90 Plant Protection Quarterly Vol.8(3) 1993 Kloot, P.M. (1987c). The naturalized flora of South Australia. 3. Its origin, intro- duction, distribution, growth forms and significance. Journal of the Adelaide Botanic Gardens 10, 99-111. Menz, K.M. and Auld, B.A. (1977). Gal- vanized burr control and public policy toward weeds. Search 8, 281-7. Pate, J.S. and Dixon, K.W. (1982). ‘Tuber- ous, cormous and bulbous plants’. (University of Western Australia Press, Nedlands). Parsons, W.T. and Cuthbertson, E.G. (1992). ‘Noxious weeds of Australia’. (Inkata Press, Melbourne, Australia). Phillips, R. and Rix, M. (1989). ‘Bulbs’. 2nd edition. (Pan Books, London, UK). Wilkinson, L. (1990). Systat, the system for statistics, version 5.0. Systat Inc., Champaign, Illinois, USA. Zar, J.H. (1986). ‘Biostatistical analysis’. 2nd edition. (Prentice-Hall, New Jer- sey, USA).

Figure 6. The response of corm production to slashing at selected intervals. The time interval between 1, 2 and 3 is two weeks; the remaining intervals one week. Labels indicate those points which are significantly different: Gumeracha (g), Balhannah (b), both of the other sites (*).

erences to organisms attacking Watsonia Prospects for biological control of Watsonia species species (Table 1). Most records are of polyphagous or pest organisms and need not be considered further. Le Maitre John K. Scott, CSIRO Division of Entomology, Private Bag, PO Wembley, (1984) observed that up to 78.1% of ova- WA 6014, Australia. ries had been bored or partly eaten by in- sects in a study of seed predation of Summary concern with the invasion of Watsonia Watsonia borbonica (Pourret) Goldblatt in Watsonia species are suitable targets for species into native vegetation has lead to South Africa. Le Maitre (1984) suggests biological control if it is accepted that the cultural and chemical control strategies that the synchronized flowering follow- threat to native vegetation far outweighs being developed in Western Australia ing fire has evolved in W. borbonica to their horticultural potential. There are (Day 1992). In South Australia, Watsonia minimize pre- and post- dispersal seed few publications on organisms associated meriana (L.) Miller cv bulbillifera (= W. predation. The insects involved were with Watsonia. Potential biological con- bulbillifera Mathews and L. Bolus) is de- probably Bruchela species (= Urodon) trol agents from the region of origin, clared a Class 5 weed (non-agricultural (Crowson 1984) which have been re- South Africa, include seed-feeding bee- weeds, the control of which is required in corded from a Watsonia species in Orange tles (Bruchela spp.). However, a better part of the State only) (Parsons and Free State (Scott 1929). The distribution of understanding of the taxonomy, distri- Cuthbertson 1992). Possibly the best long Bruchela species in South Africa includes bution and importance of Watsonia term control method for weeds in nature Table Mountain in the south west Cape species as weeds is needed to justify conservation areas is biological control. Province (Scott 1929), an area climatically starting a biological control program. Here I review the prospects for control of similar to south west Australia. The host Watsonia species by biological means. A range of the Bruchela species associated Introduction discussion of the advantages, disadvan- with Watsonia species is unknown. At least six species of Watsonia have be- tages and types of biological control can Bruchela lilii (F.) and Bruchela gladioli come naturalized in Australia (Cooke be found in Wapshere et al. (1989). Classi- (Scott) are recorded from seeds of Gladi- 1986). All are native to South Africa and cal or inoculative biological control is olus (Anderson 1947, Scott 1930). Bruchela were introduced to Australia as usually the method attempted first. In the species are also known to feed on seeds of