Factors Contributing to the Failure of the Biological Control Agent, Falconia Intermedia (Miridae: Hemiptera), on Lantana Camara (Verbenaceae) in South Africa

Session 9 Post-release Evaluation and Management 377

Factors Contributing to the Failure of the Biological Control Agent, Falconia intermedia (Miridae: Hemiptera), on Lantana camara (Verbenaceae) in South Africa

L. U. P. Heshula1, M. P. Hill1 and R. Tourle1

1Department of Zoology and Entomology, P.O. Box 94, Rhodes University, Grahamstown, South Africa, 6139 [email protected] [email protected] [email protected]

Abstract

Countrywide releases of an estimated twenty million Falconia intermedia Distant (Miridae: Heteroptera) individuals have not enhanced the biological control of the invasive Lantana camara L. (Verbenaceae) in South Africa. Although the leaf-feeder initially colonized plants at many sites, it only established at two coastal sites found in the Eastern Cape Province. Recent studies have identified additional factors that may have contributed to the failure of this agent to establish. Firstly, the mirid is incompatible with some of the hairier and tougher leafed varieties of L. camara with significantly less oviposition. Secondly, feeding by F. intermedia has been shown to induce an increase in systemic plant resistance (leaf toughness, leaf hairiness) in new leaves, leading to a significant reduction in the agent’s reproduction and impact. Thirdly, there is evidence of ant predation by two Crematogaster spp. of ants on F. intermedia. The ants removed 50% of the mobile F. intermedia nymphs while survival on ant excluded shrubs was significantly higher. Each of these factors has severely hampered the establishment and performance of this agent.

Introduction pale legs. The translucent pale green eggs are laid on the under-surface of leaves, with the emergent Attempts at the biological control of the highly green nymphs going through five instars (Baars invasive shrub, Lantana camara L. sensu lato et al., 2003). The adults and nymphs feed on the (Verbenaceae), have been on-going in South Africa intercellular tissue of lantana leaves causing entire since 1960. In what is considered a moderately shrubs to appear silvery white and leaves to abscise. successful biological control program, only three of Releases were made in a number of sites in a total 26 insect and fungal biological control agents the Mpumalanga, KwaZulu-Natal and Limpopo released against L. camara are considered relatively provinces where the bug initially established and effective (Baars and Neser, 1999; Klein, 2011). built up large populations causing considerable One of these agents, Falconia intermedia (Distant) damage. However, the large populations of F. (Hemiptera: Miridae), was released on L. camara intermedia declined and eventually disappeared at infestations in South Africa in 1999 to intensify the most sites with only minimal dispersal (Heystek, feeding pressure exerted on L. camara. The mirid 2006). In further field releases in 2001 in the Eastern is endemic to Mexico, Central America and the Cape Province, F. intermedia established at two of Caribbean Islands and its adults and nymphs are five sites found along the coast (Whitney Farm and highly active and mobile, especially when disturbed East London) (Heshula, pers. obs.; Heshula, 2005). (Palmer and Pullen, 1998). The adult is about 2 mm It is only at these sites where F. intermedia may still long and 0.9 mm wide, with a dark brown body and be found.

XIII International Symposium on Biological Control of Weeds - 2011 378 Session 9 Post-release Evaluation and Management

