OIKOS 93: 177–187. Copenhagen 2001

Minireviews provides an opportunity to summarize existing knowledge of selectedecological areas, with special emphasis on current topics where rapid and significantadvances are occurring. Reviews should be concise and not too wide-ranging. All keyreferences should be cited. A summary is required.

MINI-REVIEW

Relative roles of top-down and bottom-up forces in terrestrialtritrophic plant–insect herbivore–natural enemy systems

Matthew Walker and T. Hefin Jones

Walker, M. and Jones, T. H. 2001. Relative roles of top-down and bottom-up forcesin terrestrial tritrophic plant–insect herbivore–natural enemy systems. – Oikos 93:177–187.

Whether resources (bottom-up forces), natural enemies (top-down forces), or both,determine the abundance of insect herbivore populations in plant–insect herbivore–natural enemy systems remains a major issue in population ecology. Unlike recentsurveys of the tritrophic literature we do not seek to quantify whether top-down orbottom-up forces predominate in any given set of experimental systems. Acknowledg-ing the dearth of empirical synthesis we employ two contrasting literature surveys todetermine whether the plant–insect herbivore–natural enemy literature is currentlyadequate to form a conceptual synthesis of the relative roles of top-down andbottom-up forces.The emergence of a synthesis of the relative roles of top-down and bottom-up forcesin plant–insect herbivore–natural enemy systems appears to have been largelyprevented by (1) the absence of appropriate empirical data; (2) failure to appreciatethe merits of existing data; (3) a continued desire to emphasise either top-down orbottom-up forces to the exclusion of the other; and (4) confusion regarding whichprocesses regulate and which influence the abundance of insect herbivores.

M. Walker and T. H. Jones (correspondence), NERC Centre for Population Biology,Imperial College at Silwood Park, Ascot, Berkshire, UK SL5 7PY (present addresses:MW: Curtis+Cartwright Consulting Ltd., Surrey Technology Centre, Surrey ResearchPark, Guildford, Surrey, UK GU2 7YG; THJ: Cardiff School of Biosciences, CardiffUni�., P.O. Box 915, Cardiff, UK CF10 3TL [jonesth@cardiff.ac.uk]).

Whether resources (bottom-up forces) or natural ene-mies (top-down forces), or both, determine the abun-dance of insect herbivore populations in plant–insectherbivore–natural enemy tritrophic systems has been atopic of much debate during the past 10 years (e.g.Harrison and Cappucino 1995, Gange and Brown 1997,

Hunter et al. 1997, Fraser and Grime 1998, 1999,McMillin and Wagner 1998, Dyer and Letourneau1999, Moon et al. 1999, Schmitz et al. 2000). Whilethere have been major significant theoretical advance-ments in understanding the roles of bottom-up andtop-down forces within quite complex food-webs (Ok-

OIKOS 93:2 (2001) 177

Accepted 20 December 2000

Copyright © OIKOS 1999ISSN 0030-1299Printed in Ireland – all rights reserved

sanen et al. 1981, Menge and Sutherland 1987, Leibold1989, Schmitz 1992, Gutierrez et al. 1994) an empiricalevaluation of the relative roles of these two forces, andhow they change, remains rare, at least in terrestrialinteractions (Quinn and Dunham 1983, Weldon andSlauson 1986, Karban 1989, Underwood and Petraitis1993, Polis 1994, Jones et al. 1997).

The question of what factors determine and regulatethe population dynamics of organisms originated withthe debate between Nicholson (1933), who stated thatregulation was brought about by density dependence,and Andrewartha and Birch (1954) who postulated thatabiotic factors controlled population size (see Murdoch1994 for a review). Issues of whether top-down and/orbottom-up factors are important in determining popu-lation abundance stem from Lindeman (1942), andmore recently White (1978), who argued in favour ofresource limitation of herbivores, particularly with re-spect to nitrogen. In contrast, the belief that top-downforces are fundamentally important in determining theabundance of insect herbivores was championed byHairston et al. (1960) in what was acknowledged as acontroversial (Murdoch 1966, Ehrlich and Birch 1967)but influential paper.

The bottom-up versus top-down debate developedduring the 1980s tending to fragment into three subtlydifferent questions: (1) What factors ‘control’ trophiclevel biomass and species diversity in food-webs? (2)What regulates herbivore populations in simple, linearfood-chain modules?, and (3) What are the relativeimportance of top-down and bottom-up influences onthe abundance of herbivores in these simple linearfood-chain modules?

What factors determine trophic level biomass hasreceived much theoretical attention (Fretwell 1977,1987, Oksanen et al. 1981, Menge and Sutherland 1987,Leibold 1989, Schmitz 1992; see Polis and Strong 1996for a review). In general, these studies argue that bothresource availability and predation, together with com-petition within trophic levels, shape community diver-sity and ‘control’ trophic level biomass. Empirical testsof the strengths of top-down and bottom-up forces interrestrial systems remain rare (Strong 1992; but seeSpiller and Schoener 1990, 1994, Chase et al. 2000).This is in marked contrast to aquatic systems wherethere is support from many enclosure, small pondexperiments and whole lake manipulations of predatorsand nutrients, for the simultaneous action of both‘consumer controlled’ (top-down) and ‘producer con-trolled’ (bottom-up forces) cascades (Schindler 1978,Schindler et al. 1978, Carpenter et al. 1985, McQueen etal. 1986, Carpenter and Kitchell 1988; see McQueen etal. 1989 for a review).

While community (food-web) orientated studies ((1)above) seek to understand community level dynamics,studies conducted on simple, linear food-chain modules((2) and (3) above) examine the relative importance of

top-down (natural enemies) and bottom-up (resources)forces in determining the abundance of insect herbivorepopulations. In doing this the studies attempt to eluci-date the chemical, behavioural, spatial and temporalmechanisms that determine herbivore abundance withinthese interactions. This interdisciplinary approach wasstimulated by Lawton and McNeill (1979) and Price etal. (1980) who urged the adoption of a more pluralisticapproach that acknowledged the importance of bothtop-down and bottom-up forces in determining theabundance of insect herbivores. This represented a sig-nificant departure from the previous dichotomy preva-lent in insect ecology, by considering that threeinteracting trophic levels, plants, herbivores and naturalenemies, operate interactively in essentially all terres-trial ecosystems.

At a community level there is a growing consensus(see reviews by Polis and Strong 1996, Persson 1999,Polis 1999) that the next major challenge for terrestrialcommunity ecologists is to develop an empiricallyderived appreciation of how the complexity of food-web structure ‘controls’ trophic level biomass in terres-trial systems. At a more modular and mechanistictritrophic plant– insect herbivore–natural enemy level,and despite several reviews (Price et al. 1980, Gutierrez1986, Price 1986, Hunter and Price 1992a, b, Ohgushi1992, Harrison and Cappucino 1995, Thomas andWaage 1996) which have catalogued the findings ofnumerous tritrophic studies, no such consensus appearsto be forthcoming.

In this study we do not attempt to quantify whethertop-down or bottom-up forces predominate in anygiven plant– insect herbivore–natural enemy interac-tion. Instead we review, from a wide range of sources,what data are available. We then assess whether thedata are adequate to develop a synthesis of the relativeroles of top-down and bottom-up forces in simple,linear food-chain modules and whether these findingscan be extrapolated to more community-orientatedstudies. Finally, we consider what reasons may beresponsible for the lack of consensus regarding determi-nation of population abundance that appears to existamong terrestrial ecologists compared to aquatic ecolo-gists. We do not attempt to incorporate community orcomplex food-web structures in our consideration.

