Exp Appl Acarol (2008) 44:101–106 DOI 10.1007/s10493-008-9137-y
How do Neoseiulus californicus (Acari: Phytoseiidae) females penetrate densely webbed spider mite nests? M. Montserrat · F. de la Peña · J. I. Hormaza · J. J. González-Fernández
Received: 01 October 2007 / Accepted: 28 February 2008 / Published online: 18 March 2008 © Springer Science+Business Media B.V. 2008
Abstract The persea mite Oligonychus perseae is a pest of avocado trees that builds extremely dense webbed nests that protect them against natural enemies, including phytoseiid mites. Nests have one or two marginal entrances that are small and Xattened. The predatory mite Neoseiulus californicus co-occurs with O. perseae in the avocado orchards of the south–east of Spain. Penetration inside nests through the entrances by this predator is thought to be hindered by its size and its globular-shaped body. However, in the Weld it has repeatedly been found inside nests that were clearly ripped. Perhaps penetration of the nests has been facilitated by nest wall ripping caused by some other species or by unfavourable abiotic factors. However, to assess whether N. californicus is also able to enter the nest of O. perseae by itself, we carried out laboratory experiments and made a short Wlm. They show how this predator manages to overcome the webbed wall, and that it can penetrate and forage inside nests of O. perseae. Keywords
Persea americana · ‘Web nest’ · Prey refuge · Predatory mite · Tetranychidae
Introduction The persea mite Oligonychus perseae Tuttle, Baker & Abbatiello is a pest of avocado trees (Persea americana Mill.) that was Wrst detected in the south–east of Spain (Málaga and Granada) in 2004 (Vela et al. 2007). The life type of this web-producing spider mite is considered to be ‘web nest’ (WN) because it builds nests made of dense strands of silken webbing (Saito 1985), which have one or two marginal entrances (Aponte and McMurtry 1997). The feeding of the herbivore inside the nests results in necrotic spots that in the worst cases can occupy up to 90% of the leaf underside (Aponte and McMurtry 1997). Electronic supplementary material The online version of this article (doi:10.1007/s10493-008-9137-y) contains supplementary material, which is available to authorized users. M. Montserrat (&) · F. de la Peña · J. I. Hormaza · J. J. González-Fernández E.E. La Mayora – C.S.I.C., 29750 Algarrobo-Costa, Malaga, Spain e-mail:
[email protected]
1C
102
Exp Appl Acarol (2008) 44:101–106
Webbed nests protect both mobile and egg stages against the attack of natural enemies, including phytoseiid mites (Mori et al. 1999). The ability of phytoseiid mites to penetrate spider mite webbing is related to the setal lengths on the middorsal and marginodorsal regions of the phytoseiid soma (Sabelis and Bakker 1992). Sabelis and Bakker (1992) formulated the ‘chaetotaxy hypothesis’, stating that setae of phytoseiid mites keep the sticky silk strands away from the body, minimizing contact surface and thereby avoiding to get entangled in the sticky web. However, the chaetotaxy hypothesis of Sabelis and Bakker (1992) refers to species of spider mite that belong to Saito’s (1985) ‘complex web’ (CW) type, and it remains to be seen whether it holds for WN web-producing spider mites as well. Nevertheless, before they can or cannot cope with sticky strands within spider mite nests, it is obvious that predatory mites dealing with WN spider mite species Wrst need to penetrate the nest. Penetrating nests of WN spider mites must not be an easy task because the web is extremely dense, and although some species build nests with entrances, these are usually small and Xat. This indicates that body size or thickness may play an important role in the ability of phytoseiids to invade this type of nests. In the avocado-persea mite system of California (USA), the small species Galendromus helveolus (Chant) invades nests and attacks the persea mite (Takano-Lee and Hoddle 2002a). Another species associated with avocado trees in California, Neoseiulus californicus (McGregor), is bigger than G. helveolus and it is most commonly found foraging outside, rather than inside, protective webbed nests (Takano-Lee and Hoddle 2002a). Yet, both predators succeeded as biocontrol agents of O. persea in augmentative Weld release experiments on avocado trees in California (Hoddle et al. 1999; Kerguelen and Hoddle 1999). In the avocado orchards of the south–east of Spain, the two most abundant phytoseiid mite species co-occurring with O. perseae are Euseius stipulatus (Athias-Henriot) and N. californicus (J.R. Boyero, J.M. Vela and E. Wong, unpublished data). Whilst the Wrst is hindered by webbing and only attacks stages wandering outside the nests (M. Montserrat, unpublished data), the second has repeatedly been found in the Weld attacking persea mite stages inside nests that were clearly ripped (J.J. Gonzalez-Fernandez and F. de la Peña, personal observation). The aim of this work was to Wnd out whether N. californicus is able to penetrate intact persea mite nests and attack prey (eggs) inside, to discard that its invasion needs to be facilitated by nest wall ripping caused by other factors, such as other species or unfavourable abiotic factors (e.g. desiccation of the web).
