UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Predation, dispersal and weather in an orchard mite system Johnson, Dan Lloyd 1983

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1983_A1 J64.pdf [ 10.54MB ]
Metadata
JSON: 831-1.0095862.json
JSON-LD: 831-1.0095862-ld.json
RDF/XML (Pretty): 831-1.0095862-rdf.xml
RDF/JSON: 831-1.0095862-rdf.json
Turtle: 831-1.0095862-turtle.txt
N-Triples: 831-1.0095862-rdf-ntriples.txt
Original Record: 831-1.0095862-source.json
Full Text
831-1.0095862-fulltext.txt
Citation
831-1.0095862.ris

Full Text

Predation, Dispersal and Weather in an Orchard Mite System by Daniel Lloyd Johnson B.Sc. (High Honours in Biology), University of Saskatchewan, 1978 M.Sc, University of British Columbia, 1980 A thesis submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in The Faculty of Graduate Studies Department of Plant Science and Institute of Animal Resource Ecology We accept this thesis as conforming to the required standard The University of British Columbia March 1983 ©Daniel Lloyd Johnson, 1983 In presenting t h i s thesis i n p a r t i a l f u l f i l m e n t of the requirements for an advanced degree at the University of B r i t i s h Columbia, I agree that the Library s h a l l make i t f r e e l y available for reference and study. I further agree that permission for extensive copying of t h i s thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. I t i s understood that copying or publication of t h i s thesis for f i n a n c i a l gain s h a l l not be allowed without my written permission. Department of Plant Science  The University of B r i t i s h Columbia 1956 Main Mall Vancouver, Canada V6T IY3 ABSTRACT The history, management and ecology of the European red mite, Panonvchus ulmi Koch, and two important phytoseiid predators, Typhiodromus  caudiglans Schuster and Typhiodromus occidentalis Nesbitt were reviewed. The roles and interactions of dispersal, predation and weather in the orchard mite system were examined. Field experiments in an apple orchard with well-established phytoseiid and European red mite (ERM) populations showed that Typhiodromus rarely move among or between trees and the ground cover, either by a i r or via the trunk. They were incapable, within a single season, of repopulating trees from which phytoseiids had been removed by early-season carbaryl application, even though these trees supported high prey populations and were interspersed among unsprayed trees well-populated with Typhiodromus and the ERM. Large numbers of sticky traps captured very few aerially dispersing phytoseiids. In contrast, their ERM prey actively dispersed within trees and throughout the orchard. Mite densities were uncorrelated with leaf chlorophyll content; within-tree dispersal was not directly determined by leaf condition. Adult females were greatly over-represented in aerially-dispersing ERM emigrants in comparison with populations on the apple trees. No density threshold effect on ERM dispersal was discernible on a per-tree basis. Aerial dispersal was extensive and appeared to depend on the weather more mites disperse on warm and calm days than on cool or windy days. ERM dispersal via the tree trunk was minimal and the presence of weeds resulted in only a slight increase in ERM density on the trees. The phytoseiids affected the ERM by reducing population densities, by reducing the proportion of immatures, and by decreasing the degree of prey aggregation (as represented by frequency distribution of leaf counts). The consequences of low predator dispersal and high prey dispersal in a weather-dependent system are discussed. Results of computer simulation of the development, predation, and dispersal are presented. Dispersal (immigration and emigration) allow the phytoseiid populations in the single-tree model to persist and control the ERM. In simulations of the interaction of Typhlodromus with the ERM, the interaction of dispersal and temperature-related processes is strong and non-linear, and may operate through several processes. iv TABLE OF CONTENTS INTRODUCTION 1 0.1 The history of orchard mites and pest management in British Columbia 1 0.1.1 The European red mite 2 0.1.2 The unmanaged system 4 0.1.3 Early management 6 0.2 Understanding the ERM-Tvphlodromus system 11 CHAPTER ONE Dispersal of Typhiodromus Within the Orchard 13 1.1 Introduction 13 1.2 Scope and purpose 15 1.3 Design 15 1.3.1 Treatments and experiment units 15 1.3.2 Treatment application 18 1.3.3 Data collection 26 1.4 Results and discussion 27 1.4.1 Cultivar 30 1.4.2 Phytoseiid removal 32 1.4.3 Banding 34 1.4.4 Weeds 38 1.4.5 Other sources of variation 40 CHAPTER TWO Effect of Phytoseiid Removal on the European Red Mite 41 2.1 Introduction 41 2.2 Results and discussion 41 2.2.1 Phytoseiid removal 44 2.2.1.1 Density 44 2.2.1.2 Age structure 51 2.2.1.3 Correlation and spatial coincidence 54 2.2.1.4 Dispersion 55 2.2.1.5 Prey-to-predator ratios 63 2.2.2 Other predators and prey 68 2.2.3 Sex ratio 72 2.2.4 Other sources of variation 72 2.2.4.1 Cultivar 72 2.2.4.2 Weeds and banding 76 2.2.4.3 Position in the orchard 77 2.2.4.4 Variation between trees 78 2.2.4.5 Interactions with cultivar 79 CHAPTER THREE Dispersal of the European Red Mite 80 3.1 Introduction 80 3.1.1 Dispersal by air 80 3.1.2 Dispersal by walking 82 3.2 Design of experiments 82 3.2.1 Dispersal by air 82 3.2.2 Movement between branches 83 3.3 Results 85 3.3.1 Sticky traps 85 3.3.2 Between branches 92 3.4 Further inference on movement 92 3.4.1 Within-tree dispersal and leaf condition....... 92 v i 3.4.2 Materials and methods 94 3.4.3, Results: chlorophyll and mite density 95 CHAPTER FOUR System Simulation 97 4.1 Introduction 97 4.2 Mite population systems 98 4.3 The European red mite 99 4.3.1 Life history 99 4.3.2 Development 100 4.3.3 Fecundity 103 4.3.4 Mortality 104 4.4 Predation by Typhiodromus 104 4.5 Spatial relationships 109 4.6 Temporal considerations I l l 4.7 Behavior of the model I l l 4.7.1 The ERM sub-model I l l 4.7.2 Sensitivity to temperature 114 4.7.3 Effects of dispersal 114 4.7.4 Interaction of dispersal and temperature 116 SUMMARY 12° LITERATURE CITED 124 APPENDIX A Method of degree-day calculation 138 APPENDIX B Summary of recent weather at the Research Station, Summerland, B.C 143 APPENDIX C Map of the spatial arrangement of mite counts .. 147 v i i APPENDIX D Treatment totals of predators and prey in the orchard experiment 166 APPENDIX E Counts of miscellaneous species and stages 168 APPENDIX F Listing of the simulation model described in Chapter Four 171 Author Index 184 v i i i List of Tables Table Page 1. Analysis of variance of Typhiodromus counts made June 10 - August 20, 1982 29 2. Subdivision of the treatment sum of squares from Table 1 29 3. Results of tests (via orthogonal polynomials) of effects of treatments on the time trends of the counts . 35 4a. Analysis of variance of ERM stages in the predator removal experiment 42 4b. Breakdown of treatment effects on active stages of the ERM 43 5. Variation in numbers and age structure of the ERM as a function of date and phytoseiid density 52 53 6. Prey and predator dispersion as indicated by b of Taylor's power law 56 7. Maximum likelihood estimates of the variance-to-mean ratio 62 8. Prey-to-predator ratios in the orchard experiment, 1982 69 9a. The effect of predation on ERM sex ratio 73 9b. Sex ratio of the ERM on the two apple cultivars 74 10. Evidence for differential dispersal by age class 87 11. Weather during ERM dispersal experiment 88 12. Least-squares regressions of log(catch) on log (density) 91 13a. Mean chlorophyl content, mg/g dry weight, of Mcintosh and Delicious leaves 96 13b. Correlation (r) between number of European red mites and chlorophyll content (mg/g dry weight) of leaves 96 ix List of Figures Figure Page 1. The European red mite, Panonychus ulmi Koch 3 2. The orchard mite system and i t s environment 5 3. Map of the experimental orchard 17 4. The randomization of treatments in the Latin square.... 19 5. Some of the Delicious trees used in the study 20 6. Mcintosh trees used in the study 20 7. Application of carbaryl (Sevin) for phytoseiid predator removal 22 8a. Degree-days above 5.6°C 24 8b. Typical differences in air temperature at the four corners of the experimental orchard 24 9. Orchard air temperatures during 1982 25 10. Rainfall at the Summerland Research Station, 1982 25 11. Numbers of Typhiodromus found on leaves collected over a 24-hour period 28 12a. Average densities of Typhiodromus adults on sprayed and unsprayed Mcintosh apple trees 31 12b. Average densities of Typhlodromus adults on sprayed and unsprayed Delicious apple trees 31 13a. Average densities of Typhlodromus on banded and unhanded unsprayed trees 36 13b. Average densities of Typhlodromus on banded and unhanded sprayed trees 36 14a. Average density of larvae, nymphs and adults of the ERM on the 32 Mcintosh trees 45 14b. Average density of larvae, nymphs and adults of the ERM on the 32 Delicious trees 45 15a. ERM summer eggs on Mcintosh 46 15b. ERM summer eggs on Delicious 46 X Figure Page 16a. ERM larvae on Mcintosh 47 16b. ERM larvae on Delicious 47 17a. ERM nymphs on Mcintosh 48 17b. ERM nymphs on Delicious 48 18a. ERM adult males on Mcintosh 49 18b. ERM adult males on Delicious 49 19a. ERM adult females on Mcintosh 50 19b. ERM adult females on Delicious 50 20. Frequency distributions of ERM counts on leaves with 0, 1, 2, and 3 or more Typhlodromus adults 57 21. A typical scatterplot of ERM per leaf versus Typhlodromus per leaf 58 22. The relationship of the variance to the mean of counts of active ERM 60 23.. The effect of phytoseiid removal on dispersion of ERM adult females a. May 27,28 and June 10,11 64 b. June 24,25 65 c. July 8,9 66 d. July 22,23 67 24a. Zetzellia mali on Mcintosh 71 24b. Zetzellia mali on Delicious 71 25a. Tydeidae on Mcintosh 75 25b. Tydeidae on Delicious 75 26. Difference between weed treatments in the density of the European red mite active stages 77 27. Total dispersing adult female ERM caught on sticky traps 86 xi Figure Page 28. Correlation between dispersers and density a. July 15-16 89 b. July 19-20 89 c. August 15-16 90 d. August 21-22 90 29. Days required for ERM winter eggs to hatch as a function of temperature 102 30. The method of updating the egg (E), juvenile (J) and adult (A) developmental vectors of the ERM 104 31a. Age- and temperature-dependent oviposition rates of the ERM 105 31b. The ERM fecundity function used in the model 105 32a. The family of curves produced by the Frazer-Gilbert model (Gutierrez and Wang, 1977) 108 32b. The effect of temperature on the predation function.... 108 33. ERM population growth and age structure predicted by the model in Appendix F 113 34. The sensitivity of the ERM submodel to temperature 115 35. The consequences of including dispersal (at f i e l d temperatures) 115 36. The effect of temperature on persistance of the predator when low dispersal is allowed. a,b,c 118 d,e,f 119 ACKNOWLEDGEMENTS Dr. W.G. Wellington's insightful suggestions and cla r i t y of thought have been invaluable. The perspicacity apparent in his published work brought me to UBC, and I have not been disappointed. I appreciate the many enjoyable discussions we have had. I extend gratitude to the other members of my committee, Dr. R.H. E l l i o t t , Dr. B.D. Frazer, Dr. V.C. Runeckles and Dr. I.B. Vertinsky, for their interest and helpful suggestions. I thank Dr. R.D. McMullen and staff for provision of an experimental orchard at the Summerland Research Station. Dr. McMullen's sincere interest in my project is much appreciated; his suggestions regarding experimental techniques have been invaluable. Dr. J.H. Borden, Dr. K.K. Nair and Howard Thistlewood kindly provided me with laboratory space and equipment in the Simon Fraser University f a c i l i t i e s at the Summerland Research Station. Deborah Allain , Helene Contant, Edward McLaren and Irene Wilkin provided able technical assistance in my research at UBC. Stimulating discussions on a variety of topics were pro-vided by many friends and associates at UBC, notably Kim Baker, Dave Bernard, Dr. George Eaton, Dieter Jentsch, Tom Lester, Dr. Jon Lovett Doust, Vince Nealis and Dr. John Petkau. I extend sincere gratitude to Mr. and Mrs. Jack and Helen Ward whose generosity made my stay in the Okanagan Valley possible and very pleasant. I especially thank Cathy Johnson for sharing the excitement and weariness, and the rewards and sacrifices, of graduate research with me. Her patient support is greatly appreciated, as is her assistance in preparing the disser-tation. I thank our son, Sam Henry, for being a pleasure to be with. I am honoured and grateful for the generous support provided by a Natural Sciences and Engineering Research Council Postgraduate Scholarship and a Killam Predoctoral Fellowship during this research. INTRODUCTION 0.1 The history of orchard mites and pest management in British Columbia. Orchard agro-ecosystems require intensive management. They are based on perennial tree-fruit crops which are prone to damage from the large pest complexes that are associated with them. In order to maximize f r u i t pro-duction, the trees must be chosen, pruned, f e r t i l i z e d and cared for in ways which often favor pest production as well. The pests can be roughly divided into two groups: those which feed primarily on the leaves (affecting f r u i t production) and those which directly injure the f r u i t (affecting marketability). In British Columbia between the late 1800's and the early 1920's good control of the moths, beetles, f l i e s and fungi which damage apple plants and f r u i t was afforded by careful orchard sanitary practices and by the application of several crude poisons, primarily lime-sulfur mixtures, o i l sprays, lead arsenate and Bordeaux mixture (Marshall, 1951). Government managers, working in the B.C. Department of Agri-culture, the B.C. Board of Horticulture, the Horticultural Society and Fruit-growers' Association of B.C., and the Dominion Division of Entomology were proud of the apparent successes resulting from their diligent quarantines, spray programs, and advice to growers. Says Treherne, 1914: No praise, therefore, is too great for those whose energies have placed this Province on an independent pedestal in the f i e l d of economic entomology, causing her to stand an example of the practical and elementary principles of entomology, which other Provinces and States failed to follow until too late. But in some areas of the world, application of these same pesticides were associated with outbreaks of previously rare phytophagous mite pests (e.g. parts of Europe: Rabbinge, 1976). In most of North American orchards, however, tetranychid mites did not appear in threatening numbers until after 2 World War II. From that point through the late 1960's they were easily the most serious orchard pest in much of the world. Damage to such high unit-value products as fruits and vegetables was expensive. In California alone in 1964 tetranychid mites on f i e l d and orchard crops caused damage and loss worth $150,000,000 (Chant, 1966). They hold major pest status in most orchards of the world today and are kept in check only through vigilant multi-faceted management programs. 0.1.1 The European red mite The most troublesome tetranychid orchard mite in most regions is the European red mite fPanonychus ulmi Koch), so-called since i t is thought to have been introduced into North America from Europe in the early years of this century*. The European red mite (ERM) feeds on the leaves of apple, pear and a host of other f r u i t and ornamental trees. The mites puncture the leaf with needle-like mandibles and remove cytoplasm from mesophyll cell s , thus reducing the photosynthetic a b i l i t y of the leaves. The attacks, often causing extensive browning and even loss of the leaves, reduce f r u i t production during the year of attack and the number of flower buds the f o l -lowing year (Madsen and Arrand, 1975; van de Vrie e_£ j i l . , 1972). The ERM normally has 6 to 8 generations per year, depending on temper-ature, photoperiod and food quality. The mites overwinter as eggs. The production of these "winter eggs" occurs in August and September. In May, eggs hatch at varying times and give rise to 4 instars: larvae, protonymphs, deutonymphs and adults (Figure 1). A l l stages can be found throughout the summer season. It is generally believed, now that we have had nearly 50 hard years of 1 In Europe i t is called the fruit-tree red spider mite. Figure 1. The European red mite, Panonvchus ulmi Koch. (60 X) A: egg; B: larva; C: protonymph; D: deutonymph; E: adult male; F: adult female. (U.S.D.A. Circ. 157, 1931. Cited by Metcalf and F l i n t , 1962). experience with this pest, that the ERM is rarely limited by starvation or disease. There is a large number of known predators of the ERM (65 were recorded by Groves, 1951). In B.C. i t appears to be limited at least in part by phytoseiid mites, Typhlodromus (=Metaseiulus) spp., and a stigmaeid 2 mite, Zetzellia mali (Downing and Arrand, 1978) . ERM densities of up to 15 to 20 mites per leaf can be tolerated with relatively l i t t l e loss, especially in the late summer. Below these densities l i t t l e leaf damage 2 Predatory mites have a distinctive history in biological control. The f i r s t international introduction of a natural enemy was a mite sent from the United States to France in 1873 to control grape-vine Phylloxera ( i t fa i l e d ) . (Howard, 1930; cited in Gerson and van de Vrie, 1979). 4 results and the ERM is controlled by the predaceous mites. The position of the ERM in the orchard system is illustrated in Figure 2. The managed system, management institution and environment as I perceive them are illustrated. In the 1940's and 1950's, changing management practices (directed toward reducing "other pests" such as the codling moth and increasing fru i t production and quality) and perhaps changes in weather altered the structure and relationships of the system enough to cause severe ERM outbreaks. 0.1.2 The unmanaged system In abandoned orchards, the closest thing we have to "natural" conditions, the ERM and i t s relatives are rarely noticeable. Extensive damage by these mites i s unknown from abandoned (i.e., unsprayed and unfertilized) orchards (Chant, 1963) and from orchard records that precede the heavy spraying of the 1930's and 1940's (Huffaker et a l . , 1969; Marshall, 1951, 1952). In the seminatural system, a complex of mite and insect predators hold the ERM at low densities. In unsprayed orchards, the phytoseiids are commonly Typhlodromus caudiglans. X- Pvri. and occasionally T_. occidentalis (more common in sprayed orchards, Downing and Moilliet, 1972). The phytoseiid mites have potential rates of increase as high as or higher than their prey. This unusual situation allows the predators to quickly overtake and influence prey outbreaks in some cases. The predators never succeed in exterminating the ERM because of discrepancies in within-tree and within-leaf distributions, predator preferences, rarity effects, and the a b i l i t y of unmated females to found new pest populations. In terms of numbers of ecological relationships (e.g. predator-prey), the system is structurally more complex in the absence of spraying (Readshaw, 1971). 5 Figure 2. The orchard mite system and i t s environment. Arrows indicate effects (e.g. transfer of information or influence). Circular arrows represent internal dynamics and/or transfers. "Market conditions" and "Weather § Climate" represent the environment. "Growers" and "R t> D" represent the management system. 6 0.1.3 Early management There were few economic entomologists in B.C. in the early years of this century. The number of professionals went from a few in 1914 to a dozen by 1920 (Marshall, 1951; Treherne, 1921). The large number of pest problems (i.e., inquiries from growers) in the Okanagan Valley convinced the government managers and scientists that the headquarters of the Dominion Entomological Branch should be moved to Vernon, B.C., closer to the growers with pest problems. Now both the federal and the provincial research centers for tree f r u i t pests are at the Agriculture Canada Research Station in Summerland, B.C. In the period between 1920 and the early 1940's, four important developments had profound influences on orchard pests management. These were 1) changes in market conditions, 2) changes in pest ecology and levels of damage, 3) technological advances, and 4) effects of climatic change on both the pest and the host plant. Before this period, there were few orchard pests (mainly aphids and scale) in B.C. and these few were reported to be easily controlled by practices developed in the United States. As both the apple production areas and the consumer market grew, higher quality apples became a constant goal of management. Consumers began to develop more selective buying habits concerning f r u i t appearance. Just when consumers were becoming even more particular about f r u i t bruises, scars and "wormy" apples, B.C. was hit hard by the codling moth outbreaks that had been the scourge of some of the eastern apple-growing regions (Hoy, 1942) . At the same time, stricter regulations, such as those regarding arsenic resi-dues that shut B.C. producers off from the important U.S. market, became a concern of the managers (the government scientists and growers) (Marshall, 1952). Technological advances included improved power sprayers, new chemical combinations (e.g. nicotine-based compounds), and wiping machines to remove 7 residues from f r u i t . Unfortunately, even with innovations, control was very different from the pre-1920*s when two or three sprays controlled everything. Some growers were spraying 30 to 40 times in a summer and getting poor control. Pest management consisted of desperate attempts to organize complex spray schedules of a multitude of compounds. Small wonder that when DDT t r i a l s in 1942-1945 showed i t to be a pest cure-all i t was hailed as the salvation of orchard pest management. It proved so effective against codling moth and related pests that the entire complex of insecticide sprays and control practices became unnecessary almost overnight. Basic research on pest biology and ecology began to seem unnecessary as well (see Croft, 1978). Then came the f i r s t big surprise, the sudden elevation of the ERM to pest status. It i s well-documented that serious mite outbreaks occurred almost everywhere that DDT was used in orchards during the 1940's and 1950's (some Canadian and Pacific Northwest references are: Hussey and Huffaker, 1976; Lord § t a l . , 1958; Marshall, 1952; Pickett et a l . , 1958; van de Vrie et a l . , 1972). In fact, the ERM and i t s relatives remain unimportant in regions where pesticides are not used. The sudden change in the pest system in many areas, in which the previously rare and relatively unknown ERM displaced codling moth and remained the major pest for a decade, is explained by the destruction or behavioral modification of the predators of the ERM, the increase in fecundity and longevity caused by improvement in nutrition and the acquired pesticide resistance of the ERM. Although a l l of these hypotheses contain some useful explanation and probably interact, the f i r s t has been favored as the explanation why the phytophagous mites became a problem. Modern research efforts to solve the problem have usually involved the biology, ecology, behavior, physiology or genetics of the phytoseiid mite predators. 8 Even now, phytophagous mite outbreaks can easily be induced merely by using certain insecticides in abandoned orchards in which mites are rare. It seems that some sprays eradicate phytoseiid mites while others, such as azinphosmethyl (discussed below), allow predators such as Typhlodromus  occidentalis to survive and prosper in orchards in which i t was previously rare (Downing and Moilliet, 1971, 1972; Huffaker et a l . , 1970). It had been known very early that sprays might be detrimental to control. Although some early B.C. entomologists were hostile toward the idea of reducing sprays under any circumstances (see discussion, Proceedings of the Entomological Society of British Columbia 3: 20, 1913), Marshall (1963) an entomologist at the CD.A. Research Station in Summerland, B.C., during the early years pointed out that articles arguing the dangers of k i l l i n g natural enemies with sprays appeared in B.C. as early as 1944. When the change occurred, however, there was no looking back. The response of the managers over the next decade was, in almost every region, to search for more and more acaricides (i.e., keep the old repertoire of actions but increase i t s size without making significant changes in the type of action) which could be used against the mites, the secondary pests of earlier and ongoing spray programs (Heme' et a l . , 1979). There was even some resignation to "l i v i n g with" the mites, this being better than nothing. The alternative, cutting back on the use of DDT, seemed certain doom. Of B.C., where the relatively warm, arid conditions of the interior encouraged devastating codling moth populations, Pielou (1959) wrote: "In British Columbia a f u l l y effective material must be used for codling moth whatever the side effects may be. DDT, with i t s consequent upsurges of mite populations, had to be accepted to save the grower from disaster, and Sevin...which has the same disadvantage, w i l l have to be accepted now that DDT resistance has appeared in the codling moth in British Columbia." 