Capsicum Cultivars Utilized

‘Agriset 4108’ is a F₁ hybrid, ‘Jalapeño’-derived cultivar from the USA with 9.7 × 3.3-cm, dark green to red fruit that require 70 days from planting to harvest (Reimer Seeds 2012). ‘Fresno 6022’ is another F₁ hybrid with more visibly pubescent leaves than the other pepper cultivars; it produces 7.6 × 3.3-cm, cone-shaped, green to bright red peppers, which require 74 days from planting to harvest (Reimer Seeds 2012). ‘Hot Tormenta’, another F₁ hybrid from the USA, produces 10.2 × 3.8-cm, dark green to red peppers; it has very good disease resistance and needs 77 days from planting to harvest when green (Reimer Seeds 2012). ‘Hot San Ardo’, another F₁ hybrid, produces 15.2 × 7.6-cm, green to red peppers requiring 75 days from planting to harvest with very good disease resistance (Reimer Seeds 2012). ‘Cheyenne’, a F₁ hybrid derived from ‘Cayenne’, has 20-23-cm-long fruits, and requires 65 days (green) to 85 days (red) from planting to maturity or bloom (Johnny Seeds 2012). ‘Hot Habanero Orange’ is from the USA and has orange, 2.5 × 3.8-cm fruits that require 90 days from planting to harvest (Generic Seeds 2012). ‘Red Devil Cayenne’ is a F1 hybrid derived from ‘Cayenne’ that produces 12.7 × 0.8-cm, green-to-red fruits that require 72 days from planting to harvest (Reimer Seeds 2012). ‘Numex Big Jim’ was developed by New Mexico State University, USA, and produces green-to-red, 30.5 × 7.1-cm peppers that require 85 days from planting to harvest (Chile Pepper Institute 2012, Reimer Seeds 2012). ‘Astry’, a F1 Hungarian pepper type, has 12.7 × 9.7-cm, cream to bright red peppers, which require 70 days from planting to harvest (Reimer Seeds 2012). Capsicum chinense Jacquin (Solanaceae) cultivars include ‘Hot Fatalli’ from the Central African Republic, which produces 6.4 × 2.5-cm, green-to-yellow peppers, which take 95 days from planting to harvest (Reimer Seeds 2012) and ‘Jamaican Yellow’(Jamaican Scotch Bonnet, Yellow Strain 1) from Jamaica, with 3.8 × 3.8-cm, green to yellow fruit that require 100 days from planting to harvest (Reimer Seeds 2012). None of these 11 cultivars has been issued a certificate for legal protection in the USA (PVPO 2012). All 9 cultivars of C. annuum and 2 of C. chinense have green stems, green leaves, white flowers, are upright, and 40 to 90 cm tall when mature.

Responses of S. dorsalis to Spinetoram-Treated and Untreated Capsicum Cultivars in an Open Greenhouse (Site A).

The greenhouse was 10.4 × 18.3 m and exposed to full sun with a roof of translucent fiberglass panels resulting in 60% light blockage. The entrances, juncture to the head house, and about half the areas of the east and west walls were open, hence, it was a more open greenhouse than at Site B. The test was conducted on plastic greenhouse benches, each 2.4 m × 0.91 m × 60 cm high, held 1 replication, and with potted plants placed on inverted, 128-plant styrofoam starter trays (34 × 67 × 6 cm).

Mean temperatures at 60 cm above ground level were 24.7 °C (range 15–34 °C) for Oct 2011, 22.6 °C (range 10–31 °C) for Nov, 21.1 °C (range 9–29 °C) for Dec, and 18.4 °C (range 1–30 °C) for Jan 2012 (FAWN 2012). RH at 2 m averaged 83, 79, 78, and 75% for Oct, Nov, Dec, and Jan, respectively (FAWN 2012). To determine these data, measurements were made from 10 AM to 4 PM EST, then added to values from the Homestead FAWN station. The foregoing values were 0.6 ± 1.9 °C (n = 31) above the mean temperature and 2.6 ± 12.3% (n = 31) below the mean RH of the FAWN station at TREC in Homestead (FAWN 2012).