Materials and methods Mites were cultured in a climate chamber at 25 § 1°C, 65 § 5% r.h. and 16:8 L:D. N. californicus was supplied by Biobest (Westerloo, Belgium), and kept on detached cucumber leaves infested with Tetranychus urticae Koch that were placed on inverted pots inside water containing trays. Females of O. perseae were obtained from detached infested avocado leaves, taken from avocado orchards located in the Experimental Station La Mayora. Experiments were carried out in a climate box (600 l) at 25°C, 60% r.h. and 16:8 L:D. Intact persea mite nests were obtained by adding 10 persea mite females on avocado leaf discs (3.5 cm diameter) that were placed upside-down on a layer of water-soaked cotton wool inside plastic containers (100 ml, 6.7 cm high, lower diameter 5.1 cm, upper 6.5, Greiner nr 724201), and allowing them to feed, oviposit and produce nests for 48 h. Subsequently, the arenas were allocated to one of three treatments: (a) ‘Predator + nest manipulation’
1C
Exp Appl Acarol (2008) 44:101–106
103
(13 replicates), (b) ‘No predator + nest manipulation’ (10 replicates), and (c) ‘No predator + no nest manipulation’ (10 replicates). In the arenas allocated to treatments (a) and (b), females were disturbed by touching them with a Wne brush until they reached one of the entrances and crawled out of the nest. They were then removed without damaging the nests. Drawings with the exact location of nests were made of each leaf disc, to ease the identiWcation of nests. The nests and the eggs laid per nest on each leaf disc were then counted. Either one N. californicus female [treatment (a)], randomly taken from the rearing unit, or no predator [treatments (b) and (c)] was added to the leaf discs. After 24 h, the numbers of intact nests, broken nests, and eggs inside nests were recorded. In the arenas allocated to treatment (c) nests were not manipulated, nor were females removed. Because females could lay eggs and produce nests during the 24 h of the experiment, only the number of intact and broken nests after 24 h was recorded. This treatment was done to account for a possible eVect of experimental manipulation on nest tearing. Torn nests of all arenas were assigned to one of two categories: they either had triangular-shaped or semi-circular openings. Predation rates were compared to natural mortality comparing data from treatments (a) and (b). The initial number of eggs on the arena minus the number of eggs 24 h later was compared between treatments using a non-parametric Mann–Whitney U-test. An eVect of experimental manipulation and predator presence on the number of broken nests was assessed with Kruskal–Wallis analysis of variance, followed by pair-wise comparisons using Mann–Whitney U-tests. The sequential Bonferroni method was applied to correct for multiple comparisons. Non-parametric tests were used because of variance heterogeneity and non-normal distribution of data. Data on the shape of the openings was not analysed because the arenas of two out of the three treatments had no triangular-shaped openings. Recording of the behaviour of N. californicus on the avocado leaf discs was made in an arena similar to those used above, using a digital camera (Canon Power Shot S50) mounted on a stereomicroscope (Leica S6 D). After letting 10 persea mite females feed, built nests and oviposit for 48 h, a female N. californicus was added to the arena. Recording was done until an attack of N. californicus on a nest was observed.