9 While trying to deal with the secondary pest infestations, the managers had encountered another surprise. Codling moth, the pest whose control programs caused the mite problem, acquired resistance to DDT (Marshall, 1958). This event was not entirely unexpected but occurred more quickly than had been imagined by those willing to accept the possibility. Shortly before this, ERM had developed resistance to the organophosphates that were in use to control, or at least reduce, the mite explosions triggered by the spraying of DDT (Downing, 1954; O'Neil and Hantsbarger, 1951). The ERM and related tetranychids became notorious for developing resis-tance, often only two to four years after the introduction of a new compound (Bradner, 1977; Helle, 1965; Herne et a l . , 1979). Downing (1961) notes that by 1960 these mites had become resistant to malathion, parathion and other organophosphates; to sulfo-esters; and to new chlorinated hydrocarbons. Nitrophenols and formamidines were eventually added to the l i s t . It now seems that the ERM may eventually gain resistance to Omite and Plictran (acaricides with low predator toxicity, and popular in B.C. and.other regions), perhaps by enhancement of avoidance behavior rather than changes in frequency of resistance genes (Hall, 1980). Thus chemical control of ERM led to failure after failure, not for lack of trying new compounds and techniques, but due to the surprising a b i l i t y of the mites to acquire resistance. There seemed to be l i t t l e chance of controlling the mites by some other means, even though many economic entomologists and orchardists were beginning to look upon biological and integrated control with favor. It should be pointed out, however, that some outstanding attempts at biological control of orchard pests were made very early in Canada (Pickett et a l . , 1946, 1958). In fact, these early investi-gations ("of world reknown": Huffaker et a l . , 1970) served as a model for scientists and managers in other regions during the next decade (Hoyt and Caltagirone, 1971). The conditions surrounding these investigations differed from those elsewhere in that the secondary pests developed much earlier (1930), perhaps due to sprays for oystershell scale and apple scab (Parent and Lord, 1971). New management techniques were developed with the help of predators which could tolerate pesticides. Insecticide-resistant predators are rare. Of 234 cases of arthropod insecticide resistance, only 10 are natural enemies of pests (Croft and Brown, 1975). Behavioral (high searching activity), genetic (fewer generations per year, less inbreeding), ecological (concentration through trophic levels), and logistic ("we haven't been looking hard enough") explanations of this fact abound. Phytoseiids have, however, been quick to develop resistance (they may in fact be naturally resistant to some insecticides). Parathion-resistant forms were f i r s t noticed around 1955. Phytoseiid predators resistant to pesticides now include Typhlodromus occidentalis (azinphosmethyl), 7_. caudiglans (DDT, in some regions), J_. pyri (azinphosmethyl) and Amblyseius  fa l l a c i s (azinphosmethyl, carbaryl, DDT, parathion, methoxychlor, others) (Croft and Brown, 1975; Croft, 1978). These resistant mites were eagerly seized upon by managers who had seen years of chemical mite control failures and liked the relatively new idea of integrated pest management. The growers were a l i t t l e more d i f f i c u l t to convince (as always, and for good economic reasons) but once convinced were often quite unwilling to return to the extensive mite spray programs of the past. Resistant J_. occidentalis which appeared in B.C. around 1965 is now the basis of a strong integrated control program in the Okanagan Valley where azinphosmethyl (Guthion) is used in codling moth control (dormant o i l sprays for mites are also used) (Downing and Arrand, 1978). 11 Orchard pest management in Canada and the Pacific Northwest over the past two decades has been for the most part successful in attaining goals and adapting to minor surprises (both positive and negative). The future of control of codling moth and related lepidopteran and dipteran pests may-include new techniques, such as implementation of pheromones, sticky traps, or the sterile release programs begun in the 1950's. If so, the mite sub-system should be relatively easy to live with. On the other had, i f pest complexes and control practices change (or f a i l to change when they should), the future may be f u l l of surprises. 0.2 Understanding the ERM-Typhlodromus system Considering the length of time that we have been plagued by the ERM, i t is surprising that so much of i t s ecology is not understood. Other tetranychid orchard mites, notably Tetranychus mcdanielli McGregor, have been successfully controlled in British Columbia and in the state of Washington by integrated pest management programs designed around conser-vation and encouragement of Typhlodromus spp. The ERM, however, is often l i t t l e affected by the presence of phytoseiids which may in other cases provide good control. For a variety of reasons, mostly unknown, the ERM may remain below the conventional economic threshold, or may increase dramatically. Because of the unpredictable behavior of this system and the occasional failure to control the prey, Typhlodromus has often been considered to be an ineffective predator of the ERM (e.g. see Downing and Arrand, 1978; reviewed by Huffaker et a l . , 1970). Most of the research on tetranychid-phytoseiid relationships concerns either micro-scale processes (e.g., detailed component analysis of the functional response of predation to prey density) or gross system behavior (e.g., the results of mass release of predators). We lack basic knowledge of "meso-scale" processes. In examining the structure and behavior of this system, I chose to determine experimentally the basic features of predator and prey dispersal in the orchard, the effects of predation on prey popula-tion growth and structure, and the possible relationships with weather and microclimate. These points are considered in the following chapters. Chapter One describes the design and results of an experiment to measure the consequences of predator dispersal within the orchard. Chapter Two details the effects of the predators on the ERM population. In Chapter Three the dispersal of the ERM is investigated. System simulation is employed in Chapter Four in an examination of possible effects and inter-actions of predation, dispersal and temperature. CHAPTER ONE Dispersal of Typhlodromus Within the Orchard "These [cheeses] are never found to breed mites..., probably because the mite-fly is not to be found in Lapland." - Goldsm, 1774 (Nat. Hist. 1:405) 1.1 Introduction We have greatly increased our understanding of the importance of dispersal in ecology since Goldsm's pronouncement on tyrophilous mites. This new understanding goes beyond the observation that mites do not have wings. The ecological consequences of the types and rates of dispersal are often of fundamental importance in determining population behavior (for example see Morris, 1971; Myers and Krebs, 1971; Thompson £i fli., 1976; Wellington, 1964). It is now recognized (or should be!) that dispersal is more than leakage into and out of a population. There is no Fick's law of insect dispersal by which identical members of a population diffuse out of our detection, and out of the population that "matters". Dispersal is common to a l l l i f e strategies on this harlequin* planet, and i t is adaptive even in its uncertainty. Differential dispersal, achieved as differences among age classes, genetic groups, physiological types, or even between hav and have-nots, i s an effective means of dealing with a highly unpredictable world (Wellington, 1977a, 19791; also reviewed by Johnson, 1969; Southwood, 1962, 1977; Stearns, 1976; Stinner e_£ g ± . , 1983). Ecological experiments with mites have provided the greatest insights into the effects of dispersal and movement on predator-prey systems. 1 in the sense of Horn and MacAxthur, 1972.. Huffaker's experiments on the stabilizing effects of certain patterns of dispersion and dispersal are known to every ecologist (Huffaker, 1958; Huffaker et a l , , 1963). In those lab experiments, f a c i l i t a t i o n of prey aerial dispersal and hindrance of predator ambulatory dispersal fostered persistence of an otherwise unstable mite predator-prey system. Similar results were obtained in another mite predator-prey system by Takafuji (1977). Despite such demonstrations of the importance of dispersal in predator-prey systems, the dispersal of predaceous mites and i t s consequences in natural systems are s t i l l not well-understood. McMurtry ej£. a l . (1970) lamented "Information on dispersal powers of various mite predators is almost entirely lacking." In the decade following that assessment, few new insights have been generated, with the exceptions of studies of ambulatory dispersal, mainly of Phvtoseiulus persimilis Athias-Henriot (e.g.., Takafuji, 1977) and Amblvseius f a l l a c i s (Garman), a phytoseiid with high aerial dispersal rates (Johnson and Croft, 1976, 1979, 1981). Gadsby (1982) reviewed a number of related studies dealing with characteristics of the 2 walking behavior of phytoseiids. Some of these studies are highly detailed; e.g., Sabelis' (1981) measurements and models of walking activity, pattern The analysis of angular deviations in animal movement has been of interest for a long time. I happened across one experiment which may be the f i r s t , in fact a plethora of f i r s t s : f i r s t walking-path angle experiment, f i r s t null hypothesis, and f i r s t experiment involving tumors and mice. 1 Samuel 6:5,7-9: "So you must make images of your tumors and images of your mice that ravage the land,... Now then, take and prepare a new cart and two milch cows upon which there has never come a yoke, and yoke the cows to the cart, but take their calves home, away from them. And take the ark of the Lord and place i t on the cart, and put in a box at i t s side the figures of gold, which you are returning to him as a guilt offering. Then send i t off, and let i t go i t s way. And watch; i f i t goes up on the way to i t s own land, to Beth-shemesh, then i t is he who has done us this great harm; but i f not, then we shall know that i t is not his hand that struck us, i t happened to us by chance." (Holy Bible, Revised Standard Version, 1952). and velocity of four phytoseiids and their spider mite prey w i l l probably never be bettered. I consider characterization of dispersal to be indispensible in any thorough examination of the relationship of interacting populations under natural conditions. The following experiment describes my attempts at measuring the movements of Typhlodromus in an apple orchard. 1.2 Scope and purpose The purpose of this experiment was to determine, on the spatial scale of one apple orchard and the temporal scale of one growing season: 1) whether individual phytoseiids (Typhlodromus spp.) move between the trees and the ground cover via the trunk, and i f so, when; 2) what effects the presence of weed cover versus bare ground has on the densities of these phytoseiids in isolated trees; 3) to what extent dispersal of phytoseiid mites occurs by aerial means; 4) whether the rate of immigration is a function of predator density on the tree; 5) what effect different phytoseiid densities have on the growth and structure of ERM populations (this item is considered in a later chapter). 1.3 Design 1.3.1 Treatments and experimental units In a study of ambulatory and aerial dispersal and subsequent population density, the treatments are typically straightforward but lo g i s t i c a l l y d i f f i c u l t . The treatments were as follows: 1) phytoseiid removal (versus no removal); 2) tree trunk banding with Tanglefoot® (versus no banding); 3) allowing weeds and grass to grow directly beneath the trees (versus regular hoeing). These treatments were designed to impede or f a c i l i t a t e predator movement and thereby permit inferences concerning possible differences between resulting population densities. Immigration onto individual trees and consequent population densities were the variables of interest. Because differences among trees are often large in orchard mite experiments (these differences were verified by my own experience in preliminary experiments in 1981), I consider individual tree replicates with sub-sampling a sine qua  non in studies of this nature. Density and dispersion, as characterized by the sample mean and variance, should be measured from the point of view of the mites, that is at the level of the leaf, the tree and, ideally, the orchard ( i f more than one is available). Thus i t is essential that experi-ments be designed to obtain appropriate estimates of the variances of these factors. There are a number of ways to obtain such estimates. I chose the surest approach, basing my analysis on individual leaf counts to ensure that differences between trees could also be tested. Thus, the tree was the experimental unit, branches/tree were sub-samples, leaves/branch were sub-sub-samples, and mites/leaf was the measured variable. The experimental orchard was the east section of the Entomology Orchard, Agriculture Canada Research Station, Summerland, British Columbia. This section is separated from nearby experimental orchards. It has not received regular insecticidal sprays for years. The trees are pruned annually, and the grass between the trees mown regularly. Irrigation, via under-tree sprinklers, occurs weekly. Figure 3 depicts the physical layout of the orchard. The three treatments were applied as a 2 X 2 X 2 factorial, with eight trees per treatment combination. I.arranged the treatments in a Latin square because of possible effects of t a l l shade trees on the east and north 17 s h a d e t r e e s o o °> o 4) 0 0) TJ BJ 0 .C CO 0 0 0 0 BS 84 B3 82 81 0 0 88 87 86 75 0 0 68 78 77 76 67 66 65 58 57 56 55 0 54 74 53 64 73 52 51 72 71 63 62 61 0 0 0 0 48 47 46 45 44 43 42 41 LTJ 0 0 0 38 37 36 35 34 33 32 31 28 27 26 25 24 23 22 21 18 17 16 15 14 13 12 11 N — Figure 3 . Map of the experimental orchard. The row and column numbers of the trees included in the experiment are shown. A "0" represents a tree not included. "S" and " t " represent the positions of the Stevenson screen and temperature recorder. sides of the orchard. The trees included in the 8 X 8 Latin square are indicated in Figure 3 by their row and column numbers. Zeros represent trees not included in the experiment. Rows run east and west. Rows 1, 2, 3 and 4 are Mcintosh and 5, 6, 7 and 8 are (Red) Delicious. Rows 6, 7 and 8, at the north side of the orchard, meander to accomodate missing trees and other experiments. This causes no problem in the design; any effects would be removed as "row" and "column" variation in the analysis. Latin squares need not be physically square. Figure 4 shows the arrangement of the treatments. I arrived at this randomization by randomizing rows and columns from a standard 8 X 8 Latin square taken from Fisher and Yates (1953) . Some of the trees used in the experiment are shown in Figures 5 and 6. 1.3.2 Treatment application Removal of phytoseiids was accomplished via application of carbaryl (Sevin ). Although new strains of carbaryl-resistant Typhlodromus  occidentalis have been produced (Roush and Hoy, 1980), natural populations of T. occidentalis and T. caudiglans are highly susceptible to carbamate insecticides (Croft and Jeppson, 1970; Croft and Stewart, 1973; Flaherty and Huffaker, 1970a, b; Heme and Putman, 1966; Hoyt, 1969; McMurtry et a l . r 1970; Sanford, 1967), and spray application often annihilates them. [Growers in B.C. therefore have been advised not to use carbaryl because of i t s high toxicity to the phytoseiid mites that form the basis of integrated mite control in apple orchards (Downing and Arrand, 1978)]. Chemical removal of the phytoseiids may be cr i t i c i z e d on the grounds that some pesticides (notably DDT, f i r s t demonstrated by Hueck £iiLl., 1952) seem to increase the fecundity of tetranychid mites. No such effect of carbaryl on the ERM has been demonstrated. Experiments with Tetranvchus  urticae failed to show any effect of carbaryl on egg production (Cone, 1963; 8 c H A G E F B D 7 E A F H C G D B c 6 F B G E A D C H o 1 u m 5 G D E B F A H C 4 H F B A D C G E 3 D C H F B E A G n 2 A G C D H B E F 1 B E D C G H F A 8 7 6 5 4 3 2 1 r o w T r e a t m e n t s A B c D E F G S P R A Y : Y Y Y Y N N N B A N D : Y Y N N Y Y N W E E D S : Y N Y N Y N Y Figure 4. The randomization of treatments in the Latin square. e 6. Mcintosh trees used in the study (photographed before the second sampling date). 21 Harries, 1966; Pielou, 1962). Van de Vrie (1964, cited in van de Vrie et a l . , 1972) did not find an increase in population density of the ERM on apple. Thus i t is generally accepted that carbaryl causes increases in the ERM only when i t k i l l s the phytoseiids that previously held the prey at low densities. When these phytoseiids are removed, tetranychid mortality, especially of the larvae and nymphs, is reduced and tetranychid densities increase. Trees designated with treatments A, B, C and D (Figure 4) were indi-vidually sprayed with carbaryl (Sevin®) 50% WP applied at 1 g/L to remove the phytoseiids. Hoyt and Caltagirone (1971) report that although carbaryl is highly toxic to phytoseiids, i t s effect can be ameliorated by avoiding the low central areas of the trees. Consequently, during application of carbaryl to remove Typhlodromus, I deliberately sprayed the entire tree soaking the trunk and bark. About 11 l i t r e s of mixed insecticide was applied to each of 32 trees. The other 32 trees were sprayed with water alone since the force of spraying may knock tetranychids off the leaves (Hudson and Beirne, 1970). Before application of the insecticide, I pruned a l l the trees to a height of 2.5 to 3.0 m, and further pruned the side branches to separate individual trees from one another by at least 0.5 m, so that ambulatory dispersal would not be confounded with aerial dispersal. The individual trees were isolated during spraying with t a l l plastic screens (Figure 7) and care was taken that the low-pressure spray did not produce noticeable d r i f t . The spray treatment was applied once, on the morning of May 30, 1982, under calm conditions. The banding treatments were applied the day after spraying. The trees in treatments A, B, E and F (Figure 4) were painted with a 1 to 3 mm thick band of Tanglefoot (a sticky polybutene compound) about 0.3 to 0.5 m above the ground. This sticky band was renewed every two weeks until the end of ® the experiment, October 1, 1982. [Tanglefoot and similar products are commonly used in studies of insect dispersal, herbivory and pollination. The 22 Figure 7. Application of carbaryl (Seviir^) for phytoseiid predator removal. The screen minimized d r i f t . (Photo: Henry Woensdregt) . 23 technique has recently been applied in a study of phytoseiid dispersal fAmblyseius f a l l a c i s preying on Tetranychus urticae) on small nursery trees by Johnson and Croft (1981), and on older trees in a study of competitors of_P. ulmi (Croft and Hoying, 1977)]. In treatments A, C, E and G weeds were allowed to grow under the trees, whereas treatments B, D, F and H were hoed to a bare surface about every two weeks. Weeds grew to a height of 0.5 to 1.5 m and by mid-July touched the trunks and lower branches of the trees above them. Although spot treatments of glyphosate (Roundup®) were applied to weedy areas outside the experimental orchard, no herbicide was applied to ground below the trees involved in the experiment. The weeds present were mainly common mallow fMalva neplecta Wallr.), common mullein fVerbascum thapsus L.), lamb's-quarters fChenopodium  album L.), horsetail (Equisetum sp.), shepherd's-purse fCapsella bursa- pastoris (L.) Medic.) and various grasses. These weeds are commonly found in the Okanagan Valley of British Columbia. No insecticides other than the carbaryl used for phytoseiid removal were applied to the orchard. During the summer, temperatures were recorded by thermographs in a Stevenson screen between rows 3 and 4 (Figure 3), at various points in the trees via thermistors attached to a Digitech recorder, and by max-min thermometers placed in the trees and in the Stevenson screen. Figures 8, 9 and 10 summarize temperatures and r a i n f a l l during the experiment. Appendix B summarizes the past weather and climate of the Summerland area. Additional sticky traps were used to check for aerial dispersal. Thirty 6 X 6 cm traps coated with high-vacuum grease were posted at 10 positions in the orchard and checked periodically throughout the summer. Four times during the summer, 256 Tanglefoot -coated traps (4 under each tree) were placed in the orchard for two-day intervals. 24 25 a. E Station NW NC SW SC 1—1—i—i—i—I—I—I—i—i—I—i—I—I—i—i—r-i—i—r—i i—i 1 0 1 2 i 4 5 6 7 8 9 10 11 12 13 14 lb 16 17 18 19 20 2122 2i 24 H o u r Figure 8b. Typical differences in a i r temperature at the four corners of the experimental orchard. The temperatures were measured 2 m above the ground in trees 88, 81, 28 and 22 (see map, Figure 3 ) . 25 0> 3 V O. E Maximum, Minimum NOV Figure 9. Orchard a i r temperatures during 1982. Temperatures were recorded in the Stevenson screen shown in Figure 3. Temperatures before May 14 and after September 30 were recorded at the Research Station's main weather station. c o in 2 40 30 20 10 H Lu i i . . . LL MAR APR MAY JUN JUL AUG SCP OCT Figure 10. Rainfall at the Summerland Research Station, 1982, 1.3.3 Data collection Every two weeks, beginning May 13, I collected 10 leaves from each of the 64 trees in the experiment. Of these, 5 came from the north and 5 from the south branch of each tree. Leaves were collected randomly, with the proviso that no very young or very old leaves were taken. Each leaf was individually placed into a 15 cm diameter plastic petri dish. These were transferred to 2 to 4°C storage within 15 minutes of collection. Collections were made row by row, with the order randomized on each sampling date. This pattern ensured that any va r i a b i l i t y in the results due to time or order of collection could be accounted for by "row effect" and removed from the residual. Sampling was restricted to the period between 9 a.m. and 8 p.m. There was, however, no significant effect of time of day on the predator counts (Figure 11, Kolmogorov-Smirnow G.O.F., D = 0.060, p>0.5). In the laboratory, each leaf was examined with a binocular dissecting microscope and, where possible, the species, number, sex and stadia of mites were noted. Counts were made for the following species and categories: Phytoseiids: European red mite: Tydeid mites Typhlodromus caudiglans eggs Apple rust mite larvae nymphs3 adult males adult females A l l adult phytoseiids were mounted on permanent microscope slides (one slide per tree per date) for identification. Collection of the 640 leaves usually T. occidentalis T. spp. Typhlodromus sub-adults Typhlodromus eggs McDaniel spider mite Zetzellia mali  Campyloma verbasei Thrips 3 I counted and analyzed protonymphs and deutonymphs together. They are reasonably similar in morphology and behavior. Duration of the two stages is normally about 5 to 8 days. 27 could be accomplished in two days, but examination and counting usually required three to four days for each sampling period. After counting, the leaves were frozen and saved for further analysis. About once a week, a sample of 100 fresh leaves was weighed and dried for leaf moisture estimation, Sampling dates were: 1) May 13,14 2) May 27,28 (Treatments were applied May 30,31) 3) June 10,11 4) June 24,25 5) July 8,9 6) July 22,23 7) August 5,6 8) August 19,20 9) September 2,3 (No mid-September sample) 10) September 30, October 1 On each of these dates, 640 leaves, 80 leaves per treatment combination, were collected. Collections on sampling date #9 were not made exactly as described for the other samples. This sample was collected as 10-leaf samples which were kept in cold storage before counting. A subsample from each tree was counted later, so that estimates for this date are not as trustworthy as for the other dates. Date #'s 9 and 10 were not included in most of the following analyses. 