Between 14 Jul and 9 Sep 2011, 1–2 seeds of each cultivar were planted per cell of a 128-plant styrofoam starter tray with one cultivar per tray, 2 to 6 trays per cultivar, and 11 cultivars total. On 18–19 Oct, 60 plants of each cultivar were repotted from planter trays into 10-cm pots, except for ‘Cheyenne’, which had 57 plants. Thus there were 4 or 5 plants per treatment per cultivar. The single-plant pots and styrofoam trays used the same potting soil: a standard potting mix (Fafard Growing Mix 2, Conrad Fafard Co., Agawam, Massachusetts), which consisted of Canadian sphagnum peat moss (70%), perlite (25%), and vermiculite (5%). Each plant was manually watered once per day with 16.2 ± 7.5 mL of tap water. On 20 Oct, 8 and 28 Nov, and 6 Dec 2011, plants were fertilized with 20-20-20 liquid fertilizer (Peters water soluble fertilizer, Scotts Co., Marysville, OH) according to manufacturer instructions with a fertilizer dispenser attached to a sprayer. To encourage uniform vegetative growth while allowing a later assessment of early flowering and fruiting, flowers and fruit were pruned from all plants before 1 Dec 2011 except for their pedicels, which remained on plants to be counted later. Unwanted insect infestations included greenhouse whiteflies Trialeurodes vaporariorum (Westwood) (Hemiptera: Aleyrodidae) mainly on pepper plants and green peach aphids Myzus persicae (Sulzer) (Hemiptera: Aphididae) mainly on cotton plants. On 17 Nov 2011, plants were sprayed with a mixture of pyriproxyfen (Knack®, Valent Co., Walnut Creek, California) at 8 oz/ac (568 g/ha) for whiteflies and beta-cyfluthrin (Baythroid® XL, Bayer CropScience, Research Triangle Park, North Carolina) at 3.45 oz/ ac (245 g/ha) for aphids. Follow-up applications were made on 7 and 23 Dec, but only with pyriproxyfen (Knack®). On all 3 dates, a 4-gal (15-L) backpack sprayer (Birchmeier Iris, Birchmeier Co., Switzerland) delivered 100 gal/ac (948 liter/ ha) of pesticide solution.

Four or 5 treated plants per cultivar formed the control groups by removal of S. dorsalis by spinetoram treatment. Five experimental plants per cultivar remained untreated and infested with thrips by natural migration from 130 thrips-infested cotton plants (Gossypium hirsutum L., Malvales: Malvaceae) placed on greenhouse benches and possibly from populations outside the greenhouse. Each thrips-infested cotton plant was 23–41 cm tall in a 10-cm-wide pot that was placed on benches 0.35–16.5 m from the pepper plants. Each group of plants per cultivar was sprayed or not sprayed with spinetoram and was randomly placed among the other 21 treatment combinations within each replication (bench). To facilitate removal of treated plants for spraying, each styrofoam tray held all treated or all untreated plants. Treated and untreated trays were randomized within each replication, as were the cultivars; hence, a randomized complete block design was used. On 3, 6, 24, and 28 Nov 2011, trays of insecticide-treated plants were moved to parallel benches under the same experimental conditions as untreated plants but were separated by 92–112 cm from benches of untreated plants. This distance helped to ensure that treated plants could be sprayed without the spray drift accidentally reaching untreated plants. Spinetoram insecticide was applied with a 1-L hand-pump sprayer at 8 oz /ac (584 mL /ha) of soluble concentrate delivering 70 gal/ac (655 L/ha) total volume applied to plants followed by their return to their original places on the experimental benches.

In an initial survey (25–26 Oct 2011), the proportions of plants infested and S. dorsalis population abundances were surveyed by carefully observing 2 leaves per treatment per replication using a hand-held magnifying glass (16 X). Data were also collected on plant width, depth, height, number of fruit, number of previously cut pedicels, and flowers and flower buds per plant (at least 1 mm diam). Plant height and width were measured 1–15 Dec 2011 and pedicels of early flowers or fruit, numbers of flowers, and fruit were measured 19 Dec 2011–3 Jan 2012 with all treatments in a replication measured on the same date for each variable. Pedicels of flowers and fruit removed before 1 Dec were distinguished as suberized, cut-off stumps beginning 19 Dec when counted along with the newer flowers and fruit; hence, total plant organs counted = (pedicels + flowers + fruit). Plant canopy volume was estimated by calculating the volume of the “cylinder” formed by the mean radius of the minimum and maximum branch spreads and the plant height by modifying the formula for cylinder volume: (π r²)*h = {π × [0.25 × (smallest diam of branch spread + largest diam of branch spread)]² × (plant height)}.

