Screening of Predatory Mites as Potential Control Agentsof Pest Mites in Landscape Plant Nurseries of the Pacific

Northwest1P.D. Pratt and B.A. Croft2

U.S. Department of Agriculture, Agricultural Research ServiceInvasive Plant Research Laboratory, 3205 College Ave., Ft. Lauderdale, FL 33314

Abstract

To select a biological control agent for suppression of spider mites on landscape plants in western regions of the Pacific Northwest, wecompared life history traits of Galendromusoccidentalis Nesbitt,Neoseiulus califamicus (McGregor) and Neoseiulusfallaci~'(Garman).We also evaluated abilities of these predatory mites to suppress spider mites in 4 landscape plant species under field conditions.Comparing life history traits from the literature, intrinsic rate of increase was similar between the 2 Neoseiulus species but lower for G.occidentalis. Prey killed per day was greatest for G. occidentalis > N.fallacis > N. califomicus. For overwintering abilities, N.fallacisand G. occidentalis are indigenous to the Pacific Northwest and will survive winter assuming overwintering sites are available, butsurvival of N. califomicus is unlikely.Neoseiulus califomicus has the widest prey range, G. occidentalis the narrowest, with N.fallacisintermediate. When inoculated into spider mite infested landscape plants, N.fallacis was equally effective at suppressing spider mitesas G. occidentalis in either Malus rootstock or Acer shade trees. Further tests with N. fallacis or N. califomicus on Spiraea andRhododendron plants suggested that N.fallacis is equally or more effective at suppressing pest mites, respectively. Compared with theother candidates, N.fallacis was equally effective at controlling pest mites and has a wider prey range than G. occidentalis. Neoseiulus.,'"fallacis appears to be the best candidate for biological control of multiple spider mite species on landscape plants in these parts of thePacific Northwest.

Index words: integrated pest management, Tetranychidae, Phytoseiidae, Neoseiulus fallacis, Neoseiulus califomicus, Galendromusoccidentalis, biological control.

Species used in this study: Galendromu.~occidentalis (Nesbitt); Neoseiulus califomicus (McGregor); Neoseiulusfallacis (Garman);the two spotted spider mite, Tetranychusurticae (Koch); the southern red mite, Oligonychus illicis (McGregor);Malus (MM.I06EMLA)rootstock; Acer xfreemanii 'Jeffersred'; Spiraea bumaldo 'Crispa'; Rhododendron 'Hotie'.

Significance to the Nursery Industry

Our investigations have focused on the identification,evaluation and conservation of a predatory mite targeted forcontrol of spider mites in landscape nursery systems of thePacific Northwest. We compared 3 indigenous predatorymites and concluded that N.fallacis appears to be best suitedfor suppression of various pest mites found on landscapeplants in the region. In 3 of 4 field tests, N. fallads con-trolled spider mites below economic levels, suggesting thatinoculative releases of this predator can be an effective alter-native to pesticides.

Introduction

Spider mites of the Tetranychidae are major pests of land-scape plant nurseries worldwide. Foliar damage caused byspider mite feeding renders plants unsightly and unmarket-able (13, 22). To protect these high value plants from spidermite damage, nurserymen have traditionally relied on syn-thetic pesticides to suppress population outbreaks (35). How-ever, due to resistance of spider mites to pesticides, negativeeffects of pesticides on natural enemies, and environmentalconcerns, other control tactics besides pesticides are needed

IReceived for publication April 7, 2000; in revised form August I, 2000. Wethank J.L. Green and J.A. McMurtry (of Oregon State University) for com-ments on the manuscript. This research was funded, in part, by grants fromOregon Association of Nurseryman and Northwest Nursery Crops ResearchCenter (USDA). This is Journal Article R-0769I of the Florida AgriculturalExperiment Station.

2Department of Entomology, Oregon State University, Corvallis, OR 97331-2907.

218

to suppress spider mite pests in landscape plant nursery sys-tems (27).