1.4 Results and Discussion Most of the 1700 phytoseiids counted during the study (plus ca. 900 in other experiments) were Typhlodromus caudiglans. T. occidentalis and the 28 > o o CM c o i_ (U 50 40 H 30 A 20 10 D=0.060 p>0.5 OH—i—i—i—i—i—l—r—i—I i i i i—i—i i i.i—r—t—i—i i i 0 1 2 3 4 5 6 7 B 9 10 11 12 13 U 15 16 17 18 19 20 21 22 23 24 Hour Figure 11. Numbers of fyphlodromus found on leaves collected over a 24-hour period (June 23, 1982, at Summerland, B.C.). Counts were made immediately. The period during which leaves were normally collected, 9 a.m. to 8 p.m., is indicated. rarer _T_. pvri (identification not certain) made up a small proportion of the community. This finding agrees with the observations of Anderson and Morgan (195 8) and Downing and Moilliet (1967, 1971, 1972) in unsprayed orchards in Brit i s h Columbia. Because of the preponderance of_J_. caudiglans and taxo-nomic problems, especially associated with the immature stages, I analyzed the Typhlodromus adults as a group. The results of an analysis of variance of numbers of Typhlodromus are shown in Table 1. Variance-stabilizing transformations (such as y' = log(y+l)) usually required for count data did not alter the results or conclusions of the analysis significantly, (the frequency distributions of counts are discussed in Chapter 2) probably because of the large number of observations. Table 1. Analysis of August 20, variance 1982. of Typhlodromus counts made June 10 -Source df Mean square Test denominator J L Cultivar 1 22.357 Rows/Cultivar 0.09 Rows/Cultivar 6 5.539 Trees 0.11 Columns 7 2.384 Trees 0.58 Treatments 7 23.957 Trees < 10"! Cult X Treatment 7 3.682 Trees 0.30 Trees 35 2.925 Branches/Tree < 10"' Branches/Tree 64 0.660 Error (c) 0.43 Date 5 14.040 Error (c) <1<T Cultivar X Date 5 0.638 Error (b) 0.68 Date X Treatment 35 2.218 Error (b) <io" Cult X Date X Treat 42 1.604 Error (b) 0.013 Error (b) 553 1.013 Error (c) 3072 0.647 Total 3839 (SS = 3177.3) 5 6 6 3 (6 dates X 64 trees X 10 leaves = 3840 observations) Table 2. Subdivision of the treatment sum of squares from Table 1 (via orthogonal contrasts) . Source Spray vs. none Banding vs. none Weeds vs. none S X B S X W B X W S X B X W SS JL 162.94 <10~ 1.39 0.50 0.63 0.65 0.97 0.57 1.24 0.52 0.01 0.95 0.53 0.67 6 167.7 30 In the design of this experiment, the counts from the f i r s t two sampling dates (May 13,14 and May 27,28) were made before treatment application and were intended for use as covariates. However, adjustment for the "before" densities did not alter the mean squares enough to make a significant change, even though the covariate (the count of Typhlodromus on 20 leaves per tree, date #'s 1 and 2) had a significant slope (b = 0.034, p<0.05). In any case, the treatments tended to reverse the weak pre-treatment differences (Figs. 12 and 13).. .1 present the untransformed, unadjusted ANOVA in Table 1. Interpretation of the effects of the sources of variation are discussed below. 1.4.1 Cultivar Typhlodromus densities on Mcintosh and Delicious leaves are shown in Figures 12a and 12b respectively. The populations on the 32 carbaryl-sprayed trees exhibited roughly the same trend on the two cultivars. On the 32 unsprayed trees, the populations suffered an early season decline, and exhibited slow growth after June 10 $ 11, doubling about every 3 to 4 weeks, to peaks in early August of 0.504/leaf (SE = 0.05, n = 320) on Mcintosh and 1.08/leaf (SE = 0.17, n = 320) on Delicious. The seasonal difference in phytoseiid abundance on the two- apple cultivars is weak (p — 0.09, Table 1) and in the opposite direction from that reported by Downing and Moilliet (1967) in the same area. This finding suggests that cultivar differences are due to some factor, such as weather or prey qualities, which differs from year to year or orchard to orchard, rather than to inherent qualities of the cultivars themselves. I explain the rapid decrease experienced by Typhlodromus on Mcintosh by starvation. No McDaniel spider mites fTetranychus mcdanieli McGregor) or apple rust mites (Aculus schlechtendali (Nalepa)) were available on either cultivar until August 5 § 6 and June 24 § 25, respectively, and 0.8 O ve 0.6 H II N o « a. a> 3 C c o « 5 0.4 0.2 0.0 Treatment: O C o r b o r y l  A C e n t r a l 1 r*—I 1—I 1 1 1 1 1 1 1 T" 7 14 21 21 4 II II iS 1 t l« « 30 t U 70 X7 3 10 17 MAY JUN JUL AUG SEP Figure 12a. Average densities of Typhlodromus adults on sprayed and unsprayed Mcintosh apple trees. The arrow indicates the date of carbaryl application. (The smooth curves between the means are cubic splines). Treatment: O C a r b a r y l  A C e n t r e ! 10 17 24 SEP Figure 12b, Average densities of Typhlodromus adults on sprayed and unsprayed Delicious apple trees. The arrow indicates the date of carbaryl application. Standard errors are shown on this and the following figures where applicable (i.e., f o r cases in which treatments d i f f e r significantly and in which variances can be calculated). then only in small patches. The preferred ERM prey stages, larvae and nymphs, were unavailable in late May because the bulk of the f i r s t previous year's winter eggs hatched during the f i r s t week of May. Thus, by May 27, most of the prey generation were in the less vulnerable stage. In addition to the fact that i t s stadia did not overlap sufficiently to provide the predators with younger stadia, the density of the f i r s t generation of ERM was also relatively sparse. At the time of the predators' decline, total ERM prey of a l l active stages was 0.824/leaf (SE = 0.08) on Mcintosh and 2.58/leaf (SE = 0.186) on Delicious. Both values are low, but the fewer prey than leaves on Mcintosh represents a serious food shortage for the phytoseiids on this cultivar. Similarly, ERM summer egg density (winter eggs are relatively unpalatable) was lower on Mcintosh (6.26/leaf, SE = 0.71) than on Delicious (21.02/leaf, SE = 1.90) on May 27 § 28. Differences on May 13 § 14, the sampling date before the decline, also showed the Mcintosh with fewer active prey available (3.2/leaf, SE = 0.59; 4.6/leaf, SE =0.58).. Since the i n i t i a l May 13 § 14 sample) densities of phytoseiids were not dissimilar (0.31/leaf, SE = 0.05; 0.32/leaf, SE = 0.04), the phytoseiids on Mcintosh had a significantly smaller food supply, suffered a demonstrably higher disappearance rate (Figs. 12a and b), and presumably also had a reduced fecundity. The numerical response of the related phytoseiid T. occidentalis has been shown to depend greatly on spider mite consumption (Chant, 1961) and especially on the availability of tetranychid eggs and larvae (Croft, 1972). 1.4.2 Phytoseiid removal On both cultivars, the Typhlodromus were virtua l l y annihilated by the carbaryl application (see carbaryl vs. control, Figs. 12a and b). The three counts during the six weeks following spray application show conclusively that the predator removal treatment was successful: on a total of 960 leaves collected from the 32 sprayed trees I found only 20 adults and 1 nymph versus 243 adults and 117 nymphs on 960 leaves from the 32 unsprayed trees. This successful removal of T_. caudiglans by carbaryl i s in good agreement with Putman and Heme (1960) who report changes of 67 to 6 and 64 to 330 J_. caudiglans per sample on sprayed and unsprayed trees respectively (at twice the concentration I used). Population increase was delayed until late July when a small number of immigrants, and possibly some survivors and descendants of survivors, increased the populations to about half the i n i t i a l counts on the sprayed trees, and about one-eighth of the predator population on the untreated trees. The rate of degradation of carbaryl, like many insecticides, i s often assumed to be f a i l y constant, so that "disappearance" can be modelled by exponential decay. The h a l f - l i f e in the f i e l d is 3 to 7 days; that i s , the 4 decay rate i s between 23% and 9.9% per day . Thus rough estimates of average concentrations as a percentage of concentration on the day of application are as follows: D a t e davs since application remaining May 30 June 10 June 24 July 8 July 22 0 100% 11 7.8 to 33.6% 25 0 . 3 to 8.4% 39 0 . 0 1 to 2.1% 53 0 to 0.5% Rainy weather in June and July (Figure 10) and the growth of leaves probably C e o r t log (0.5)/3 = -0.231, log (0.5)/7 -0.099, reduced the actual concentrations faster than these figures suggest. The period over which carbaryl remains toxic to Typhlodromus also may be shorter than these residual concentrations suggest. Flaherty and Huffaker (1970a) treated grapevines with carbaryl on July 27, August 16 and September 1 to remove phytoseiids, and then reintroduced T_. occidentalis to half the vines on October 9 with no apparent i l l effects. 1.4.3 Banding Unhanded trees did not acquire significantly higher numbers of phyto-seiids than those with movement on the trunk restricted (Table 2; Figs. 13a and b). This surprising result suggests that there is virtu a l l y no ambula-5 tory movement among trees or between ground cover and trees . It is well-known that other Typhlodromus species do not generally overwinter in the ground cover, as does, for example, Amblyseius f a l l a c i s . It seems that X, caudiglans does not move between the tree and the ground cover to any great extent, even during the growing season funlike A. f a l l a c i s : Penman and Chapman, 1980). X- caudiglans overwinters on the bark of the trunk and 6 branches, near pruning damage and s p l i t t i n g . In summer i t ranges over the entire tree, and can be found i n small numbers even on the bark. There was a very small influx of phytoseiids onto the trees in early August, resulting in a significant (p — 0.008) interaction of the banding treatment with the shape of the trends in the counts (Table 3; Figs. 13a and b). Consequently the peak density appears to have been shifted to late 5 From observations of banded trunks and branches, i t appears that sticky barriers may often prevent mites from dispersing without actually trapping and k i l l i n g them. It is possible to contain the mites without catching them in the sticky band. 6 By April 13, 1982, Typhlodromus adult females in the experimental orchard had not yet emerged from overwintering sites. Table 3. Results of tests (via orthogonal polynomials) of effects of treatments on the time trends of the counts. Effects are left blank when p> 0.01, and denoted by * when 0.001< p< 0.01, and ! when p < 0.001. Small values of p indicate a significant increase in the variance explained by adding a given term (i.e., linear, quadratic, cubic, quartic) and thus indicates an effect of the treatment (spraying carbaryl, banding trunks or allowing weeds) on the shape of the "date" trend. European Red Mite E* L N F M JL+N+F+M| Typhlodromus Zetzellia Tydeidae Date 'linear' X Spraying 1 * ! * • Date 'quadratic' X Spraying ! ! ! ! ! * Date 'cubic' X Spraying ! * * ! ; ! Date 'quartic' X Spraying ! ! | Date 'deviations' X Spraying ! ! * ! Date 'linear* X Banding Date 'quadratic' X Banding Date 'cubic' X Banding Date 'quartic' X Banding Date 'deviations' X Banding Date 'linear' X Weeds Date 'quadratic' X Weeds Date 'cubic' X Weeds Date 'quartic' X Weeds Date 'deviations' X Weeds *E = summer eggs; L = larvae; N = protonymphs and deutonymphs; F = adult females; M = adult males o IT c o V v. «l Si E 3 C C o CD 2 H 0.6 0.6 04 0.2 T — i — r -7 14 II 2 » MAY , — . — | r—I 1—I—i—>— 1 — 1 — 1 — 1 — 1 1 ; ; i » i ; w w *«« » « » » » » * 1 JUN JUL AUG SEP Treatment: D B o n d e d  A U n b o n d e d Figure 13a. Average densities of Typhlodromus on banded and unhanded unsprayed trees. o I D o V V Q. k_ 0 ) XI E 3 C C a 0.30 0.25 0.20 0.15 0.10 0.05 Treatment: O B o n e e d A U n b o n d e d 0.00 t . . -1—, , 1 1 r—I 1 1 1 1 ' 1 1 r • • ; /« i,,. ; t « 2 S 2 t « » 3 0 * » « 2 7 J » » 2 4 1 MAY JUN JUL AUG SEP Figure 13b. Average densities of Typhlodromus on banded and unhanded sprayed trees. Note the scale relative to Figure 13a. July or early August (Figs. 13a and b).t Given the lack of an overall effect of banding (Table 2), this apparent effect should be considered weak and can be taken cum grano sa l i s . Possibly there was a slight tendency for the phytoseiids to move off the trees in early July and onto them again in late July and early August, but i t must be emphasized again that observed and recorded movement up and down the trunk was virtua l l y n i l over the entire summer. The above result would be less surprising i f Typhlodromus were capable of relatively high rates of aerial dispersal; e.g., like Aj. f a l l a c i s (Johnson and Croft, 1976, 1979). But I found only 10 Typhlodromus on 30 6 X 6 cm grease plates (10 horizontal, 20 vertical) monitored in the orchard over the summer. A total of 1024 8 X 9 cm horizontal sticky.traps set out in groups of 256 on each of four dates (July 15 & 16, July 19 $ 20, August 15 § 16, August 21 $ 22) yielded only a further 22. If aerial dispersal of Typhlo- dromus were a regular feature of i t s l i f e history, these sticky traps should have captured more airborne mites, even i f a l l such dispersal were unsuccess-ful (a hypothesis which would explain this mite's apparent i n a b i l i t y to quickly reestablish i t s e l f ) . The results of the banding experiment provide further evidence that this mite in fact has a low aerial dispersal rate. Once removed from trees with carbaryl, Typhlodromus did not reach even the early season levels i t exhibited on the unsprayed trees (Figs. 12a and b). Reasonably high rates of aerial dispersal would have decreased these differences by the end of the summer, particularly since the sprayed trees had higher prey densities. In comparison with the behavior of another phytoseiid, this virtual absence of aerial dispersal suggests that Typhlodromus has a relatively poor dispersive capacity. Johnson and Croft (1976) described behavior of A.  fa l l a c i s which i s conducive to i t s dispersal by wind: individuals orient themselves to the wind and raise six of their eight legs which, in the right wind conditions, allows them to become airborne. I observed no such behavior of T_. caudiglans. Increased wind instead stimulated a low, clinging stance. This behavior supports Wellington's (1979b) contention that arthropods are in control of their dispersal and do not normally become airborne "accidentally". Although starved T. caudiglans individuals C increased their rate of searching, they did not show any further proclivity for catching the wind or f a l l i n g off the leaf. Johnson and Croft (1979) found an average of ~50 A_. f a l l a c i s individuals/grease plate/week during mid-August in an apple orchard in Michigan. Johnson and Croft (1981) verified that aerial dispersal was important in i n i t i a l colonization of trees by this species, although not as important as movement up the trunk. However, in other experiments (Croft and Hoying, 1977) sticky bands were unable to prevent A J f a l l a c i s from attaining levels of 1 to 3/leaf, thus attesting to significant aerial dispersal onto the trees. Virtual absence of aerial dispersal does not mean that _T. caudiglans cannot colonize an orchard. For example, J_. caudiglans has been shown to reappear in an orchard, displacing resistant _J_. occidentalis. once organo-phosphate insecticide spraying ceased (within two years in an experiment performed by Downing and Moilliet, 1972), but i t is not clear whether insecticides reduced the_T_. caud ig lans popu la t i ons on those sprayed trees to zero or merely below detection. Even extremely small numbers could repopulate an orchard in which trees touched. I examine the theoretical consequences of the low rate of predator dispersal in a later chapter. 1.4.4 Weeds There is some evidence from other studies that the presence of weeds or grass cover below f r u i t trees may have an effect on phytoseiid population dynamics. This i s particularly true of A. f a l l a c i s . a phytoseiid which overwinters in the ground cover and interacts with resident prey populations (Putman and Heme, 1966; Croft and McGroarty, 1973; McGroarty and Croft, 1978; Penman et a l . , 1979). Phytoseiids in ground cover may even gain protection from pesticide sprays applied to the trees above (Meyer, 1974) . Populations of tetranychid prey, e.g., Tetranychus urticae Koch, in the ground cover have been shown to allow early build-up and movement into the trees of _, f a l l a c i s . although i t i s not always so: in some cases, _ f a l l a c i s may appear earlier and attain significantly higher densities on trees above bare ground (Penman and Chapman, 1980). Weeds may also affect Typhlodromus species, which are often found on a variety of herbs and shrubs (Schuster and Pritchard, 1963). Flaherty (1967, 1969) found that the presence of johnsongrass as a vineyard weed improved the a b i l i t y of T_. occidentalis to control spider mites on grapes. Huffaker et a l . (1970) report that "Fleschner (unpublished data) observed lower citrus red mite Panonychus c i t r i (McGregor) and higher phyto-seiid populations in plots of citrus having a cover crop." There has been some interest in British Columbia in possible effects of weed control practices on integrated mite control programs which depend on encouragement of Typhlodromus spp. (Dr. E. Hogue, Agriculture Canada, during questions following a seminar at the University of British Columbia, 1981). Although Typhlodromus caudiglans overwinters on the trees, I had hypothesized that weed cover could have one of three effects on their numbers. Fi r s t , movement between the tree and the ground cover could stabilize predator populations by providing an alternate prey source and a moderated microclimate. This hypothesis would predict higher late season predator populations, and significant spray X weed and spray X banding X weed interactions, since the resurgence of reduced populations would be facili t a t e d , at least on unhanded trees. Second, weeds could allow Typhlodromus which had fallen or blown out of the trees access to banded trees later in the season, when weeds have grown into the lower branches of the trees. This hypothesis would predict a significant weed X banding interaction, with the greatest differences in late summer. Third, weeds could "trap" phytoseiid mites which might otherwise move or return to the tree via the trunk. This third hypothesis would predict significantly fewer phytoseiid mites on trees with underlying weeds, and significant spray X weed and spray X banding X weed interactions, but in the opposite direction from the prediction of the f i r s t hypothesis. Tables 1 and 2 show that the experiment has reproduced the a l l too common "tragedy of science" ("the slaying of a beautiful hypothesis by an ugly fact.", Huxley, 1870). One cannot conclude that the presence of weeds and grass directly below trees has any effect, positive or negative, on Typhlodromus caudiglans. Even when the analysis i s restricted to only the July and August samples (i.e., the period of most vigorous growth and greatest height of weeds) main and interaction effects of weeds provide no evidence of any differences in phytoseiid density with ground cover. Nor are there any interactions with time (Table 3). This result is further evidence of the relatively low dispersive capacity of Typhlodromus caudiglans. 1.4.5 Other sources of variation Although a weak row effect (p — 0.11) is apparent, the rows and columns did not account for a significant portion of the variation in Typhlodromus counts. The treatments affected the phytoseiids in the same way on the two cultivars, and both cultivars showed the same trend over time. Hence these interactions were not significant. Tree-to-tree variation was considerable (Table 1, p < 10"^) and emphasizes the importance of careful design in orchard experiments. The difference between north and south sides of the trees (branches nested in trees) was not significant (p — 0.43). CHAPTER TWO Effect of Phytoseiid Removal on the European Red Mite 2.1 Introduction In the last chapter, I described an experiment designed to measure dispersal of the phytoseiid Typhlodromus occidentalis, a common predator of tetranychid mites in B.C. orchards. I showed that, because of the absence of significant dispersal onto trees, the early-season removal of the phyto-seiids amounted to predator exclusion*, at least until August (Figs. 12a, b and 13a, b). This chapter details the effects of phytoseiid removal on the population dynamics of the European red mite. The design of this experiment is described in Chapter One. 2.2 Results and Discussion There were five f u l l generations of the ERM. A weak sixth generation consisted of the progeny of a small proportion of the f i f t h generation which seem to have opted for "summer" egg production rather than "diapause" egg production. Unlike Herbert (1970), I found that the generations overlapped considerably, so that the generation peaks, though clear, were not sharp (Figs. 14-19). This result may be due to the separation of samples by 14 days, although my conclusions take into account other samples taken at, odd intervals between sampling dates. Chronologically spaced sampling dates w i l l usually be asynchronous with arthropod populations because of tempera-ture effects. The results of analyses of variance are shown in Table 4. As was the 1 Other predators and possible effects on the ERM are summarized in a later section of this chapter. Table 4a. Analysis of variance of ERM stages in the predator removal experiment, treatments are discussed in the text. Effects of the eggs Source MS _ MS cultivar 7.08 •C.OOOl 4.40 rows/cultivar 0.29 .76 0.72 columns 0.44 .65 0.34 treatments* 3.15 <.001 4.56 cult X treatment 1.07 .12 0.92 trees 2.99 <.0001 2.17 branches/trees 1.79 < .0001 1.39 date 14.46 < .0001 20.18 cultivar X date 2.74 <.0001 2.36 date X treatment 4.56 <.0001 10.08 cult X date X treat 3.95 <.0001 3.59 error (b) 23.70 <.0001 21.37 error (c) 33.78 27.91 larvae _ _p_ nymphs  MS _p_ adult females _MS JP_ < .001 3, .15 * .0001 6.26 <.001 2.25 .10 0, .20 .83 1.08 .012 0.47 .60 0 .38 .63 0.36 .50 0.35 <.0001 4 .78 < .0001 1.81 <.001 5.42 .07 1 .12 .05 0.44 .36 0.64 < .001 2 .50 <.0001 1.95 .002 2.59 < .0001 1 .04 <.0001 1.57 <.0001 1.68 < .0001 24 .49 <.0001 25.81 < .0001 20.46 < .0001 3 .49 <.0001 1.18 <.0001 1.72 < .0001 11 .03 •C.0001 3.65 <.0001 9.77 < .0001 5 .27 <.0001 3.44 <: .oooi 3.29 < .0001 17 .60 •e.0001 19.81 -c.oooi 21.80 24 .95 32.64 29.58 adult males MS _ .002 .41 .70 <: .oooi .31 <.001 <.0001 <.0001 <.0001 <.0001 < .001 <.0001 total sum of squared deviations: 20,308,659 390373 5,936,688 179793.6 119654.5 *Nearly a l l of the treatment effect was due to predator removal. The other treatments i the factorial, weeds and banding, are discussed in section 2.2.3.2. See Table 4b. Table 4b. Breakdown of treatment effects on active stages of the ERM (via orthogonal contrasts). Numerical changes were almost entirely due to the phytoseiid removal treatment. Analyses for separate instars yield similar results, except that the weak B X W interaction is not significant for adult males or females. Contrast SS F Sevin vs. none 535225.0 60.53 <0.00001 Banding vs. none 10570.2 1.19 0.28 Weeds vs. none 1001.5 0.11 0.74 S X B 52.5 0.1 0.94 S X W 20650.8 2.34 0.14 B X W 45961.2 5.20 0.03 S X B X W 19988.3 2.26 0.14 case with the phytoseiid counts, variance-stabilizing transformations did not significantly affect the results or conclusions. The analyses shown here are of the untransformed data. Adjustment via analysis of covariance with the "before treatment" counts as concomitant variables offered no substantial improvement, presumably because of the similar prey densities measured before application of the treatments. Hence the analyses include dates 3 through 8; that i s , the treatment period June 20 through August 20. The major sources of variation are discussed in the following sections. 