The terrestrial plant– insect herbivore–naturalenemy interaction literature

Data on terrestrial plant– insect herbivore–natural en-emy interactions are derived from three major sources.The mainstream literature consists of experiments de-signed specifically to untangle the relative roles of top-down and bottom-up forces. These experiments, ingeneral, ask fundamental ecological questions. Addi-

178 OIKOS 93:2 (2001)

tional sources of data include agricultural field studiesassessing the compatibility of host-plant resistance andnatural enemies, and life tables of long-term observa-tional data. The apparent absence in all three datasources of a synthesis of the relative roles of top-downand bottom-up forces may be, at least partially, at-tributed to a failure (1) to distinguish regulation fromother population determining dynamics (see below),and (2) to appraise the suitability of data from the threesources for determining the relative importance of top-down and bottom-up factors.

The distinction between population regulation andother population dynamic behaviour is more than se-mantic, as both the inference required to interpretresults and the methodology used to gather data differsubstantially. Population regulation is the return of apopulation to an equilibrium density as the result ofdensity dependent processes (Dempster and McClean1998) and is distinct from the consequences of non-den-sity dependent processes called influences (Hassell et al.1998) that determine abundance. Questions of influ-ences are most similar to those frequently posed bycommunity ecologists concerning the ‘control’ oftrophic level biomass. All three sources of data men-tioned above may be utilised to develop a synthesis ofthe relative roles of top-down and bottom-up forces ininfluencing terrestrial plant– insect herbivore–naturalenemy interactions. In contrast, determining the relativeroles of top-down and bottom-up forces in regulatingpopulations of herbivorous insects requires a moremultidisciplinary protocol, such as the three-prongedapproach of Hassell et al. (1998). This approach com-bines the analysis of long-term observational data bytime-series analysis with mechanistic population mod-els, and field and laboratory experiments.

Survey of tritrophic experiments

Literature surveys that rely exclusively upon keywordsearches are extremely restrictive, as only a limitednumber of authors use potential search terms such as‘top-down’, ‘bottom-up’ or ‘tritrophic’ in their titles,abstracts or keywords. In this study we took the 1107papers and book chapters cited by the ten most recentreviews of plant– insect herbivore–natural enemy inter-actions (Price et al. 1980, 1997, Gutierrez 1986, Price1986, 1992, Hunter and Price 1992a, b, Ohgushi 1992,Harrison and Cappucino 1995, Thomas and Waage1996) as our initial database.

A preliminary survey of titles and abstracts (whereavailable) was made to determine which of these studiesreported primary experimental and field observationresults for systems involving three trophic levels. Of the1107 citations only 299 were retained for further analy-sis. After detailed consideration of their content a fur-ther 243 papers, which on reading focused almostexclusively on interactions between two rather thanthree trophic levels, were discarded. Those papers thatcontained data from all three trophic levels but thenlater focused on a ditrophic interaction in their discus-sions were retained, as despite the limited conclusionsdrawn by the authors, the experimental data collectedwere tritrophic. This left us with 56 studies (see Table 1)– a surprisingly small number of primary papers con-sidering the relatively large number of reviews that havebeen written and the size of the initial database.

The 56 studies reporting experimental results in ter-restrial plant– insect herbivore–natural enemy interac-tions were then scored according to five criteria todiscover the nature of the experiment performed (seeTable 1): (1) Was the interaction studied ‘fundamental’(i.e. systems not involved in biocontrol research) or

Table 1. Summary of results of the literature survey of studies cited in the ten recent reviews of plant–insect herbivore–naturalenemy interactions. B, M and T refer to bottom, middle and top trophic levels, respectively, with studies classified as B-M(bottom and middle), M-T (middle and top), B-T (bottom and top), or All trophic levels. G/h represents greenhouse studies (seetext for more details). As four papers used sub-experiments in different locations, the column and row totals in the Locationcolumn exceed the expected totals.

‘Fundamental’ or ‘Applied’? Location Manipulated? Trophic levels considered? Relative roles?

NoYesLab.G/h NoField YesAllB-TM-TB-M

026‘Fundamental’ (n=35) 32328340102512‘Applied’ (n=21) 11 7 5 20 1 0 0 210183

37 7 17 45 11 0 4Total (n=56) 6 46 3 53

Data were taken from the following sources. Annis and O’Keefe (1987), Arthur (1962), Ball and Dahlsten (1973), Brower et al.(1967, 1968), Campbell and Duffey (1981), Clancy and Price (1986, 1987), Craig et al. (1990), Denno et al. (1990), Duffey (1970),Eickwort (1977), Faeth (1985), Gross and Price (1988), Hamm and Wiseman (1986), Hulspas-Jordaan and van Lenteren (1978),Kahn and Cornell (1989), Karban (1989), Kareiva and Sahakian (1990), Kartohardjono and Heinrichs (1984), Katanyukul andThurston (1973), Laine and Niemela (1980), Marquis (1994), Monteith (1955), Nadel and van Alphen (1987), Natarajan (1990),Obrycki and Tauber (1984), Obrycki et al. (1983), Ohsaki and Sato (1990), Orr and Boethel (1985), Pimentel and Wheeler(1973), Power et al. (1985), Price (1988), Price et al. (1987), Rabb and Bradley (1968), Rahman (1970), Salto et al. (1983),Shahjahan and Streams (1973), Simmons et al. (1975), Slansky (1978), Smith (1957), Starks et al. (1972), Stary (1978), Stilinget al. (1982), Streams et al. (1968), Thorpe and Barbosa (1986), Tilman (1978), Turlings et al. (1990, 1991), van de Merendonkand van Lenteren (1978), van den Berg et al. (1990), Weis and Abrahamson (1985), Weseloh (1974), Woodman and Price (1992),Yanes and Boethel (1983), Zohdy (1976).

OIKOS 93:2 (2001) 179

‘applied’?, (2) Was the study conducted in the field, orunder controlled conditions in the greenhouse or labo-ratory?, (3) Were one or more trophic levels manipu-lated?, (4) Did the authors evaluate results from alltrophic levels and make some estimate of the relativeimportance of top-down and bottom-up forces?, and (5)Did the authors concentrate on a particular di-trophicinteraction?

Results from ‘fundamental’ systems

Within the ‘fundamental’ systems’ category the major-ity of studies were manipulative field experiments andconsidered all possible interactions between plant, in-sect herbivore and natural enemy trophic levels. Twomajor trends were apparent in approximately half of allstudies and this may potentially reduce the success withwhich data from such ‘fundamental’ systems can beused to resolve the relative roles played by top-downand bottom-up forces in influencing the size of insectherbivore populations.

First, a quarter of all these studies investigate thetritrophic interactions associated with chemical ecology,either of plant volatiles (Monteith 1955, Turlings et al.1990, 1991) or food quality and, in particular, palatabil-ity (Brower et al. 1967, 1968, Duffey 1970, Zohdy 1976,Slansky 1978, Campbell and Duffey 1981). For exam-ple, Turlings et al. (1991) used gas chromatographicanalysis and flight tunnel tests to investigate the mecha-nisms associated with the release of terpenoid volatilesby corn seedlings after they had been fed upon bycaterpillars, and the subsequent orientation of parasiticwasps to these volatiles. Although chemical ecologymay be incorporated into a population dynamics’framework (see Stiling et al. 1982), the studies listedabove merely elucidate the mechanisms of top-downand bottom-up interactions and do not contribute tothe debate regarding population level processes. Inmany cases, extrapolating to food-web/community con-siderations is impossible! Such studies are also morelikely to concentrate on interactions between only twotrophic levels.