The failure of biological control agents released of two treatments, un-induced test treatment (T) against lantana in South Africa has mainly been on which F. intermedia adults were allowed to feed, attributed to climatic incompatibility, genetic and control (C) with no feeding. After eight weeks variability of native Lantana versus naturalized L. of feeding, and a recovery period in which plant camara varieties, the varied performance of agents and insects parameters were measured, the mirids on L. camara varieties, and predation and parasitism were re-released onto the test plants, now named test (Day and Neser, 2000; Day and Zalucki, 2009). induced (Ti), and the control plant (C), now named In one set of studies F. intermedia showed varied control induced (Ci). The populations of Falconia reproductive performance and higher preference for intermedia adults (Figure 1), nymphs (Figure 2), and South African L. camara varieties over Australian in oviposition (Figure 3) on Whitney Farm L. camara no-choice tests (Urban and Simelane, 1999; Urban plants was higher than on Lyndhurst Farm L. camara. et al., 2004). Baars (2002) found however that F. This suggested a superior suitability of Whitney intermedia did not show difference in performance Farm plants for F. intermedia performance. The on some major varieties tested in South Africa, and un-induced plants (Ci) had significantly higher F. consequently suggested that varietal preference and intermedia feeding damage (21.4% higher), number performance had been overestimated as a reason for of adults (increases of 187.5%), number of nymphs failure of this biological control agent. (110% more), and oviposition (99.8% more) than None of the studies has ever investigated plants previously fed on (Figures 1-3). The reduction the mechanism by which the plants resist the in the performance parameters of F. intermedia on agents however, despite the fact that it is known plants previously fed on from Lyndhurst Farm show that insect feeding may elicit responses in plants that this L. camara variety possesses factors that (Karban and Baldwin, 1997). Related to that was enable it to induce resistance to feeding. The different the lack of knowledge into whether some varieties varieties of lantana have different physiological induced responses after feeding that may affect the and morphological features that may contribute performance of F. intermedia. Further, the role of ant towards the observed responses, such as leaf predation on this agent had never been studied. toughness, leaf hairs/trichomes. Varietal differences The aim of this article was thus to provide a brief in responses by other plant species, such as soybean report on recent post release studies conducted in Glycine max (L.) Merrill, have also been shown to South Africa in an effort to further understand the affect plant responses due to previous feeding by matrix of factors contributing to the failure of F. the same herbivore, Spodoptera litura (Fabricius) intermedia. Three factors in particular have emerged (Lepidoptera: Noctuidae) (Endo et al., 2007). as important: the effect of previous feeding on performance, incompatibility with feeding induced Incompatibility with feeding induced plant plant quality factors, and the effect of ant predation. quality factors

The effect of previous feeding on Plants may effectively respond to herbivory performance by increasing or decreasing the quality of their leaves (Karban and Baldwin, 1997). In this way The observed release, colonization, population an increase or decrease in herbivore performance increase, and population crash pattern in F. may be achieved (Karban and Baldwin, 1997). This intermedia releases provided the rational for an is the alternative to responses that are directed investigation into whether feeding has an effect towards decreasing the impact of feeding damage on the subsequent feeding and performance of F. by increasing plant tolerance, resulting in the plant intermedia on L. camara. Plants from two of the initiating some level of compensatory growth (Strauss Eastern Cape sites, Lyndhurst Farm where the agent and Agrawal, 1999). The nature of the response by failed to establish, and Whitney Farm where a small Lyndhurst Farm L. camara variety observed in the population of the agent persisted for a number of years study summarized above, was determined in eight before being chemically excluded by the landowner, week long bioassays (Heshula and Hill, 2011). The were tested in no choice trials (Heshula, 2009). The adults of F. intermedia were allowed to feed on the first phase of the trial (induction) initially consisted plants from Lyndhurst and Whitney Farm varieties