2.2.1 Phytoseiid removal 2.2.1.1 Density The presence of the phytoseiids had a strong effect on mid-summer ERM densities. In early July sprayed trees had 3 to 4 times as many active ERM prey (larvae, nymphs and adults) as unsprayed trees (Figs. 14a and b). While the differences were highly significant for a l l prey stadia, the presence of the phytoseiids had the strongest effects on the immature stages (Figs. 16a, b and 17a, b) and numbers of summer eggs (Figs. 15a and b). Midsummer densi-ties of adult females were only about 2 1/2 times higher on the sprayed trees than on those with untreated phytoseiid populations. Adult females are too large and active to be easy prey for Typhlodromus (they are larger than their predators in most cases), which is one of the reasons for the poor control of the ERM provided by Typhlodromus in this experiment, and in integrated mite management programs . [In this respect the predator is similar to 2 Since the mid-1960's, Typhlodromus T especially the organophosphate-resistant T_. occidental i s . has formed the basis of a successful Tetranychus mcdanieli control program, but is inefficient in con-tr o l l i n g _P_. ulmi (Downing and Arrand, 1978; Hoyt and Caltagirone, 1971). However, Downing and Moilliet (1972) consider J_, caudiglans to be a more efficient predator of the ERM than is X- occidentalis. v>o^ Figure 14a. Average density of larvae, nymphs and adults of the ERM on the 32 Mcintosh trees. The arrow indicates the date of treatment application (described in Chapter One). 200 n 7 14 21 2* 4 11 11 25 I t It 21 50 t IS 20 27 3 10 1' 24 I MAY JUN JUL AUG SEP Figure 14b. Average density of larvae, nymphs and adults of the ERM on the 32 Delicious trees. 46 Figure 15b. ERM summer eggs on Delicious. 47 Figure 16b. ERM larvae on Delicious. Phytoseiids: D R * m B » U - i — r - 1 — i — i — i — i — i — i — i — < ~ 14 11 21 4 11 11 25 2 1 1« 23 30 t 13 20 27 3 10 17 24 J MAY JUKI JUL AUG SEP Figure 17a. ERM nymphs on Mcintosh. ISO o C O Ti o C D 100 O ) o. 01 Si E 3 c c o s. P h y t o s e i i d s : O R t m « v d A P r 4 l 4 « * T 14 II 2» MAY 11 II 2» 2 JUN » 1« 23 50 t JUL 15 20 27 J 10 17 24 AUG SEP Figure 17b. ERM nymphs on Delicious. 49 Figure 18b. ERM adult males on Delicious. Figure 19b. ERM adult females on Delicious. 51 some insect control specialists who are more interested in k i l l i n g insects indiscriminately than in selectively interfering with those population attributes and qualities which create pest problems]. 2.2.1.2 Age structure Since Typhlodromus caudiglans tends to prefer ERM larvae and nymphs (Putman and Herne, 1964) to adults (or at least finds the younger stages easier to overcome), and since this predator does not move rapidly between leaves, the age structure of prey on a leaf should be modified by the presence and number of phytoseiids. To determine whether the data were in agreement with this hypothesis I divided the prey counts from 5760 leaves (640 from each of nine dates) into four categories based on the number of Typhlodromus per leaf: 0, I, 2, and 3 or more. The age class frequencies of the 193,626 summer eggs and the 122,164 active ERM are shown in Table 5. The differences among dates are large, as might be expected when sampling i s not synchronized with generations. The mere presence of the predator also has a major effect, but once the predators are present, further differences due to increases in their numbers per leaf are not so striking. These slight (but s t a t i s t i c a l l y significant) additional differences are in most cases due to the further small decreases in the proportion of active forms, especially larvae and nymphs, on leaves with additional phytoseiids. Such modifications of age structure consequently are mainly a function of differential predation risk between stadia, but i t is interesting to find that this differential effect of predation is not so strong as the apparent preferences of the predator would lead one to expect. 3 Precision of the counts was determined by repeated calibration to be within ± 5%; doubling this to ± 10% gives likely and acceptable bounds, Table 5. Variation in numbers and age structure of the ERM as a function of date and phytoseiid density. The number in parentheses is the number of leaves that the corresponding column represents. A large likelihood ratio s t a t i s t i c (G2) indicates significant differences in age structure between predator density classes for a given date. Number of Typhlodromus Date ERM stadium 0 1 2 3 summer egg 0 0 0 0 larva 105 4 3 1 May 13,14 nymph 1871 265 60 107 adult female 32 7 3 5 adult male 20 5 0 0 (516) (93) (15) (16) Test G2=26.24 (n=2488) p < .01 summer egg 6713 1309 633 73 larva 3 0 0 0 nymph 11 2 1 0 adult female 846 144 54 10 adult male 18 3 1 0 (534) (73) (28) (5) 2 May 27.28 G =11.74 3 ~ (n=9821) p =i .5 summer egg 9032 624 121 22 larva 1348 116 10 2 nymph 148 15 2 0 adult female 237 8 2 0 adult male 11 0 0 0 1 (593) (39) (4) (4) June 10,11 nymph 148 15 2 0 ^-11698) p cs . 15 summer egg 37881 1460 204 375 larva 472 37 3 1 nymph 876 44 8 8 adult female 3528 165 30 40 adult male 584 21 0 3 (588) (36) (7) (9) June 24,25 G2=40.28 (n=45740) p < .001 July 8,9 summer egg larva nymph adult female adult male 29797 1494 764 416 8643 299 95 65 33013 1627 353 205 3524 183 95 22 4443 150 58 19 (569) (51) (13) (7) G =419.1 (n=84995) p < .001 53 Table 5., continued. Number of Typhlodromus Date July 22,23 ERM stadium 0 _1_ _2_ 3 summer egg 52742 5293 945 2967 larva 2808 239 34 39 nymph 8462 589 94 133 adult female 6563 683 150 187 adult male 2472 198 35 32 (541) (63) (19) (17) Test G =640.4 (n=84665) p < .001 summer egg 21785 4031 1190 2029 larva 2773 537 126 246 nymph 13644 2611 734 1409 adult female 3280 618 244 348 adult male 1395 268 88 129 (496) (85) (24) (35) 2 August 5,6 G =39.16 B (n=57485) p < .001 August 19,20 summer egg 8630 1368 441 880 larva 679 135 46 66 nymph 2014 333 135 237 adult female 1618 350 115 182 adult male 273 60 27 38 (531) (59) (19) (31) G =49.96 (n=17627) p < .001 summer egg 350 26 13 18 larva 3 2 0 0 nymph 132 16 6 11 adult female 489 50 17 36 adult male 78 10 6 8 (548) (52) (20) (20) 2 September 30, nymph 132 16 6 11 G =12.58 October 1 adult female 489 50 17 36 (n=1271) p ^ .3 54 2.2.1.3 Correlation and spatial coincidence The correlation coefficient has been applied as a s t a t i s t i c of spatial coincidence of phytoseiids and tetranychids on leaves. For example, Nachman (1981) found significant positive correlation between Phvtoseiulus persimilis and Tetranvchus urticae. £. .persimilis. unlike most Typhlodromus species, i s a voracious, fast-moving phytoseiid noted for i t s a b i l i t y to quickly find and extirpate tetranychid prey (e.g., when used as a biological control agent in greenhouses: reviewed by Huffaker et a l . , 1970; Hussey and Huffaker, 1976; Sabelis, 1981). Consequently, P. persimilis i s either found i n close association with i t s spider mite prey (e.g., Chant, 1961; Oatman and McMurtry, 1966), or by i t s e l f on the verge of extinction. Typhlodromus has been shown to be uncorrelated with tetranychids on leaves and leaf parts (upper vs. lower surfaces, proximity to midvein) and to assume more aggregative d i s t r i -butions than the ERM (Anderson and Morgan, 1958; Chant, 1958, 1959). This lack of spatial coincidence has been misinterpreted as evidence that occidentalis. JT. caudiglans and £. pyri are not very efficient predators. It would, of course, be equally incorrect to assume the converse: that significant correlations are proof of strong influence and a firm basis for prediction. However, now that the biological relationships between phytoseiids and tetranychids are better understood, i t is at least possible to interpret their interesting and sometimes unexpected numerical relation-ships . In this study, the relationship between Typhlodromus and ERM counts was never linear; scatterplots of counts of the predators versus counts of the prey were hyperbolic in shape. Phytoseiids tended to occur in relatively high numbers on leaves with few (but rarely zero) prey and in low numbers on leaves with many prey (e.g.,-see Figs20...and 21).. Leaves without any phyto-seiids had the highest counts of ERM active stages (commonly 10 - 20, but as 55 high as 460 on one leaf). Thus the correlation was non-significant and negative throughout the summer (except on the last sampling date, when phytoseiids were plentiful and prey scarce). (Similar unexpected negative correlations are common in host-parasitoid systems; see Morrison and Strong, 1980). This relationship could be misinterpreted as indicating poor overlap between predator and prey and therefore the predator's limited capacity to influence prey numbers. In reality, Typhlodromus is unlikely to remain on a leaf with no prey, but w i l l eliminate many of the prey with which i t shares a leaf. Consequently, leaves without predators have either very high or very low (zero or near zero) prey densities, while leaves with predators usually have low to medium prey densities. This effect of predators on the d i s t r i -bution of prey counts is apparent in Figures 20 and 21. 2.2.1.4 Dispersion It is well known that the ERM is not randomly dispersed among leaves. The negative binomial distribution has been applied in most studies as a model of ERM dispersion (Croft et a l . , 1976; Mowery et a l . , 1980; Pielou, I960) 4. This "under-dispersed" pattern is characteristic of insects and mites which are not highly mobile, reproduce parthenogenetically, produce a relatively large number of progeny, and spend a great deal of time eating or resting. Although they do move between leaves, ERM adult females do not move far or quickly unless leaf quality deteriorates. Both the ERM prey and Typhlodromus were aggregated, as is evidenced by b of Taylor's power law (Table 6). The variance of counts is proportional to some power b of the arithmetic mean (Taylor 1961; Taylor and Woiwod, 1982). 4 In fact, Bliss and Fisher (1953) chose ERM adult females on apple leaves as an example data set to i l l u s t r a t e the f i t t i n g of the negative binomial distribution. Table 6. Prey and predator dispersion as indicated by b of Taylor's power law. Each slope was calculated by regression of loge**" on log^ x from n samples of 10 leaves each. The counts for the prey and predator came from the same samples. ERM eggs ERM active stages Typhlodromus Date b (S h, r 2 , n) b (S h > r 2 , n) b (S b, r 2 , n) May 13,14 1.718 1.330 (no eggs) (0.063, 0.93, 59) (0.073, 0.87, 54) May 27,28 1.617 1.285 1.163 (0.053, 0.94, 63) (0.069, 0.86, 60) (0.048, 0.93, 46) June 10,11 1.723 1.551 1.494 (0.051, 0.95, 62) (0.043, 0.96, 56) (0.083, 0.93, 28) June 24,25 1.736 1.424 1.489 (0.063, 0.93, 64) (0.070, 0.87, 64) (0.084, 0.93, 27) July 8,9 1.550 1.737 1.301 (0.071, 0.88, 64) (0.085, 0.87, 64) (0.079, 0.90, 31) July 22,23 1.823 1.513 1.369 (0.118, 0.79, 64) (0.117, 0.73, 64) (0.077, 0.90, 38) Aug. 5,6 1.568 1.750 1.517 (0.092, 0.82, 64) (0.075, 0.90, 64) (0.059, 0.94, 47) Aug. 19,20 1.588 1.581 1.330 (0.079, 0.87, 64) (0.074, 0.88, 64) (0.060, 0.93, 39) Sept. 30, 1.497 1.368 1.439 Oct. 1 (0.057, 0.94, 46) (0.086, 0.82, 58) (0.075, 0.91, 39) 57 woo-tooo -too 5 too-\ XT 200 H 0^ 20 H CRM on loov«« with 0 phr»oM"<ta 111! rtTTrv|J1 0 30 tOO 150 200 250 300 350 CRM on Uov«« with 2 phftoniidl . i l l J M , , r-0 50 100 ISO 200 250 300 350 C o u n t c l o s s 20 0 40 cr CRM on loovoi with I ph/toniid to iath n ,n oDQ_ 0 50 100 130 200 250 300 350 CRM on l«o«oi » i lh S or moro phftosoiidt n n Q 0 50 100 0 0 JOO 250 300 330 C o u n t c l o s s Figure 20. Frequency distributions of ERM counts on leaves with 0, 1, 2, and 3 or more Typhlodromus adults. The distributions are of separate mite counts (larvae, nymphs and adults) from 3317, 334, 85 and 104 leaves, respectively, collected from June 10 to August 20, 1982. Classes are centered on 0 and on multiples of 10, i.e., 0-4, 5-14, 15-24, 25-34, etc. 58 4 0 0 3 0 0 -> • 2 cc Ul 200 100 •••••• • •••••• ••••••••• •••• 3 4 5 T y p h l o d r o m u s Figure 21. A typical scatterplot of ERM per leaf versus Typhlodromus per leaf. These data are from July 8 and 9, 1982. For simplicity, only leaves with both predator and prey present are shown (76 of 640 leaves collected on that date). The parameter /3 is an index of dispersion of the population, and can be 2 estimated by calculation of b via least squares regression of l o g g s on lo g g x (Figure 22). Samples from a Poisson distribution, given by P[X = x] = XX e'A, x! would have values of b not significantly different from 1, since X is both the mean and variance of this distribution. Values of b significantly 2 greater than 1 (i.e.,.s > x) indicate an aggregated distribution of counts. Table 6 shows that ERM eggs, ERM and Typhlodromus adults were aggregated throughout the summer. Larvae, nymphs, adult females and adult males a l l had regression slopes significantly greater than 1 (p < 0.0001). Given that the dispersion of the ERM is non-random, can predation increase or decrease i t s deviation from randomness? Predators, depending on their predation and dispersal behavior and on qualities of the environment, can effect the dispersion of other prey species. Ladybird beetle predation has been shown to decrease (Frazer and Gilbert, 1976) and increase (Gutierrez et a l . , 1980a) the degree of aggregation of aphids under different circum-stances. I hypothesized that the phytoseiids in the experimental orchard could modify the dispersion of adult female ERM through the following behaviors: 1) Predators are highly mobile within a tree and seek out high-prey-density leaves. Upon finding such a leaf they stay, feed and reproduce. Mortality of prey immature stages i s high on such a leaf and results in production of few or no adults. 2) Predators move onto leaves at random and, i f even a few prey are present, do not readily move off the leaf in search of higher density. 3) Predators move on and off leaves often and at random, and do not base their dispersal and oviposition decisions on prey availability. 60 20.000 2000 u £ 200 > "a E «D •> 20 0.1 V * • .• • • i v • • • • • • » • • *_»:vr • •••• .•••YT • •• «• ••• • • • /v J .523 S 2 =0 .683 X r2 =0 .94 = 0 . 0 2 5 , n = 2 4 8 D 0.1 30 300 s a m p l e m e a n Figure 22 The r e l a t i o n s h i p o f the v a r i a n c e to the mean o f counts o f F igu re 22 . The ^ V equat ion i s based on the s t a t i s t i c s from 248 1 0 - l e a f samples c o l l e c t e d between June 10 and J u l y 23 , 1982 (8 o f the 256 1 0 - l e a f samples c o l l e c t e d du r ing t h i s p e r i o d had no m i t e s } . The f i r s t s c e n a r i o would r e s u l t i n d i s p r o p o r t i o n a t e m o r t a l i t y on h i g h - d e n s i t y leaves and reduce the w i t h i n - t r e e v a r i a b i l i t y o f counts o f prey per l e a f r each ing m a t u r i t y more than the mean, thus dec reas ing the depar ture from randomness. The second and t h i r d types o f behav io r would reduce o v e r a l l prey d e n s i t y wi thou t r ega rd to numbers per l e a f and e i t h e r i n c r e a s e the depar ture from randomness o r leave i t unchanged. To t e s t whether p r e d a t i o n would a l t e r the depar ture o f p rey counts per l e a f from P o i s s o n e x p e c t a t i o n s , I compared es t imates o f the var iance- td-mean r a t i o s o f ERM e x p e r i e n c i n g p h y t o s e i i d p r e d a t i o n w i t h those which were f ree 61 from phytoseiid predation^. The phytoseiids were removed via carbaryl application May 30, 1982 from 32 trees, and allowed to remain on 32 trees. The randomization, other features of the design and methods of sampling are described in Chapter One. Estimates of the variance-to-mean ratios of ERM adult females from the with- and without-phytoseiid treatments on the three sampling dates following treatment application are shown in Table 7. Since 2 the method of moments estimate, s^/x^, is inefficient, the maximum likelihood estimates of the variance-to-mean ratio for unde^dispersed counts, 9^ = m^A^ » are shown. Values of 9 and k, estimates of the parameters of the negative binomial distribution parameterized as were estimated by Newton-Raphson iteration. A likelihood ratio test of H : 0 = ft, o 1 2 equivalent to (Johnson and Petkau, 1983) was performed for each date to assess differences between with- and without-phytoseiid samples (Table 7). Variance-to-mean ratios were significantly reduced by phytoseiid predation, indicating that predators may be seeking out and severely reducing high counts of ERM prey. This i s supported by behavioral observations I made in the orchard: Typhlodromus adults seem to be able to assess prey density since they do I do not recommend the variance-to-mean ratio, or any of its relatives, as a general index of clumping. However, for samples from populations with the same distribution (e.g., negative binomial), this ratio i s a relative measure of degree of departure from Poisson expectations. A ratio of 6 indicates a "less random" distribution than a ratio of 3. Inference for and power of variance-to-mean ratios are described by Johnson and Petkau (1983) and Perry and Mead (1979) respectively. Table 7. Maximum likelihood estimates of the variance-to-mean ratio, 0='7k/or counts of ERM females with and without phytoseiid predators (Figure 23). The ratios are compared before treatment and on six dates after (320 leaves per treatment per date). The result of likelihood ratio test of equality of variance-to-mean ratios i s shown for each date. (The method of moments estimates, s 2/x, are shown in parentheses) The parameters were estimated via Newton-Raphson iteration. phytoseiids removed phytoseiids present Date May 27,28 June 10,11 June 24,25 July 8,9 July 22,23 August 5,6 August 19,20 September 30, October 1 2.93 (4.02) common 0 = 2.95 2.56 (3.56) 6.09 (7.45) 3.60 (5.09) 3.91 (5.35) common 0 = 4.16 6.50 (7.24) common 0 = 6.54 5.70 (7.76) common 0 = 5.32 1.97 (3.11) 2.98 (3.60) 0.95 (1.75) 3.33 (4.19) 2.27 (3.24) 4.42 (5.26) 6.58 (6.68) 4.96 (6.07) 2.87 (3.64) -2 log X 0.003 p ~0.9 6.29 p=*0.012 13.07 p<0.001 7.27 p<0.01 0.68 p> 0.5 0.001 p>0.9 0.58 p > 0.5 1.93 p^0.16 63 not remain on "clean" leaves as long as on leaves with prey. This assessment may be t a c t i l e , olfactory (Hoy and Smilanik, 1981), or both, and may interact with hunger. The change in the shapes of the distributions is apparent in Figure 23. The presence of Typhlodromus results in fewer high-density leaves as well as more leaves free from prey. Distributions of counts did not d i f f e r s i g n i f i -cantly for the month before the treatment application (Figure 23a). It may be argued that, since the experimental units were trees and the samples consisted of 10 leaves from each of the 64 trees, the counts per 10-leaf sample should be analyzed rather than the counts per leaf. However, the leaf provides the natural sampling unit from the point of view of the mites. I would in some cases accept "spurs" or leaf-clusters as the sampling unit, but feel that the count of mites in a sample of n (say 10) leaves individually chosen at random is suitable for inferences about density changes (although variance among leaves would be unknown) but not about dispersion and local predator-prey interaction. 2.2.1.5 Prey-to-predator ratios The ratio of numbers of prey to predator is not always useful and is often misleading since: 1) the average ratio is not the ratio experienced by most individuals; 2) the degree of and differences in spatial coincidence between predators and prey are masked; 3) the ratio is unrelated to disper-sion of either species; 4) other variables important to system behavior are unrepresented; 5) subsystems which are out of phase tend to be averaged by the ratio, and 6) the ratio does not define any unique system state, and consequently has l i t t l e descriptive, explanatory or predictive power. ISO wo H 30 H 2i0-\ 200 ^ R ( b e f o r e ) ISO 100 -- R -isoH 100 0 10 20 0 10 Number of L"RM females per leaf The effect of phytoseiid removal on dispersion of ERM adult females. The frequency distributions of counts per leaf are shown for two treatments: phytoseiids removed (R) and phyto seiids present (P). May 27,28 (left) and June 10,11 (right) 65 60-, Figure 23b. The effect of phytoseiid removal on dispersion of ERM adult females. The frequency distributions of counts per leat are shown for two treatments: phytoseiids removed (R) and phyto-seiids present (P). June 24,25. 40 A 3 0 A 20 A « H 0 50 A AOA 30 A 20A io H 0 ERM females per leaf The effect of phytoseiid removal on dispersion of ERM adult females. The frequency distributions of counts per leaf ar shown for two treatments: phytoseiids removed (R) and phyt seiids present (P). July 8,9. 4 ( H 30 A R io A 10 10 20 30 40 >50 20 A io H 0 0 10 20 30 40 30 £RM females per leaf The effect of phytoseiid removal on dispersion of ERM adult females. The frequency distributions of counts per leaf are shown for two treatments: phytoseiids removed (R) and phyto-seiids present (P). July 22,23. However, since they are easy to estimate" and provide rough "go/no go" decision c r i t e r i a , . prey-to-predator ratios often serve as a gross index for growers interested in integrated management of orchard sites. I present the ratios here as a comparison between my experimental populations and those normally encountered in apple pest management. Apple-growers in British Columbia are recommended to apply an acaricide " i f there is an average of more than fifteen red mites per leaf and more than ten red mites for every phytoseiid mite" (B.C.M.A., 1980). Up to thirty red mites per leaf are allowed after mid-July. The populations on the sprayed trees in my dispersal experiment, reaching a peak in early July of over 129 ERM active stages/leaf (SE = 5.8, n = 320), were disastrously high, as was expected. The effects on the trees were apparent: the leaves on many of the sprayed trees were chlorotic and leathery. The unsprayed trees had ERM densities of up to about 40/leaf (SE = 2.9, n = 320) in early August (see Figs. 14-19). A grower who might not spot the tiny larvae and lighter-colored males would have observed densities of up to 32/leaf and probably would not have applied acaricide. Nevertheless, the prey-to-predator ratios would have indicated a need for chemical control according to the rules given above: even on the unsprayed trees, prey-to-predator ratios were usually well above 10.0 (Table 8). The densities of Typhlodromus were similar to densities in many unsprayed orchards, but well-managed 1PM orchards often have even more phytoseiids. 2.2.2 Other predators and prey The other predators of the ERM which were encountered in the 1982 orchard experiment were Zetzellia mali (Ewing) and Campyloma verbasei But the qualities of the estimate (bias, precision, covariance) are rarely given adequate attention. Table 8. Prey-to-predator ratios in the orchard experiment, 1982. Calculated as (ERM active stapesl/fTvphlodromus adults + nymphs). Mcintosh Delicious Date spraved May 13,14 645/(35+0)=18.4 May 27,28 150/(21+0)=7.14 (Treatment application) June 10,11 June 24,25 July 8,9 July 22,23 August 5,6 August 19,20 September 30, October 1 187/(1+0)=187 1264/(2+0)=632 12578/(6+0)=2096 5107/(10+0)=511 5166/(25+0)=207 1328/(10+0)=133 168/(13+0)=12.9 unsprayed . 376/(63+0)=5.97 117/(17+2)=6.16 158/(14+8)=7.18 530/(20+7)=19.6 4510/(38+5)=105 2368/(52+31)=28.5 2706/(103+3)=25.5 597/(69+4)=8.18 99/(62+0)=1.60 sprayed 645/(49+0)=13.2 459/(57+0)=8.05 975/(4+0)=244 2923/(3+0)=974 28576/(4+l)=5715 unsprayed 822/(83+0)=9.90 367/(55+9)=5.73 579/(45+33)=7.42 1102/(70+47)=9.42 6864/(56+17)=94.0 11309/(15+2)=665 3934/(109+30)=28.3 10371/(14+0)=741 10205/(172+29)=50.8 1744/(7+0)=249 2638/(141+10)=17.5 296/(17+0)=17.4 301/(95+0)=3.17 ON (Meyd.