An alternative to pesticides for suppression of spider mitesis the inoculative release of predatory mites from the familyPhytoseiidae. Predatory mites are important biological con-trol agents of spider mites in many agricultural systems (15).Phytoseiids are effective at suppressing pests in landscapeplants grown in greenhouse sy~tems (32), but few studieshave tested their ability to suppress pest mites on landscapeplants grown outdoors. Recent studies evaluated the use ofPhytoseiulus persimilis Athias-Heiuiot for the suppressionof Tetranychus urticae Koch in landscape species grown insemi-tropical regions of the United States (2, 3). Results sug-gest thatP.persimilis can reduce T.urticaeon landscapeplantsbelow economic levels. However, this predator is not an idealbiological control agent for all production regions. For in-stance, P.persimilis may not overwinter in important tem-perate growing regions of the United States (e.g., the PacificNorthwest; 19). In addition, P.persimilis is a specialist ofTetranychus species and may not control other pests in thespider mite complex (16).

McMurtry and Croft (21) classified specialist and gener-alist phytoseiid life styles into 4 types according to life his-tory and morphological traits: Type I phytoseiids are spe-cialized predators of Tetranychus species, Type II includesselective predators of spider mites in the familyTetranychidae, Type III phytoseiids are generalist predatorsand Type IV species are specialist on pollen but may alsofeed on mites. The objective of this study was to identify aphytoseiid mite from the Type II predator classification forbiological control of multiple spider mites in outdoor land-scape nurseries of the Pacific Northwest. We first compared

J. Environ. Hart. 18(4):218-223. December 2000

life history traits from the literature for three species thatappeared to be suitable candidates. From among the candi-dates that possessed the most promising characteristics, weperformed field tests to compare their effectiveness at con-trolling spider mites when released into landscape plants of4 representative types.

..

Materials and Methods

Selection criteria. Three Type II selective predators oftetranychids that are indigenous to the western United Statesare Galendromus occidentalis Nesbitt, Neoseiuluscalifomicus (McGregor)andNeoseiulusfallaci.~Garman(12,21). For these species, we comparedintrinsic rate of increase,number of prey killed per day, ability to overwinter in thePacific Northwest, humidity toleranceand commercial avail-ability. Although Tetranychus species, and especiallyTetranychus urticae Koch, are majorpests in landscapenurs-eries, we were also interested in comparing prey range ofselected phytoseiids for non-Tetranychus diets (Le.,Oligonychus, SchizotetranychuS, Panonychus, etc.). Unfor-tunately, comparative data for prey range among phytoseiidsis limited (6, 7). Therefore, we estimated prey range by cal-culating the sum of non-Tetranychus prey references fromthe literature citation index (LCI) of Croft et aI. (6). Our as-sumption that the LCI adequately describes the prey rangeof Type II phytoseiids is based on recent comparisons be-tween LCI and actual prey range tests for a limited numberof species (26).

Predaceous mite cultures. Laboratory cultures of G.occidentalis, N. califamicus andN.fallacis used in this studywere originally collected from agricultural crops in theWillamette Valley,OR (12). These cultures have been main-tained for 3-6 years with regular additions from field-col-lected mites. Cultures were held at 25 :t:5C, 16:8 light dark(L:D), and 80:t: 10% relative humidity (RH), and mites werefed mixed life stages of T. urticae 3 times per wk. Predatorsfor inoculative releases in field tests originated from labora-tory cultures and were mass reared at Oregon State Univer-sity: phytoseiids were produced on lima beans (Phaseoluslunatus L.) infested with T.urticae under greenhouse condi-tions of 26:21 (:t:5)C day:night (D:N), 75% (:t:10)RH and aphotoperiod of 16:8 L:D h (31).

To compare the 3 candidate phytoseiid species under fieldconditions, we performed tests of 2 or more of these speciesin 4 spider mite infested landscape plant types. These planttypes consisted of stoolbed rootstock. shade tree, deciduousshrub and evergreen shrub. These types were selected be-cause they represent a range of plant architectural types andharbor a diversity of mite pest species.

t

Malus rootstock. The study site was located near Gervis,OR (45.1N and I22.8W).Malus rootstocks(MM.106EMLA)were cultivated in a 7.3 ha stoolbed field with 400 (:t:22)plants per m2and 1 m between each row (for a general de-scription of stoo1beds see 14). Plants emerged from the pe-rennial roots in early spring, and by May continuous densecanopies of leaves were created within rows and nearly be-tween rows. Plants were sprinkler irrigated as needed ac-cording to soil moisture sensors.

