THRIPS AND TOSPOVIRUSES: PROCEEDINGS OF THE 7TH INTERNATIONAL SYMPOSIUM ON THYSANOPTERA 157

Thrips: the primeval pollinators?

Irene Terry University of Utah, Department of Biology, 257 South 1400 East, Salt Lake City, UT 84112, USA E-mail: [email protected]

Introduction essential for pollination of any plant species. As From their presumed beginnings as fungal hyphae more attention is being paid to thrips activities feeders in detritus, thrips have diversified into and behaviors, more discoveries are being grass, herb and floral herbivory, pollenivory, made that speak to their potential as pollinators. carnivory, and sporophagy. Among these, Thien et al (2000) reviewed the pollenivory is quite common (see review by characteristics of the pollination biology of 29 Kirk 1997). Pollen of many plant species basal angiosperm families, including all the is very nutritional, containing essentials for basal dicots (e.g., Magnoliaceae, Lauraceae, development such as starches (up to 22% by Monimiaceae, Annonaceae, and Nymphaeaceae) dry weight), proteins (2.5-61%), lipids (1- and one basal monocot (Araceae). One common 20%) and vitamins (Roulston and Cane 2000). feature of these basal families is that most of their Certainly with so many species of species are pollinated, with wind pollination Thysanoptera known to utilize pollen, the a rarity. The dominant pollinators in these 29 basal potential for thrips to effect pollination is possible. angiosperm families are members of Coleoptera However, thrips have been systematically and Diptera, these two orders being involved overlooked by pollination biologists. These small in pollination of species in 17 and 14 families, have generally been considered only a respectively. Hymenoptera (mostly bees) and minor or secondary contributor to pollination of thrips are secondary pollinators, found in 7 and 9 of these families, respectively. One implication of some plants, many of which are crops. In part these results is that Coleoptera and Diptera, being this attitude is due to thrips lacking a number associated commonly with basal angiosperms, as of characters that are deemed essential to be an well as pollinators of some gymnosperms, are “efficient pollinator” (Kirk 1997): 1) thrips are possibly the more primitive pollinators and have tiny and have no specific organs or structures shifted to and became more specialized on higher that carry pollen; 2) they carry only a small angiosperms. Other pollinators, such as moths, number of pollen grains per individual; and 3) butterflies, birds, and bats, likely evolved later. they are assumed to be poor fliers with little However, Thien et al. (2000) caution that thrips directed flight, and rarely leave their flowers. importance in pollination is often overlooked. But these traits are not always true for thrips, nor are these traits always limitations. Thrips Pollination syndrome do have some directed flight, and some species Although the utility of the term “pollination do move between flowers very often. Members syndrome” (i.e., plants, sometimes unrelated, of both Frankliniella and Thrips genera are with similar floral traits and similar types of considered important pollinators of some crops. pollinator) has been questioned (Johnson and Finally, even though individual thrips only carry Steiner 2000), it may serve as a good starting a few grains, sometimes up to hundreds of grains point for analyzing particular associations of (Kirk 1997), they can move between plants and pollinators with their floral hosts. Flowers that flowers in high numbers. Even Darwin (1876, tend to be associated with ‘thripophily’ (as 1877) observed the movement of thrips carrying reviewed by Kirk 1997) are medium sized, with pollen between convolvulus flowers and noted white to yellow color, they are sweetly scented how they could interfere with pollination with or without nectar, their structure is compact, experiments. Until recently, though, there have globose, or urceolate, with pollen chamber or been no definitive studies showing that thrips are shelter, and their pollen grains are small and dry. 158 THRIPS: THE PRIMEVAL POLLINATORS? THRIPS AND TOSPOVIRUSES: PROCEEDINGS OF THE 7TH INTERNATIONAL SYMPOSIUM ON THYSANOPTERA 159