-D). I assume that ,C. verbasei. a mirid bug predaceous on the ERM (Lord, 1949), had limited impact on the tetranychid population, since i t s density was zero to very low during most of the summer (see Appendix D). Z. mali is a stigmaeid predator of eriophyids and tetranychids. This mite predator may effect significant predation on populations of the apple rust mite (Hoyt, 1969) . It is a weak predator of the ERM, feeding mainly on eggs and immature mites (Santos, 1976, 1982; White and Laing, 1977 a, b), and may have a useful but secondary role in integrated mite control (Downing and Arrand, 1978; Parent and Lord, 1971). I found that ,Z_. mali confined on leaves in the orchard consumed up to two ERM summer eggs per day, but often did not eat for several days. ERM egg counts per leaf and egg/female ratios were not significantly correlated with _Z. mali density, but they were never high when more than about 3-4 Z_. mali were present on a leaf. Covariance analysis suggested that Z. mali had l i t t l e effect on the ERM, except at very low densities. Zj mali populations were not strongly affected by the carbaryl, banding and weed treatments applied in the study of phytoseiid dispersal (p — 0.17 for differences between insecticide treatments), so they are apparently resistant to the insecticide carbaryl and are not in strong competition with Typhlodromus caudiglans (Figs. 24a andtx, and Appendix C) . However, on both Mcintosh and Delicious, Z_, mali populations on unsprayed trees had strong peaks in early August. The differences are slight, but perplexing. It is not clear why this difference would be delayed more than two months after application of treatments. Before the experiment, I expected Z. mali, i f i t survived the carbaryl application, to show increased population growth in the absence of phytoseiid competitors. However, since apple rust mites fAculus schlechtendali (Nal.), the prey species preferred by Z. mali and readily eaten by phytoseiids, did not appear (Appendix E), Z_. mali did not Figure 24b. Zetzellia mali on Delicious. 72 gain from removal of the phytoseiids. Other common mites on the apple leaves were members of the family Tydeidae. Tydeids may be predators of eriophyids (Downing and Arrand, 1978), but some more li k e l y feed on cast cuticle and dead mites, insects and eggs (Brickhill, 1958). I found these tiny mites at relatively high densities throughout the experiment (Figs. 25a, b), usually nestled, with their eggs attached to leaf hairs, along the midrib on the lower surfaces of leaves. One one occasion I observed _• caudiglans feeding on a tydeid, but am inclined to believe that I witnessed a rare event. Tetranvchus mcdanieli McGregor was rare in the orchard (Appendix E), probably due to predation by phytoseiids. This spider mite is readily suppressed by Typhlodromus and is easily controlled in an integrated mite control program. 2.2.3 Sex ratio The female proportion of the adult European red mites ranged between 34% and 99%, depending on treatment and date. The trees with phytoseiids present had a significantly lower proportion of males (p < .0001), probably due to differences in predation risk. Adult males are smaller than adult females and make easier prey. Males are also much more active, and may have a higher probability of encounter with the predator. Differences in sex ratio between the two cultivars are small and inconsistent. It is unlikely that differences in density or phenology of the mite populations on Mcintosh and Delicious are related to sex ratio. 2.2.4 Other sources of variation 2.2.4.1 Cultivar A l l stages of the ERM were about twice as high on Delicious as on Mcintosh. This difference is well-known to growers and entomologists 73 Table 9a. The effect of predation on ERM sex ratio. Phytoseiids present greater predation risks to the males and thus raise the female proportion significantly. The numbers are the count on 320 leaves. Treatment: Phytoseiids removed Phytoseiids present Date Q proportion f/ proportion May 13,14 T 32 16 .67 15 9 .63 May 27,28 588 14 .98 466 8 .98 (May 30: treatment appl ication) June 10,11 144 10 .94 103 1 .99 June 24,25 2574 506 .84 1189 102 .92 July 8,9 2232 3669 .34 1322 1001 .57 July 22,23 4346 2116 .67 3238 621 .84 August 5,6 2144 1025 .68 2346 855 .73 August 19,20 1070 191 .85 1195 207 .85 September 30, October 1 299 66 .82 293 36 .89 74 Table 9b. Sex ratio of the ERM on the two apple cultivars. No trend or important differences are apparent. Cultivar: Delicious Mcintosh nat.fi May 13,14 9 18 cf 6 proportion .75 29 6 19 proport .60 May 27,28 246 9 .97 808 13 .98 (May 30: treatment application) June 10,11 70 1 .99 177 10 .95 June 24,25 ,1167 157 .88 2596 451 .85 July 8,9 1181 1638 .42 2373 3032 .44 July 22,23 3059 1149 .73 4525 1588 .74 August 5,6 1538 499 .76 2952 1381 .68 August 19,20 650 100 .87 1615 298 .84 September 30, October 1 158 33 .83 434 69 .86 Figure 25b. Tydeidae on Delicious. 76 (e.g., Downing and Moilliet, 1967). Delicious often requires more acaricide applications than other apple varieties (B.C.M.A., 1980). The difference can be attributed to some quality of the leaves, and not to indirect effects via phytoseiid densities: J\ caudiglans populations were about twice as dense on Delicious as on Mcintosh in this study. On the 32 trees from which phytoseiids were removed, the same difference in prey density between c u l t i -vars is apparent (Figs. 14-19). Cultivar X treatment interactions were negligible (Table 4). 2.2.4.2 Weeds and banding As I showed in the last chapter, the presence of ground cover had no effect on the densities of Typhlodromus caudiglans or on i t s a b i l i t y to move between trees. Consequently no indirect effects of weeds via predator numbers could have been significant in determining parameters of ERM popu-lations . There was no significant effect (p > 0.5) over the summer as a whole but this is not surprising, since the weeds did not attain significant growth and biomass until late July. After this date, however, the presence of weeds seemed to result in an increase in ERM numbers on the trees above (Figure 26). The time courses of most ERM stages were altered in shape by the weed effect (Table 3). By August 5 and 6, the trees "with weeds" had reversed from being slightly lower in density to significantly higher (active stages per leaf: 53.80, SE = 3.97, vs 35.11, SE = 2.48; p < .001). A l l stadia showed similar differences, although these may be due to effects on one stage, i.e., adult females. Dispersal of the ERM during July and August was quite high, and i t may be that the presence of weeds below a tree increased density some-what by returning mites to the tree, or by modifying microclimate and increasing disperser success in moving into a new tree. Dispersal of the European red mite is described in the next chapter. 100 04 o with A Without Weeds: 14 11 2* 4 11 II 25 2 * 1« 2 5 50 ( 15 20 27 3 10 1/ 24 1 MAY JUN JUL AUG SEP Figure 26. Difference between weed treatments in the density of the European red mite active stages. Trees with undergrowth had significantly more ERM, but only during early August. Banding had v i r t u a l l y no effect on any stage of the European red mite. 7 There seems to be no t r a f f i c on the tree trunks . In related experiments ® adhesive tape traps coated with vaseline or Tanglefoot wrapped around various parts of test trees showed l i t t l e evidence of ERM movement on trunks or large branches, although movement on twigs and small branches was substantial. 2.2.4.3 Position in the orchard I double-blocked this experiment (i.e., randomized according to the constraints of a Latin square) because of possible gradients running across 7 The few ERM individuals found on the trunks are usually adult females. 78 rows (N-S) and across columns (E-W). Because of the shade trees (Figure 3) the sun broke on the westernmost trees every morning. Shade trees, a stream and two rows of apple trees on the north side could have altered microclimate or introduced dispersing mites and insects. Temperatures recorded at various times over the summer showed that the northeast corner was cooler than the rest of the orchard (Figure 8b). In the end, "columns", the east-west gradient, had no significant effect on the prey or predators. The "row" effect remaining after accounting for differences due to cultivar was small and significant only for ERM adult females (p — 0.012). This small difference may be due to the differences in microclimate shown. It may also be due to dispersal gradients, although the gradient directions are not clear. 2.2.4.4 Variation between trees Tree-to-tree differences are often a nuisance in experiments of this type. They may arise because of the different histories of the pest popu-lations and of the trees themselves. Physiological condition of the trees may vary greatly, depending on the types and densities of other pests, root condition, f r u i t production, s o i l factors and random events. Since the individual trees were the treated units in this experiment, I tested the significance of treatments and the components of the Latin square against tree-to-tree variation. Subsampling within trees (branches/tree and leaves/ branch) allowed assessment of this variation. Differences between branches on a tree were small but highly significant. They did not follow a definite pattern for a l l age classes. The moderate to high variation between trees and leaves (Table 4) verified the need for well-designed experiments of this type in orchard pest management. The relationship between the variance and the mean of single tree samples was described in section 2.2.1.4. 2.2.4.5 Interactions with cultivar The phytoseiid removal, banding and weed treatments affected the ERM populations in the same way, over the summer as a whole, on the two cultivars hence cultivar X treatment interactions were not significant (Table 4). However, the differences between the with- and without-phytoseiid groups disappeared on Delicious much earlier than on Mcintosh (Figs. 14-19), resulting in highly significant cultivar X date X treatment and cultivar X date interactions (Table 4). The differences may be due to the generally higher prey-to-predator ratios on the Delicious trees (Table 8), although this difference in ratios may be a consequence or coincidence rather than a cause. It seems more likely that the lower-prey-density trees gain ERM migrants from high-density trees, and that this effect is more pronounced on the Delicious side of the orchard due to the higher ERM densities. In the next chapter I present evidence that European red mites disperse by air in high numbers, and that this dispersal is an approximately linear function of density, modified by weather and age class. 80 CHAPTER THREE Dispersal of the European Red Mite 3.1 Introduction 3.1.1 Dispersal by air Spider mites, being tiny and parthenogenetic, have great potential for air dispersal, and have evolved fascinating behavioral and anatomical adaptations to increase the ease, distance and probability of success of travel and subsequent colonization. These modes of dispersal have been under investigation for some time. Orchard (1927) described the way in which Tetranychus urticae climbed to the top ridge of a greenhouse and blew away in large numbers on the wind. The ERM, unlike most tetranychids, spins a fine s i l k thread ( f i r s t described by Steer, 1940) and hangs from i t u n t i l , at a time loosely determined by the length of the strand, the strength of the breeze, and the humidity, the thread breaks and allows the mite to float away. This general method of aerial dispersal is common to phylogenetically diverse groups of arthropods: some caterpillars "balloon" on s i l k (the gypsy moth is a well-known example), as do some spiders (even between continents: Platnick, 1976). The ERM tends to spin under calm conditions. Marie (1951) found that in wind tunnels the mites were slower to descend after exposure to high winds than to low to moderate breezes. I f i r s t observed spinning in the Summerland experimental orchard on July 10, 1982, under clear conditions in a dead calm. From this date on, I observed occasional spinning, but almost always on warm, calm evenings. It is likely that dispersal of the mites is inhibited, not enhanced by wind. Putman (1970) found that air blasts greater than 20 m/s are needed to dislodge them from peach leaves. As Rabbinge (1976) points out, such wind speeds would occur rarely i f at a l l in an orchard. Wolfenbarger (1975) considers control by insects and mites over their dispersal timing, distance and even direction to be the norm. He even observed a population of another species of phytophagous mite before and after a hurricane and found that many individuals remained. The dispersing class of the ERM is the adult female (Marie, 1951). This i s generally the case with other tetranychid species (e.g., T. urticae. the two-spotted spider mite; Hussey and Parr, 1963). In fact, male ERM are incapable of spinning s i l k ; their aerial dispersal probably depends on accidental dropping off the leaves and occasional entanglement in the s i l k of the females. ERM populations generally produce more females than males (see Table 9) and, since the females are also much larger, more than 90% of a population's biomass goes into potential dispersers (Mitchell, 1970). Although females are usually mated at the time of their last ecdysis*, an unfertilized ERM female would be able to found a colony because of haplo-diploid gender determination. Males are haploid (arrhenotoky) and a female could live long enoughXo mate with her sons. These then are the accepted characteristics of ERM dispersal: adult females disperse on silk strands, probably under calm conditions, and passively disperse to other vegetation. It has also been suggested, with resort to l i t t l e data, that dispersal rate is a function of density, such that i t does not occur below a threshold, or occurs at an increased rate at higher densities. Thus predators might reduce prey dispersal rate by 1 ERM males seem to find female deutonymphs nearing emergence by detection of a sex pheromone. After finding a female ready to emerge, they exhibit guarding behavior and agression toward other males. These observations are based on lab and f i e l d examination with a portable microscope. The same general behavior has been reported for other tetranychid species (Potter et a l . , 1976a, b). 82 reducing prey density (Marie, 1951). Although obligatory dispersal during a particular l i f e stage seems to be common among insects, there is evidence that density-dependent dispersal does occur in some cases (e.g., Murai, 1977) and may serve as a means of population regulation. From the individual's point of view, i t s likely fitness may be increased by dispersing from a host plant in poor condition. 3.1.2 Dispersal by walking European red mites hatch from diapause eggs on bark in the f i r s t gener-ation and from summer eggs on leaves in subsequent generations. Upon hatch-ing, larvae seek out a feeding site (often only millimeters from their hatch-ing place). Larvae, protonymphs and deutonymphs do not tend to move between leaves to the same extent as adults, but even the adults are not very active. Much of their time is spent either feeding or resting (or, in the case of the highly active males, attempting to mate). Movement between leaves may be stimu-lated by leaf deterioration and in this sense is a function of density and age structure. My observations in the laboratory and orchard suggest that movement between touching trees can be appreciable. There is l i t t l e information on the behavior of these mites on the ground or on other vegetation. Other tetranychid mites have been shown to move over the s o i l in accordance with the plane of polarized light (Hussey and Parr, 1963). This a b i l i t y would presumably help them to increase their dispersal distance and find other food plants, rather than wandering without guidance, and perhaps returning to the original host. However, I rarely found tetra-nychids or their predators on s o i l or ground cover in the orchard. 3.2 Design of experiments 3.2.1 Dispersal by air In order to measure prey dispersal within the orchard and assess i t s 83 relationship to density, I performed an experiment in which I captured dispersers under trees of known density, and repeated the experiment in different kinds of weather and at different phenological stages. Because of the nature of the randomization (Latin square) of the predator-removal experiment, and since row and column effects showed no clear gradients, by June I had an 8 X 8 random patchwork of prey densities that could be used for the dispersal experiment. On four dates, July 15, July 19, August 15 and August 21, I set out 256 8 X 9 cm white cardboard sticky traps (coated with Tanglefoot®) . The cards were placed horizontally, four beneath each tree about 0.5 m above the ground and as far from the trunk. The traps were set out at noon and picked up 48 hours later. I counted the ERM caught on the top of the cards and determined their stadium.and sex where possible. The few mites found on the bottom or very edge of the cards were not counted, since these did not arrive by air, but more likely crawled up the supporting poles. Counts were made at 10X and 25X magnification. The July counts were made immediately, but the August traps were stored for three months before counting. As a consequence, the males and some immatures were darkened and misshapen, so that only female red mites were counted on the August samples. Males and nymphs were both rarer on the August traps, but no precise estimates of numbers were made. Density of the mites on the trees was estimated by the July 8 § 9, July 22 § 23, August 5 § 6 and August 19 5 20 counts described in Chapter One. 3.2.2 Movement between branches In order to measure dispersal of walking migrants between the branches, I fumigated individual branches to remove a l l mites and then measured the rate of return. In a preliminary experiment (June 17) I fumigated with nicotine sulfate but found that most of the adults and a smaller proportion 84 of the nymphs and eggs survived the effects of 20 minutes of this fumigant. Repeating the experiment with the fumigant HEPT was successful. The experi-ment was performed as a one-factor RBD, s p l i t for sampling time. There were three blocks, each consisting of one Mcintosh and one Delicious. On each ® tree I selected six branches, three of which were painted with Tanglefoot about 0.5 m from the t i p . On the day of fumigation (August 8, 1982), I slipped a black plastic garbage bag with both ends open over each branch and taped i t at the bottom. I sprayed about 100 ml of fumigant (@ 2 ml/1 water) into each bag and sealed the upper end with twisted wire. Each branch was fumigated for 30 minutes before removal of the bag. HEPT dissipates within a short time (-'hours) and leaves no residue. I examined the leaves of the branches with a hand lens during the two days following the fumigation and then collected the leaves from one branch with and one without Tanglefoot on each of the six trees 4, 8 and 16 days after fumigation and counted the mites. I intended to use the difference between the sticky and unpainted branches to adjust for air dispersal between branches. 85 3.3 Results 3.3.1 Sticky traps Large numbers of red mites were captured on the sticky traps, indicating that emigration from the trees can be extensive. On the four dates, repre-senting a total of eight days of dispersal and about 7.4 m of horizontal sur-face, the traps caught nearly 8000 adult female dispersers (Figure 27). Differences between cultivars were large and significant (ANOVA, p > .0001): about 2.4 times as many females f e l l below Delicious as below Mcintosh. These differences partially reflect differences in the densities on the trees (see Figs. 19a and b): during July and August there were 1.8 times as many females on Delicious as on Mcintosh trees (based on samples of 320 leaves from each cultivar on each of four dates). Most of the emigrants were females, and they were overrepresented on the traps. In July, the population on the trees consisted of about 17.7% adult females, while 56.3% and 69.9% of the emigrants in the July experiments were adult females (Table 10). Weather during the dispersal experiment is outlined in Table 11. As predicted, warm, dry, clear conditions resulted in higher catches. The number of dispersers caught below trees was highly correlated (r — 0.5 on a l l dates) with estimates of densities on corresponding trees (Figure 28). It seems that many dispersers from the trees do not travel very far, and may represent mites which accidentally f e l l or were shaken off the leaves. Since these trees had about 3000 to 6000 leaves each 2 5 during this period, there were densities from 10 to about 10 female ERM 4 per tree, and dispersal of 0 to 30 females/trap, or 0 to 10 under a tree, per two-day period. At these population levels, i t seems that from one quarter- to a half-million mites are showering down each day during late summer in an orchard of this size. 86 2500 n Cultivar: E 2 M c i n t o s h CD D t l i c i o u e July 15,16 July 19,20 Aug 15,16 Aug 21,22 Figure 27. Total dispersing adult female ERM caught on sticky traps. Dispersal, as estimated by trap catches beneath trees, did not increase disproportionately under trees of high density. There is no evidence for a threshold density for dispersal, or even for a relationship between the rate (i.e. proportion emigrating) and the density on the tree (see Figs. 28a, b, c, d). The relationships in a l l cases are nearly linear. Models f i t t e d to the data in search of upward curvature indicate that the relationships are either linear or actually reaching asymptote. Fit t i n g log-log regressions (power curves) gives the best f i t and none have powers significantly greater than 1 (Table 12). It may be argued that the effects of density on dispersal rate would be apparent i f the trap catches and mite density estimates were less variable. However, the density-dependent dispersal rate hypothesis can be rejected for lack of evidence on the basis of high v a r i a b i l i t y alone. If other factors are obscuring this relationship, i t cannot be a strong and influential one. Table 10. Evidence for differential dispersal by age class. Adult females are the active dispersers and are overrepresented in the sticky trap samples. Dispersal of nymphs and males is also substantial, however. Age structure of the ERM population, ...on the apple leaves: July 8,9 July 22,23 nymphs adult females adult males 35198 3554 4670 (on 640 leaves) (81.1%) (8.2%) (10.7%) 9278 7583 2737 (on 640 leaves) (47.3%) (38.7%) (14.0%) ...dispersing by air: nymphs adult females adult males ( traps) July 15,16 891 ' 1421 213 on 256 8 X 9 cm (35.3%) (56.3%) (8.4%) horizontal sticky July 19.20 703 (16.0%) 3070 (69.9%) 617 (14.1%) (on 256 8 X 9 cm horizontal sticky traps) Table 11. Weather during ERM dispersal experiment. date July 15. July 16 Temperature °C (± 0.5) max min 13.3 21.1 11.1 8.9 'D ID. 2.2 5.1 rain (mm) General conditions 5.2 partly cloudy; windy 0 clear Mean catch per trap (and SE) Mr.Tnt.osh Delicious 2.77 (0.22) 8.34 (0.46) July 19 July 20 25.6 11.1 20.0 15.6 8.4 7.8 3.0 sunny and partly cloudy overcast part of the day 9.09 (0.59) 14.90 (0.70) August 15 August 16 21.1 21.1 5.6 10.0 4.4 0 sunny 5.6 6.0 sunny; overcast and windy part of the day 0.52 (0.08) 4.23 (0.36) August 21 August 22 24.4 25.6 13.3 11.1 8.9 8.4 0 0 sunny sunny 6.00 (0.45) 16.52 (0.88) * Traps were 8 cm X 9 cm 89 5 0 4 0 a a - 3 0 H 0) a 20H E 10 ui J u l y 15,16 r = 0.513 (p< .0001) — f • • •« • • *• • • • • t | • • I # * • • • • • • I. •* • • h • • •: J |»-» l'« M »• 1 0 0 2 0 0 3 0 0 E R M f e m a l e s o n 2 0 - l e a t s a m p l e 4 0 0 Figure 28a. Correlation between dispersers and density, July 15-16, (discussed in text). 5 0 i 4 0 H J u l y 19,20 r = 0 . 4 8 8 (p< .0001) a ID 30H 0 a 20H 104 • • » M • • * • | "•*••" |t ~ " l i t * ( I f | .. • t • | • • • • • | | • • " I 1 0 0 2 0 0 3 0 0 E R M f e m a l e s o n 2 0 - l e a f s a m p l e 4 0 0 Figure 28b. Correlation between dispersers and density, July 19-20. 5 0 n 4 0 a 3 0 ^ 2 0 9 a , •> c a E A u g u s t 15,16 r = 0 . 5 2 7 ( p < . 0 0 0 1 ) 1 0 0 2 0 0 3 0 0 E R M f e m a l e s o n 2 0 - l e a f s a m p l e 4 0 0 90 Figure 28c. Correlation between dispersers and density, August 15-16, 5 0 4 0 3 0 2 0 10 . A u g u s t 2 1 , 2 2 r =0.512 • ( p < . 0 0 0 1 ) I • • •• - • • !• • • • • • • • • 1 0 0 2 0 0 3 0 0 E R M f e m a l e s o n 2 0 - l e a f s a m p l e 4 0 0 Figure 28d. Correlation between dispersers and density, August 21-22. Table 12. Least-squares regressions of log(catch) on log(density). The hypothesis of increasing dispersal rate with increasing density would predict b > 1 in ln(y) = b ln(x) + a. Date fit t e d model 2 r F ' l . , 254 Sb July 15,16 ln(y) = 1.067 ln(x) - 3.822 0.342 131.9 0.092 July 19,20 ln(y) = 0.701 ln(x) - 1.166 0.267 92.31 0.073 August 15,16 ln(y) = 0.527 ln(x) - 1.470 0.271 94.58 0.054 August 21,22 ln(y) = 0.622 ln(x) - 0.562 0.299 108.35 0.060 These regressions are for adult female European red mites: y = (catch per trap) +1, x = count on 20 leaves, collected on two dates, 10 leaves per date. 92 3.3.2 Between branches The results of the branch fumigation experiment were highly variable, but indicate that adults aire the most active stage, capable of recolonizing a small (about 0.5 m in length) branch within 2 to 4 days. Between 4 and 8 days after colonization, densities of adults on the branches without Tanglefoot were not significantly different from the tree as a whole. Recolonization time of the sticky-based branches varied widely: some branches were recolonized after only 2 days, while other remained relatively free of mites even after 16 days. It may be that proximity to other branches is.the important factor in migration to and from a branch. Branches may touch in the wind, or directional air currents may aid recolonization of branches in some positions but not others. This experiment shows that adult mites are very capable of movement within a tree, by crawling along branches (easily observed with a hand lens), and perhaps by drifting between branches between which there is no direct contact. 