In 1995 we tested the ability of G. occidentalis and N.fallacis to control T. urticae in small plots. Fifteen 1000 m1plots were randomly assigned one of three treatments: re-

J. Environ. Hort. 18(4):218-223. December 2000

lease. .of 210 (:t: 8) adult females of either N. fallacis, G.occidentalis or no release of predators (control). We moni-tored mite populations in each replicate plot by removing 50leaves in an 'X' type pattern from each replicate every 14days. By June 27, spider mite populations had increased to0.60 (:t:0.12) per apple leaf and predatory mites were re-leased into plots by placing a bean leaf containing 3 adultfemales every 6 m along each of 12 rows per replicate. Leafsamples wereplaced in an ice cooler, transported to the labo-ratory and a 40x microscope wasused to count pest and preda-tor mites. All predators found on sampled leaves weremounted on glass slides and identified by morphologicalcharacteristics (29). To normalize data of density estimates,we performed a log(x+l) transformation prior to analysis.To adjust for sampling the same populations over time wecompared treatmentswith repeatedmeasures analysis ofvari-ance (ANOVA)(33).

Acer xfreemanii 'Jeffersred'. After initial success of thepredatory mites in the Malus rootstock system, in 1996 wecompared the ability of G. occilkntalis, N. califomicus, andN.fallacis to suppress populations of T.urticae in deciduousshade trees.The study site was located near Dayton, OR (45.2Nand 123.1W) and consisted of a 2 ha field of l-yr-old (sincebudding) Acer xfreemanii 'Jeffersred' saplings. Trees wereplanted in rows spaced approximately 1m apart and 0.76 mwithin the row.Acer trees were irrigated via overhead sprin-klers as needed and measured approximately 0.9 (:t:0.3) min height with 10(:t:4) leaves per tree at the initiation of theexperiment. Due to the sparse canopy we questioned if themore humid-adapted Neoseiulus species would be as effec-tive at suppressing spider mites as the arid-tolerant G.occidentalis.

A single, randomly selected row from the 2 ha A. xfreemanU planting was used for this test. One hundred andfifty trees were randomly assigned 1of 4 treatments: releaseof G. occidentalis, N. califomicus, N. fallacis or no release(contro]), with 3 trees used as a border between each of thereplicates. On July 2, T. urticae populations reached 1.1 (:t:0.2) per leaf and 2 adult females of a single species wereadded to a basal leaf of each tree within release blocks. Mitepopulations were estimated by scanning 3 randomly selectedleaves from basal, intermediate and apical portions of eachtree with an optical visor at lOx magnification. Predator iden-tification and data apalysis were performed as describedabove.

In our initial tests, N. fallacis was equally or more effec-tive than G. occidentalis. Thus, we proCeeded with only N.fallads and N. califomicus in the final 2 plant assessments.

Spiraea bwnalda 'Crispa'. The test site was also locatednear Dayton, OR (see above) and consisted of 0.3 ha plot of2 year old Spiraea humalda 'Crispa' plants that were plantedin rows spaced 0.76 m apart. The plants were contiguouswithin row and inigated with overhead sprinklers as needed.Again, a single row, measuring approximately 35 m, wasrandomly selected for this test. Four 1 m replicated blocksper treatment were randomly assigned along the row with a0.5 m border between each replicate. Borders did not receiverelease of predaceous mites but were sampled to indicate thedegreeof predator movementbetweenplot<;.Treatments wererelease of N.fallacis, N. calijomicus or no release (control).On July 2, 1996, T. IIrticaepopulations reached 1.9 (:t:0.3)

219

per leaf and 5 adult female predators were released into thecenter of the canopy of each release block. Samples of bor-ders and replicates were taken every wk to estimate spidermite and predator mite populations. Samples consisted of 10randomly selected leaves removed from each replicate andborder. Processing of samples, predator identification anddata analysis were performed as described above.