XIII International Symposium on Biological Control of Weeds - 2011 Session 9 Post-release Evaluation and Management 379 in no choice caged trials. Falconia intermedia nymph and/or forager densities. This represents inoculated plants from Lyndhurst Farm showed a large proportion of the population removed by a significant increase in the toughness of its new predation activity, suggesting that ant species would leaves compared to control plants with no feeding. put agent populations under immediate pressure and Feeding did not have a similar effect for plants of affect establishment. Survival of F. intermedia and Whitney Farm variety (Figure 4). Additionally, T. scrupulosa nymphs in particular was low on ant- plants from Lyndhurst Farm significantly increased accessed shrubs in choice experiments and high on leaf trichome density on new leaves after prolonged ant-excluded shrubs. Tourle (2010) suggested that feeding by F. intermedia (Figure 5). This increase ants were likely to significantly depress F. intermedia was significantly correlated to a decrease in F. populations in the field, aided by Falconia intermedia’s intermedia survival (Figure 6). The defensive fast movement that triggered a predatory response responses were systemic and rapidly induced about by ant species. In contrast, the relatively immobile eight weeks after insect feeding. Although more behaviour of nymphs of the most successful agent work needs to be conducted on more varieties in in South Africa, T. scrupulosa, was identified as South Africa, these results suggest that there may a highly effective predator avoidance strategy. be a role that leaf quality responses play in the poor performance of F. intermedia in South Africa. Conclusion Effect of ant predation The biological control agent F. intermedia has Similar to other agents, predation by ants and had negligible benefit to the biological control spiders has been reported to occur in the field and programme of L. camara in South Africa. The under laboratory conditions (Heystek and Olckers, studies on induced resistance above have been 2003; Heshula, 2009). In mass rearing facilities, ant conducted using only two varieties, and there is a predation of eggs and adults has been reported, and need to extend the work to cover more varieties in is likely linked to the availability of the agent as a South Africa to ascertain the extent to which they food resource. These reports have been based on are important. Certainly under field conditions individual observations however, and until recently there is a negative relationship between F. intermedia no studies have sought to quantify the effect of abundance and leaf hairiness of L. camara (Heystek, predation on F. intermedia. The effect of ant predation 2006). It is more likely however that these factors in by two Crematogaster species on F. intermedia and concert with any of the other factors, such as varietal two biological control agents, Hypena laceratalis incompatibility and inclement climatic conditions, (Walker) (Lepidoptera: Noctuidae) and Teleonemia have been responsible for the current status. scrupulosa Stål (Hemiptera: Tingidae), was recently Localized outbreaks of the agent have been reported investigated on L. camara varieties in the Eastern in the Eastern Cape Province in recent summers, and Cape Province (Tourle, 2010). Crematogaster sp. it appears that this species is likely to be a boom and 1 and 2 colonies were offered the immature stages bust species. This is similar to its status in Australia, of the three agents in choice and no-choice trials. where F. intermedia has established only in a few Crematogaster sp.1 foragers removed 50% of F. sites in north Queensland (Day and Zalucki, 2009). intermedia nymphs, followed by 45% of H. laceratalis larvae and only 9% of T. scrupulosa nymphs (Figure References 7). Density dependent predation on F. intermedia was also observed with more predation at high Baars, J-R & Neser, S. (1999) Past and present

XIII International Symposium on Biological Control of Weeds - 2011 380 Session 9 Post-release Evaluation and Management

5.0 a Mean 4.5 Mean±SE Mean±1.96*SE a ab 4.0

3.5

3.0 bc c 2.5

Number of adults per leaf per adults of Number 2.0

1.5

1.0 d

0.5 LF T LF Ti LF Ci WF T WF Ti WF Ci

Figure 1. Number of adults of Falconia intermedia on two Lantana camara varieties (Lyndhurst Farm – LF, Whitney Farm – WF) subjected to three treatments (T – un-induced, Ti – induced, Ci – control induced). Plots with the same letters are not significantly different (p > 0.05, T - tests) (Heshula 2009).

XIII International Symposium on Biological Control of Weeds - 2011 Session 9 Post-release Evaluation and Management 381

7 a Mean Mean±SE Mean±1.96*SE a 6

ab 5 bc b

Number of nymphs per leaf per nymphs of Number c 2

0 LF T LF Ti LF Ci WF T WF Ti WF Ci

Figure 2. Number of nymphs of Falconia intermedia on two Lantana camara varieties (Lyndhurst Farm –LF, Whitney Farm – WF) subjected to three treatments (T – un-induced, Ti – induced, Ci – control induced). Plots with the same letter are not significantly different (p > 0.05, T - tests) (Heshula 2009).