3.4 Further inference on movement 3.4.1 Within-tree dispersal and leaf condition Most spider mites are phyllophagous, that i s , they feed on leaf tissue and not on other plant parts. European red mites feed by inserting cheliceral stylets into mesophyll cells and sucking out the c e l l contents. The mouth-parts penetrate to a depth of about 50-100 ny< (Avery and Briggs, 1968), damaging the palisade mesophyll and, to a lesser extent, the spongy meso-phyll. Parenchyma is not damaged. Fluid loss results in a c e l l death, and cells adjacent to damaged cells exhibit aberrant organelle structure (Tanigoshi and Browne, 1981) and reduce their activity or die. Leaf surfaces are characteristically "speckled" with dead and weakened cells at low to 93 moderate levels of damage, and become "chlorotic" (light-coloured due to chlorophyll loss) i f damage i s extreme. The t e l l - t a l e bronzing from the loss of f l u i d and pigments i s recognizable from a distance as an indicator of high mite population density. There i s some evidence that European red mites move from damaged leaves. ASquith et a l . (1980) report the effects of leaf damage caused by rust mites on the ERM. Although developmental time and survival were unaffected by the degree of damage, 60% of young adult females moved from damaged to undamaged leaves. I have found, based on laboratory and f i e l d observations, that young adult females have a tendency to Walk away even from fresh, undamaged leaves. Within-tree dispersal, leaf condition and the reproductive success of mites are interrelated. Mite damage affects leaf condition, and leaf quality may influence mite behavior. Mites may or may not be highly mobile and sensitive to food-quality differences within a tree. They may disperse in response to food, randomly, or in response to the behavior they adopt during dispersive phases. The empirical relationship between mite numbers and chlorophyll content, an index of leaf damage, can provide indirect evidence that allows inference on the hypothetical relationships between dispersal and feeding. Chlorophyll concentration is both an indicator of leaf damage and of leaf quality, since healthy, green, active leaves are generally believed to be the most nutritious (e.g., females on damaged leaves suffer a 90% decrease in fecundity, Asquith et a l . , 1980). Three hypotheses and corresponding predictions of the nature of the empirical relationship between mite numbers and chlorophyll content are detailed below. H^ : European red mites are highly mobile and choose the best leaves, i.e., the leaves which are relatively undamaged and highly palatable. Prediction: a positive correlation between leaf chlorophyll content and the number of mites on the leaves, since mites search out the best feeding sites and remain until food quality degrades. H^ : The mites are relatively incapable of movement between leaves and of habitat selection. They remain on the leaf even after i t has sustained heavy feeding damage. Prediction: a negative correlation between chlorophyll content and mite numbers, since the degree of leaf damage reflects i t s present mite population. H^ : Mites are capable of movement between leaves, but do so in a way which is not determined by food condition. They may move i f the food quality declines, or they may move even i f i t remains high. Prediction: no correlation between chlorophyll and mites. 3.4.2 Materials and methods The sampling design for the collection of leaves is described in Chapter One. After the mites were counted, each of 64 10-leaf samples per date was immediately bagged and frozen. The samples were kept frozen for 2 to 5 months. The mite counts on the 10 leaves per tree were pooled; thus one 10-leaf sample represents one observation on mite density and chlorophyll content. From each leaf in a sample a 1.6 cm diameter disk was cut with a brass cork-borer and used for quantitative determination of chlorophyll a and b by the spectrophotometry method of Bruinsma (1963).. Similar disks were cut from the leaves for dry weight determination. From each sample, the 10-leaf disks were finely chopped in a blender (with sharpened blades) in 95 50 ml of cold (1-4°C) 80% acetone, 20% d i s t i l l e d water. The slurry was suction-filtered over ice, and washed with 20 ml of cold 80% acetone solution. Determinations of total chlorophyll were made in a spectrophotometer by measuring absorbance at 652 m/j against an 80% acetone standard. This peak absorbance wavelength, given by Bruinsma (1963) was verified by a scan of sample f i l t r a t e on a Unicam SP.800 spectrophotometer. As a check on total chlorophyll, absorbances at 663 and 645 mja were measured for calculation of chlorophyll a and b content respectively. Concentrations of chlorophyll were determined using the equations of Bruinsma (1963). These were converted to mg/g dry weight, based on the second sub-sample of leaf disks from each sample, and to mg/cm leaf area. Determinations were made for a total of 192 10-leaf samples, collected on three sampling dates: June 10 § 11, June 24 5 25, and August 5 § 6, 3.4.3 Results: chlorophyll and mite density Delicious had a higher chlorophyll content than Mcintosh on a l l three dates (p < .0001). Chlorophyll content (mg/g dry weight) was higher in June than in August in both cultivars (p < .0001) (Table 13). Chlorophyll per unit area (not shown) gave similar results. Correlations between mite counts and chlorophyll determination were not significant (p > .05) for either cultivar on any of the three sampling dates., There does not seem to be any evidence for the ERM seeking out the least damaged leaves, at least as indicated by chlorophyll content. Nor can leaf damage be attributed to the mites present on the leaves. This lack of correlation supports H^, not or H2; i.e., mites are not consistently influenced by food quality during their movements between leaves. 96 Table 13a. Mean chlorophyll content, mg/g dry weight, of Mcintosh and Delicious leaves. Each mean (and standard error shown in parentheses) is based on 32 samples. Date Mcintosh Delicious June 10 5 11 3.05 (0.095) 3.81 (0.078) June 24 § 25 3.22 (0.094) 3.63 (0.090) August 5 § 6 2.82 (0.065) 3.37 (0.066) Table 13b. Correlation (r) between number of European red mites and chloro-phyll content (mg/g dry weight) of leaves. Each s t a t i s t i c is based on 32 paired Delicious observations. Date Mcintosh Delicious June 10 $ 11 0.130 0.260 June 24 § 25 0.326 -0.239 August 5 5 6 -0.234 -0.141 None of the correlations are significant (p > .05), implying no support for hypotheses 1 and 2 discussed i n the.text. 97 CHAPTER FOUR System Simulation 4.1 Introduction It i s often d i f f i c u l t to predict the consequences of ecological processes such as dispersal, predation and weather. The roles of various factors and processes in population systems can be examined through computer simulation. Simulation has two main advantages in ecological analysis. First, i t provides a dynamic context in which to examine hypotheses regarding components of the perceived system. Our hypotheses and ideas often do not produce the results we might expect, and simulation can act as a sieve that removes errors in logic from a l i s t of conjectures and beliefs. I n i t i a l selection and modification of hypotheses can be done much more quickly by simulation than by experimentation, leaving fewer possible relationships to be subjected to more c r i t i c a l empirical tests. Secondly, simulation not only reveals which parts of our mental models do not f i t ; i t also can indicate neglected relationships which should be more closely examined and perhaps expanded. Indeed, the most revealing lesson from a simulation model may emerge from the reasons for i t s inadequacy (Gilbert et a l . , 1976). Reexamination of the system of interest often leads to f r u i t f u l changes in the direction of research. In simulating a complex dynamic system such as an arthropod population, we are forced to formalize the relationships we perceive (or believe) to exist. Thus the model acts as a deductive machine which, given the premises we supply, deduces the con-sequences. In such a model, the usual weakness of deduction—the fact that the conclusion is true only i f the premises are true—becomes an asset : the consequences of our assumptions and hypotheses can be quickly ascertained. This aspect of modeling is discussed by Watt (1961). The general features of the application of systems analysis and simulation to insect populations are reviewed by Berryman and Pienaar (1974), Getz and Gutierrez.(1982), Gilbert et a l . (1976), Holling (1978), Ruesink (1976) and Stark (1973). 4.2 Mite population systems Phytoseiid-tetranychid predator-prey systems have received a great deal of attention because of the economic damage attributed to tetranychids and the potential for their control by phytoseiids. Logan (in press) and Welch (1979) have summarized mite predator-prey simulation models. Two models of the interaction of the European red mite and phytoseiid predators have been published. Dover et a l . (1979) simulated the relationship of the ERM to Amblvseius f a l l a c i s in Michigan orchards, with innovative approaches to spatial distribution and i t s effect on predation. Rabbinge (1976) produced a highly-detailed state-variable simulation of the ERM and phytoseiid mites, principally Amblyseius potentillae. He provided a detailed sensitivity analysis and f i e l d validation. Typhlodromus occidentalis has received considerable attention in computi simulation, though not in relation to the ERM. Fransz (1974a, b) simulated the functional response of J \ occidentalis to Tetranvchus urticae prey via a detailed component analysis. His findings and the CSMP model of Rabbinge (1976) have been adapted by Rabbinge and Hoy (1980) to a simulation of these species. Logan (1977) simulated the interaction of_J_. occidentalis with Tetranvchus mcdanieli. via a specialized simulation language (SIMBUG) and a general model for temperature-dependent rate phenomena. Further effects of differential temperature-dependent growth in this system were modelled by Wollkind and Logan (1978). Rabbinge and Sabelis (1980) and Sabelis (1981) published the results of simulations of several species of tetranychids and phytoseiids, and provided detailed models of walking behavior, reproduction 99 and predation by phytoseiids. Population growth of the ERM in the absence of phytoseiids has been modelled by Heme and Lund (1979) . Although they are based in a variety of computer and simulation lan-guages, these models have the same basic form. A l l follow a state-variable structure in which state descriptions of the populations of interest are changed in relation to internal and external variables. Within this frame-work a wide variety of approaches to particular functional relationships is apparent. I constructed a computer simulation model in order to summarize the ecology and interaction of the ERM and Typhlodromus, and to explore the potential effects of prey dispersal in the absence of predator dispersal. The results suggest the existence of interesting interactions, and identify areas for which more empirical information on the form and behavior of components is required. The components of the model are described in the following sections. A complete l i s t i n g of the present version, written in FORTRAN, is provided in Appendix F. 4.3 The European red mite 4.3.1 Life history The ERM hatches from diapausing "winter eggs" in the spring. The larvae begin feeding on young apple leaves and within days become protonymphs, then deutonymphs, and f i n a l l y adult males and females. One generation (egg to egg) takes about three weeks, depending on the weather and time of year. Five to nine generations a year have been reported. Under decreased photoperiod in August and September, winter eggs, which require c h i l l i n g before they w i l l complete development and eclosion, are lai d . This l i f e cycle is modelled by beginning a year with a development vector of winter eggs from the previous year. In a one-year run of the model, this 100 vector i s i n i t i a l i z e d at 1000 winter eggs. The eggs may be of similar developmental status, or may be distributed over 20 levels of completed development (described in next section). 4.3.2 Development As in a l l poikilotherms, mite development is temperature-dependent. This relationship for mites has been shown to be non-linear within the temperature range experienced under f i e l d conditions (Tanigoshi et. a l . , 1975a, b; Logan et jgl,, 1976). In cases in which development rate is a reasonably linear function of temperature, degree-day summations can be used to simulate development (Frazer and Gilbert, 1976). In general, however, a different approach to integrating developmental rate i s required. Since developmental rate i s a function of temperature and temperature is a function of time, we must integrate R (T(t)) dt The integral can be approximated by assuming that temperature is constant for small time intervals, say one hour. This general approach is commonly employed in models of arthropod development (e.g., Ruesink, 1976; Logan, 1977; Herne and Lund, 1979; Regniere, 1982). The rate of development of winter eggs of the ERM was estimated by measuring the number of days required to hatch at three fluctuating temper-atures. A group of 316 winter eggs, collected December 29, 1981, from the UBC Experimental Orchard, was randomized among three Percival growth cabinets programmed to reproduce smooth sine temperature functions with an amplitude of 5°C, period of 24 hours, and mean temperatures of 10, 15 and 20°C. The number hatched was counted once a day and the larvae removed. Hatching time as a function of temperature (Figure 29) was similar to the results of Trottier and Heme (1979), but times required to hatch were considerably longer than those reported by Herbert and McRae (1982). Time in days required to hatch is described by a least-squares f i t to the 316 hatch times: Days = [exp (5.0326 - 0.10225 (Temp)] r 2 = 0.836 F : _ 1 4 = 1598. (p < .0001) The experiment was repeated and the equation above verified by comparison with the hatch times of over 13,000 winter eggs collected on April 13, 1982, at the Summerland Research Station. The same non-linear functional relation-ship was apparent, with adjustment for heat experienced during early April. Trottier and Heme (1979) give physiologic time to 50% hatch as 155 degree-days above 5.6°C. Although this linear estimate gives a reasonable predic-tion (Figure 8a correctly predicts a hatch around the end of April to the f i r s t week of May) the f i t t e d equation as applied in the model gives a better dynamic description of hatching. Developmental functions of summer eggs and active mites from hatch to maturity did not differ significantly from those reported by Heme and Lund (1979), so their exponential equations were used in the model. Field temp-eratures measured in the orchard in 1982 are used in the model to calculate hourly development rate from a sine function fit t e d to the daily maximum and minimum temperatures. Developmental vectors consist of 20 positions, and contain as elements the number of mites that have completed up to i/20 of their development, that i s 0-5%, 5-10%, and so on up to 95-100% of the development for that stage. The daily development, calculated from the integration discussed above, is summed and the developmental vector updated. For example, i f the weather allows 17% of summer egg development to be completed on a particular day (one time-step), then the elements of the 102 15 c Figure 29. Days required for ERM winter eggs to hatch as a function of temperature. Times until f i r s t hatch, 50% hatch and last hatch are shown. Data are from Trottier and Heme (1979); my observations, described in the text, are indicated by stars. 103 summer egg vector are advanced forward 3 positions (i.e., INT(.17/.05)), leaving 3 positions free at the bottom and advancing the top 3 positions into hatcher status. The remainder (2% in this case) is saved and added to the next day's development. Once development is calculated for the next stage (juveniles), they are moved forward the appropriate number of positions, and the hatching eggs are merged into the open positions available at the bottom of the juvenile vector, preserving the age structure of the hatchers. Thus the problems of different rates of growth by different stages and of hatching occurring at different times of the day are accounted for (see Figure 30). A similar result is achieved by a slightly different method in the simulation system of Logan (1977). 4.3.3 Fecundity ERM adults are parthenogenetic (arrhenotokous) but apparently rarely go unmated in nature, since all-male families are never seen. Females lay up to 5-6 eggs per day, but usually 1-3 per day, depending on temperature and female age. Females usually live 18 to 22 days. After a preovipositional period of 2.5 - 3.5 days, they lay 10 to 46 eggs over a period of 12 - 16 days (Cagle, 1946; Herbert, 1981) . In the model, the f i r s t two positions in the adult age vector are pre-ovipositional adults; egg-laying begins in the third position. Data from Rabbinge (1976) were f i t t e d by eye to the fecundity function shown in Figure 31. Data from my laboratory colonies were not as extensive, but indicated that fecundity did not increase much above 25°C, so an asymptotic effect of temperature on fecundity was used in the model. Field temperatures used in the model rarely exceeded 30°C. Winter egg production begins August 1. A small proportion of females are devoted to laying winter eggs in early August, and this proportion increases in a typical logit manner to 50% of the females by mid-September 104 A-EL1 EGGS Figure 30. The method of updating the egg (E), juvenile (J) and adult (A) developmental vectors of the ERM. The method of merging transfers is described in the text and in Appendix F. and nearly 100% by mid-October. Females of later generations lay fewer eggs than females earlier in the year. The sex ratio used in the model was 70% female through the summer. 4.3.4 Mortality A multiple regression equation fitted to data from Rabbinge (1976) was used to give approximate adult mortality rate as a function of age and temperature. Winter eggs, summer eggs and immature stages suffer daily mortality of 1%, 1% and 2% respectively. 4.4 Predation by Typhlodromus Most of the phytoseiids present in the 1982 f i e l d experiments were T.  caudiglans. with some JT. occidentalis and rarer species present. Since their l i f e histories are reasonably similar (Putman and Heme, 1964; Rabbinge, 1976), l i f e history parameters for both species are freely mixed. Data and/or 105 EGGS PER Ott AGE CLASSES Figure 31a. Age- and temperature-dependent oviposition rates of the ERM. (Figure from Rabbinge, 1976) . o 5 10 15 20 Age c l a s s Figure 31b. The ERM fecundity function used in the model. The function is calculated in Appendix F, lines 391-393. The f i r s t three age classes represent pre-ovipositional adult females. 106 parameter estimates on development, predation and fecundity were obtained from Laing (1969), Lee and Davis (1968), Logan (1977), Putman (1962), Tanigoshi e_t a l . (1975a) and Sabelis (1981), and verified where possible by experiments in growth cabinets or plastic mite cages on apple leaves in the orchard. In some cases, assumptions as to the effects of temperature and other factors were made (e.g., search rate is an unknown function of temper-ature, so a function is assumed in order to give a reasonable functional response rather than assume that temperature has no effect). Logan's (1977) developmental function for immature Typhlodromus was incorporated into the model. An exponential function describing aging of the adults was f i t t e d to data from Tanigoshi et a l , (1975a). The Frazer and Gilbert (1976) model of the functional response of predation to prey density, as formalized by Gutierrez and Wang (1977)*, was used in the model. The number eaten during one interval of time is descended by: E = N (1 - exp ( - (bP/N)(l - exp(-aN/b)))) where N = prey available E = total number eaten in one timestep (1 day) P = number of predators a = predator search rate (a f i t t e d constant) b = maximum predator demand per timestep (1 day) Good data on the functional response of Typhlodromus predation to ERM density are lacking. In the laboratory, the largest number of protonymphs eaten in one day by a single adult Typhlodromus was 14 at 25°C. The mean predation 1 When this paper was republished as Gutierrez et a l . (1980b), a typo-graphical error was introduced into the equation. The 1977 version is correct. 107 (n = 8) at this temperature was 8.5 per day. At 10°C, predation rates were highly variable, and adults ate between 0 and 7 per day. Predation rates on other tetranychids reported in relevant literature are near this range (Logan, 1977; Putman and Herne, 1964; Sabelis, 1981). I assume search rate (a) and predator demand (b) to be asymptotic functions of temperature, reaching practical maxima of ~13 and 0.03 near 30°C. Figure 32a illustrates estimated predation at 20°C for a range of predator densities. Figure 32b illustrates the increase in predation rate with temperature. The predation parameters may also be functions of predator age and experience (Eveleigh and Chant, 1981a, b), but these are not included in the model in i t s present form. It is clear that better functional response estimates are needed. However, these preliminary estimates are sufficient for exploring some of the relationships of processes in the system. Typhlodromus caudiglans larvae do not feed, and nymphs and adults rarely eat eggs or adult ERM (Putman and Herne, 1964). Winter eggs are not eaten. The model assumes that the effects of predation by Typhlodromus act on the juvenile stages of the ERM, and that the numerical response of the predator depends on the availability of juvenile prey (ERM larvae, protonymphs and deutonymphs). A two-day running average of the number eaten per predator is used to determine predator fecundity and survival rates. Adult female Typhlodromus are mated in the f a l l and overwinter on the trees. A low fecundity (0.2 eggs per day per adult) is possible in the early season even without prey. _T. caudiglans normally lays up to 1 egg per day (Putman, 1962; Putman and Herne, 1964), depending on the number eaten. Fecundity of 0.5 eggs per day occurs at 5 prey eaten per day, and approaches 1.0 as the number of prey eaten increases. Mortality of adult predators is based on the number of prey eaten, and time required to starve to death depends on the temperature (predators live longer at cooler temperatures). 