Some of these traits do fit the floral clade among extant gymnosperms (Bowe et al. morphologies of some recently reported subtropical 2000, Chaw et al. 2000). Cycads were most plants (and some are in basal angiosperm diverse during the Mesozoic Era - Jurassic period families), where thrips play a role in pollination. (144-213 MYA), when the first flowering plants 1) Webber and Gottsberger (1995) found thrips were evolving (Friis et al. 1987, Thomas and visitors carrying pollen between flowers on two Spicer 1987). During most of the 20th century, species of Amazonian Annonaceae, Bocageopsis cycads were thought to be wind-pollinated like multiflora and Oxandra euneura. Flowers are other gymnosperms. The first study to show that small and white (< 4 mm across) with a tiny cycads were insect pollinated was by Norstog et pollen chamber formed by overlapping petals al. (1986). They observed pollen laden weevils, and stamens. Other members of this family have Rhopalotria mollis (Sharp), moving from male medium to large sized flowers and are pollinated cones to female cones of Zamia furfuracea, and by . The sweet odors were stronger during they demonstrated that seed set significantly the day. 2) Two species of Macaranga (Family declined when beetles were excluded from Euphorbiaceae) trees are likely pollinated by female cones. Strong mutualisms between thrips: M. velutiniflora, a newly described species specialist beetles in the weevil superfamily in Borneo (Davies 1999), and M. hullettii, a (Coleoptera, Curculionoidea) (Oberprieler 1995a, common species in southeast Asia (Moog et Oberprieler 1995b), and their cycad hosts are now al. 2002). A phlaeothripid species of the genus known across most cycad-bearing continents. Neoheegeria has been found on both male and The cycad genus Macrozamia (Zamiaceae) female flowers of M. hullettii carrying pollen. is found only in Australia, and there are The trees are dioecious, and both male and female approximately 40 extant species. Most of these inflorescences are tiny and hidden. The staminate species are found near the east coast, with flowers are about 1 mm long, and multiples of three species being found in the southwestern these flowers are enclosed by a greenish bracteole. coastal area near Perth. However, one species, Female flowers are slightly larger but subglobose. M. macdonnellii, is found in the central desert A vanilla-like odor is emitted from flowers of M. interior, in scattered isolated populations hullettii. 3) Thrips setipennis were the only insects throughout the Macdonnell and Hart Ranges, found on both male and female flowers, and are at least 1400 km in all directions from all other the likely pollinators, of Wilkiea huegeliana Macrozamia species. Numerous insect species (Monimiaceae), a rainforest tree in Queensland, have been found on male and female cones of Australia (Williams, et al. 2000). Individual Macrozamia (Forster, et al. 1994), but only flowers are tiny (<4.5mm), white and have only a beetles, primarily Tranes spp. (Coleoptera, tiny ostiole by which these insects enter the flower. ), and wind were thought to be In these studies, thrips use tiny and pollen vectors, even though thrips in the genus sometimes hidden flowers rather than medium to Cycadothrips (Terebrantia: Aeolothripidae) had large flowers, but globose and urn-like shapes and been observed by the thousands in male cones small openings are common. Because thrips have of several species. Cycadothrips chadwicki been widely overlooked, pre-conceived notions Mound was reported on male cones of M. about thripophily should not be taken as complete, communis (Mound 1991, Chadwick 1993) and and the focus should be placed on searching C. emmaliami Mound & Marullo was found for thrips on plant species where pollinators on both male and female cones of M. riedlei in are not known, regardless of their floral traits. southwestern Australia (Mound, et al. 1998). During October through December 1999, a Discovery of thrips pollination of cycads study was undertaken to examine the potential One unusual thrips pollination system has been for C. chadwicki to pollinate Macrozamia discovered on cycads in Australia that should communis, a cycad found on the southeastern bring much more focus on thrips as pollinators. coast of New South Wales, using exclusion Cycads (Cycadales) are dioecious plants of experiments as well as determining pollen Paleozoic origin, and are considered the basal loads and behavioral observations (Terry 2001). 158 THRIPS AND TOSPOVIRUSES: PROCEEDINGS OF THE 7TH INTERNATIONAL SYMPOSIUM ON THYSANOPTERA 159