In the initial survey of infested and control plants (Site A), proportions infested, numbers of larvae, adults, and sum of larvae and adults were compared by one-way ANOVAs followed by Waller-Duncan multiple range tests for mean separation. For numbers of flowers, fruit, cut pedicels, total counted organs, and estimated plant volume, there were 2 pesticide treatments: treated (non-infested) and non-treated (infested) and 11 cultivars, thus, a 2-treatment × 11-cultivar factorial design. Each insecticide-cultivar treatment combination had 6 replications and 5 plants, except for replications 4-6 of treated ‘Cheyenne”, which had 4 plants. A 2-way factorial analysis of variance (ANOVA) was used to determine if there were significant interactions between pesticide and cultivar treatments. For estimated canopy volume and fruit counts, cultivar and treatment data were not pooled, but for pedicels, flowers, and total counted organs, cultivar data were pooled to compare pesticide treatments and pesticide treatments were pooled to compare cultivars. Effects of pesticide and cultivar on plant canopy volume, cut pedicels, flowers, fruit, and total counted organs were compared using non-paired T-tests between pesticide treatments or one-way ANOVAs followed by Waller-Duncan multiple range tests for mean separation among cultivars.

Choice Test with Small Capsicum Plants in a Closed Greenhouse (Site B).

A 4.1 × 9.4-m glass-panel greenhouse was used with an A-shaped roof (2.0 m to 2.6 m tall) covered by shade cloth and exposed to full sun resulting in 70% light blockage. For ventilation, the north door of the greenhouse was left open, and vents along the top of the roof formed 2 parallel openings 7.6 cm × 9.4 m separated by 1.2 m. The same kind of plastic greenhouse benches with the same kind of inverted styrofoam starter trays were used as in Site A, but each tray held one replication.

Mean temperatures 60 cm above ground level were 30.3 °C (range 21–39 °C) for Oct 2011, 28.2 °C (range 16–37 °C) for Nov, 26.7 °C (range 15–34 °C) for Dec, and 24.0 °C (range 7–35 °C) for Jan 2012. RH at 2 m averaged 93, 89, 88, and 85% for Oct, Nov, Dec, and Jan, respectively (FAWN 2012). To determine these data, measurements were made from 10 AM to 5 PM EST, then added to values from the FAWN station in Homestead resulting in greenhouse temperatures 6.2 ± 3.0 °C (n = 36) above the mean temperature and 7.2 ± 17.3% (n = 36) above the mean humidity of the FAWN station at TREC (FAWN 2012).

There were 10 cultivars (all except ‘Cheyenne’) and 6 replications: 3 replications with 2 leaves per plant and 3 replications with 4 leaves per plant. Plants with 4 leaves per plant probably would provide a larger amount of material to lure more S. dorsalis than 2 leaves per plant. However, plants with 2 leaves provided less material possibly allowing a given number of S. dorsalis to aggregate on a smaller number of leaves rendering them easier to find then with more leaves per plant.

Plants in Site B were obtained from the same starter trays, had the same potting soil, fertilizer regime, and control of unwanted pests as in Site A, and were manually watered once a day with 11.4 ± 2.9 mL tap water. On 27–31 Oct 2011, 10 plants of each cultivar were repotted from the starter trays into the final 10-cm-wide pots used in the experiment. For each replication (inverted starter tray), there were 10 potted pepper plants with one plant (5–23 cm tall) per cultivar. Each pepper plant was randomized within the circumference of an oval 30 × 40 cm between centers of pepper plants. Centers of pepper plant pots were 10 to 15 cm apart in the circumference of the oval, and each plant was spaced 10 to 40 cm from the other 9 pepper plants and 10 to 30 cm from the center of each cotton plant in the oval. The oval surrounded 5 cotton plants each 25 to 48-cm tall, growing in 10 to 11-cm-wide plastic pots, and infested with thrips. Also there were about 500 cotton plants in the greenhouse 10–38 cm tall, spaced 0.1 to 6 m from the pepper plants, and infested with S. dorsalis.

On 4 Nov 2011, an initial survey of numbers of S. dorsalis on all leaves of 6 cotton plants helped to determine infestation levels. On 7, 15, 21–22, and 29 Nov, numbers of S. dorsalis adults and larvae were recorded by carefully observing the topmost leaf of each pepper plant using a handheld magnifying glass (16 X). Scirtothrips dorsalis were sampled from the topmost leaf because of the preference by larvae and adults for newer growth (Dev 1964; Shibao et al. 1993; Seal et al. 2006, 2010). Numbers of adults, larvae, and total larvae and adults were analyzed by repeated measures ANOVAs. Measurements on dates with significant results were also analyzed individually by one-way ANOVA, but only the resulting Waller-Duncan mean separation tests, which are not available with repeated measures ANOVAs, were used in the results and tables.