ornamentals from as early as 1842 (Par- context of Watsonia species, this would plants from the families Resedaceae and sons and Cuthbertson 1992). Watsonia mean the importation into Australia of Cruciferae (Crowson 1984). A seed preda- species were acknowledged as weeds of host-specific biotic control agents from tor would be useful to slow the spread of roadsides in Western Australia long be- South Africa, the region of origin. weeds that reproduce by seed, but they fore the current awareness of the threat would have no effect on the major weedy weeds pose to nature conservation (Carne Potential biological control agents form of watsonia in Australia, Watsonia and Gardner 1928, Meadly 1968). Like- for Watsonia species meriana cv bulbillifera, which only repro- wise, Watsonia species have been included The plant has never been surveyed for as- duces asexually by cormils and corms on the noxious weed list in Victoria from sociated pathogens and fauna. An exten- (Goldblatt 1989, Parsons and Cuthbertson as early as 1922 (Parsons 1973). Continued sive search of the literature found few ref- 1992). Plant Protection Quarterly Vol.8(3) 1993 91 Table 1. Arthropods and pathogens associated with Watsonia species. Organism Host Watsonia spp. Source Comments A mosaic disease Watsonia marginata (L.f.) Ker Smith and Brierley (1944) Possibly turnip mosaic virus (Moore 1979). Wide host range. Fungi: Fusarium sp. Watsonia spp. McClellan (1945) Pest of horticulture Uromyces transversalis (Thuem.) Wint. W. densiflora Baker, Doidge (1926, 1948) Also on Gladiolus and Tritonia (= Uromyces watsoniae Syd.) W. angusta Ker and species (Doidge 1948) W. meriana (L.) Miller Penicillium gladioli McCull. and Thom Watsonia sp. Anon. (1943) Storage rot disease of Iridaceae bulbs present in Australia (Moore 1979) Pestalotia watsoniae Verwoerd and W. borbonica (Pourret) Verwoerd and Dippenaar Only known from Watsonia Dippenaar Goldblatt (1930) spp. (Guba 1961, Laughton 1948) Insects: A weevil W. borbonica Le Maitre (1984) Probably Bruchela sp. Bruchela lilii (F.) (Urodontidae) Watsonia sp. Scott (1929) Also recorded from Gladiolus sp. (Anderson 1947) Imbrasia cytherea (F.) (Saturniidae) Watsonia sp. Annecke and Moran (1986) Polyphagous pest Macrosiphum euphorbiae (Thomas) Watsonia sp. Blanchard (1922) Polyphagous pest (Aphididae) Rhopalosiphum rufiabdominalis (Sasaki) Watsonia sp. Millar and Durr (1985) Polyphagous pest (Aphididae) Sitobion africanum (Hille Ris Lambers) Watsonia sp. Millar and Durr (1985) Polyphagous pest (Aphididae)

The host range of the rust fungus, Asia. None is native to Australia. The other be ‘critical’ test plants in any assessment Uromyces transversalis (Thuem.) Wint., three subfamilies of Iridaceae have Aus- of potential biological control agents. warrants study as rust fungi are often host tralian representatives (Goldblatt 1991). specific and have proved to be successful Watsonia species have no close relatives Research priorities biological control agents. However, that are crops, indeed the only significant Too little is known about Watsonia species Doidge (1948) states that U. transversalis crop in the Iridaceae is saffron, Crocus to justify a biological control project in the causes a serious disease on gladioli and sativus L. In contrast, many Iridaceae are immediate future. At least 251 species of that there appears to be no difference be- important garden and cut flower species, plants from southern Africa are natural- tween this fungus and U. watsoniae Syd. for example, crocus, freesia, gladiolus and ized in Australia (Scott and Delfosse 1992). which was described from a rust found iris (Goldblatt 1991). However, they are Many of these are serious weeds and are on Watsonia species. Furthermore, placed in the Tribe Ixieae (Goldblatt 1991), being considered for biological control. Viennot-Bourgin (1978) describes this whereas