In this study, thrips were observed moving a very strong and pungent odor that humans between cones, including from male to female could detect at least 10 m away from cones. cones, during the day time only. Average pollen Thrips moved out of M. macdonnellii male cones loads of thrips were determined by counting daily en masse in the late afternoon. Female M. the pollen grains around thrips bodies caught macdonnellii cones were receptive (had tiny gaps on sticky traps as they were leaving male cones between sporophylls, emitted odor similar to male or arriving on female cones. Those leaving cone, attracted thrips) for perhaps only one day. male cones had slightly higher loads than those Over 5000 thrips were caught on one sticky trap arriving at female cones (Table 1). Thrips visited collar (2cm wide and 45 cm diameter) around a and carried pollen to female cones. Seed set receptive female cone in a single afternoon. Pollen was high in cycad cones where weevils were loads per thrips leaving male cones averaged excluded and was not significantly different around 20 grains per thrips, and averaged slightly from open controls (62% versus 59.9%, controls over 15 grains on thrips arriving at female cones versus exclusion, respectively). Further, (Table 1). Pollen loads and estimates of total excluding wind from vectoring pollen did not thrips visitation at the female cone indicated a significantly reduce seed set (62% versus 57.7%, pollen delivery of >5700 grains per ovule in a control versus wind exclusion, respectively). single afternoon (Table 1). Thus C. albrechti Total pollen grains delivered to each female appears to be the sole pollinator of this species. ovule was estimated to be over 1000 grains, more than sufficient to achieve fertilization. Implications of thrips pollination of cycads While this study was underway, a new Surveys of insects on Macrozamia cones (see species, Cycadothrips albrechti Mound and Terry, was discovered on male cones of Macrozamia review by Terry 2001) suggest that at least four macdonnellii, the desert cycad of central Macrozamia species are pollinated by only Australia. Mound and Terry (2001) examined Cycadothrips spp.; eight species are pollinated the interaction between this thrips species and only by Tranes spp. weevils (Coleoptera: its cycad host, to determine the potential for Cuculionidae); and three species have both the thrips to pollinate M. macdonnellii. A brief insects. More than 20 other Macrozamia species summary of the results of the observations have not been surveyed for cone visitors. On on this desert cycad and a discussion of the other continents, only beetles are associated with significance of other cycad studies follows. cycads, although researchers have not specifically Cycadothrips albrechti was the only looked for thrips and may have overlooked them. potential pollinator species found on cones in If a thrips/cycad association is found only on a survey of several different M. macdonnellii Australian Macrozamia, then thrips association populations. As many as 50,000 thrips per male with cycads may be recent. Macrozamia genus cone were estimated on some male cones during is at least late Cretaceous in origin (Pole and pollen dehiscence, based on subsamples of Douglas 1999) based on fossil records, but individual sporophylls from male cones. Thrips fossils of thrips on cycads are lacking. However, mated and oviposited on male cones, and both biogeographical information can be used in adults and larvae fed on pollen inside sporangia. lieu of fossil evidence to establish a possible During pollen dehiscence, male cones emitted age of thrips associated with Macrozamia.

Pollen grains per thrips or ovule M. communis M. macdonnellii Leaving male plant 41.9 (6.3) 20.5 (3.8) Arriving at female plant 20.5 (2.8) 15.1 (3.2) Ovule 1 1218 5700

1 Pollen grains per ovule estimate based on average estimate of thrips arriving at female cones (number caught on sticky traps corrected for trap size relative to cone size) multiplied by the pollen load per individual thrips arriving at female cones, all divided by the average number of ovules per cone.

Table 1. Pollen loads of Cycadothrips spp. (SE) and estimate of pollen grains per ovule in Macrozamia communis and Macrozamia macdonnellii 160 THRIPS: THE PRIMEVAL POLLINATORS? THRIPS AND TOSPOVIRUSES: PROCEEDINGS OF THE 7TH INTERNATIONAL SYMPOSIUM ON THYSANOPTERA 161

The cycad genus Macrozamia is found on fossil is dated at 65 MYA. This is from New both east and west coasts of Australia as well as Zealand, which separated from Australia around in a small area within the central desert region. 85 MYA. The earliest fossil of the weevil These cycads probably had a continent wide superfamily, Curculionoidea, is from the Late distribution during a previous geological period, Jurassic, represented by Nemonychidae, and the and species now survive only in a few areas as modern families (Brentidae and Curculionidae) relicts. Cycadothrips is the only pollinator found in the early Cretaceous, although the modern in all three regions, and this thrips is not found on weevil genera such as Tranes spp. that are other plants. Marine incursions and subsequent associated with cycad pollination did not evolve drying in the southern part of Australia during the until the Cenozoic Era, in the Paleocene or Eocene 50 MYA may have isolated the eastern Eocene (Oberprieler 1995b). Evidence suggests and western species of the southern Macrozamia that weevil pollinators of modern cycads are populations and their pollinators (Ladd and derived from angiosperm dwelling ancestors Connell 1993), but even earlier vicariance events that developed in reproductive organs or bored are possible. The massive marine intrusion during in wood, rather than from the older gymnosperm the Cretaceous ~114-119 MYA (Cranston and feeding weevil lineage, Nemonychidae, which has Naumann 1991, Beynon, et al. 1992) fragmented never been found on either fossil or extant cycads the continent into eastern, central, and western (Oberprieler 1995a, Oberprieler 1995b). It is now islands. These islands match the current believed that each cycad bearing continent has had Macrozamia and Cycadothrips distributions and an independent evolution of its weevil pollinators. endemism. Thus, one possibility is that thrips If this is true, then other pollinators of cycads mutualism with Macrozamia existed before these presumably existed before these weevils evolved. events. Because this argument is circumstantial, The insect order Thysanoptera is Paleozoic however, further corroborating evidence is in origin (Kukalova-Peck 1991, Labendeira and needed to give more validity to this time frame. Seposki 1993); thus ancestors in the basal groups of Thysanoptera predate angiosperms and some It is possible that thrips were involved in of the modern genera of cycads. Finally, the early cycad pollination systems before weevil genus Cycadothrips has been placed in its own involvement. Cycads are at least Permian in sub-family, Cycadothripinae within one of the origin (Table 2) and the earliest Macrozamia basal thysanopteran families, Aeolothripidae.