Rhododendron 'Hotie'. As mentioned above, we ques-tioned if the predatory mites would also suppress other, non-Tetranychusspider mites foundin the landscape nursery sys-tem. Oligonychus illicis (McGregor) is a majorpest of rhodo-dendrons in the United States (20). In addition, leaves ofmost rhododendrons are smooth and lack morphologicalcharacteristics known to affect phytoseiid behavior (such aspronounced hair on the underside of leaves or domatia; 34).Therefore, we compared the ability of N. fallacis and N.califomicus to suppress O. illicis on rhododendron plants.

The study site was located near Corvallis, OR, (44.5 Nand I23.3W). Four-yr-old Rhododendron 'Hotie' plants weregrown in a 5 x 10 m plot with plants spaced approximately0.5 m apart. Fifteen replicate plants were randomly assignedone of three treatments: release of N. califomicus, N.fallacisor no release. Predatory mites were inoculated into releaseplants on April 17. A one-plant border surrounded each rep-licate. Irrigation, sampling, identification of predators anddata analysis were exactly like previous studies except only5 leaves were sampled per replicate.

Results and Discussion

Selection criteria. When comparing life history data fromthe literature, values of intrinsic rate of increase were similaramong N. fallacis and N. califomicus when feeding onTetranychus prey, but lower for G. occidentalis (Table 1). Incontrast, Friese and Gilstrap (11) found the number of preykilled per day was higher for G. occidentalis than N.califomicus. In an unrelated study,prey killed per day for N.fallacis appeared to be intermediate to that of the otherphytoseiids (1). With respect to overwintering, N. fallacisand G. occidentalis are indigenous to the Pacific Northwestand are expected to persist assuming adequate hibernationsites are present (12). While N. califomicus occurs in Cali-fornia, it has not been found i~ the Pacific Northwest and itssuccessful overwintering In this region is unlikely (12, 19).

Assuming the LCI adequately describes prey range, N.califomicus appears to feed on the widest range of prey, G.occidentalis the narrowest, with N. fallacis intermediate(Table 1).While commercial availability from producers issimilar for all 3 species, their humidity tolerances are differ-ent. Galendromus occidentalis eggs are tolerant of muchlower humidities than either Neoseiulus species (Table I).

When comparing life history traits, our findings suggestthat each species has one or more traits that may be limitingand that selection based on literature alone is difficult (10).For instance, G. occidentalis has the lowest intrinsic rate ofincrease yet kills more prey items per day (Table I). Simi-larly, N. califomicus appears to feed on a wider range oflandscape plant pests but its potential to overwinter in theregion is unlikely (Table 1). In addition, sparsely canopiedornamental plants (shade tree saplings) that grow severalmeters above ground may have humidities below the toler-ances of humid adapted Neoseiulus species.

Introductionof thepredators G. occidentalisandN.fallacisintoMalus rootstockplants significantly reduced populationsof T. urticae when compared to the control (P < 0.001; Fig.1).No significant differences were found among populationsof spider mites in either of the 2 predator release treatments(P > 0.05). Release of either G. occidentalis or N. fal/adsresulted in a reduction of T. urticae populations as much as95% when compared to the control plots (Fig. I). For in-stance,in N.fallacis releaseplots spidermite populationlevelspeaked in early August at 1.34 (:f 0.32) per leaf with preda-tor populations peaking at 0.48 (:t 0.10) per leaf 16 dayslater. Tetranychus urticae population levels in control plotsreached 6.45 (:t 0.35) per leaf in early September.

Release of the predatory mites into spider mite infestedAcerxfreemaniio'Jeffersred' plants significantly reduced thepopulations of T. urticae when compared to control treat-ments (P< 0.001; Fig. 1).Although treatments were not sig-nificantly different (P-value > 0.05), spider mite densitieswere slightly higher in G. occidentalis treatments than testswith the 2Neoseiulus species. While some dispersal of preda-tory mites did occur from the release treatments into the con-trol treatment late in the experiment, biological control wasnot realized in the control treatment. 0 .