Watsonia is grouped in the Tribe Available resources for biological control fungus as a major introduced pest of Watsonieae with Lapeirousia, Savannosiphon research will limit this type of work to the gladioli in Morocco and representing a and Thereianthus, none of which are impor- most prominent species. At present this major quarantine threat to Europe. tant horticultural genera. Watsonia species does not include Watsonia. There appears Pestolatia watsoniae Verwoerd and are extensively planted in South Africa to have been no studies of Watsonia species Dippenaar is another fungus found on (Goldblatt 1989), but less so in Australia, reported in Weed Abstracts between 1976 leaves of Watsonia species (Verwoerd and possibly because the plant has been de- and 1992 (Scott and Delfosse 1992) which Dippenaar 1930, Laughton 1948). Fungi in clared a noxious weed for some time. is one indication of the low level of concern this genus are mostly saprophytes (Gupa Watsonia species are considered to be engendered by these weeds in the past. A 1961) and it is not known if P. watsoniae is potentially important horticultural species first priority is to clarify the taxonomy of a primary pathogen. that are being bred for selection of cultivars the plant. Which species are threats to na- (Goldblatt 1991). The importance of this tive vegetation? Where are they found? Ac- Prospects for biological control potential use would need to be resolved curate taxonomy using techniques such as Taxonomic relatedness is an important fac- before Watsonia species could be de- isozyme analysis and cytogenetics are es- tor influencing whether or not to start a clared targets for biological control. sential. Detailed studies of the ecology of biological control project. A comparatively Scott and Panetta (1993) developed a Watsonia species are needed so that there is wide host range i.e. within the host genus method for predicting the weed status of an informed basis for control techniques. or between closely related host genera, southern African plants. Watsonia meriana A preliminary survey in South Africa is may be permitted for biological control was one species found to be weedy in Aus- also needed to ascertain the extent of poten- agents of weeds that have no close relatives tralia but not so in southern Africa. One ex- tial biological control agents. Together, these in Australia. Watsonia species belong in the planation is control by biotic factors, some types of studies would greatly increase the Ixioideae, the largest subfamily in the of which are known (Table 1), in the native chance of successful biological control. Iridaceae, which has three tribes, habitat. Thus the prospects for biological Pillansieae, Watsonieae and Ixieae includ- control are positive. While the plant is Acknowledgments ing at least 27 genera. The Ixioideae are al- taxonomically isolated from Australian I thank J. Matthiessen, S. Morrison and P. most exclusively African with some spe- crops and native plants, it is related to im- Yeoh for comments on drafts of the cies found in Madagascar, Europe and portant horticultural species which would manuscript. 92 Plant Protection Quarterly Vol.8(3) 1993 References systematic monograph. Annals of (1992). ‘Noxious weeds of Australia’. Anderson, W.H. (1947). Larvae of some Kirstenbosch Botanic Gardens 19, 1-148. 692 pp. (Inkata Press, Melbourne). genera of Anthribidae (Coleoptera). Goldblatt, P. (1991). An overview of the Scott, H. (1929). Biological note on the Annals of the Entomological Society of systematics phylogeny and biology of South African beetle Urodon lilii, America 40, 489-517. the African Iridaceae. Contributions Fahraeus. Proceedings of the Entomo- Anon. (1943). Service and regulatory an- from the Bolus Herbarium 13, 1-74. logical Society of London 4, 99-100. nouncements. List of intercepted plant Guba, E.F. (1961). ‘Monograph of Scott, H. (1930). A new East African spe- pests 1942. SRA, BEPQ, US Dept. Agric., Monochaetia and Pestalotia’. 