ERA Period MYA1 Plant evolution Insect evolution Rise of true/modern Cenozoic Eocene 38 weevils, including the genus Tranes 2 Rise of modern weevil Paleocene 55 genera? Diversification of 1st record of Macrozamia Mesozoic Cretaceous 65 weevils; rise of modern 1st record of a modern cycad family weevil families Jurassic 144 1st angiosperms? 1st Cucurlionoidea Triassic 213 Cycad dominance 1st beetles 1st thrips / Paleozoic Permian 244 Many extinct cycads endopterygote insects Carboniferous 280 1st cycads?

1 approximate first year of the period, million years ago 2 Tranes is one of the modern weevil genera whose members are pollinators of Lepidozamia spp. and some Macrozamia spp. Cycads

Table 2. Time of appearance of particular insect and plant groups 160 THRIPS AND TOSPOVIRUSES: PROCEEDINGS OF THE 7TH INTERNATIONAL SYMPOSIUM ON THYSANOPTERA 161

Although the exact relationships among these Friis EM, Chalconer WG and Crane PR (eds). basal families and of the Cycadothripinae 1987. The origins of angiosperms among its sister taxa are unresolved, this genus and their biological consequences. is among the basal clades of Thysanoptera Cambridge University Press, Cambridge. (Marullo and Mound 1995), suggesting that Johnson SD and Steiner KE. 2000. Generalization versus ancestors of this clade were likely around during specialization in plant pollination systems. the evolution of some gymnosperms and before Trends in Ecology and Evolution 15, 140-143. angiosperms. Based on this information, thrips Kirk WDJ. 1997. Feeding, pp. 119-174. In Lewis, T. (ed.), Thrips as Crop Pests. may be among the oldest pollinators of plants. CAB International, Wallingford. Kukalova-Peck J. 1991. Fossil history and the evolution References of Hexapod structures, pp. 141-179. In Beynon RM, Struckmeyer HIM and Totterdell CSIRO (ed.), The Insects of Australia, 2 ed. JM. 1992. Australia – evolution of a Melbourne University Press, Melbourne. continent. Bureau of Mineral Resources Labendeira CC and Seposki JJ. 1993. Insect diversity Palaeogeographic-Group, Australian in the fossil record. Science 261, 310-315. Government Publishing Service, Canberra. Ladd PG and Connell SW. 1993. Vicariant Macrozamia Bowe LM, Coat G and de Pamphilis C. 2000. Phylogeny species in southern Australia?, pp. 225-240. of seed plants based on all three genomic In Vorster, P. (ed.), Proceedings of the Third compartments: Extant gymnosperms are International Conference on Cycad Biology: monophyletic and Gnetales closest relatives Conservation through cultivation, vol. 3. are conifers. Proceedings of the National Cycad Society of South Africa, Stellenbosch. Academy of Science 97, 4092-4097. Marullo R and Mound LA. 1995. Su una Chadwick CE. 1993. The roles of Tranes lyterioides classificazione sopra-generica della and T. sparsus Boh (Col., Curculionidae) famiglia Aeolothripidae (Thysanoptera), in the pollination of Macrozamia communis pp. 87-90, Atti del XVII Congresso Nazionale Italiano di Entomologia, Udine. (Zamiaceae), p. 77-80. In Stevenson, DW and Moog U, Fiala B, Feerle W and Maschwitz Norstog NJ (eds), Proceedings of CYCAD U. 2002. Thrips Pollination of The 90, Second International Conference on Dioecious Ant Plant Macaranga hullettii Cycad Biology. Palm and Cycad Societies of (Euphorbiaceae) in Southeast Asia. Australia Ltd., Milton, Brisbane, Australia. American Journal of Botany 89, 50-59. Chaw SM, Parkinson CL, Cheng Y, Vincent TM and Mound LA. 1991. The first thrips species (Insecta, Palmer JD. 2000. Seed plant phylogeny Thysanoptera) from cycad male cones inferred from all three plant genomes: and its family level significance. Monophyly of extant gymnosperms and origin Journal of Natural History 25, 647-652. of Gnetales from conifers. Proceedings of the Mound LA, den Hollander E. and den Hollander L. National Academy of Science 97, 4086-4091. 