When compared to the control treatment, the predatory. 0mites N. fallacis and N. califomicus significantly reduced

populations of T. urticae in the landscape shrub S. bumalda

'Type n selective predator of tetranychid spider mites (24).

'Average r.. calculated from 6 estimates as reported in Sabelis and Janssen (30).

'as reported in reference 21.

"as reported in reference 8.

"Fed excess eggs of T. cinabarinus.

"Fed excess eggs of T. urticae.'Winter survival in Pacific Northwest.

'Prey range estimated as a summation of the literature citation index of non- Tetranychus diets from Croft et al. (9).'Reference number for literature cited.

220 J. Environ. Hort. 18(4):218-223. December 2000

Table 1. Characteristics of potential biological control agents targeted for spider mites infesting landscape nursery systems of the Pacific Northwest.

Intrinsic Humidity HumidityPredator rate of Prey killed! Over- Prey Commercially tolerance tolerance

Phytoseiid predator type increase femaleld wintering' range' available (RH50of eggs) reference'

Galendromus occidentalis n' 0.228' 14.39. Yes 0.373 Yes 28.41 12Neoseiulu.f californicus n 0.287' IO.OS" Unknown 0.564 Yes 73.73 5Neoseiulus fallacis II 0.298" 11.40" Yes 0.419 Yes 69.65 12

j

Ma/us Rootstockr-- ~

I . G. occidentalis I --q..-.-.-----

.-~ II N.fallacis i q ----.-

1 D Control ~- .-~ , ,----

-~ ~ N. californicus f < , qq-., ' m___.____

--- 0. ... .---..--

18-Jul 19-Aug 6-Sep26-Aug31-Jul 1D-Aug

30 Acer x freemanii 'Jeffersred'

25 ~---...._-_..-------.....---------..-.-._.---------.-......-.-..--

CD 20~t:: 15 _, ~__m..__....___.---.-......::s...: 10

5

o

Fig. 1. Biological control of spider mite pests by inoculative releases of phytoseiid mite species in four ornamental nursery plant systems.

J. Environ. Hort. 18(4):218-223. December 2000 221

7

6

5'I CD4

i 3...:

2

1

0-,

10-Jul 16-Jul 23-Jul 3D-Jul 6-Aug

25Spiraea buma/da 'Crisps'

20

CD5 15i:::s...: 10

5

02-Jul 9-Jul 17-Jul 22-Jul 3D-Jul 6-Aug 13-Aug 20-Aug

Rhododendron 'Hotie'50

40C/)13 30--: 200

10

0

17-Apr 1-May 15-May 3-Jun 18-Jun

Sample Dates

'Crispa' (P < 0.001). Although not significantly different atall dates, spider mite densities were often lower in releaseplots of N. fallacis as compared to those of N. cali/amicus(Fig. 1). In contrast, N. cali/amicus dispersed from releaselocations to border plants earlier than did N. fallads (P <0.05).

The introduction of N.fallads significantly reducedpopu-lations of O. illicis on Rhododendron when compared to ei-ther the control or release of N. cali/omicus (P < 0.001; Fig.I). Two wks after the release date no individuals of N.cali/omicus'were recovered. No differences were foundamong O. illicis densities in control or release of N.cali/omicus treatments (P > 0.05). On May 15, N. faUaciswas collected from control and N. cali/omicus treatments.The decrease in pest densities in control and N. cali/omicustreatments on the last sampling date may have been due tothe arrival of N.fallacis or declining host suitability (Fig. 1).

Results from our field studies demonstrated thatN.faUadswas equally effective as G. occidentalis at suppressing thetwo spotted spider mites in either Malus rootstock or Acershade trees. Further comparisons among N. fallacis and N.cal!fomicus in Spiraea and Rhododendron suggested that N.fallacis is equally or more effective at suppressing the twospotted spider mite or southern red mite, respectively (Fig.1).The reason for the disappearance of N. cali/omicus fromRhododendron plants was unclear. Possible explanations in-clude incompatibility with the hostplant or theprey, O. illicis.Thus, considering that N. fallacis will reproduce on a widerrange of pest mites than G. occidemalis and is as effective inthe field as the other 2 species, we suggest that it is the bestcandidate for biological control of multiple spider mite pestsof landscape plants in the region.

