyrodidae). Fla. Entomol. 79(2):206­221.

Mutschler, M.A. and O.H. Pearson. 1987. The origin, inheritance, and instability of Butternut squash (Cucurbita moschata Duchnesne). HortScience 22(4):535­539.

Paris, H.S., P.J. Stoffella, and C.A. Powell. 1993b. Differential susceptibility to leaf silvering in Cucurbita pepo. HortScience 28:657­658.

Paris, H.S., P.J. Stoffella, and C.A Powell. 1993a. Susceptibility to leaf silvering in the cultivar groups of summer squash. Euphytica 69:69­72.

Segarra-Carmona, A., J. Bird, J. Escudero, G. Fornaris-Rullán, and R. Franqui. 1990. Silvering of Cucurbita moschata (Duchesne) Poir. associated with Bemesia tabaci Genn. (Homoptera: Aleyrodidae) in Puerto Rico. J. Agr. Univ. P.R. 74(4):477­478.

Wessel-Beaver, L. 1997. Screening for silverleaf resistance in Cucurbita accessions. Cucurbit Genet. Coop. Rpt. 20:54­56.

Yokomi, R.K., K.A. Hoelmer, and L.S. Osborne. 1990. Relationships between the sweetpotato whitefly and the squash silverleaf disorder. Phytopathology 80:895­900.

Cucurbitaceae '98