The second major trend observed is that almost halfof all field studies rely upon the natural variability ofun-manipulated populations. This is particularly preva-lent in those short-term studies investigating top-downand bottom-up forces in gallmaker or leafminer systems(Clancy and Price 1986, 1987, Price et al. 1987, Price1988, Kahn and Cornell 1989). Whilst it is possible todetermine the relative roles of top-down and bottom-upforces in an un-manipulated system, natural variabilitymust be sufficiently great at the herbivore trophic level(see Price et al. 1987) to assess the changing balance oftop-down and bottom-up forces over as large as possi-ble a range of herbivore densities. Manipulative experi-

ments, in which the relative importance of top-downand bottom-up influences are assessed over a widerange of herbivore densities, do enable the comparisonof populations at extremes of herbivore abundance andmake it relatively easy to distinguish the roles of re-sources and natural enemies in influencing herbivoreabundance.

Of the 35 ‘fundamental’ studies, only three can beconsidered to assess the relative roles of top-down andbottom-up forces in detail. These studies, by Eickwort(1977) on the milkweed leaf beetle Labidomera cli�icol-lis, Price et al. (1987) on the bud galling sawfly Euuramucronata, and Kareiva and Sahakian (1990) on thepea aphid Acryhrosiphon pisum, are exceptional in thatthey provide, in a balanced manner, qualitative esti-mates of the relative roles of top-down influences of asuite of natural enemies and the bottom-up influencesof the host plant. All three studies were performed inthe field; Eickwort, and Kareiva and Sahakian usedmanipulative approaches whilst Price et al. relied uponnatural variability of both top-down and bottom-upforces. Interestingly, although the authors acknowledgethat both top-down and bottom-up forces can actsimultaneously in a single system both Price et al. andEickwort indicate in their concluding remarks that ei-ther top-down or bottom-up forces are important indetermining herbivore population size. In particular,Price et al. note that the survival of larvae after estab-lishment is influenced by plant resistance and naturalenemy attack, but that the most important factor wasthe success of larval establishment which is determinedby the plant. Eickwort, on the other hand, states that,at least for her particular study system, bottom-upforces are unlikely to be limiting as the host plant wasabundant in the study area. Only Kareiva and Sahakianretain a pluralistic view, concluding that host-planteffects were only detectable in the presence of predatorsand that they accounted for a 50% reduction in aphidnumbers.

Results from ‘applied’ studies

Despite only two (Price 1986, Thomas and Waage1996) of the ten reviews used in this survey as a sourcefor papers being of an ‘applied’ nature, non-‘fundamen-tal’ studies were well represented (21 papers) in thetotal sample of 56 studies. ‘Applied’ studies almostuniversally employ a manipulative approach (but seevan den Berg et al. 1990) and are usually performed inthe field. The majority of the studies considered interac-tions between all trophic levels, although two studies(Nadel and van Alphen 1987, Rabb and Bradley 1968)did focus primarily on insect herbivore–natural enemytrophic levels, despite specifically mentioning the role ofthe host plant in their titles.

180 OIKOS 93:2 (2001)

Whilst most ‘applied’ papers did not mention thewords ‘tritrophic’, ‘top-down’ or ‘bottom-up’ in theirabstracts or titles, they were nevertheless more likelythan the more ‘fundamental’ papers to consider fullyinteractions between all trophic levels. Such a balancedappraisal by the ‘applied’ literature is perhaps unsur-prising considering that many studies represent at-tempts to assess the likely practical implications of theinteractions between host-plant resistance and para-sitoids or entomopathogens for biocontrol pro-grammes. For example, Starks et al. (1972) performed amanipulated greenhouse trial and determined that resis-tant varieties of barley and sorghum complemented theactivity of the parasitoid Lysiphlebus testaceipes in re-ducing damage caused by the greenbug Schizaphisgraminum.

Whilst authors of ‘applied’ work frequently assess thecompatibility of host-plant resistance and predators orparasitoids for the purposes of biocontrol, they do notattempt to quantify the relative contributions of thesetop-down and bottom-up forces in controlling pestpopulation size. However, by virtue of their balancedmanipulative approach, data collected by such studiesrepresent a large, currently under-utilised but poten-tially highly profitable dataset that could be used toestimate the quantitative contributions of bottom-upand top-down forces in influencing insect herbivorepopulation dynamics.

Survey of life-table studies

Multi-generational life history data offer an excellentopportunity to estimate the relative roles of variousmortality factors at different life stages for herbivorousinsects (Stiling 1988, Cornell 1990, Cornell andHawkins 1995, Cornell et al. 1998). Such data arecommonly analysed using key-factor analysis (Morris1959, Varley and Gradwell 1960). As a technique todetect density dependent regulation, and despite a num-ber of potential analytical defects (Maelzer 1970,Royama 1977, 1984, Hassell 1986, Hassell et al. 1987,May 1989), key-factor analysis remains a useful methodto appraise the relative roles of top-down and bottom-up forces.

Key-factor analysis determines the stage-specific fac-tor, or factors, which make the greatest contribution tothe total variation in mortality. For example, in theclassic study of the population dynamics of the wintermoth, Operophtera brumata, in Wytham Woods, Ox-ford, Varley and Gradwell (1968) determined that win-ter disappearance caused by asynchrony between theinsect and its food plant was the key factor. Theiranalysis revealed that density dependent predation ofpupae in the soil made a substantial contribution to-wards overall mortality whilst assorted parasitism of

Fig. 1. Frequency-based analysis of five herbivore key factorsat four developmental stages in 52 life tables. Key: Bottom-up(black); fecundity (cross-hatched); top-down (open); migrationand dispersal (grey); weather (hatched).

larval and pupal stages made up the residual mortality.This represents an appreciation of the relative roles ofmortality factors from a variety of sources which is rarein the mainstream plant– insect herbivore–natural en-emy literature.

Results from key-factor analysis survey

To gain an appreciation of the relative importance ofvarious factors influencing populations of herbivorousinsects we re-analysed the papers listed in Stiling’s(1988) compendia of key-factor analysis data. Key fac-tor(s) recorded (if any) were attributed to top-down,bottom-up, weather, dispersal or migration, and the lifehistory stage on which they acted was recorded (seeFig. 1). Miscellaneous or unattributed mortality factorswere excluded from the analysis.

Unlike previous life-table surveys (e.g. Cornell andHawkins 1995, Cornell et al. 1998), reduced fecundityin adults (i.e. ‘‘the extent to which females emergingfrom one generation fail to contribute a maximumpotential number of eggs to the next generation’’ (Ben-son 1973)) has been included. This was assumed to bepredominantly bottom-up in origin (Taylor 1975,Leather 1984, Sopow and Quiring 1998, Hirschberger1999, Hopkins and Ekbom 1999). In some species,realised adult fecundity may also be determined bymortality, migration and weather (Southwood andReader 1976, Hayes 1981; see Leather 1988 for a reviewof factors affecting fecundity).

The data show that both top-down and bottom-upforces may act as key factors in influencing the popula-tions of herbivorous insects (see Fig. 1). Top-downforces predominate during egg, larval and pupal stages(19 of 31 reported key factors during immature stages(Hughes and Mitchell 1960, Klomp 1965, Harcourt1966)). Harcourt (1966)), for example, found that gran-ulosis of larval Pieris rapae by the capsule virusBergoldia �irulenta was the key factor. Bottom-upforces are much rarer during immature stages, and wererecorded in five instances. When they are importantthey act as key factors through either antibiotic effects,

OIKOS 93:2 (2001) 181

for example Latheef and Harcourt (1974) in the Colo-rado potato beetle Leptinotarsa decimlineata feeding onan alternative host (tomato) or through starvation, forexample Dempster (1982) in the cinnabar moth Tyriajacobeae.