Results and discussion The number of broken nests and egg mortality rate were statistically diVerent among treatments (H = 13.12, P = 0.0014, and Z = 2.75, P = 0.006, respectively). Both variables were higher when the predator was present in the arenas than otherwise (Table 1). Four of the 13 predatory females tested did not feed on O. perseae eggs. Because females were taken from the rearing unit, neither age nor hunger level was standardized among individuals. Older and/or satiated females will be less motivated to attack nests. In these four replicates, all the nests were found intact 24 h after the introduction of the predator. Alternatively, all nests found broken in the rest of the arenas with a predator present (a total of 23 nests in 9 arenas) had been ripped. Ripped nests were characterized by having a triangular-shaped opening (Fig. 1, T-s, also clearly visible on the video). The openings of the few nests found broken in absence of predators (a total of 3 in 20 arenas) were all semi-circular and Xat, and resulted from part of the wall being detached from the leaf surface (Fig. 1, Sc-s). Altogether, the results indicate that nest ripping was done by predators. Nests of the persea mite have one or more small entrances (Aponte and McMurtry 1997). The shape of the entrances is semi-circular and persea mite females, with their slightly Xattened body (soma), still need to move to the surface and crawl to get in and out
1C
104
Exp Appl Acarol (2008) 44:101–106
Table 1 Average (§SE) of the variables measured in the experiment to determine the ability of N. californicus females to rip nests of the persea mite Treatments
Initial no. of nests
Initial no. of eggs
No. of broken nests
Egg mortality
No. of triangular-shaped openings
Predator + nest manipulation
4.1 § 0.3
24.1 § 1.50
3.7 § 0.30
20.4 § 1.81
12.3 § 2.45 A 17.8 § 0.78 0.2 § 0.13 B
1.8 § 0.41
No predator + nest manipulation No predator + no nest manipulation
1.8 § 0.41 A 2.6 § 0.29 0.1 § 0.10 B
5.1 § 0.43
0.2 § 0.13 B
0 0
DiVerent letters within a column indicate signiWcant diVerences among treatments, after sequential Bonferroni correction, if applicable. Numbers in italics are values when only females that ripped nests were considered
Lv e WN
pm
Sc-s T-s o P
Fig. 1 Schematic drawing of a broken nest of O. perseae: Lv, leaf vein; WN, nest; e, eggs; pm, persea mite; o, nest entrance; P, phytoseiid mite; T-s, triangular-shaped (ripped) nest opening; Sc-s, semicircular-shaped (detached) nest opening
of a nest (M. Montserrat, personal observation). Predatory mite intrusion of a nest through the entrances is most likely hindered by its globular-shaped body, although some species are known to be capable of Xattening and crawling to penetrate through entrances as narrow as the spaces between bulb scales or under the perianth of a coconut (Aratchige et al. 2004, 2007). Small size seems to be the key to G. helveolus’ successful penetration through the persea mite nest entrances. N. californicus females are bigger and cannot enter the nests of O. perseae via the entrances. Instead, they invade nests by ripping the web and thus create much bigger openings. Our video caught the moment in which a N. californicus female was ripping the nest, using her Wrst pair of legs. It is not clear from the Wlm whether she started ripping the web from one of the entrances or from any part of the wall. However, nest entrances made by the persea mite are mainly located marginally (Aponte and McMurtry 1997), whereas in most of the replicates of our experiment the gash was found in the central part of the nest. This suggests that the predator does not need the ‘nest doors’ as a starting point for the ripping. Unfortunately, we cannot be absolutely certain, because we did not record the exact location of the doors of the nests before predators were added, and therefore we could not compare them with the location of the triangular-shaped openings. The
1C
Exp Appl Acarol (2008) 44:101–106
105
fact that no nests with semi-circular openings were seen when N. californicus was in the arena suggests that the predator may proWt from points of the nests that are detached from the leave surface to start ripping; yet not exclusively, because the number of nests with semi-circular openings in absence of predators was much smaller that the number of triangular-shaped openings in presence of predators (0.10 § 0.10 and 0.20 § 0.13 vs. 1.77 § 0.41; Table 1). In the avocado orchards of California N. californicus is mostly found wandering outside nests (Takano-Lee and Hoddle 2002a). In the south–east of Spain, however, it is common to Wnd adults of N. californicus inside nests. Predation and competition are the two main factors determining the spatial distribution of individuals (Fretwell and Lucas 1970; Lima 1998). In California, N. californicus co-occurs with a complex of Galendromus species (Takano-Lee and Hoddle 2002b) that are good at crawling under the protective webbing of the persea mite, and hence are probably competitively superior at exploiting O. perseae inside nests. That in the avocado trees of the south–east of Spain N. californicus commonly invades persea mite nests, as the triangular-shaped openings observed in the Weld suggest, is perhaps due to the lack of phytoseiid mites specialized on web-nest building spider mites in this area. Indeed, N. californicus mainly co-occurs with E. stipulatus, a species that is hindered by the web and mostly forages on mobile stages wandering outside nests. Additionally, given that both phytoseiid mites are potentially engaged in intraguild-predation, the nests of the persea mite may act as refuges for N. californicus eggs and juveniles. Alternatively, bigger nest openings made by N. californicus could facilitate the invasion of E. stipulatus inside nests. Whether these interactions among predators occur in the Weld, as well as their eVect on the control of the persea mite populations, is a subject that is currently being investigated. Mori and Saito (2004, 2005) deWned two anti-predator strategies evolved in WN spider mite species of the genus Stigmaeopsis that inhabit bamboo plants, the so called ‘defense by many individuals in larger nests’ and ‘protection by smaller nests’. Studying four species that diVer in nest size (size ranging between ca. 0.4 and 1.2 cm2), the authors demonstrated that counterattack by many prey individuals is more eVective than counterattack by few individuals, with group size being determined by nest size (Saito 1990). On the other hand, they also showed that smaller nests are more eVective in preventing predator intrusion than larger nests. The authors suggested that the explanation could be that nest size is positively related to entrance size (Mori and Saito 2004). O. perseae builds nests that are much smaller (0.02 § 0.002 cm2, N = 30) than those built by the bamboo mites, and probably are highly protected from predator intrusion in general. Yet, females of N. californicus can overcome that protection by making bigger entrances. Nevertheless, we are aware that smaller stages of this species might be able to enter the nests through the entrances, and that this aspect needs further investigation. In conclusion, females of the predatory mite N. californicus are able to invade the persea mite nests by ripping the densely webbed walls. Subsequently, they attack and feed on the stages inhabiting inside. Because of its occurrence in avocado orchards and its ability to penetrate the densely webbed nests of the persea mite, N. californicus should be considered as candidate for the biological control of the O. perseae in the Spanish avocado orchards. Acknowledgements This work has been partially Wnanced by Biobest Biological Systems SL.U. M.M. was employed by the C.S.I.C. within the framework of a post-doctoral I3P position. The comments of three anonymous referees substantially improved the manuscript. A copy of the digital video can be requested from the senior author.