342 pp. cies of Urodon (Coleoptera) bred from 41 pp. (not seen, Review of Applied My- (Harvard University Press, Cambridge, Gladiolus, with notes on the biology of cology 22, p. 416). UK). the genus. Entomologist Monthly Maga- Annecke, D.P. and Moran, V.C. (1986). Laughton, E.M. (1948). Some South Afri- zine 66, 104-9. ‘Insects and mites of cultivated plants can Pestalotia species. Bothalia 4, 821-36. Scott, J.K. and Delfosse, E.S. (1992). in South Africa’. 383 pp. (Butterworths, Le Maitre, D.C. (1984). A short note on Southern African plants naturalized in Durban). seed predation in Watsonia pyramidata Australia: a review of weed status and Blanchard, E.E. (1922). Aphid notes, Part (Andr.) Stapf in relation to season of biological control potential. Plant Pro- I. Argentine species of the subtribe burn. Journal of South African Botany 50, tection Quarterly 7, 70-80. Macrosiphina (Homoptera). Physis, 407-15. Scott, J.K. and Panetta, F.D. (1993). Pre- Buenos Aires 5, 184-214. (not seen, Re- Meadly, G.R.W. (1968). Watsonia dicting the Australian weed status of view of Applied Entomology A 10, p. 606). (Watsonia species and Chasmanthe southern Africa plants. Journal of Bioge- Carne, W.M. and Gardner, C.A. (1928). aethiopica (L.) N.E. Brown). Journal of the ography 20 (in press). “Watsonia.” (Antholyza aethiopica.). Department of Agriculture of Western Smith, F.F. and Brierley, P. (1944). Pre- Journal of Agriculture of Western Aus- Australia 4th Series 9, 380-3. liminary report on some mosaic dis- tralia 2nd Series 5, 356-7. McClellan, W.D. (1945). Pathogenicity of eases of iridaceous plants. Phytopathol- Cooke, D.A. (1986). Iridaceae. Flora of the vascular fusarium of gladiolus to ogy 34, 593-8. Australia 46, 1-66. some additional iridaceous plants. Verwoerd, L. and Dippenaar, B.J. (1930). Crowson, R.A. (1984). On the systematic Phytopathology 35, 921-30. Descriptions of some new species of position of Bruchela Dejean (Urodon Millar, I.M. and Durr, H.J.R. (1985). A list South African fungi and of species not Auctt.) (Coleoptera). The Coleopterists and host plant index of the aphids previously recorded from South Africa. Bulletin 38, 91-3. (Homoptera: Aphidiodea) of South Af- South African Journal of Science 27, 326- Day, P. (1992). “Watsonia” and its con- rica. Entomology Memoir Department of 30. trol. Wildflower Society of Western Aus- Agriculture and Water Supply, Republic of Viennot-Bourgin, M.G. (1978). Uromyces tralia, Eastern Hills Branch Occasional South Africa 62, 1-38. transversalis (Thüm.) Wint. parasite Publication 4. 13 pp. Moore, W.C. (1979). ‘Diseases of bulbs’. dangereux des cultures de glaieuls. Doidge, E.M. (1926). A preliminary study 205 pp. (Her Majesty’s Stationary Of- Compes Rendus des Seances de l’Academie of the South African rust fungi. Bothalia fice, London, UK). d’Agriculture de France 64, 880-5. 2, 1-228. Parsons, W.T. (1973). ‘Noxious weeds of Wapshere, A.J., Delfosse, E.S. and Cullen, Doidge, E.M. (1948). South African rust Victoria’. 300 pp. (Inkata Press, Mel- J.M. (1989). Recent developments in fungi. Part V. Bothalia 4, 919-37. bourne). biological control of weeds. Crop Protec- Goldblatt, P. (1989). The genus Watsonia a Parsons, W.T. and Cuthbertson E.G. tion 8, 227-50.

Watsonia, weed or optimal habitat?

Garry Connell, Ecologia Environmental Consultants, 120 McKenzie Street, Wembley, WA 6014, Australia.

Summary Editors’ note Research at Mt Barker reveals Watsonia Details of this poster and further research to be an optimal habitat for the rare and are available from the author at the above endangered Southern Brown Bandicoot address. Isoodon obesulus. Management of Watsonia must consider all interactions, positive and negative, within the bio- logical communities where it is present.