Laboratory Choice Test with Leaf Disks from Capsicum Cultivars (Site C)

The growth chamber for the test was maintained at 25 ± 1 °C, 90 ± 1% humidity, 14:10 h L:D, and 29 umoles/m²/s light intensity. There were 6 replications and ten cultivars in the first test with large larvae (‘Cheyenne’ was omitted) and 9 replications and 9 cultivars in the second test with small larvae (‘Cheyenne’ and ‘Hot Habanero Orange’ were omitted). Each replication involved a large plastic Petri dish (138 mm interior diam × 14 mm interior height) lined with a layer of cotton 3 ± 2 mm thick saturated with tap water. On top of the cotton layer was placed a 124-mm-diam disk of Parafilm M laboratory film (American National Can, Chicago, Illinois). About 5–10 mm of the underlying white cotton was folded over the top of the parafilm to overlap and thus prevent the larvae from escaping. The underlying cotton had extra reinforcement so it was not thinner than the rest of the cotton beneath the parafilm. Leaf disks 25 mm in diam were collected from plants in a greenhouse, refrigerated for 1–11 days at 6.1–7.2 °C, then placed on top of the parafilm. The cotton leaf disk was in the center surrounded by the disks of Capsicum cultivars spaced with edges 2–7 mm apart and randomized in a circle with closest edges 17–24 mm from the edge of the cotton leaf disk. Replications 1–9 began with large larvae (mostly second instar) while replications 10–18 used small larvae (mostly first instar). We used first and second instars to better represent the species than if only one instar was tested: data trends may occur with one instar that are not evident with the other instar. For each test, we released 33–34 larvae onto the central cotton leaf disk, then numbers of larvae per leaf disk were recorded at 24 h and 48 h after infestation. For the third measurement, however, data were not available for the test with large larvae at 72 h or with small larvae at 120 h. Thus, we used data for 120 h after infestation for the test with large larvae and 72 h after infestation for the test with small larvae. Larval counts per leaf disk were analyzed by one-way ANOVA followed by the Waller-Duncan multiple range test for mean separation to determine responses of S. dorsalis to different pepper cultivars.

Statistical analyses

Data for proportion infested (Site A) were transformed by arcsin [square root (n)] and all other data (Sites A–C) by [square root (n + 0.25)], but non-transformed data are shown in the results and tables. All statistical analyses were performed with SAS statistical software (SAS Institute 2012).

Responses of S. dorsalis to Spinetoram-Treated and Untreated Capsicum Cultivars in an Open Greenhouse (Site A).

In the initial survey of S. dorsalis on 2 leaves per plant, significant differences were found between cultivars in proportion infested (Table 1) but not in numbers of larvae, adults, or total larvae and adults. ‘Agriset 4108’, ‘Fresno 6022’, ‘Hot Habanero Orange’’, ‘Red Devil Cayenne’, ‘Numex Big Jim’, ‘Astry’, and ‘Jamaican Yellow’ each had significantly higher proportions infested than ‘Hot Tormenta’ or ‘Hot Fatalli’ (Table 1). There were significant interactions (P < 0.05) between treatment and cultivar for numbers of fruit per plant (F = 4.14, df = 10, P = < 0.0001) and estimated canopy volume (F = 2.19, df = 10, P = 0.023), but not for cut pedicels (F = 1.80, df = 10, P = 0.068), numbers of flowers (F = 0.91, df = 10, P = 0.529), or total counted organs per plant (F = 1.20, df = 10, P = 0.302) (Tables 1 and 2). There were significant differences between cultivars for pedicels, flowers, and total counted organs (Table 1). ‘Fresno 6022’, ‘Hot San Ardo’, ‘Cheyenne’, and ‘Astry’ each produced significantly more flowers per plant than ‘Agriset 4108’, ‘Hot Tormenta’, ‘Hot Habanero Orange’, ‘Numex Big Jim’, ‘Jamaican Yellow’, or ‘Hot Fatalli’. ‘Agriset 4108’, ‘Hot Tormenta’, and ‘Cheyenne’ were “early bloomers” with significantly more cut-off pedicels per plant than ‘Hot San Ardo’, ‘Hot Habanero Orange’, ‘Numex Big Jim’, or ‘Hot Fatalli’ with ‘Agriset 4108’ significantly higher than all other cultivars for pedicels. ‘Astry’, ‘Fresno 6022’, ‘Hot San Ardo’, and ‘Cheyenne’ produced significantly more total counted organs than ‘Hot Habanero Orange’, ‘Numex Big Jim’, ‘Jamaican Yellow’, or ‘Hot Fatalli’; ‘Astry’ ranked highest in total counted organs and flowers. However, ‘Hot Habanero Orange’ and ‘Hot Fatalli’ had significantly fewer flowers and total counted organs than all other cultivars, and for pedicels, they were in the lowest statistical grouping and had the fewest (Table 1). When cultivars were pooled for spinetoram treatment, there were no significant differences between treated and untreated plants for flowers, but pedicels and total counted organs were each significantly higher for treated than untreated plants (Table 1).

