Seed Production in Watermelon: A Comparison between Two Commercially Available Pollinators
Abstract
The number of honey bees (Apismellifera L.) continues to decline due to parasitic mite pests and other factors. Honey bees and bumble bees (Bombus impatiens Cresson) were therefore compared for their effects on the seed set of watermelon [Citrulluslanatus (Thunb.) Matsum.&Nakai] in a 2-year field experiment. The experiment was a 2 x 4 + 2 factorial, comparing bee type (honey bee or bumble bee) at four visitation levels (1, 6, 12, and 18 bee visits) to pistillate flowers, with two controls: a no-visit treatment and an open-pollinated treatment. Bee visitation level had a strong positive influence on seed set (P ≤ 0.0001). All flowers bagged to prevent insect visitation aborted, demonstrating the need for active pollen transfer between staminate and pistillate watermelon flowers. Flowers visited by B. impatiens consistently contained more seed than those visited by A. mellifera, when compared at equal bee visitation levels (P ≤ 0.0001). We conclude that bumble bees have great potential to serve as a supplemental pollinator for watermelon when honey bees available for rental are in limited supply.
The use of bee pollinators in crop production is not new; since 1947, the U.S. Department 0f Agriculture, Agricultural Research Service, has supported research programs devoted to the development and management of native and imported bee species for use as alternative pollinators of specific crops (Torchio, 1990). For example, commercial seed production of alfalfa (Medicago Saliva L.) is now largely accomplished by the alkali bec, Narnia melanderi (Hymenoptera:
Received for publication 1 May 1997.Accepted for publication 26 Aug. 1997. This research was funded in part by the North Carolina Agricuitural Research Service (NCARS Projects NC 6355 and 6187) Raleigh, N.C., Koppert Biological Systems, 2856 S. Main Street, Ann Arbor, MI 48103, and the North Carolina Stute Beekeepers Association. Use of trade names in this publication does not imply endorsement by the N CARS of products named or criticism of similar ones not mentioned. We gratefully acknowledge Drs. Cavell Brownie and Mary Peet for their advice, and Dennis Adams, Leslie Newton, Jason Doheny, Lane Tabor, Mark Scepanski, Pau] Lineberger, Clifton Ryan, and Watson Hall for their technical assistance, The cost of publishing this paper was defrayed in part by the payment of page charges. Under postal regulations, this paper there~ fore must be hereby marked advertisement solely to indicate this fact.
lPhD. student; to Whom reprint requests should be addressed.
2Professor.
3Associate Professor,
28
Halictidae) and the alfalfa leafcutting bee, Megachile rotunda (Hymenoptera: Megachilidae), Several species of (Hymenoptera: Megachilìdae) have provento be efficient and manageable pollinators of orchard crops such as almonds (PrunusamygdalusBatsch), apples (MalusxdomesticaBorkh.), and pears (PyrusCommunisL.) (Free, 1993; Kuhn and Ambrose, 1984). More recently, humble bee colonies [Bombas spp. (Hymenoptera: Apidae)] have become commercially available in the United States and have had a dramatic effect on the greenhouse tornato (Lycopersiconesculentum Mill.) industry, replacing costly and laborintensive hand pollination and/or mechanical vibration (Kueneman, 1995).
In many cases, development of manageable non»honey bee pollinators has focused on the few crops that are inadequately pollinated by honey bees (Torchio, 1990). However, honey bees, which were introduced to North America from Europe during the colonial period, are generally recognized as lhe most important pollinators for commercial production of the vast majority of U.S. crops that require insect pollination (Free, 1993; McGregor, 1976; Oertel, 1980). Honey bee colonies have certain attributes that make them very efficient and reliable pollinators for most crops: they have large population sizes (up to 60,000 individuals per colony), exist on a perennial basis, can be managed using established practices, and can be transported over
considerable distances to selected locations. ln the United States, the number of honey bee colonies rented for crop pollination exceeds 2 million annually and the value added to crop production by honey bees is estimated to be U.S. $9.3 billion (Robinson et al., 1989).
Unfortunately, in the last several years the availability of honey bee colonies has become a serious limiting factor in the pollination of many bee-pollinated crops. The accidental introductions of two parasitic mite species [Acarapz'swoodi (Rennie) (Tarsonemidae) and Varroajacobsom' Oudemans (Varroidae)] have decimated honey bee populations in North Carolina, reducing the feral and managed honey bee colonies by 90% and 30%, respectively, and other states have experienced similar reduetions (Ambrose, 1997; Harbo and Hoopingarner, 1997). These recent losses to the pollinator force available for crop production are compounded by steadily declining populations of wild bees and managed honey bees, which has been attributed to increased herbicide and pesticide use over the last halfcentury and depletion of hibernation and nesting sites due to large-scale monoculture and urban sprawl (Martin and McGregor, 1973).