1998. Do thrips help pollinate Macrozamia Cranston PS and Naumann ID. 1991. Biogeography, cycads? Victorian Entomologist 28, 86-88. pp. 180-197. In CSIRO (ed.), The Mound LA and Terry I. 2001. Pollination of the Insects of Australia, 2 ed. Melbourne central Australian cycad, Macrozamia University Press, Melbourne. macdonnellii, by a new species of basal Darwin C. 1876. The Effects of Cross and Self Fertilisation clade thrips (Thysanoptera). International in the Vegetable Kingdom. Murray, London. Journal of Plant Sciences 162, 147-154. Darwin C. 1877. The Different Forms of Flower on Norstog K, Stevenson DW and Niklas KJ. 1986. The role of beetles in the Plants of the Same Species. Murray, London. pollination of Zamia furfuracea L. fil. Davies SJ. 1999. A New MyrmecophyticThrip- (Zamiaceae). Biotropica 18, 300-306. pollinated Species of Macaranga Oberprieler RG. 1995a. The weevils (Coleoptera: from the Highlands of Sarawak. Curculionoidea) associated with cycads Harvard Papers in Botany 4, 433-437. 1. Classification, relationships and Forster PI, Machin PJ, Mound LA and Wilson biology, pp. 295-365. In Vorster, P. GW. 1994. Insects associated with (ed.), Third International Conference on reproductive structures of cycads in Cycad Biology: Conservation through Queensland and North-east New South cultivation, vol. 3. Cycad Society of Wales, Australia. Biotropica 26, 217-222. South Africa, Stellenbosch, South Africa. 162 THRIPS: THE PRIMEVAL POLLINATORS? THRIPS AND TOSPOVIRUSES: PROCEEDINGS OF THE 7TH INTERNATIONAL SYMPOSIUM ON THYSANOPTERA 163

Oberprieler RG. 1995b. The weevils (Coleoptera: Thien LB, Azuma S and Kawano S. 2000. New Curculionoidea) associated with cycads. perspectives on the pollination biology 2. Host specificity and implications for of basal angiosperms. International cycad , pp. 335-365. In Vorster, Journal of Plant Sciences: 161, 225-235. P. (ed.), Third International Conference Thomas BA and Spicer RA. 1987. The on Cycad Biology: Conservation through Evolution and Palaeobiology of cultivation, vol. 3. Cycad Society of Land Plants. Croom Helm, London. South Africa, Stellenbosch, South Africa. Webber AC and Gottsberger G. 1995. Floral Pole M and Douglas B. 1999. Plant macrofossils biology and pollination of Bocageopsis of the Upper Cretaceous Kaitangata multiflora and Oxandra euneura in Coalfield, New Zealand. Australian Central Amazonia, with remarks on the Systematic Botany 12, 331-364. evolution of stamens in Annonaceae. Roulston TH and Cane JH. 2000. Pollen nutritional Feddes Repertorium 106, 515-524. content and digestibility for . Plant Williams G, Adams P and Mound LA. 2000. Systematics and Evolution 222, 187-209. Thrips (Thysanoptera) pollination in Terry LI. 2001. Thrips and weevils as dual, Australian subtropical rainforests, with specialist pollinators of the Australian particular reference to pollination of cycad Macrozamia communis Veiny Wilkiea, Wilkiea huegeliana (Zamiaceae). International Journal (Tul.) A. DC. (Monimiaceae). of Plant Sciences 162, 1293-1305. Journal of Natural History 35, 1-21.