TABLE 1.

PROPORTION OF PLANTS INFESTED WITH SCIRTOTHRIPS DORSALIS AND ITS DAMAGE EFFECTS ON 11 PEPPER CULTIVARS IN THE GREENHOUSE (SITE A): INITIAL SURVEY AND FACTORIAL DATA WITH NO INTERACTIONS.

Canopy volume was significantly higher for treated than untreated plants for ‘Hot Fatalli’, ‘Hot San Ardo’, ‘Hot Habanero Orange’, and ‘Red Devil Cayenne’ (Table 2), while numbers of fruit were significantly higher for treated than untreated plants for ‘Agriset 4108’. Neither canopy volume nor fruit differed significantly between treated and untreated plants for ‘Numex Big Jim’, though both canopy volume and fruit were significantly higher for treated than untreated plants for ‘Fresno 6022’, ‘Hot Tormenta’, ‘Cheyenne’, ‘Astry’, and ‘Jamaican Yellow’. Among the 11 cultivars, canopy volume differed significantly for treated and for untreated plants, and numbers of fruit differed significantly for treated but not for untreated plants (Table 2). For treated canopy volume, ‘Hot San Ardo’, ‘Hot Tormenta’, ‘Cheyenne’, ‘Red Devil Cayenne’, and ‘Astry’ were each significantly higher than ‘Hot Fatalli’, ‘Agriset 4108’, ‘Hot Habanero Orange’, ‘Numex Big Jim’, or ‘Jamaican Yellow’ with ‘Hot San Ardo’ highest and ‘Hot Fatalli’ lowest. ‘Red Devil Cayenne’, ‘Agriset 4108’, ‘Hot Tormenta’, ‘Hot San Ardo’, ‘Cheyenne’, and ‘Astry’ were each significantly higher than ‘Hot Habanero Orange’, ‘Fresno 6022’, ‘Jamaican Yellow’, or ‘Hot Fatalli’ in untreated canopy volume with ‘Red Devil Cayenne’ highest and ‘Hot Habanero Orange’ lowest. For treated fruit, ‘Fresno 6022’, ‘Agriset 4108’, ‘Hot Tormenta’, and ‘Cheyenne’ were each significantly higher than ‘Hot Fatalli’, ‘Hot Habanero Orange’, or ‘Numex Big Jim’ with ‘Fresno 6022’ highest and ‘Hot Fatalli’ lowest.

Greenhouse choice test with small plants (Site B). An initial survey of S. dorsalis (4 Nov 2011) on 6 randomly chosen plants found 479 ± 157 thrips per cotton plant indicating abundant thrips for inoculation. There were no significant differences between cultivars in numbers of adults or total larvae and adults for any measurement, or in numbers of larvae on the first, third, or fourth measurements. However for the second measurement (15 Nov), there were significant differences between cultivars in larvae per pepper plant (Table 3). ‘Astry’ had significantly more larvae than ‘Hot Fatalli’, ‘Fresno 6022’, ‘Hot Tormenta’, ‘Hot San Ardo’, ‘Hot Habanero Orange’, ‘Numex Big Jim’, or ‘Jamaican Yellow’ with ‘Hot Fatalli’ lowest.

Laboratory Choice Test with Leaf Disks from Capsicum Cultivars (Site C)

There were no significant differences between cultivars in numbers of larvae per leaf disk at 24 h after infestation for either instar or 48 h after infestation for first instars. However, for second instars at 48 h, 120 h, and first instars at 72 h, there were significant differences between cultivars (Table 4). At 48 h, ‘Red Devil Cayenne’, ‘Agriset 4108’, ‘Fresno 6022’, ‘Hot San Ardo’, ‘Numex Big Jim’, and ‘Astry’ each had significantly more larvae than ‘Hot Fatalli’ or ‘Hot Habanero Orange’ with ‘Red Devil Cayenne’ highest and ‘Hot Fatalli’ lowest. ‘Jamaican Yellow’ at 120 h had significantly more larvae than ‘Hot Fatalli’, ‘Fresno 6022’, and ‘Hot Habanero Orange’ with ‘Jamaican Yellow’ highest and ‘Hot Fatalli’ lowest. For first-instar larvae at 72 h, ‘Agriset 4108’ had significantly more than ‘Hot Fatalli’, ‘Hot San Ardo’, or ‘Fresno 6022’ with ‘Agriset 4108’ highest and ‘Hot San Ardo’ and ‘Hot Fatalli’ lowest.