50 predators T o t a l number eaten Prey a v a i l a b l e Figure 32a. The family of curves produced by the Frazer-Gilbert model (Gutierrez and Wang, 1977) as parameterized at 20°C in the model shown in Appendix F. 109 4.5 Spatial relationships The theoretical aspects of dispersal and i t s effects on populations were f i r s t addressed by Skellam (1951). Huffaker's (1958) experiments stimulated interest in the effects of animal movement and spatial relation-ships on predator-prey systems. Subsequently, a great deal of attention has been given to the problem (reviewed by Hassell, 1978, 1980; Levin, 1976, 1977; Murdoch and Oaten, 1975). Most of this attention has been theoretical in nature. A large number of recent studies concern the effects of predator and prey dispersal along gradients or between patches on the s t a b i l i t y of model populations (e.g., Comins and Blatt, 1974; Crowley, 1981; Gurney and Nisbet, 1978; Hastings, 1977, 1978; Hilborn, 1975, 1979; Hogeweg and Hesper, 1981; Mimura and Murray, 1978; Renshaw, 1982; Roff, 1974; Smith, 1974; Vandermeer, 1973). Most of these models either ignore within-patch dynamics and consider only the rates of prey-patch appearance and disappearance, or are variants of typical Lotka-Volterra models, e.g.: A -dt 1 - r l N l T W l + n 2N 2 " n l N l 4p -o T i - e i N l P l - diPi + m2P2 - m l P l dt 2 - r 2 N 2 - a 2N 2P 2 + n l N l - n 2N 2 _ P = dt 2 = e2 N2 P2 " d2 P2 + V l - m2P2 Here N and P represent prey and predator numbers and r, a, e and d are the usual Lotka-Volterra constants: prey growth rate, an attack constant, an assimilation factor and predator death rate. Dispersal between the coupled systems is determined by the rates n and m. More complicated versions can be constructed by adding prey self-limitation, predator handling time, more subsystems, complex or restricted dispersal, gradients, age structure or more predator and prey species. Although these models are quite a r t i f i c i a l , they usually support one generalization: increasing spatial complexity and/or prey dispersal increases persistence of the system, given a low predator dispersal rate and a heterogeneous environment. Very low relative rates of dispersal of the predator lead to predator extinction in most models Unfortunately, adding spatial components and animal movement to even moderately complex simulations can increase their complexity and running-time enormously (as in the case of the spruce-budworm model: Clark, 1979; Holling, 1978). Other ways of adding dispersal are needed. Logan (in press) emphasizes this need: "Development of methods to effic i e n t l y model the interactive dynamics of age-structured populations dispersing through a heterogeneous environment may result in a third generation of prey-predator models. Such models...will provide insights which are at least as great as the previous two generations of models..., the analytic models of population ecology and computer simulation models." Rather than model a single isolated tree as a representative "average" of system behavior, and thereby exclude dispersal, or ignore the admonitions of Clark (1979) and plunge into the large-scale computation and complexity needed to model an entire orchard, I chose to compromise with a simplified approach to dispersal. A single tree is modelled within which dispersal is unrestricted and homogeneity is assumed. A switch is supplied which wi l l allow random dispersal of female red mites to and from the tree. Under conditions of no dispersal, the ERM population on the tree is determined by the i n i t i a l numbers of winter eggs in the spring, the daily temperatures supplied over the year, and the action of the predator, i f present. When dispersal occurs (i.e., when the dispersal switch is "on"), each day n migrants (-16 n ^ 1) are added to m randomly chosen adult ERM age classes (1 to 20). The number of migration events, m, can be varied. I originally used 10-20 such transfers per day, but found that even low numbers of dis-persers had striking results on the interaction of Typhlodromus and the ERM. I l l The effects of within-tree dispersion (e.g., clumped or random distributions) are not considered in the model. 4.6 Temporal considerations The time-step is one day for a l l processes other than development, for which hourly rates are summed. Variable time-steps are not useful in this case; there are no long periods of l i t t l e change, since generations are fre-quent and part i a l l y overlapping. One run of the model is one growing season, April 1 to October 31. Field temperatures for 1982 are used in the model. Temperatures from other years could be used i f adjustments were made for differences in orchard and weather station temperatures. Since winter eggs are laid in the f a l l of each run, runs can easily be strung together to simulate successive years, though the model has not been put to this purpose. The model has not been "tuned" to reproduce f i e l d observations, since i t s purpose is to examine gross behavior of the system while maintaining realis-t i c age-structure and development. It i s a vehicle for preliminary examin-ation of the ecology of orchard mites, and in i t s present form is not meant for use in management, although with better parameter estimates, f i e l d validation and adjustment i t could serve this purpose. 4.7 Behavior of the model 4.7.1 The ERM sub-model The ERM submodel has the i n i t i a l values of the state variables set to simulate the mite population on one branch, for example i t begins April 1 with 1000 winter eggs. The scale is not so important as the general behavior, but the following graphical output can be considered to represent the mite population on one moderate-size branch, or 50-100 leaves. Given a vector of maximum and minimum orchard temperatures, the unadjusted ERM sub-model behaves very much like unrestricted populations might (Figure 33). The oscillations are the result of the multivoltine l i f e history of the ERM (the model contains no stochastic terms or predation at this point). Winter eggs begin hatching after mid-April and continue to do so until the end of May, with the 50% hatch occurring around the second week of May. Juveniles appear by March 1, and result in a f i r s t generation of adults peaking near the end of May. This result is in good agreement with the f i e l d data from 1982 (Figs. 14-19). Since the only connection between the model and the f i e l d observations is the vector of daily temperatures, i t can be concluded that the simulation of ERM development is successful. With minor modification and verification, this part of the model could be used to predict early season numbers, and perhaps age-structure as well. The model output exhibits 6 complete generations, with a weaker 7th in October. Adult peaks lag juvenile maxima by 1-2 weeks. There are 6 genera-tions of summer eggs; the 7th generation of females, and a proportion of the 6th, are devoted to production of winter eggs. Winter eggs are produced from early August to October, and by the end of the summer more than enough have been laid to produce a similar population the following year. Parameters determining winter egg production and mortality would need refinement i f this model was to be used to simulate several successive years. The late-season decline does not begin until September in the model out-put. Actual ERM populations exhibit a decline after July or August (Figs. 14 19; see also Downing and Moilliet, 1967, 1972; Hoyt, 1969). Moderate decline in female productivity over the summer is included in the model. The discrep ancy may be due to any of several factors: 1) fecundity may be reduced after mid-summer much more than expected; 2) other predators, such as Zetzellia  mali and Campyloma verbasei. may cause significant late-season declines; and 3) emigration from the trees may be extensive in late summer (as shown in 113 a) summer eggs (- ) winter eggs ( ) . _ , 1 ^ , , , APR MAY JUN JUL AUG SEP t>) juveniles (— — ) adults ( ) Figure 33. ERM population growth and age structure predicted by the model in Appendix F. No predation or dispersal are included in this run. 114 Chapter Three). The last hypothesis has been invoked to explain a similar decline in Panonychus c i t r i by.Nishino (1976) (cited in Takafuji, 1980). 4.7.2 Sensitivity to temperature In order to explore the basic ERM submodel's sensitivity to temperature, three runs at temperatures above and three runs at temperatures below the standard (1982) f i e l d temperatures were made. In the six runs, -3, -2, -1, 1, 2, or 3 was added to the daily maximum and minimum temperatures. The results of runs with "T + 3", "T + 1", "T + 0", "T - 1", and "T - 3" are shown in Figure 34. The "T + 3" run produced an unrealistically large ERM population by midsummer with no apparent separation of generations after June. The model in i t s present form contains no density-dependent self regulation of the ERM, and thus population density rises to extreme values under such warm conditions. Since a 3°C increase across the year is unnatur-all y high, and since the model has been constructed with the British Columbia climate in mind, this high temperature treatment is omitted in the following analyses and discussions. 4.7.3 Effects of dispersal The importance of dispersal on stabilizing predator-prey systems is well-established (Hilborn, 1975). I include i t here to allow examination of it s effect on predation at the meso-scale of a single patch; that i s , one tree with immigration and emigration. When predation is added to the ERM submodel, the Typhlodromus population starts at typically moderate to high levels (adult females overwinter on the tree and emerge from cracks and scars on the bark in early spring), but dies out by the f i r s t of June. This trend occurs over a wide range of justifiable predation, fecundity and mortality parameter values. At f i r s t this result seems unrealistic, but i t is in keeping with the early-season decline apparent 115 Figure 34. The se n s i t i v i t y of the ERM submodel to temperature. The "T + 3" run is unreasonably high (discussed in the text). This run does not include predation. , 1 r—1—^-i 1 1 APR MAY JUN JUL AUG SEP Figure 35. The consequences of including dispersal (at f i e l d temperatures). This run includes predation. The predators survive and provide mid- and late-season control of the ERM when dispersal is added. 116 in the 1982 f i e l d data (Figs. 12 and 13). It would be possible to "force" the predator population to survive through this period, but a better solution of this modelling d i f f i c u l t y can be achieved by incorporating prey dispersal. When the dispersal switch is "on", adults migrate to and from the modelled system after May 20. Figure 35 illustrates the resulting effect of dispersal (-1 ^  n s £ 1; m = 10) on the ERM adult female populations: the predator sur-vives into July, and significantly reduces prey populations over this longer period. The a b i l i t y of the predator to suppress the prey during this period is quite sensitive to numerical and functional responses. In standard runs, maximum predator fecundity does not exceed 1.0 eggs per female per day, and maximum realized predation is less than 10 prey juveniles per adult day. Refinement of the functions and adjustments of the parameters (based on experimental estimates) would be required before this model could be a reliable management tool. Incorporating the dispersal (immigration and emigration) of adult ERM females in the model smooths the otherwise violent fluctuations in juvenile prey early in the season (Figure 35) and thus reduces predator starvation and/or losses in potential fecundity. Since a l l these processes are temperature-dependent, dispersal probably would have different effects under different temperature regimes. 4.7.4 Interaction of dispersal and temperature At extremely low rates of adult prey dispersal (-1 t£ n ^  1; m = 3, i.e., an average of 0, and up to 3, adult prey invade the branch each day), predator populations cannot quite survive through the c r i t i c a l early .period. The out-come, however, is temperature-dependent. Figure 36 shows the results of 6 runs with low dispersal at "T + 2", "T + 1", "T + 0", "T - 1", "T - 2" and "T - 3" °C. In two cases, "T - 1" and "T - 3" °C, the predator population persisted and significantly reduced the ERM. At "T - 1" (Figure 36d) Typhlodromus enjoyed a good supply of prey through May and June, due to slower ERM development, and thus was able to r a l l y sufficiently in July and August to reduce the ERM population. At "T - 3", fewer prey were available in May than at "T - 1", but survival of the starving predators was sufficiently improved by the cool conditions for them to persist long enough to attain low to moderate levels by late summer. The "T - 2" run had neither the benefit of increased prey densities in May nor especially cool temperatures, so the predator died out. However, the prey was limited by the cooler temperatures and did not attain the densities exhibited under higher temperatures. At higher temperatures, the phytoseiids failed (Figs. 36a, b and c) and the prey increased unimpeded under favorable growing conditions. Such breakdowns of the biological control of the ERM by Typhlodromus are well-known. + 0 « n HAY JUN JUL AUG V P 0C1 + 0 a » 1 A M UAT J U N AUG SEP OCT Figure 36a, b, c. The effect of temperature on persistence of the predator when low dispersal is allowed (explained in the text). 119 Figure 36d, e, f. The effect of temperature on persistence of the predator when low dispersal is allowed (explained in the text). 120 SUMMARY Say what the use, were finer optics giv'n, T' inspect a mite, not comprehend the heav'n? Pope, 1733 Essay on Man Typhlodromus caudiglans is capable of suppressing population growth of the European red mite even at relatively low predator densities. The phytoseiids reduce the density of the ERM by predation on the immature stages, and cause a general decrease in prey aggregation. Variation among trees tends to be high for both the predator and prey populations, thus emphasizing the need for careful sampling methods and well-designed experiments. In addition, microclimate gradients within the orchard, position on the tree and cultivar characteristics have significant effects on the ERM. The phytoseiids are capable of controlling the ERM, but the predator and prey phenologies are not well-synchronized. The adult phytoseiids emerge from diapause before an adequate or stable supply of ERM immatures is avail-able, and may suffer extensive early-season starvation. Orchard mite pred-ator-prey systems tend to be very sensitive to temperature. It is well-known that warm weather favors ERM build-up. Given the inherent patchiness and discontinuity of orchards and the importance of predator success in finding prey, dispersal can have important effects on the outcome of the predator-prey interaction. For example, in the absence of alternate prey, as was the case in these experiments, easy dispersal of the predator through-out the orchard might f a c i l i t a t e prey-finding and thus enable the predator to survive and significantly reduce ERM populations later in the summer. It is apparent from these experiments that Typhlodromus caudiglans. and probably other Typhlodromus species present in the Okanagan Valley of British Columbia, are very poor dispersers in comparison with their prey. They rarely move among the trees or between the trees and the ground cover, by a i r or via the trunks. Even after several months, they were incapable of repopulating isolated carbaryl-sprayed trees interspersed among unsprayed, well-populated trees. Neither banded nor unhanded trees regained phytoseiids, even though prey populations were high on the sprayed trees. In contrast, their prey, the European red mite, actively disperses within trees and throughout the orchard. Young adults move more than other stages, indicating a high dispersive tendency during this stage. Field experiments showed that adult females were greatly over-represented in the ERM emigrants from apple trees. No density threshold effect on dispersal was discernible on a per-tree basis. Aerial dispersal may be extensive over much of the summer, depending on weather conditions. Such dispersal did not seem to be accidental: dispersal rates were considerably lower on windy days. The theoretical consequence of high prey dispersal and extremely low predator dispersal in a discontinuous environment is increased system per-sistence due to the ab i l i t y of prey to disperse to and colonize patches, and thus avoid massive prey decline and subsequent predator extinction. This result is complicated in natural systems (as opposed to theoretical models) by effects and interactions of population growth rates, dispersal rate and timing, predator starvation and weather. Field experiments showed that prey dispersal is weather-dependent. Under warm and calm weather, high prey dispersal and growth rates would tend to reduce differences in prey densities among trees within an orchard, and perhaps between adjacent orchards. Warm and windy conditions, however, could result in local predator extinction and subsequent ERM build-up. Cooler weather could prevent predator extinction 122 by allowing predators to live u n t i l weather conditions more favorable to prey dispersal and population growth increased the food supply. Thus "hotspots" in an orchard may not be directly due to differences in micro-climate or host plant condition, but may represent areas of local predator extinction, caused by spray practices or starvation due to low prey avail-a b i l i t y during a crucial period. These data suggest that such areas would be slow to regain their Typhlodromus populations i f trees were separated. In such cases addition of predators to augment the populations present in an orchard could prove, very successful, especially i f the additions were made after the period of high starvation risk had passed. Although this study was not concerned with laboratory-reared Typhlodromus, i t is likely that phytoseiids in this condition would be more prone to dispersal between trees, but would soon assume the low dispersive tendencies apparent in the phytoseiid dispersal experiment described in Chapter One. In simulations of the interaction of Typhlodromus with the ERM, the interaction of dispersal and temperature-related phenomena is strong and discontinuous, and may operate through several processes. These processes and their interaction do not merely alter scale or phenology, but change the fundamental behavior of the system. Dispersal of prey allowed the predator to persist and provide control later in the season. Low tempera-tures decreased the amount of prey dispersal needed to allow predator persistence, but intermediate temperatures did not. The model suggests that prey availability, and perhaps age distribution, in May and early June is the important component. Since there is l i t t l e or no predator dispersal, dispersal of the prey between trees becomes crucial. The powerful interaction of dispersal, temperature, and timing emphasizes the need for further investigation of these effects so they can be included in formal and informal descriptions of this predator-prey system. Such descriptions should lead to two further, somewhat different improvements a better understanding of the predation process in the f i e l d and i t s related numerical response (including development, fecundity and starvation), and a more detailed phenology of prey dispersal, including rates of movement, distances travelled, establishment rates, and the relationships to weather, I believe that similar improvements are required to further our understanding of most kinds of arthropod predator-prey systems, not just those of the orchard mites. Therefore in this endeavor I hope that I have responded to Pope's appeal, and begun to comprehend that larger ecological heaven. 124 LITERATURE CITED Anderson, N.H. and C.V.G. Morgan. 1958. The role of Typhlodromus spp. in British Columbia apple orchards. Proceedings of the Tenth Inter-national Congress of Entomology, 1956, 4: 659-665. Asquith, D., B.A. Croft, S.C. Hoyt, E . H. Glass and R.E. Rice. 1980. The systems approach and general accomplishments toward better insect control in pome and stone f r u i t s . In: New Technology of Pest Control. C.B. Huffaker (ed.), John Wiley, N.Y., pp.249-317. Avery, D.J. and J.B. Briggs. 1968. Damage to leaves caused by fr u i t tree red spider mite, Panonychus ulmi (Koch). Journal of Horticultural Science 43: 463-473. Baskerville, G.L. and P. Emin. 1969. Rapid estimation of heat accumulation from maximum and minimum temperatures. Ecology 50: 514-517. Berryman, A.A. and L.V. Pienaar. 1974. Simulation: a powerful method of investigating the dynamics and management of insect populations. Environmental Entomology 3: 199-207. Bliss, C.I. and R.A. Fisher. 1953. Fitting the negative binomial d i s t r i -bution to biological data. Biometrics 9: 176-200. Bradner, L. 1977. Resistance in mites and insects affecting orchard crops. In: Pesticide Management and Insecticide Resistance. D.L. Watson and A.W.A. Brown (eds.), Academic Press, N.Y., pp.353-376. Br i c k h i l l , CD. 1958. Biological studies on two species of tydeid mites from California. Hilgardia 27: 601-620. British Columbia Ministry of Agriculture. 1980. Tree f r u i t production guide for interior d i s t r i c t s . Province of British Columbia. Ministry of Agriculture. 63p. Bruinsma, J. 1963. The quantitative analysis of chlorophylls a and b in plant extracts. Photochemistry and Photobiology 2: 241-249. Brunner, J.F., S.C. Hoyt and M.A. Wright. 1982. Codling moth control — a new tool for timing sprays. Washington State University, Cooperative Extension Bulletin 1072. 4p. Cagle, L.R. 1946. Life history of the European red mite. Virgiha Agri-cultural Experiment Station, Technical Bulletin 98. 19p. Chant, D.A. 1958. On the ecology of typhlodromid mites in Southeastern England. Proceedings of the Tenth International Congress of Entomology, 1956, 4: 649-658. 125 Chant, D.A. 1959. Phytoseiid mites. Part I. Bionomics of seven species in southeastern England. Part II. A taxonomic review of the Phyto-seiidae, with description of 38 new species. Canadian Entomologist (Supplement 12). 166p. Chant, D.A. 1961. An experiment in biological control of Tetranvchus  telarius (L.) (Acarina: Tetranychidae) in a greenhouse using the predacious mite Phytoseiulus persimilis Athias-Henriot (Phytoseiidae). Canadian Entomologist 93: 437-443. Chant, D.A. 1963. Some mortality factors and the dynamics of orchard mites. Memoirs of the Entomological Society of Canada 32: 33-40. Chant, D.A. 1966. Integrated control systems. In: Scientific Aspects of Pest Control. Washington, D.C., Natural Science Publication 1402, p.193-218. Clark, W.C. 1979, Spatial structure and population dynamics in an insect ecosystem. Ph.D. dissertation, University of British Columbia. 505p. Comins, H.N. and D.W.E. Blatt. 1974. Prey-predator models in spatially heterogeneous environments. Journal of Theoretical Biology 48: 75-83. Cone, W.W. 1963. Insecticides as a factor in population fluctuations of mites on a l f a l f a . Washington Agricultural Experiment Station Bulletin 41. 55p. Croft, B.A. 1972. Prey stage distribution, a factor affecting the numerical response of Typhlodromus occidentalis to Tetranychus mcdanieli and Tetranychus pacificus. Great Basin Naturalist 32: 61-75. Croft, B.A. 1978. Potentials for research and implementation of integrated pest management on deciduous tree-fruits. In: Pest Control Strategies. E.H. Smith and D. Pimentel (eds.), Academic Press, N.Y., p.101-115. Croft, B.A. and A.W.A. Brown. 1975. Responses of arthropod natural enemies to insecticides. Annual Review of Entomology 20: 285-335. Croft, B.A. and S.A. Hoying. 1977. Competitive displacement of Panonychus  ulmi (Acarina: Tetranychidae) by Aculus schlechtendali (Acarina: Erio-phyidae) in apple orchards. Canadian Entomologist 109: 1025-1034. Croft, B.A. and L.R. Jeppson. 1970. Comparative studies on four strains of Typhlodromus occidentalis. II. Laboratory toxicity of ten compounds common to apple pest control. Journal of Economic Entomology 63: 1528-1531. Croft, B.A. and D.L. McGroarty. 1973. A model study of acaricide resistance, spider mite outbreaks, and biological control patterns in Michigan apple orchards. Environmental Entomology 3: 633-638. Croft, B.A. and P.G. Stewart. 1973. Toxicity of one carbamate and six organ-ophosphorus insecticides to 0-P resistant strains of Typhlodromus  occidentalis and Amblyseius f a l l a c i s . Environmental Entomology 2: 486-488. 126 Croft, B.A., S.M. Welch and M.J. Dover. 1976. Dispersion statistics and sample size estimates for populations of the mite species Panonychus  ulmi and Amblyseius f a l l a c i s on apple. Environmental Entomology 5: 227-234. Crowley, P.H. 1981. Dispersal and the s t a b i l i t y of predator-prey inter-actions. American Naturalist 118: 673-701. Dover, M.J., B.A. Croft, S.M. Welch and R.L. Tummala. 1979. Biological control of Panonychus ulmi (Acarina: Tetranychidae) by Amblyseius f a l l a c i s (Acarina: Phytoseiidae) on apple: A prey-predator model. Environmental Entomology 8: 282-292. Downing, R.S. 1954. Strains of the European red mite, Metatetranychus ulmi (Koch), resistant to parathion and malathion in British Columbia. Proceedings of the Entomological Society of British Columbia 51: 10-11. Downing, R.S. 1961. Experiments in British Columbia with acricid, a new dinitro miticide. Proceedings of the Entomological Society of British Columbia 58: 22-25. Downing, R.S. and J.C. Arrand. 1978. Integrated control of orchard mites on apple orchards in British Columbia. British Columbia. Ministry of Agriculture Publication 78-1. 