During the adult stage, top-down forces are absentand bottom-up forces, usually a reduction in fecundity,are dominant. Reduction in fecundity, or infecundity asit is sometimes known (Berryman 1973), was the keyfactor in 11 of 21 instances (Grimble and Knight 1970,Benson 1973, Southwood and Reader 1976, Hirose etal. 1980, Warren et al. 1986). Benson (1973) calculatedthat the failure of adult cabbage root flies Erioischiabrassicae (now Delia radicum) to lay their full comple-ment of eggs was the key factor, assuming a maximumpotential natality of 80 eggs per female (Swailes 1961)and equal numbers of males and females.

Over all life stages, the number of studies reportingtop-down (19) and bottom-up (16) key factors areroughly equal at one third each, with migration,weather and co-limitation accounting for the remainingthird. The balance between studies showing top-downand bottom-up key factors varies depending on the lifestage being examined.

Partitioning of top-down and bottom-up forces inlife-table studies

While the analysis above allows the determination ofthe key factors causing population change, life-tabledata can also be used to determine the changing bal-ance between top-down and bottom-up factors in influ-encing insect populations. Attempts to assess therelative importance of top-down and bottom-up forcesin the life-table of a single species have generally provedcontroversial (but see McMillin and Wagner 1998). Forexample, Hunter et al. (1997) used 16 years of life-tabledata for the winter moth, Operophtera brumata, andfound that top-down forces explained 34.2% of thepopulation variance, whilst bottom-up forces and unex-plained factors accounted for 17.2% and 48.6% of thepopulation variance, respectively. This analysis howeverwas criticised by Hassell et al. (1998) for (1) failing toseparate density dependent (regulatory) effects andnon-density dependent (influencing) effects; and (2)making the assumption that immediate density depen-dent responses are caused by bottom-up factors anddelayed density dependent responses are caused bytop-down factors.

Life-table data can be used to develop a synthesis ofthe relative roles of top-down and bottom-up influencesat the level of individual host plant– insect herbivore–natural enemy interaction. If stage specific ‘mortality’ ispartitioned into top-down and bottom-up components,and is plotted for each life stage together with the total

overall ‘mortality’, the changing balance of top-downand bottom-up ‘mortality’ factors can be readily ob-served (see Fig. 2 for two examples of suchpartitioning).

As Fig. 2 illustrates, the cinnabar moth experiencesconsiderable variation in the relative roles of top-downand bottom-up ‘mortality factors’ between years. Dur-ing 1967, the availability of the cinnabar moth’s hostplant (ragwort) exerted substantial stage-specific bot-tom-up effects through starvation of late larval instarsand most prominently by determining fecundity realisedby adults compared with the potential maximum natal-ity of 600 eggs per female (Dempster 1971). The top-down effects of arthropod predators, which attackedearly instars before the larvae became distasteful, andpupal predation by moles (Talpa europea) were alsoparticularly important stage-specific factors, with para-sitism of late instar larvae by Apanteles popularis beingof lesser importance. During 1968, the same populationtrends were also exhibited. The large population during1967 consumed much of the plant resources leading todefoliation. This meant that adults were less fecund.Ragworts were substantially smaller in 1968 and bot-tom-up mortality increased dramatically from the thirdand fourth instar stages onwards.

The light brown apple moth, a pest of pome fruit,grapes and horticultural crops in Australia and NewZealand (Danthanarayana 1975, 1983), also exhibitstemporal variation in the relative importance of top-down and bottom-up forces. This variation was partic-ularly striking during the immature life stages; inparticular, the large disappearance of the first instarlarvae and the mortality of second to sixth larval instarsfrom parasitoids, predators and nuclear polyhedrosisviruses. Comparison between the two study orchards atLa Trobe University and Warrandyte revealed that therelative roles of top-down and bottom-up forces alsochanged between sites, despite their geographical prox-imity (within 2 km from the centre of Melbourne,Victoria, Australia). Such a demonstration of the neces-sity of a pluralistic approach and the contingent effectsof life-history strategy, time and location on the relativeimportance of top-down and bottom-up forces is un-usual in the mainstream plant– insect herbivore–natu-ral enemy literature.

Discussion

Empirical synthesis of the relative roles of top-downand bottom-up forces in terrestrial plant– insect herbi-vore–natural enemy interactions is rare, despite theemergence of a more pluralistic conceptual approachthat recognises the importance of both types of forcesin determining the abundance of insect herbivores. Thisreview has considered the level of synthesis currentlyexisting in the ecological literature using two surveys of

182 OIKOS 93:2 (2001)

the three major sources of plant– insect herbivore–nat-ural enemy data: the ‘fundamental’ experiments of themainstream literature, agricultural ‘applied’ field studiesassessing the compatibility of host-plant resistance andnatural enemies, and life tables of long-term data.

The survey of plant– insect herbivore–natural enemyinteractions from the more ‘fundamental’ literature re-vealed two major trends that may have substantiallyreduced both the quality of empirical data and thepotential of such data to assess the relative roles oftop-down and bottom-up forces: (1) the large numberof chemical ecology studies that have not consideredpopulation level processes, and (2) the reliance uponthe natural variability of un-manipulated populations.The limited number of studies available suggests that‘fundamental’ studies are more likely than ‘applied’studies to attempt to make qualitative estimates of therelative roles of top-down and bottom-up forces but ofthe three studies (Eickwort 1977, Price et al. 1987,Kareiva and Sahakian 1990) that did attempt to dothis, those of Eickwort (1977) and Price et al. (1987)still emphasised either top-down or bottom-up forces tothe exclusion of the other.

’Applied’ studies, in their efforts to assess the com-patibility of host-plant resistance, and predator or para-sitoid interactions for the purposes of biocontrol, areunlikely to even mention the words ‘tritrophic’, ‘top-down’ or ‘bottom-up’ in their titles, abstracts and key-words. They are also less likely to attempt quantitativeestimates of the contributions of top-down and bottom-

up forces. Despite this, the balanced, manipulative ap-proach employed by ‘applied’ studies means that thisliterature represents a large, currently under-utilised butpotentially highly profitable source of data that couldbe employed to provide quantitative estimates of therelative roles of top-down and bottom-up forces ininfluencing insect herbivore population dynamics. Itwould appear that the potential of ‘applied’ studies todevelop a synthesis of the relative importance of top-down and bottom-up forces has yet to be appreciated.

In common with ‘applied’ studies, multigenerationallife-history data also represent a currently unexploitedsource of plant– insect herbivore–natural enemy data.The survey of key factors acting on populations ofherbivorous insects, and our life-table data analysis ofthe cinnabar moth and the light brown apple moth,strongly suggest that both top-down and bottom-upforces are important in determining the populations ofinsect herbivores. Partitioning of top-down and bot-tom-up forces for each life-history stage demonstratesthat argument about whether top-down or bottom-upforces are most important in determining the abun-dance of insect herbivore populations is overly simplis-tic. Their relative importance is contingent on a numberof factors including the herbivore’s life history, thelife-history stage, year and study site. Provided that noattempt is made to determine the relative roles oftop-down and bottom-up forces in regulating popula-tions of insect herbivores, and that the role of the plantis fully considered by inclusion of the factors determin-

Fig. 2. Frequency-based analysis of ‘mortality’ factors at six developmental stages over two years for two species of herbivorousinsects; the cinnabar moth, Tyria jacobeae, and the light brown apple moth, Epiphyas post�ittana at two field sites, Warrandyteand La Trobe. Key: Bottom-up (black); top-down (white); unknown mortality factor(s) (hatched); survivors (grey).

OIKOS 93:2 (2001) 183

ing adult fecundity, life-history data provide an excel-lent opportunity to develop a synthesis of the relativeroles of top-down and bottom-up influences.