As described by Raupp et a1. (27), laboratory and fieldstudies are needed to provide acceptable alternatives to theuse of pesticides for control of pests of outdoor landscapeand nursery plants. This is the first report of the use of aphytoseiid for control of multiple mite pests in outdoor or-namental plants. As shown in these studies, inoculative re-leases of N. fallacis into Malus, Rhododendron and Spiraeaplants provided suppression of spider mite populations be-low damaging levels, resulting in marketable plants withoutthe use of pesticides.

Unlike most agricultural systems, outdoor landscape nurs-eries are complex polycultures with many pests. When de-veloping biological control in such systems, one must iden-tify the range of pests to be controlled. We sought a Type IIselective predator of tetranychids that would suppress many,if not all, pest mites in the system. Our findings suggestedthat N.fal/acis would numerically respond to T. urticae andO. illicis (Fig. I). In related studies, Pratt et al. (26) mea-sured the ability of N.faUacis to survive, feed and reproduceon a range of landscapeplantpests and alternativefoods underlaboratory conditions. Their findings show that measuredattributes were highest when held with tetranychid mites, butalternative foods (i.e. other mites, pollen, thrips, etc.) alsoprovided for greater survival and reproduction than whenstarved. Studies are needed to determine how alternativefoods may enhance predator conservation in landscape nurs-ery systems (20).

Compatibility with plant types and microhabitats also maybe as important to biological control success as are the abili-ties of a predator to feed and reproduce on a pest (10). Forinstance, morphological differences in pea plants can affect

222

the ability of Coccinel/a septempunctata to suppress aphids(17). Similarly, tritrophic level effects have been describedamong phytoseiid, their prey and leaves possessing domatiaand nectaries (34). In this study, N.fallacis was less effectivein shade trees than on lower profile plants (rootstocks andshrubs). These findings are consistent with recent studiescomparing biological control of spider mites by N. fal/acisamong 30plant varieties ranging in morphologicaltypes (24).In general, limitations in control occurred mostly in tall, ver-tical growing plants that had little foliar canopy (24). Modi-fied cultural techniques may be used to improve these re-strictions.

Our data also have relevance to dispersalof N. cal!fomicusand N./allacis. When comparing within-plant movement inthe presence of excess prey, Pratt et al. (25) observed that N.cali/omicus dispersed over a greater distance than N.fallacis.In the studies reported herein, N. cali/omicus dispersed intocontiguous Spiraea border plants earlier than N. fal/acis,suggesting that N. califamicus dispersed more throughoutthe plots as compared to N.fallacis, which remained more inthe area of release. With respect to N. califomicus, biologi-cal control was realized over a larger area but suppression of.pest mites was slower for this predator than for N. fal/acis(Fig. I). Dispersal of biological control agents influences therate of pest control, area of pest control and sampling proto-cols (7, 25).

Our decision to focus on N.fallacis is consistent with otherstudies that sought a predator of spider mites in agriculturalsystems of inland valley regions of western Oregon. Strong(30) selected N.fal/acis over G. occidentalis for control of T.urticae in hops in western Oregon and Washington. Simi-larly, introductions of N. fallacis into strawberry fields re-sulted in marked reductions of T. urticae and the cyclamenmite Phtyonemus pallidus (Banks) in this region (8). Morris(23) selected N.fal/acis for biological control of T.urticae inpeppermint systems in both humid and arid regions of thewestern United States. The use of a single biological controlagent in multiple agricultural systems within a common re-gion may facilitate area-wide conservation strategies for N.fal/ads.

Literature Cited

I. Ball, J.C. 1980. Development, fecundity, and prey consumption offour species of predaceous mites at two constant temperatures. Environ.Entomol. 9:298-303.