XIII International Symposium on Biological Control of Weeds - 2011 382 Session 9 Post-release Evaluation and Management

35 Mean a Mean±SD Mean±1.96*SD a a 30

25 b b

c 15 Number of eggs per leaf per eggs of Number 10

0 LF T LF Ti LF Ci WF T WF T i WF Ci

Figure 3. Numbers of eggs laid by Falconia intermedia on two Lantana camara varieties (Lyndhurst Farm – LF, Whitney Farm – WF) subjected to three treatments (T – un-infested, Ti – induced, Ci – control induced). Plots with the same letter are not significantly different (p > 0.05, T - tests) (Heshula 2009).

XIII International Symposium on Biological Control of Weeds - 2011 Session 9 Post-release Evaluation and Management 383

Leaf toughness (g) toughness Leaf 15

11 Lyndhurst Whitney Farm

Variety

Figure 4. Mean leaf toughness for Falconia intermedia infested test (dark bars) and un-infested control treatments (light bars) for Lantana camara varieties. Means within one variety followed by different letters are significantly different (p < 0.05, Mann-Whitney U Test). Whiskers denote standard errors (Heshula and Hill 2011).

XIII International Symposium on Biological Control of Weeds - 2011 384 Session 9 Post-release Evaluation and Management

2 38

30 Leaftrichome density per 2mm

24 Lyndhurst Whitney Farm

Figure 5. Mean number of leaf trichomes in 2 mm2 for Falconia intermedia infested test (dark bars) and un-infested control treatments (lighter bars) for Lantana camara varieties. Means in one variety followed by different letters are significantly different (p < 0.05, Mann-Whitney U Test). Whiskers denote 95% confidence intervals (Heshula and Hill 2011).

XIII International Symposium on Biological Control of Weeds - 2011 Session 9 Post-release Evaluation and Management 385

50 Y = 41.799 - 21.994*x R2 = 0.786

. F = 80.849, p < 0.001 2 45 (1, 22)

25 Trichome densityper 2 mm

15 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 0 4 8 Falconia intermedia individuals per leaf pair

Figure 6. Relationship between Falconia intermedia population numbers and trichome density (Heshula and Hill 2011).

XIII International Symposium on Biological Control of Weeds - 2011 386 Session 9 Post-release Evaluation and Management

Figure 7. Predation rates (No. agents removed/hour) on Falconia intermedia, Hypena laceratalis and Teleonemia scrupulosa by Crematogaster sp.1 colonies (Tourle, 2010).

XIII International Symposium on Biological Control of Weeds - 2011 Session 9 Post-release Evaluation and Management 387