The assertion by McQueen et al. (1986) that thefailure of researchers in terrestrial systems to develop asynthesis of the relative roles of top-down and bottom-up forces may be explained by the absence of empiricaldata of sufficient quality is only part of the story. Thisreview has demonstrated that the emergence of a syn-thesis, at least in terrestrial systems, has been largelyprevented by researchers, particularly those researchingplant– insect herbivore–natural enemy interactions inwhat we have defined as traditionally more ‘fundamen-tal’ systems, continuously emphasising either top-downor bottom-up forces to the exclusion of the other (seecomments by Karban 1997). The failure to appreciatethat the majority of top-down and bottom-up forcesplay a role in influencing population dynamics (but notnecessarily in regulating populations) and the apparentunwillingness to recognise the merits of data from‘applied’ systems and life-table studies have also con-tributed to the current absence of synthesis. Both top-down and bottom-up forces can, and have been shownto influence the abundance of insect herbivore popula-tions in plant– insect herbivore–natural enemy systems.Only through a greater appreciation of how complexity,and the contingent effects of life-history strategy, timeand location and the adoption of a more pluralisticapproach can a synthesis of the relative roles of top-down and bottom-up forces develop.

Acknowledgements – We are grateful to John Lawton and IainWilliams for comments on an early draft of the manuscript.MW and THJ are funded by the Natural EnvironmentalResearch Council Grants No. GT 04/97/164/TS and GR3/11209, respectively.

ReferencesAndrewartha, H. G. and Birch, L. C. 1954. The distribution

and abundance of animals. –Univ. of Chicago Press.Annis, B. and O’Keefe, L. E. 1987. Influence of pea genotype

of parasitisation of the pea weevil, Bruchus pisorum(Coleoptera: Bruchidae) by Eupteromalus leguminis, (Hy-menoptera: Pteromalidae). – Environ. Entomol. 16: 653–655.

Arthur, A. P. 1962. The influence of host tree on abundance ofItoplectis conquisitor (Say) (Hymenoptera: Ichneumondiae),a polyphagous parasite of the European pine shoot moth,Ryacionia buoliana (Schiff.) (Lepidoptera: Olethreutidae).– Can. Entomol. 4: 337–347.

Ball, J. C. and Dahlsten, D. L. 1973. Hymenopterous parasitesof Ips paraconfusus (Coleoptera: Scolytidae) larvae andtheir contribution to mortality. I. Influence of host tree andtree diameter on parastisation. – Can. Entomol. 105:1453–1464.

Benson, J. F. 1973. Population dynamics of the cabbage rootfly in Canada and England. – J. Appl. Ecol. 42: 105–124.

Berryman, A. A. 1973. Population dynamics of the fir en-graver, Scolytus �entralis (Coleoptera: Scolytidae). I. Anal-ysis of population behaviour and survival from 1964 to1971. – Can. Entomol. 105: 1465–1488.

Brower, L. P., van Zandt Brower, J. and Corvino, J. M. 1967.Plant poisons in a terrestrial food chain. – Proc. Natl.Acad. Sci. USA 57: 893–898.

Brower, L. P., Ryerson, W. N., Coppinger, L. L. and Glazier,S. C. 1968. Ecological chemistry and the palatability spec-trum. – Science 161: 1349–1351.

Campbell, B. C. and Duffey, S. 1981. Alleviation of �-tomatine-induced toxicity to the parasitoid, Hyposoter ex-iguae, by phytosterols in the diet of the host Heliothis zea.– Chem. Ecol. 7: 927–946.

Carpenter, S. R. and Kitchell, J. F. 1988. Consumer control oflake productivity. – BioScience 38: 764–769.

Carpenter, S. R., Kitchell, J. F. and Hodgson, J. R. 1985.Cascading trophic interactions and lake productivity. –BioScience 35: 634–639.

Chase, J. M., Leibold, M. A., Downing, A. L. and Sharin, J.B. 2000. The effects of production, herbivory and plantspecies turnover in grassland food webs. – Ecology 81:2485–2497.

Clancy, K. M. and Price, P. W. 1986. Temporal variation inthree-trophic-level interactions among willows, sawflies andparasites. – Ecology 67: 1601–1607.

Clancy, K. M. and Price, P. W. 1987. Rapid herbivore growthenhances enemy attack: sublethal plant defences remain aparadox. – Ecology 68: 733–737.

Cornell, H. V. 1990. Survivorship, life history and conceal-ment: a comparison of leaf miners and gall formers. – Am.Nat. 136: 581–597.

Cornell, H. V. and Hawkins, B. A. 1995. Survival patterns andmortality sources of herbivorous insects: some demo-graphic trends. – Am. Nat. 145: 563–593.

Cornell, H. V., Hawkins, B. A. and Hochberg, M. E. 1998.Towards an empirically-based theory of herbivore demog-raphy. – Ecol. Entomol. 23: 340–349.

Craig, T. P., Itami, J. K. and Price, P. W. 1990. The windowof vulnerability of a shoot-galling sawfly to attack by aparasitoid. – Ecology 71: 1471–1482.

Danthanarayana, W. 1975. Factors determining variation infecundity of the light brown apple moth, Epiphyas post�it-tana (Walker) (Tortricidae). – Aust. J. Zool. 23: 439–451.

Danthanarayana, W. 1983. Population ecology of the lightbrown apple moth, Epiphyas post�ittana (Walker) (Lepi-doptera: Tortricidae). – J. Anim. Ecol. 52: 1–33.

Dempster, J. P. 1971. The population ecology of the cinnabarmoth, Tyria jacobeae L. (Lepidoptera, Arctiidae). – Oe-cologia 7: 26–67.

Dempster, J. P. 1982. The ecology of the cinnabar moth, Tyriajacobeae L. (Lepidoptera: Artiidae). – Adv. Ecol. Res. 12:1–36.

Dempster, J. P. and McClean, I. F. G. 1998. Definition ofterms. – In: Dempster, J. P. and McClean, I. F. G. (eds),Insect populations: in theory and in practice. Kluwer, pp.1–4.

Denno, R. F., Larsson, S. and Olmstead, K. L. 1990. Role ofenemy-free space and plant quality in host-plant selectionby willow beetles. – Ecology 71: 124–137.

Duffey, S. R. 1970. Cardiac glycosides and distastefulness:some observations on the palatability spectrum of butterfl-ies. – Science 169: 78–79.

Dyer, L. E. and Letourneau, D. K. 1999. Relative strengths oftop-down and bottom-up forces in a tropical forest com-munity. – Oecologia 119: 265–274.

Ehrlich, P. R. and Birch, L. C. 1967. The ‘‘balance of nature’’and ‘‘population control’’. – Am. Nat. 101: 97–107.

Eickwort, K. R. 1977. Population dynamics of a relatively rarespecies of milkweed beetle (Labidomera). – Ecology 58:527–538.

Faeth, S. H. 1985. Host leaf selection by leaf miners: interac-tions among three trophic levels. – Ecology 66: 870–875.

Fraser, L. H. and Grime, J. P. 1998. Top-down control and itseffect on the biomass and composition of three grasses athigh and low fertility in outdoor microcosms. – Oecologia113: 239–246.

184 OIKOS 93:2 (2001)

Fraser, L. H. and Grime, J. P. 1999. Experimental tests oftrophic dynamics: towards a more penetrating approach. –Oecologia 119: 281–284.

Fretwell, S. D. 1977. The regulation of plant communities byfood chains exploring them. – Perspect. Biol. Med. 20:169–185.

Fretwell, S. D. 1987. Food chain dynamics: the central theoryof ecology? – Oikos 50: 291–301.