2. Brushwein, J. 1991. Spider mite IPM: Using the predatory mite,Plzytoseiulus persimilis, to control two-spotted spider mite, Tetranychusurticae. Florida Foliage 6-8 p.

3. Cashion, GJ., H. Bixler, and J.E Price. 1994. Nursery IPM trialsusing predatory mites. Proc. Fla. State Hort. Soc. 107:220-222.

4. Castagnoli, M. and S. Simoni. 1994. The effect of different constanthumidities on eggs and larvae of Amblyseius californicus. RediaLXXVII:349-359.

5. Croft, B.A. and Z.-Q. Zhang. 1999. Assessing roles of individualspecies in a phytoseiid complex using life history traits of female adult andimmature miles. Chpt. 13.5 p.p. 8. Acarol. Intern. Congo Symp., Vol II Eds.G.R. Needham, R. Michell, DJ. Horn. w.e. Welboum. Ohio BioI. Surv.,Columbus, OH.

6. Croft, B.A., I.A. McMurtry, and H.-K. Luh. 1998. Do Literaturerecords of predation reflect food specialization and predation types amongphytoseiid mites? Exp. Appl. Acarol. 22:467-480.

7. Croft, B.A., L.N. Monetti, and P.D. Pnlt!. 1998. Comparative lifehistories and predation types: Are Neoseiulus califomicus and N. falladssimilar type II selCl..'livepredators of spider mites? Environ. Entomol. 27:531-538.

J. Environ. Hart. 18(4):218-223. December 2000

8,. . Croft, B.A., Pratt. P.O., Koskela, G., and Kaufman, D. 1998. Predation,

reproduction and impact of phytoseiid mites (Acari:Phytoseiidae) oncyclamin mite, Phytonemus pallidus (Acari:Tarsonemidae) on Strawberry.J. Econ. Entomo!. 91:1307-1314.

9. Croft, B.A., W.B. Strong, R.H. Messing, and J.E. Dunley. 1993.Effects of humidity on eggs and immatures of NeoseiuJusJallacis, Amblysieusandersoni, Metaseiulus occidenlalis and Typhlodromus pyri: implications

for biological control on apple, caneberry, stntwberry and hop. Exp. Appl.Acaro!. 17:451-459.

10. DeBach, P., C.B. Huffaker, and A. W. MacPhee. 1976. Evaluation of

the impact of natural enemies, pp. 255-285. Ill: C.B. Huffaker and P.S.Messenger [eds.], Theory and Practice of BiologicaJ Control. Academic Press,New York.

I I. Friese. D.O. and EE. Gilstrap. 1982. Influence of prey availabilityon reproduction and prey consumption of PhylOseiu/us persimilis. Amblyseiu.\"califomicus, and Metaseiulus occidentali.\". Intern. l. AcaroL 8:85-89.

12. Hadam, U., M.T. Aliniazee, and B.A. Croft. 1986. Phytoseiid mitesof major crops in Willamette Valley, Oregon, and pesticide resistance in1Yphlodromus pyri. Environ. EntomoL 15:1255-1263.

13. Hamlen, R.A. and R.K. Lindquist. 1981. Comparison of twoPhytoseiulus species as predators of two-spotted spider mites Tetrallychusurticae on greenhouse ornamentals. Environ. EnlOmol. 10:524-527.

14. Hartmann, H.T. and D.E. Kester. 1983. Plant Propagation: Principlesand practices. Prentice Hall Inc., NI. p. 727.

15. Helle, Wand M.W Sabelis. 1985. Spider Mites: Their Biology.Natll.rdl Enemies and ControL VoUB. Elsevier, Amsterdam. 458 pp.

16. Johnson, W.T. and H.H. Lyon. 1991. Insects that feed on trees andshrubs. Cornell Univ. Press. Ithaca, NY. 465 pp.

17. Kareiva, P. and R. Sahakian. 1990. Tritrophic effects of a simplearchitectural mutation in pea plants. Nature 345:433-434.