Due to the detrimental impact the parasitic mites have on colony survival, obtaining rental honey bee colonies for pollination may become a limiting factor in the production of many crops that are dependent upon bee pollination for fruit and seed set. With the advent of commercial bumble bee hives; a potential alternative 0r supplemental pollinator is now available when the numbers of honey bee colonies are in limited supply. Our objective in this 2-year study was to compare the effects of honey bee and bumble bee pollination on seed and subsequent fruit production in watermelon under conditions. A previous article (Stanghellini et al., 1997) dealt with the effects on fruit set and abortion. In this paper We present data for seed production.
Materials and Methods
Cultivation practices and experimental design. From mid-April to early October in 1995 and 1996, studies were conducted to compare the effects of honey bee and bumble bee Visitation and bee Visit number on seed set in Watermelon. Bee type was also compared with two controls; a no-visit (no entomophilous pollination) treatment and an unrestrictedVisit (open pollination) treatment, The field. located at the Horticultural Field Laboratory in Raleigh, NC., measured 28 50m. Sequential plantings of ‘Royal Jubilee”, a recommended cultivar in North Carolina (Schultheis, 1995), were made each year on raised, plasticcovered beds 50 m in length. Soil type in the field was an Appling Sandy Loam (Clayey Kaolinitic Thermic TypicKanhapludult). The plants were drip irrigated and fertilized in accordance with standard commercial practices (Sanders et al., 1995).
The experiment was a randomized cornplete block design with four replicates. Containerized transplants were grown in a green~ house for 21 d, hardened for about a week, then
HORTSCIENCE, VOL. 33(1), FEBRUARY 1998
planted into the field with a row spacing 0f L5 m and in-row spacing of 0.9 m, giving 53 plants per bed (two beds and 106 plants per replicate). In 1995, the field plantln g dates for the four replicates were 18 May, 30 May, 13 June, and 27 June, respectively. Bee visitation dates for the four replicates were as follows: 3 July through 20 July (replicate l); 24 July through 26 July (replicate 2); 28 July through 3 Aug. (replicate 3); and 4 Aug. through 6 Aug. (replicate 4). ln 1996, another four repliCates were planted on 21 Apr., 6 May, 22 May, and 7 June. The dates of treatments (bee visitation) were 7 June through l July and 7 July through 18 July for replicates l and 2, respectively.
Treatment. In mid-May 1995 and 1996, two honey bee colonies (=20,()00 bees per colony) and two commercial bumble bee colonies (250-75 bees per colony) were placed along the edge 0f the field when most of the watermelon plants had reached anthesis; hives of both species were replaced periodically throughout the season. The honey bee colonies were from the North Carolina State Univ. (NCSU) apiary. The bumble bee colonies contaìned a common native American species (Bambusimpatíens Cresson) and were provided by Koppert Biological Systems, Ann Arbor, Mich. The bumble bee colony housing was designed for in«greenhouse use (cardboard exteriors); therefore, they were placed in protective Wooden stands l m ín height. Voucher specimens for both bee species used in these studies were placed in the NCSU Insect Collection.
Pistillate flowers were bagged in the early evening before anthesis to control bee visits the following day. Bags (26 26 cm Closed) were fashioned from nylon pollination cage netting. The bags were edged by Velcro-like strips to enclose the pistillate flower and thus prevent unwanted visitation prior to treatment. Aluminum Wire was shaped to keep the sides of the bag from contacting the blossom.
Bee visitation treatments began when temperatures reached 21 oC on the day of anthesis; bags were removed and l7 6, 12, or 18 bee visits were allowed on each ñower by either honey bees or humble bees. Each time a bee landed on the flower, it was counted as a Visit. An individual bee could be responsible for more than one visit. An approaching forager of any other species was prevented from landing on the treatment flower by using an aluminum pole with plastic forestry ribbon attached to the end. Flowers were discarded if they were visited by an insect other than the one designated for that particular flower. Foragers from Wild bumble bee colonies were present in the field and had a variable range in body size. These foragers were sometimes not readily distinguishable from the Koppert (commercial) colonies; thus, uniformity of Bambus treatments was established by restricting flower visitation to those humble bees that had a body