Gange, A. C. and Brown, V. K. (eds) 1997. Multitrophicinteractions in terrestrial systems. – Blackwell Scientific.

Grimble, D. G. and Knight, F. B. 1970. Life tables andmortality factors for Spaerda inornata (Coleoptera: Ceram-bycidae). – Ann. Entomol. Soc. Am. 63: 1309–1319.

Gross, P. and Price, P. W. 1988. Plant influences on parasitismof two leafrniners: a test of enemy-free space. – Ecology69: 1506–1516.

Gutierrez, A. P. 1986. Analysis of the interactions of hostplant resistance, phytophagous and entomophagous spe-cies. – In: Boethel, D. J. and Eikenbary, R. D. (eds),Interactions of plant resistance and parasitoids and preda-tors of insects. Ellis Horwood, pp. 198–215.

Gutierrez, A. P., Mills, N. J., Schreiber, S. J. and Ellis, C. K.1994. A physiologically-based tritrophic perspective onbottom-up– top-down regulation in populations. – Ecol-ogy 75: 2227–2242.

Hairston, N. G., Smith, F. E. and Slobodkin, L. B. 1960.Community structure, population control and competition.– Am. Nat. 92: 421–425.

Hamm, J. J. and Wiseman, B. R. 1986. Plant resistance andnuclear polyhedrosis virus for suppression of the fall army-worm (Lepidoptera: Noctuidae). – Fla. Entomol. 69: 541–549.

Harcourt, D. G. 1966. Major factors in survival of the imma-ture stages of Pieris rapae (L.). – Can. Entomol. 98:653–662.

Harrison, S. and Cappucino, N. 1995. Using density-manipu-lation experiments to study population regulation. – In:Cappucino, N. and Price, P. W. (eds), Population dynam-ics: new approaches and synthesis. Academic Press, pp.131–147.

Hassell, M. P. 1986. Detecting density dependence. – TrendsEcol. Evol. 1: 90–93.

Hassell, M. P., Southwood, T. R. E. and Reader, P. M. 1987.The dynamics of the virburnum whitefly (Aleurotrachelusjelinekii ): a case study of population regulation. – J. Anim.Ecol. 56: 283–300.

Hassell, M. P., Crawley, M. J., Godfray, H. C. J. and Lawton,J. H. 1998. Top-down versus bottom-up and the Rurita-nian bean bug. – Proc. Natl. Acad. Sci. USA 95: 10661–10664.

Hayes, J. L. 1981. The population ecology of a natural popu-lation of the pierid butterfly Colinas alexandra. – Oecolo-gia 49: 188–200.

Hirose, Y., Suzuki, I., Takagi, M. et al. 1980. Populationdynamics of the citrus swallowtail Papiho xuthus Linne(Lepidoptera: Papilionidae). Mechanisms stabilising itsnumbers. – Res. Popul. Ecol. 21: 260–285.

Hirschberger, P. 1999. Larval population density affects fe-male weight and fecundity in the dung beetle Aphodiusater. – Ecol. Entomol. 24: 316–322.

Hopkins, R. J. and Ekbom, B. 1999. The pollen beetle,Meligethes aenus, changes egg production rate to matchhost quality. – Oecologia 120: 274–278.

Hughes, R. D. and Mitchell, B. 1960. The natural mortality ofErioschia brassicae (Bouche) (Diptera, Anthomyiidae): lifetables and their interpretation. – J. Anim. Ecol. 29: 359–374.

Hulspas-Jordaan, P. M. and van Lenteren, J. C. 1978. Therelationship between host-plant leaf structure and parasti-sation efficiency of the parasitic wasp Encarsia formosaGahan (Hymenoptera: Aphelinidae). – Meded. Fac. Land-bouw. 43: 431–440.

Hunter, M. D. and Price, P. W. 1992a. Playing chutes andladders: heterogeneity and the relative roles of bottom-upand top-down forces in natural communities. – Ecology73: 724–732.

Hunter, M. D. and Price, P. W. 1992b. Natural variability inplants and animals. – In: Hunter, M. D., Ohgnshi, T. andPrice, P. W. (eds), Effects of resource distribution onanimal–plant interactions. Academic Press, pp. 1–11.

Hunter, M. D., Varley, G. C. and Gradwell, G. R. 1997.Estimating the roles of top-down and bottom-up forces oninsect herbivore populations: a classic study revisited. –Proc. Natl. Acad. Sci. USA 94: 9176–9181.

Jones, T. H., Hassell, M. P. and Godfray, H. C. J. 1997.Host-multiparasitoid interactions. – In: Gange, A. C. andBrown, V. K. (eds), Multitrophic interactions in terrestrialsystems. Blackwell Scientific, pp. 257–276.

Kahn, D. M. and Cornell, H. V. 1989. Leafruiners, early leafabscission and parasitoids: a tritrophic interaction. – Ecol-ogy 70: 1219–1226.

Karban, R. 1989. Community organisation of Erigeron glaucusfolivores: effect of competition, predation and host plant.– Ecology 70: 1028–1039.

Karban, R. 1997. Introductory remarks. – In: Gange, A. C.and Brown, V. K. (eds), Multitrophic interactions in terres-trial systems. Blackwell Scientific, pp. 199–200.

Kareiva, P. and Sahakian, R. 1990. Tritrophic effects of asimple architectural mutation in pea plants. – Nature 345:433–434.

Kartohardjono, A. K. and Heinrichs, E. A. 1984. Populationsof the brown planthopper, Nilapar�ata lugens (Stal) (Ho-moptera: Delphacidae), and its predators on rice varietieswith different levels of resistance. – Environ. Entomol. 13:359–365.

Katanyukul, W. and Thurston, R. 1973. Seasonal parasitismand predation of eggs of the tobacco hornworm on varioushost plants in Kentucky. – Environ. Entomol. 2: 939–945.

Klomp, H. 1965. The dynamics of a field population of thepine looper, Bupalus piniarius (Lepidoptera: Geom.). –Adv. Ecol. Res. 3: 207–305.

Laine, K. J. and Niemela, P. 1980. The influence of ants on thesurvival of mountain birches during an Oporinia autumnata(Lep., Geometridae) outbreak. – Oecologia 47: 39–42.

Latheef, M. A. and Harcourt, D. G. 1974. The dynamics ofLeptinotarsa decemlineata populations on tomato. – Ento-mol. Exp. Appl. 17: 67–76.

Lawton, J. H. and McNeill, S. 1979. Between the devil and thedeep blue sea: on the problem of being a herbivore. – In:Anderson, R. M., Turner, B. D. and Taylor, L. R. (eds),Population dynamics. Blackwell, pp. 223–244.

Leather, S. R. 1984. The effect of adult feeding on thefecundity, weight loss and survival of the pine beautymoth, Panolis flammea (D&S). – Oecologia 65: 70–74.

Leather, S. R. 1988. Size and reproductive potential andfecundity in insects. Things aren’t as simple as they seem. –Oikos 51: 386–389.

Leibold, M. A. 1989. Resource edibility and the effects ofpredators and productivity on the outcome of trophicinteractions. – Am. Nat. 134: 922–949.

Lindeman, R. L. 1942. The trophic-dynamic aspect of ecology.– Ecology 23: 399–418.

Maelzer, D. A. 1970. The regression of log Nn+1 on log Nn asa test of density dependence: an exercise with computer-constructed density-independent populations. – Ecology51: 810–820.

Marquis, R. J. 1994. Insectivorous birds increase growth ofwhite oak through consumption of leaf-chewing insects. –Ecology 75: 2007–2014.

May, R. M. 1989. Detecting density dependence in imaginaryworlds. – Nature 338: 16–17.