initiatives on the biological control of Lantana Rhodes University, Grahamstown, South Africa. camara (Verbenaceae) in South Africa. In Heshula, L.U.P. & Hill, M.P. (2011) The effect of Biological Control of Weeds in South Africa Lantana camara leaf quality on the performance (1990-1998) (eds Olckers, T. & Hill, M.P.). African of Falconia intermedia. BioControl, 56(6), 925– Entomology Memoir 1, pp. 21–33. 933. Baars, J-R. (2002) Biological control initiatives Heystek, F. & Olckers, T. (2003) Impact of the lantana against Lantana camara L. (Verbenaceae) in mirid in South Africa. In Proceedings of the XIth South Africa: An assessment of the present International Symposium on Biological Control status of the programme, and an evaluation of of Weeds, April 27- May 2, 2003 (eds Cullen, Coelocephalapion camarae Kissinger (Coleoptera: J.M., D.T. Briese, D.T., Kriticos, D.J., Lonsdale, Brentidae) and Falconia intermedia (Distant) W.M. Morin, L. & Scott, J.K.). CSIRO, Canberra, (Heteroptera: Miridae), two new candidate Australia, p. 606. natural enemies for release on the weed. Ph.D. Heystek, F. (2006) Laboratory and field host thesis. Rhodes University, Grahamstown, South utilization by established biological control agents Africa. of Lantana camara L. in South Africa. MSc thesis. Baars, J-R., Urban, A.J. & Hill, M.P. (2003) Biology, Rhodes University, Grahamstown, South Africa. host range, and risk assessment supporting release Karban, R. & Baldwin, I.T. (1997) Induced Responses in Africa of Falconia intermedia (Heteroptera: to Herbivory. Chicago: University of Chicago Miridae), a new biocontrol agent for Lantana Press. camara. Biological Control, 28, 282–292. Klein, H. (2011) A catalogue of the insects, mites Day, M.D. & Neser, S. (2000) Factors influencing the and pathogens that have been used or rejected, or biological control of Lantana camara in Australia are under consideration, for the biological control and South Africa, In Proceedings of the Xth of invasive alien plants in South Africa. African International Symposium on Biological Control Entomology 19(2), 515–549. of Weeds (ed Spencer, N.R.). July 4–14 1999, Palmer, W.A. & Pullen, K.R. (1998) The host range Montana State University, Bozeman, Montana, of Falconia intermedia (Distant) (Hemiptera: USA, pp. 897–908. Miridae): A potential biological control agent for Day, M.D. & Zalucki, M.P. (2009) Lantana camara Lantana camara L. (Verbenaceae). Proceedings Linn. (Verbenaceae). In Biological Control of the Entomological Society of Washington 100: of Tropical Weeds using Arthropods (eds 633–635. Muniappan, R., Reddy, G.V.P. & Raman, A.), Strauss, S.Y. & Agrawal, A.A. (1999) The ecology and Cambridge University Press, 211–246. evolution of plant tolerance to herbivory. Trends Endo, N., Hirakawa, I., Wada, T., & Tojo, S. (2007) in Ecology & Evolution 14, 179–185. Induced resistance to the common cutworm, Tourle, R. (2010) Effects of ant predation on the Spodoptera litura (Lepidoptera: Noctuidae) in efficacy of biological control agents: Hypena three soybean cultivars. Applied Entomological laceratalis Walker (Lepidoptera: Noctuidae), Zoology, 42(2), 199–204. Falconia intermedia Distant (Hemiptera: Miridae) Heshula, U.L.P. (2005) Establishment and impact and Teleonemia scrupulosa Stål (Hemiptera: of the sap-sucking mirid, Falconia intermedia Tingidae) on Lantana camara (Verbenaceae) (Distant) (Hemiptera: Miridae) on Lantana in South Africa. MSc thesis. Rhodes University, camara (Verbenaceae) varieties in the Eastern Grahamstown, South Africa. Cape Province, South Africa. MSc thesis. Rhodes Urban, A.J. & Simelane, D.O. (1999) Performance of University, Grahamstown, South Africa. a candidate biological agent, Falconia intermedia Heshula U.L.P. (2009) Induced plant responses (Hemiptera: Miridae), on selected biotypes of of different Lantana camara L. (Verbenaceae) Lantana camara (Verbenaceae). In Proceedings varieties to herbivory by Falconia intermedia of the Twelfth Entomological Congress (eds Van (Distant) (Hemiptera: Miridae). PhD thesis. Rensburg, J.B.J. & van den Berg, J.), Potchefstroom,

XIII International Symposium on Biological Control of Weeds - 2011 388 Session 9 Post-release Evaluation and Management

South Africa, p 120. Fourteenth Entomological Congress (ed Stals, R), Urban, A. J., Neser, S., Baars, J-R., Simelane, D.O., July 6–9, 2003, Pretoria, South Africa, 106. den Breëyen, A., Mabuda, K., Klein, Williams, H.E., Heystek, F., Phenye, M.S., Samuels, G.A., & Madire., L.G., (2004) New biocontrol agents developed for lantana. In Proceedings of the

XIII International Symposium on Biological Control of Weeds - 2011