1C
106
Exp Appl Acarol (2008) 44:101–106
References Aponte O, McMurtry JR (1997) Damage on ‘Hass’ avocado leaves, webbing and nesting behaviour of Oligonychus perseae (Acari: Tetranichidae). Exp Appl Acarol 21:265–272 Aratchige NS, Lesna I, Sabelis MW (2004) Below-ground plant parts emit herbivore-induced volatiles: olfactory responses of a predatory mite to tulip bulbs infested by rust mite. Exp Appl Acarol 43:97–107 Aratchige NS, Sabelis MW, Lesna I (2007) Plant structural changes due to herbivory: do changes in Aceriainfested coconut fruits allow predatory mites to move under the perianth? Exp Appl Acarol 33:21–30 Fretwell SD, Lucas HL (1970) On territorial behavior and other factors inXuencing habitat distribution in birds. Acta Biotheor 19:16–36 Hoddle MS, Aponte O, Kerguelen V, Heralty J (1999) Biological control of Oligonychus perseae (Acari: Tetranychidae) on avocado: I. Evaluating release timings recovery and eYcacy of six commercially available phytoseiids. Int J Acarol 25:211–219 Kerguelen V, Hoddle MS (1999) Biological control of Oligonychus perseae (Acari: Tetranychidae) on avocado: II. Evaluating the eYcacy of Galendromus helveolus and Neoseiulus californicus (Acari: Phytoseiidae). Int J Acarol 25:221–229 Lima SL (1998) Nonlethal eVects in the ecology of predator-prey interactions. BioScience 48:25–34 Mori K, Saito Y (2004) Nest size variation reXecting anti-predator strategies in social spider mites of Stigmaeopsis (Acari: Tetranychidae). Behav Ecol Sociobiol 56:201–206 Mori K, Saito Y (2005) Variation in social behavior within a spider mite genus, Stigmaeopsis (Acari: Tetranychidae). Behav Ecol 16:232–238 Mori K, Saito Y, Sakagami T (1999) EVects of the nest web and female attendance on survival of young in a subsocial spider mite, Schizotetranychus longus (Acari:Tetranychidae). Exp Appl Acarol 23:411–418 Sabelis MW, Bakker FM (1992) How predatory mites cope with the web of their tetranychid prey: a functional view on dorsal chaetotaxy in the Phytoseiidae. Exp Appl Acarol 16:203–225 Saito Y (1985) Life types of spider mites. In: Helle W, Sabelis MW (eds) Spider mites. Their biology, natural enemies and control, vol 1A. Elsevier, Amsterdam, pp 253–264 Saito Y (1990) ‘‘Harem’’ and ‘‘non-harem’’ type mating systems in twospecies of subsocial spider mites (Acari: Tetranychidae). Res Popul Ecol 32:263–278 Takano-Lee M, Hoddle M (2002a) Predatory behaviours of Neoseiulus californicus and Galendromus helveolus (Acari: Phytoseiidae) attacking Oligonychus perseae (Acari: Tetranychidae). Exp Appl Acarol 26:13–26 Takano-Lee M, Hoddle MS (2002b) Oligonychus perseae (Acari: Tetranychidae) population responses to cultural control attempts in an avocado orchard. Fla Entomol 85(1):216–226 Vela JM, González-Fernández J, Wong E, Montserrat M, Farré JM, Boyero JR (2007) El ácaro del aguacate (Oligonychus perseae): Estado actual del problema e investigación en Andalucía. Agrícola Vergel 306:301–308
1C