McMillin, J. D. and Wagner, M. R. 1998. Influence of host-plant vs. natural enemies on the spatial distribution of apine sawfly, Neodiprion autumnalis. – Ecol. Entomol. 23:397–408.

OIKOS 93:2 (2001) 185

McQueen, D. J., Post, J. R. and Mills, E. L. 1986. Trophicrelationships in freshwater pelagic ecosystems. – Can. J.Fish. Aquat. Sci. 43: 1571–1581.

McQueen, D. J., Johannes, M. R. S. and Post, J. R. 1989.Bottom-up and top-down impacts on freshwater pelagiccommunity structure. – Ecol. Monogr. 59: 289–309.

Menge, B. A. and Sutherland, J. P. 1987. Community regula-tion: variation in disturbance, competition and predationin relation to environmental stress and recruitment. – Am.Nat. 130: 730–757.

Monteith, L. G. 1955. Host preferences of Drino bohemicaMesn. (Diptera: Tachinidae), with particular reference toolfactory responses. – Can. Entomol. 12: 509–530.

Moon, D. C., Stiling, P. and Cattell, M. V. 1999. Experimentaltests of trophic dynamics: taking a closer look. – Oecolo-gia 119: 275–280.

Morris, R. F. 1959. Single-factor analysis in population dy-namics. – Ecology 40: 580–588.

Murdoch, W. W. 1966. Community structure, populationcontrol, and competition – a critique. – Am. Nat. 100:219–226.

Murdoch, W. W. 1994. Population regulation in theory andpractice. The Robert H. MacArthur Award Lecture. –Ecology 75: 271–287.

Nadel, H. and van Alphen, J. J. M. 1987. The role of host-and host-plant odours in the attraction of a parasitoid,Edinocarsis lopezi, to the habitat of its host, the cassavamealybug, Phenacoccus manihoti. – Entomol. Exp. Appl.45: 181–186.

Natarajan, K. 1990. Effect of leaf pubsecnce in cotton,Gossypium hirsutum on the parasitism of whitefly, Bemisiatabaci (Gennadius). – J. Biol. Control 4: 57–58.

Nicholson, A. J. 1933. The balance of animal populations. – J.Anim. Ecol. 2: 132–178.

Obrycki, J. J. and Tauber, M. J. 1984. Natural enemy activityon glandular pubescent potato plants in the greenhouse: anunreliable predictor of effects in the field. – Environ.Entomol. 13: 679–683.

Obrycki, J. J., Tauber, M. J. and Tingey, W. M. 1983.Predator and parasitoid interaction with aphid-resistantpotatoes can reduce aphid densities: a two year field study.– J. Econ. Entomol. 76: 456–462.

Ohgushi, T. 1992. Resource limitation on insect herbivorepopulations. – In: Hunter, M. D., Ohgushi, T. and Price,P. W. (eds), Effects of resource distribution on animal–plant interactions. Academic Press, pp. 199–241.

Ohsaki, N. and Sato, Y. 1990. Avoidance mechanisms of threePieris butterfly species against the parasitioid wasp Apante-les glomeratus. – Ecol. Entomol. 15: 169–176.

Oksanen, L., Fretwell, S. D., Arruda, J. and Niemela, P. 1981.Exploitation ecosystems in gradients of primary productiv-ity. – Am. Nat. 118: 240–261.

Orr, D. B. and Boethel, D. J. 1985. Comparative developmentof Copidosoma truncatellum (Hymenoptera: Encurtidae)and its host, Pseudoplusia includens (Lepidoptera: virovi.Noctuidae), resistant and susceptible soybean genotypes. –Environ. Entomol. 14: 612–616.

Persson, L. 1999. Trophic cascades: abiding heterogeneity andthe trophic level concept at the end of the road. – Oikos85: 385–397.

Pimentel, D. and Wheeler, A. G., Jr. 1973. Influence of alfalfaresistance on a pea aphid population and its associatedparasites, predators and competitors. – Environ. Entomol.2: 1–11.

Polis, G. A. 1994. Food webs, trophic cascades and commu-nity structure. – Aust. J. Ecol. 19: 121–136.

Polis, G. A. 1999. Why are parts of the world green? Multiplefactors control productivity and the distribution ofbiomass. – Oikos 86: 3–15.

Polis, G. A. and Strong, D. R. 1996. Food web complexityand community dynamics. – Am. Nat. 147: 813–846.

Power, M. E., Matthews, W. J. and Stewart, A. J. 1985.Grazing minnows, piscovorous bass, and stream algae:

dynamics of a strong interaction. – Ecology 66: 1448–1456.

Price, P. W. 1986. Ecological aspects of host plant resistanceand biological control: Interactions among three trophiclevels. – In: Boethel, D. J. and Eikenbary, R. D. (eds),Interactions of plant resistance and parasitoids and preda-tors of insects. Ellis Horwood, pp. 11–30.

Price, P. W. 1988. Inversely density-dependent parasitism: therole of plant refuges for hosts. – J. Anim. Ecol. 57: 89–96.

Price, P. W. 1992. Plant resources as the mechanistic basis forinsect herbivore population dynamics. – In: Hunter, M.D., Ohgushi, T. and Price, P. W. (eds), Effects of resourcedistribuhon on animal–plant interactions. Academic Press,pp. 139–173.

Price, P. W., Bouton, C. E., Gross, P. et al. 1980. Interactionsamong three trophic levels: influence of plants on interac-tions between insect herbivores and natural enemies. –Annu. Rev. Ecol. Sys. 11: 45–65.

Price, P. W., Roininen, H. and Tahvanainen, J. 1987. Whydoes the bud-galling sawfly, Euura mucronata, attack longshoots? – Oecologia 74: 1–6.

Price, P. W., Fernandes, G. W. and Declerck-Floate, R. 1997.Gall-inducing insect herbivores in multitrophic systems. –In: Gange, A. C. and Brown, V. K. (eds), Multitrophicinteractions in terrestrial systems. Blackwell Scientific, pp.239–255.

Quinn, J. F. and Dunham, A. E. 1983. On hypothesis testingin ecology and evolution. – Am. Nat. 122: 602–617.

Rabb, R. L. and Bradley, J. R. 1968. The influence of hostplants on parasitism of eggs of the tobacco horuworm. – J.Econ. Entomol. 61: 1249–1252.

Rahman, M. 1970. Effect of parasitism on food consumptionof Pieris rapae larvae. – J. Econ. Entomol. 63: 820–821.

Royama, T. 1977. Population persistence and density depen-dence. – Ecol. Monogr. 47: 1–35.

Royama, T. 1984. Population dynamics of the spruce bud-worm Choristoneura fumiferana. – Ecol. Monogr. 54: 429–462.

Salto, C. E., Eikenbary, R. D. and Starks, K. J. 1983. Com-patibility of Lysiphlebus tesiceipes (Hymenoptera: Bra-conidae) with Greenbug (Homoptera: Aphididae) Biotypes‘‘C’’ and ‘‘E’’ reared on susceptible and resistant oatvarieties. – Environ. Entomol. 12: 603–604.

Schindler, D. W. 1978. Factors regulating phytoplankton pro-duction and standing crop in the world’s fresh-waters. –Limnol. Oceanogr. 23: 478–486.

Schindler, D. W., Fee, E. J. and Ruszcynski, T. 1978. Phos-phorus input and its consequences for phytoplanktonstanding crop and production in the experimental lakesarea and in similar lakes. – J. Fish. Res. Board Can. 35:190–196.

Schmitz, O. J. 1992. Exploitation in model food chains withmechanistic consumer reosurce dynamics. – Theor. Popul.Biol. 41: 161–183.