18. Ma, W-L. and J.E. Laing. 1973. Biology, potential for increase andprey consumption of Ambly.\"eius chilellensis. Entomophaga 18:47-60.

19. McMurtry, J.A. 1982. The use ofphytoseiids for biological control:progress and future prospects. pp. 291-321. Ill: M.A. Hoy (ed.) RecentAdvances in Knowledge of the Phytoseiidae., Div. Agric. Sci. Univ. CaliforniaPub. 3284.

20. McMurtry, J.A. 1992. Dynamics and pmential impact of 'generalist'phytoseiids in agroecosystems and possibilities for establishment of exoticspecies. Exp. Appl. Acarol. 14:371-382.

21. McMurtry, J.A. and B.A. Croft. 1997. Life-styles ofphytoseiid mitesand their roles in biological control. Annu. Rev. EntomoL 42:291-321.

22. Mizell, R.F. and D.E Short. 1992. Seasonal occurrence andmanagement of landscape and ornamental pests in north Florida and SouthGeorgia. Proc. Annu. Meet. Fla. State Hort. Soc. 105:204-210.

J. Environ. Hort. 18(4):218-223. December 2000

23. Morris, M.A., B.A. Croft, and R.E. BeTl)'. 1996. Overwintering andeffects of autumn habitat manipulation and carbofuran on Neoseililus Jallacisand Tetranychus urticae in peppermint. Exp. Appl. Acarol. 20:249-258.

24. Pratt, P.O. 1999. Biological control of spider mites by the predatorymite /IIeoseiulus Jallacis in ornamental nursery systems. Ph.D. dissertation.Department of Entomology, Oregon State University. 175 p.

25. Pratt, P.D., L.N. Monetti, and B.A. Croft. 1998. Within- and between-

plant dispersal and distributions of Neo.5eiulus califomicus and N. Jallaci.\'in simulated bean and apple branch systems. Environ. Entomol. 27:148-153.

26. Pratt, P.D., P. Schausberger, and B.A. Croft. 1999. Prey-food typesof Neo.5eiulusfallacis and literature versus experimentally-derived prey-foodestimates for five phytoseiid species. Exp. Appl. Acarol. 23:551-565.

27. Raupp, MJ., C.S. Koehler, and l.A. Davidson. 1992. Advances inimplementing integrated pest management for woody landscape plants. Annu.Rev. Entomo!. 37:61-85.

28. Sabelis, M.W. and A. Janssen 1994. Evolurion oflife-history patternsin the Phytoseiidae., pp. 70-98./n: M.A. Houck [ed.] Mites. Ecology andEvolutionary Analyses ofUfe-history Patterns. Chapman & Hall, New York& London.

29. Schuster, R.O. and A.E. Pritchard. 1963. Phytoseiid mites ofCalifornia. Hilgardia 34:191-194.

30. Strong, W. 1995. Biological control of Tetranychus urticaii Koch inhops by phytoseiid mites: Feasibility, spatial aspects of interactions, andmanagement. Dissertation, Oregon State University.

31. Strong, W.B. and B.A. Croft. 1995. Inoculative release ofphytoseiidmites into the rapidly expanding canopy of hops for control of Tetrallychusurticae. Environ Entomol 24:446-453.

32. Van de Vrie, M. 1985. Greenhouse ornamentals., pp. 273-282.111: WHelle and M. Sabelis [eds.], Spider mites: Their Biology, Natural Enemiesand Control., Vol. lB. Elsevier, Amsterdam.

33. von Ende, C.N. 1996. Repeated-measures analysis: growth and othertime-dependent measures., pp. 113-137. III: S.M. Scheiner and J. Gurevitch[cds.] Design and Analysis of Ecological Experiments. Chapman & Hall,New York, NY.

34. Walter, D.E. and DJ. O'Dowd. 1992. Leaf Morphology and predators:effect of leaf dornatia on the abundance of predatory mites. Environ. Entomo!.21:478-483.

35. Weidhaas, I.A. 1979. Spider mites and other acarina on trees andshrubs. I. Arboriculture 5:9-15.

223

4.. . - : - ...:"-("'" ... _.-

_.