Schmitz, O. J., Hamback, P. A. and Beckerman, A. P. 2000.Trophic cascades in terrestrial systems: a review of theeffects of carnivore removals on plants. – Am. Nat. 155:141–153.

Shahjahan, M. and Streams, F. A. 1973. Plant effects onhost-finding by Leiophron pseudopallipes (Hymenoptera:Braconidae) a parasitoid of the tarnished plant bug. –Environ. Entomol. 2: 921–925.

Simmons, G. A., Leonard, D. E. and Chen, W. E. 1975.Influence of tree species density and composition on para-sitism of the spruce budworm, Choristoneura jumiferana(Clem.). – Environ. Entomol. 4: 832–836.

Slansky, F., Jr. 1978. Utilisation of energy and nitrogen bylarvae of the imported cabbageworm Pieris rapae, as af-fected by parasitism by Apanteles glomeratus. – Environ.Entomol. 2: 179–185.

Smith, J. M. 1957. Effects of the food plant of california redscale, Aonidiella aurantii (Mask.) on reproduction of itshymenopterous parasites. – Can. Entomol. 14: 219–230.

186 OIKOS 93:2 (2001)

Sopow, S. L. and Quiring, D. T. 1998. Body size and spruce-galling adelgids is positively related to realised fecundity innature. – Ecol. Entomol. 23: 476–479.

Southwood, T. R. E. and Reader, P. M. 1976. Populationcensus data and key factor analysis for viburnum whitefly,Aleurotrachelus jelinekii (Frauenf.) on three bushes. – J.Anim. Ecol. 45: 313–325.

Spiller, D. A. and Schoener, T. W. 1990. A terrestrial fieldexperiment showing the impact of eliminating top preda-tors on foliage damage. – Nature 347: 469–472.

Spiller, D. A. and Schoener, T. W. 1994. Effects of top andintermediate predators in a terrestrial food web. – Ecology75: 182–196.

Starks, K. J., Muniappan, R. and Eikenbary, R. D. 1972.Interaction between plant resistance and parasitism againstthe greenbug on barley and sorghum. – Ann. Entomol.Soc. Am. 65: 650–655.

Stary, P. 1978. Seasonal relationships between lucerne, redclover, wheat and barley agro-ecosystems through theaphids and parasitoids (Homoptera, Aphididae; Hy-menoptera, Aphidiidae). – Acta Entomol. Bohemoslov. 75:296–311.

Stiling, P. 1988. Density-dependent processes and key factorsin insect populations. – J. Anim. Ecol. 57: 581–593.

Stiling, P. D., Brodbeck, B. V. and Strong, D. R. 1982. Foliarnitrogen and larval parasitism as determinants of leafminerdistribution patterns on Spartina alterniflora. – Ecol. Ento-mol. 7: 447–452.

Streams, F. A., Shahjahan, M. and LeMasurier, H. G. 1968.Influence of plants on the parasitisation of the tarnishedplant bug by Leiophron pallipes. – J. Econ. Entomol. 61:996–999.

Strong, D. R. 1992. Are trophic cascades all wet? Differentia-tion and donor-control in speciose ecosystems. – Ecology73: 747–754.

Swailes, G. E. 1961. Laboratory studies on mating and ovipo-sition of Hylemya brassicae (Bouche) (Diptera: Anthomyi-dae). – Can. Entomol. 93: 940–943.

Taylor, L. R. 1975. Longevity, fecundity and size; control ofreproductive potential in a polymorphic migrant, Aphisfabae Scop. – J. Anim. Ecol. 44: 135–159.

Thomas, M. and Waage, J. 1996. Integration of biologicalcontrol and host-plant resistance breeding. – TechnicalCentre for Agricultural and Rural Cooperation (ACP-EEC), Wageningen.

Thorpe, K. W. and Barbosa, P. 1986. Effects of consumptionof high and low nicotine tobacco by Manduca sexta (Lepi-doptera: Sphingidae) on survival of gregarious endoparas-tioid Cotesia congregata (Hymenoptera: Braconidae). – J.Chem. Ecol. 12: 1327–1337.

Tilman, D. 1978. Cherries, ants and tent caterpillars: timing ofnectar production in relation to susceptibility of caterpil-lars to ant predation. – Ecology 59: 686–692.

Turlings, T. C. J., Tumlinson, J. H. and Lewis, W. J. 1990.Exploitation of herbivore induced plant odors by host-seeking parasitic wasps. – Science 250: 1251–1253.

Turlings, T. C. J., Tumlinson, J. H., Eller, F. J. and Lewis, W.J. 1991. Larval-damaged plants: source of volatile syn-omones that guide the parasitoid Cotesia margini�entris tothe micro-habitat of its hosts. – Entomol. Exp. Appl. 58:75–82.

Underwood, A. J. and Petraitis, P. S. 1993. Structure ofintertidal assemblages in different locations: how can localprocesses be compared? – In: Ricklefs, R. E. and Schluter,D. (eds), Species diversity in ecological communities. Univ.of Chicago Press, pp. 39–51.

van de Merendonk, S. and van Lenteren, J. V. 1978. Determi-nation of mortality of greenhouse whitefly Trialeurodes�aporariorum (Westwood) (Homoptera: Aleyrodidae) eggs,larvae and pupae on four host-plant species: eggplant(Solanum melongena L.), cucumber (Cucumis sati�us L.),tomato (Lycopersicum esculentum L.) and paprika (Cap-sicum annuum L.). – Meded. Fac. Landbouw. 43: 421–429.

van den Berg, H., Nyambo, B. T. and Waage, J. K. 1990.Parasitisation of Helico�erpa armigera (Lepidoptera: Noc-tuidae) in Tanzania: analysis of parasitoid-crop associa-tions. – Environ. Entomol. 19: 1141–1145.

Varley, G. C. and Gradwell, G. R. 1960. Key factors inpopulation studies. – J. Anim. Ecol. 29: 399–401.

Varley, G. C. and Gradwell, G. R. 1968. Population modelsfor the winter moth. – In: Southwood, T. R. E. (ed.),Insect abundance, Blackwell, pp. 132–142.

Warren, M. S., Pollard, E. and Bibby, T. J. 1986. Annual andlong-term changes in population of the wood white but-terfly Leptidea sinapis. – J. Anim. Ecol. 55: 707–719.

Weis, A. E. and Abrahamson, W. G. 1985. Potential selectivepressures by parasitoids on plant-herbivore interaction. –Ecology 66: 1261–1269.

Weldon, C. W. and Slauson, W. L. 1986. The intensity ofcompetition versus its importance: an overlooked distinc-tion and some implications. – Q. Rev. Biol. 61: 23–44.

Weseloh, R. M. 1974. Host-related microhabitat preferences ofthe gypsy moth larval parasitoid, Parasetigena agilis. –Environ. Entomol. 3: 363–364.

White, T. C. R. 1978. The importance of a relative shortage offood in animal ecology. – Oecologia 33: 71–86.

Woodman, R. L. and Price, P. W. 1992. Differential larvalpredation by ants can influence willow sawfly communitystructure. – Ecology 73: 1028–1037.

Yanes, J., Jr. and Boethel, D. J. 1983. Effect of a resistantsoybean genotype on the development of the soybeanlooper (Lepidoptera: Noctuidae) and an introduced para-sitoid, Microplitis demolitor Wilkinson (Hymenoptera: Bra-conidae). – Environ. Entomol. 12: 1270–1274.

Zohdy, N. 1976. On the effect of the food of Myzus persicaeSulz. on the hymenopterous parasite Aphelinus asychisWalker. – Oecologia 26: 185–191.

OIKOS 93:2 (2001) 187