8p. Downing, R.S. and T.K. Moilliet. 1967. Relative densities of predacious and phytophagous mites on three varieties of apple tree. Canadian Entomologist 99: 738-741. Downing, R.S. and T.K. Moilliet. 1971. Occurrence of phytoseiid mites (Acarina: Phytoseiidae) in apple orchards in south central British Columbia. Journal of the Entomological Society of British Columbia 68: 33-36. Downing, R.S. and T.K. Moilliet. 1972. Replacement of Typhlodromus occidentalis by X- caudiglans and X- pvri (Acarina: Phytoseiidae) after cessation of sprays on apple trees. Canadian Entomologist 104: 937-940. Eveleigh, E.S. and D.A. Chant. 1981a. Experimental studies on acarine predator-prey interactions: effects of predator age and feeding history on prey consumption and the functional response (Acarina: Phytoseiidae). Canadian Journal of Zoology 59: 1387-1406. Eveleigh, E.S. and D.A. Chant. 1981b.. Experimental studies on acarine predator-prey interactions: the numerical response of immature and adult predators (Acarina: Phytoseiidae). Canadian Journal of Zoology 59: 1407-Fisher, R.A. and F. Yates. 1953. Statistical tables for biological. agri-cultural and medical research. Oliver and Boyd, Edinburgh, 126p. Flaherty, D.L. 1967. The ecology and importance of spider mites on grape-vine in the southern San Joaquin Valley, with emphasis on the role of Metaseiulus occidentalis (Nesbitt). Ph.D. dissertation, University of California, Berkeley. 1418. 127 Flaherty, D.L. 1969. Ecosystem trophic complexity and densities of the Willamette mite, Eotetranvchus willamettei Ewing (Acarina: Tetranychidae). Ecology 50: 911-916. Flaherty, D.L. and C.B. Huffaker. 1970a. Biological control of Pacific mites and Willamette mites in San Joaquin Valley vineyards. I. Role of Metaseiulus occidentalis. Hilgardia 40: 267-308. Flaherty, D.L. and C.B. Huffaker. 1970b. Biological control of Pacific mites and Willamette mites in San Joaquin Valley vineyards. II. Influence of dispersion patterns of Metaseiulus occidentalis. Hilgardia 40: 309-330. Fransz, H.G. 1974a. The functional response to prey density in an acarine  system. PUDOC, Wageningen, the Netherlands, 143p. Fransz, H.G. 1974b. Simulation of predator-prey relationships in an acarine system. Pesticide Science 5: 305-317. Frazer, B.D. and N. Gilbert. 1976. Coccinellids and aphids: a quantitative study of the impact of adult ladybirds (Coleoptera: Coccinellidae) preying on f i e l d populations of pea aphids (Homoptera: Aphididae). Journal of the Entomological Society of British Columbia 73: 33-56. Gadsby, M.C. 1982. 'Dispersal in Phvtoseiulus persimilis Athias-Henriot (Acarina: Phytoseiidae) and i t s importance to biological control programs. M.Sc. thesis, University pf British Columbia. Gerson, U. and M. van de Vrie. 1979. The potential of mites in the biological control of mite and insect pests. Proceedings of the Fourth International Congress of Acarology, 1974, p.629-635. Getz, W.M. and A.P. Gutierrez. 1982. A perspective on systems analysis in crop production and insect pest management. Annual Review of Entomology 27: 447-466. Gilbert, N., A.P. Gutierrez, B.D. Frazer and R.E. Jones. 1976. Ecological  relationships. W.H. Freeman, San Francisco. Groves, J.R. 1951. A synopsis of the world literature on the fr u i t tree red spider mite Metatetranychus ulmi (C.L. Koch, 1835), and i t s predators. Commonwealth Institute of Entomology, London, 180p. Gurney, W.S.C. and R.M. Nisbet. 1978. Predator-prey fluctuations in patchy environments. Journal of Animal Ecology 47: 85-102. Gutierrez, A.P. and Y. Wang.. 1977. Applied population ecology: models for crop production and pest management. In_: Proceedings of a Conference On Pest- Management. G.A. Norton and CS. Holling (eds.), International Institute for Applied Systems Analysis, Laxenburg, Austria, CP-77-6, p.255-280. Gutierrez, A.P., C.G. Summers, J. Baumgaertner. 1980a. The phenology and distribution of aphids in California a l f a l f a as modified by ladybird beetle predation (Coleoptera: Coccinellidae). Canadian Entomologist 112: 489-495. 128 Gutierrez, A.P., D.W. DeMichele, Y. Wang, G.L. Curry, R. Skeith and L.G. Brown. 1980b. The systems approach to research and decision making for cotton pest control. In: New Technology of Pest Control, C.B. Huffaker (ed.), John Wiley § Sons, N.Y., p.155-186. Hall, F.R. 1980. Organotins and mite control on apples in Ohio. In: Fruit Crops — 1980: A Summary of Research. Ohio Agriculture and Research Development Center, Research Circular 259, p.25-27. Harries, F.H. 1966. Reproduction and mortality of the two-spotted spider mite on f r u i t seedlings treated with chemicals. Journal of Economic Entomology 59: 501-506. Hassell, M.P. 1978. The dynamics of arthropod predator-prey systems. Princeton University Press, Princeton, New Jersey, 237p. Hassell, M.P. 1980. Some consequences of habitat heterogeneity for popu-lation dynamics. Oikos 35: 150-160. Hastings, A. 1977. Spatial heterogeneity and the s t a b i l i t y of predator-prey systems. Theoretical Population Biology 12: 37-48. Hastings, A. 1978. Spatial heterogeneity and the s t a b i l i t y of predator-prey systems: predator-mediated coexistence. Theoretical Population Biology 14: 380-395. Helle, W. 1965. Resistance in Acarina: mites. Advances in Acarology 2: 71-93. Herbert, H.J. 1970." Limits of each stage in populations of the European red mite, Panonychus ulmi. Canadian Entomologist 102: 64-68. Herbert, H.J. 1981. Biology, l i f e tables and intrinsic rate of increase of the European red mite, Panonychus ulmi (Acarina: Tetranychidae). Canadian Entomologist 113: 65-71. Herbert, H.J. and K.B. McRae. 1982. Predicting eclosion of overwintering eggs of the European red mite, Panonychus ulmi (Acarina: Tetranychidae), in Nova Scotia. Canadian Entomologist 114: 703-712. Herne, D.C. and C.T. Lund. 1979. Simulation model of European red mite population dynamics developed for a mini-computer. Canadian Entomologist 111: 499-507. Herne, D.C. and W.L. Putman. 1966. Toxicity of some pesticides to predacious arthropods in Ontario peach orchards. Canadian Entomologist 98: 936-942. Herne, D.H.C., J.E. Cranham and M.A. Easterbrook. 1979. New acaricides to control resistant mites. Recent Advances in Acarology 1: 95-104. Hilborn, R. 1975. The effect of spatial heterogeneity on the persistence of predator-prey interactions. Theoretical Population Biology 8: 346-355. 129 Hilborn, R. 1979. Some long term dynamics of predator-prey models with diffusion. Ecological Modelling 6: 23-30. Hogeweg, P. and B. Hesper. 1981. Two predators and one prey in a patchy environment: an application of MICMAC modelling. Journal of Theoretical Biology 93: 411-432. Holling, CS. 1978. Adaptive environmental assessment and management. John Wiley $ Sons, N.Y., 377p. Horn, H.S. and R.H. MacArthur. 1972. Competition among fugitive species in a harlequin environment. Ecology 53: 749-752. Howard, L.O. 1930. A history of applied entomology (somewhat anecdotal). Smithsonian Miscellaneous Collections 84. 564p. Hoy, B. 1942. The advance of the codling moth in British Columbia. Proceedings of the Entomological Society of British Columbia 39: 16-19. Hoy, M.A. and J.M. Smilanick. 1981. Non-random prey location by the phyto-seiid predator Metaseiulus occidentalis: differential responses to several spider mite species. Entomologia Experimentalis et Applicata 29: 241-253. Hoyt, S.C 1969. Integrated chemical control of insects and biological control of mites on apple in Washington. Journal of Economic Entomology 62: 74-86. Hoyt, S.C. and L.E. Caltagirone. 1971. The developing programs of integrated control of pests of apples in Washington and peaches in California. In: Biological Control. C.B. Huffaker (ed.), Plenum Press, N.Y., p.395-421. Hudson, W.B. and B.P. Beirne. 1970. Effects of sprinkler irrigation on McDaniel and European red mites in apple orchards. Journal of the Entomological Society of British Columbia 67: 8-13. Hueck, H.J., D.J. Kuenen, P.J. Van den Boer, E. Jaeger-Draafsel. 1952. The increase of egg production of the frui t tree red spider mite under influence of DDT. Physiologia comp. Oecol. 2: 371-377. Huffaker, C.B. 1958. Experimental studies on predation: dispersion factors and predator-prey oscillations. Hilgardia 27: 343-383. Huffaker, C.B., K.P. Shea and S.G. Herman. 1963. Experimental studies on predation: complex dispersion and levels of food in an acarine predator-prey interaction. Hilgardia 34: 305-330. Huffaker, C.B., M. van.de Vrie and J.A. McMurtry. 1969. The ecology of tetranychid mites and their natural control. Annual Review of Entomology 14: 125-174. Huffaker, C.B., M. van de Vrie and J.A. McMurtry. 1970. Ecology of tetra-nychid mites and their natural enemies: a review. II. Tetranychid popu-lations and their possible control by predators: an evaluation. Hilgardia 40: 391-458. 130 Hussey, N.W. and C.B. Huffaker. 1976. Spider mites. In: Studies in Biological Control. V.L. Delucchi (ed.), International Biological Programme 9, Cambridge University Press, Cambridge, p.179-286. Hussey, N.W. and W.J. Parr. 1963. Dispersal of the glasshouse red spider mite Tetranvchus urticae Koch (Acarina, Tetranychidae). Entomologia Experimentalis et Applicata 6: 207-214. Huxley, T.H. 1870. Biogenesis and abiogenesis. Reprinted in: Collected  Essays of T.H. Huxley. Volume VIII, Discourse: biological and geological. 1902. Macmillan and Co., London. Ito, Y., K. Miyashita and H. Yamada. 1968. Biology of Hyphantria cunea Drury in Japan. V. Effect of temperature on development of immature stages. Applied Entomology and Zoology 3: 163-175. Johnson, C.G. 1969. Migration and dispersal of insects by f l i g h t . Methuen, London. 763p. Johnson, D.L. and A.J. Petkau. 1983. Inference for variance-to-mean ratios with applications in ecology. Institute of Applied Mathematics and Statistics, University of British Columbia, Technical Report No.83-1. 31p. Johnson, D.T. and B.A. Croft. 1976. Laboratory study of the dispersal behavior of Amblyseius f a l l a c i s (Acarina: Phytoseiidae). Annals of the Entomological Society of America 69: 1019-1023. Johnson, D.T. and B.A. Croft. 1979. Factors affecting the dispersal of Amblyseius fa l l a c i s in an apple tree ecosystem. Recent Advances in Acarology 1: 477-483. Johnson, D.T. and B.A, Croft. 1981. Dispersal of Amblyseius f a l l a c i s (Acarina: Phytoseiidae) in an apple ecosystem. Environmental Entomology 10: 313-319. Laing, J.E. 1969. Life history and l i f e table of Metaseiulus occidentalis. Annals of the Entomological Society of America 62: 978-982. Lee, M.S. and D.W. Davis. 1968. Life history and behaviour of the predatory mite Typhlodromus occidentalis in Utah. Annals of the Entomological Society of America 61: 251-255. Levin, S.A. 1976. Spatial patterning and the structure of ecological communities. In.: Some Mathematical Questions in Biology. S.H. Levine (ed.), Volume VII, American Mathematical Society,' Providence, "Rhode Island. Levin, S.A. 1977 Population dynamic models in heterogeneous environments. Annual Review of Ecology and Systematics 7: 287-310. Logan, J.A. 1977. Population model of the association of Tetranychus mcdanieli (Acarina: Tetranychidae) with Metaseiulus occidentalis (Acarina: Phytoseiidae) in the apple ecosystem. Ph.D. dissertation, Washington State University, 137p. 131 Logan, J.A. (in press). Recent advances and new directions in phytoseiid population models. In: Recent Advances in Knowledge of the Phytoseiidae, M.A. Hoy (ed.), University of California Press. Logan, J.A., D.J. Wollkind, S.C. Hoyt and L.K. Tanigoshi. 1976. An analytic model for description of temperature dependent rate phenomena in arthropods. Environmental Entomology 5: 1133-1140. Lord, F.T. 1949. The influence of spray programs on the fauna of apple orchards in Nova Scotia. III. Mites and their predators. Canadian Entomologist 81: 202-214, 217-230. Lord, F.T., H.J. Herbert and A.W. MacPhee. 1958. The natural control of phytophagous mites on apple trees in Nova Scotia. Proceedings of the Tenth International Congress of Entomology, 1956, 4: 617-622. Lund, C.T. and D.C. Heme. 1980. Relationships between degree-days and computer simulated events within a population of the European red mite, Panonychus ulmi (Acarina: Tetranychidae). Proceedings of the Entomo-logical Society of Ontario 111: 33-37. Madsen, H.F. and J.C. Arrand. 1975. The recognition and life-history of the major orchard insects and mites in British Columbia. British Columbia. Department of Agriculture. Publication. 32p. Marie, G. 1951. Observations on the dispersal of the f r u i t tree red spider mite, Metatetranychus ulmi (Koch). Report of the East Mailing Research Station for 1950, p.155-159. Marshall, J. 1951. Applied entomology in the orchards of British Columbia, 1900-1951. Proceedings of the Entomological Society of British Columbia 48: 25-31. Marshall, J. 1952. A decade of pest control in British Columbia orchards. Proceedings of the Entomological Society of British Columbia 49: 7-11. Marshall, J. 1958. DDT-tolerant strain of codling moth. British Columbia Fruit Grower's Association Quarterly Report 3: 13-14. Marshall, J. 1963. Background for integrated spraying in the orchards of British Columbia. Proceedings of the Entomological Society of British Columbia 60: 26-29. McGroarty, D.L. and B.A. Croft. 1978. Sampling the density and distribution of Amblyseius f a l l a c i s (Acarina: Phytoseiidae) in the ground cover of Michigan apple orchards. Canadian Entomologist 110: 785-794. McMurtry, J.A., C.B. Huffaker and M. van de Vrie. 1970. Ecology of tetra-nychid mites and their natural enemies: a review. I. Tetranychid enemies: their biological characters and the impact of spray practices. Hilgardia 40: 331-390. Metcalf, C.L. and W.P. F l i n t . 1962. Destructive and useful insects, their  habits and control. revised by R.L. Metcalf. 4th edition. Tata McGraw-H i l l Publ. Co., New Delhi. 132 Meyer, R.H. 1974. Management of phytophagous and predatory mites in I l l i n o i s orchards. Environmental Entomology 3: 333-340. Mimura, M. and J.D. Murray. 1978. On a diffusive prey-predator model which exhibits patchiness. Journal of Theoretical Biology 75: 249-262. Mitchell, R. 1970. An analysis of dispersal in mites. American Naturalist 104: 425-431. Morris, R.F. 1971. The influence of land use and vegetation on the popu-lation density of Hyphantria cunea. Canadian Entomologist 103: 1525-1536. Morrison, G. and D.R. Strong, Jr. 1980. Spatial variations in host density and the intensity of parasitism: some empirical examples. Environmental Entomology 9: 149-152. Mowery, P.D., L.A. Hull and D. Asquith. 1980. Two new sampling plans for European red mite surveys on apple u t i l i z i n g the negative binomial distribution. Environmental Entomology 9: 159-163. Murai, M. 1977. Population studies of Cavelerius saccharivorus Okajima (Heteroptera: Lygaeidae): adult dispersal in relation to the density. Researches on Population Ecology 18: 147-159. Murdoch, W.W. and A. Oaten. 1975. Predation and population s t a b i l i t y . Advances in Ecological Research 9: 1-131. Myers, J.H. and C.J. Krebs. 1971. Genetic, behavioral, and reproductive attributes of dispersing f i e l d voles Microtus pennsvlvanicus and Microtus  ochrogaster. Ecological Monographs 41: 53-78. Nachman, G. 1981. Temporal and spatial dynamics of an acarine predator-prey system. Journal of Animal Ecology 50: 435-451. Nishino, M. 1976. Population trends of the citrus red mite and i t s control. Noyaka Tsushin 96: 23-29. Oatman, E.R. and J.A. McMurtry. 1966. Biological control of the two-spotted spider mite on strawberry in southern California. Journal of Economic Entomology 59: 433-439. O'Neil, W.J. and W.H. Hantsbarger. 1951. Apple and pear pest control experiments for 1951. Proceedings of the 47th Annual Meeting Washington State Horticultural Association, p.134-137. Orchard, 0. 1927. Red spider investigations. Report of the Experimental Research Station, Chestnut, p.62-68. Parent, B. and F.T. Lord. 1971. Panonychus ulmi (Koch), European red mite, (Acarina: Tetranychidae) and other phytophagous mites on f r u i t trees in Canada, In: Biological Control Programs against Insect and Weeds in  Canada. 1959-1968. Technical Communication, CIBC, 4, 28-30. 133 Penman, D.R. and R.B. Chapman. 1980. Integrated control of apple pests in New Zealand. 17. Relationships of Amblyseius f a l l a c i s to phytophagous mites in an apple orchard. New Zealand Journal of Zoology 7: 282-287. Penman, D.R., C.H. Wearing, E. Collyer and W.P. Thomas. 1979. The role of insecticide-resistant phytoseiids in integrated mite control in New Zealand. Recent Advances in Acarology 1: 59-69. Perry, J.N. and R. Mead. 1979, On the power of the index of dispersion test to detect spatial patterns. Biometrics 35: 613-622. Pickett, A.D., W.L. Putman and E.S. LeRoux. 1958. Progress in harmonizing biological and chemical control of orchard pests in eastern Canada. Proceedings of the Tenth International Congress of Entomology, 1956, 3: 169-174. Pickett, A.D., N.A. Patterson, N.T. Stulte and F.T. Lord. 1946. The influence of the fauna of apple orchards in Nova Scotia. I. An appraisal of the problem and a method of approach. Science Agriculture 26: 590-600. Pielou, D.P. 1959. Problems in the compilation of a spray calendar for orchards in British Columbia. Proceedings of the Entomological Society of British Columbia.56: 3-12. Pielou, D.P. 1960. Contagious distributions in the European red mite, Panonychus ulmi (Koch), and a method of grading population densities from a count of mite-free leaves. Canadian Journal of Zoology 38: 645-653. Pielou, D.P. 1962. The ineffectiveness of Sevin in increasing oviposition in the spider mite, Tetranychus telarius (L.). Canadian Journal of Zoology 40: 9-11. Platnick, N.I. 1976. Drifting spiders or continents?: vicariance biogeo-graphy of the spider subfamily Laroniinae (Araneae: Gnaphosidae). Systematic Zoology 25: 101-109. Potter, D.A., D.L. Wrensch and D.E. Johnston. 1976a. Aggression and mating success in male spider mites. Science 193: 160-161. Potter, D.A., D.L. Wrensch and D.E. Johnston. 1976b. Guarding, aggressive behavior, and mating success in male twospotted spider mites. Annals of the Entomological Society of America 69: 707-711. Putman, W.L. 1962. Life-history and behavior of the predacious mite Typhlodromus (T.) caudiglans Schuster (Acarina: Phytoseiidae) in Ontario, with notes on the prey of related species. Canadian Entomologist 94: 163-177. Putman, W.L. 1970. Some effects of wind on the European red mite, Panonychus ulmi (Acarina: Tetranychidae). Canadian Entomologist 102: 659-667. 134 Putman, W.L. and D.C. Herne. 1960. Effects of Sevin on phytophagous mites and predators in an Ontario peach orchard. Canadian Journal of Plant Science 40: 198-201. Putman, W.L. and D.C. Herne. 1964. Relations between Typhlodromus caudiglans Schuster and phytophagous mites in Ontario peach orchards. Canadian Entomologist 96: 925-943. Putman, W.L. and D.H.C. Herne. 1966. The role of predators and other biotic agents in regulating the population density of phytophagous mites in Ontario peach orchards. Canadian Entomologist 98: 808-820. Rabbinge, R. 1976. Biological control of fruit-tree red spider mite. PUDOC, Wageningen, the Netherlands. 234p. Rabbinge, R. and M.A. Hoy. 1980. A population model for two-spotted spider mite Tetranychus urticae and it s predator Metaseiulus occidentalis. Entomologia Experimentalis et Applicata 28: 64-81. Rabbinge, R. and M.W. Sabelis. 1980. Systems analysis and simulation as an aid to the understanding of acarine predator-prey systems. In: Integrated Control of Insect Pests in the Netherlands. PUDOC, Wageningen, Chapter 10.1, p.281-290. Readshaw, J.L. 1971. An ecological approach to the control of mites in Australian orchards. Journal of the Australian Institute of Agricultural Science 37: 226-230. Regniere, J. 1982. A process-oriented model of spruce budworm phenology (Lepidoptera: Tortricidae). Canadian Entomologist 114: 811-825. Renshaw, E. 1982. The development of a spatial predator-prey process on interconnected sites. Journal of Theoretical Biology 94: 355-365. Roff, D.A. 1974. Spatial heterogeneity and the persistence of populations. Oecologia 15: 245-258. Rosenberg, N.J. 1974. Microclimate: the biological environment. John Wiley 5 Sons, N.Y. 315p. Roush, R.T. and M.A. Hoy. 1980. Selection improves Sevin resistance in spider mite predator. California Agriculture 34: 11-14. Ruesink, W.G. 1976. Status of the systems approach to pest management. Annual Review of Entomology 21: 27-44. Sabelis, M.W. 1981. Biological control of two-spotted spider mites using  phytoseiid predators. Part I. Modelling the predator-prey interaction  at the individual level. PUDOC, Wageningen (Agricultural Research Reports 910). 242p. Sanford, K.H. 1967. The influence of spray programs on the fauna of apple orchards in Nova Scotia. XVII. Effects on some predacious mites. Canadian Entomologist 99: 197-201. 135 Santos, M.A. 1976. Evaluation of Zetzellia mali as a predator of Panonychus ulmi and Aculus schlechtendali. Environmental Entomology 5: 187-191. Santos, M.A. 1982. Effects of low prey densities on the predation and oviposition of Zetzellia mali (Acarina: Stigmaeidae). Environmental Entomology 11: 972-974. Schuster, R.O. and A.E. Pritchard. 1963. Phytoseiid mites of California. Hilgardia 34: 191-285. Shelford, V.S. 1927. An experimental study of the relationship of the codling moth to weather and climate. Bulletin of the I l l i n o i s Natural History Survey 16: 311-440. Skellam, J.G. 1951. Random dispersal in theoretical populations. Biometrica 38: 196-218. Smith, J.M. 1974. Models in ecology. Cambridge University Press, Camb r i dge. 14 5p. Southwood, T.R.E. 1962. Migration of terrestrial arthropods in relation to habitat. Biological Reviews 37: 171-214. Southwood, T.R.E. 1977. Habitat, the templet for ecological strategies. Journal of Animal Ecology 46: 337-365. Stark, R.W. 1973. Systems analysis of insect populations. Annals of the New York Academy of Sciences 217: 50-57. Stearns, S.C. 1976. Life-history tactics: a review of the ideas. Quarterly Review of Biology 51: 3-47. Steer, W. 1940. Migration of the red spider. Gardeners' Chronicle, 20th July, p.28-29. Stinner, R.E., CS. Barfield, J.L. Stimac and L. Dohse. 1983. Dispersal and movement of insect pests. Annual Review of Entomology 28: 319-335. Takafuji, A. 1977. The effect of the rate of successful dispersal of a phytoseiid mite, Phytoseiulus persimilis Athias-Henriot (Acarina: Phyto-seiidae) on the persistence in the interactive system between the predator and i t s prey. Researches on Population Ecology 18: 210-222. Takafuji, A. 1980. Population dynamics of some phytophagous mites and their predators on goldenrod Solidago altissima L. I. Seasonal trends in population numbers and spatial distributions. Researches on Population Ecology 21: 197-216. Tanigoshi, L.K. and R.W. Browne. 1981. Coupling the cytological aspects of spider mite feeding to economic injury levels on apple. Protection Ecology 3: 29-40. Tanigoshi, L.K., S.C. Hoyt, R.W. Browne and J.A. Logan. 1975a. Influence of temperature on population increase of Metaseiulus occidentalis (Acarina: Phytoseiidae). Annals of the Entomological Society of America 68: 979-986. Influence arina: Tanigoshi, L.K., S.C. Hoyt, R.W. Brovme and J.A. Logan. 1975b. In f l u of temperature on population increase of Tetranychus mcdanieli (Ac Tetranychidae). Annals of the Entomological Society of America 68. 972-978. Taylor, L.R. 1961. Aggregation, variance and the mean. Nature, London 189: 732-735. Taylor, L.R. and I.P. Woiwod. 1982. Comparative synoptic dynamics. I. Relationships between i n t e r - and i n t r a - s p e c i f i c s p a t i a l and temporal variance/mean population parameters. Journal of Animal Ecology 51: 879-906. Thompson, W.A., P.J. Cameron, W.G. Wellington and I.B. Vertinsky. 1976. Degrees of heterogeneity and the survival of an insect population. Researches on Population Ecology 18: 1-13. Treherne, R.C. 1914. A review of applied entomology i n B r i t i s h Columbia. Proceedings of the Entomological Society of B r i t i s h Columbia 4: 67-71. Treherne, R.C. 1921. A further review of applied entomology i n B r i t i s h Columbia. Proceedings of the Entomological Society of B r i t i s h Columbia 13: 135-146. T r o t t i e r , R. and D.H.C. Heme. 1979. Temperature relationships to forecast hatching of overwintered eggs of the European red mite, Panonychus ulmi (Acarina: Tetranychidae). Proceedings of the Entomological Society of Ontario 110: 53-60. Vandermeer, J.H. 1973. On the regional s t a b i l i z a t i o n of l o c a l l y unstable predator-prey relationships. Journal of Theoretical Biology 41: 161-170 van de V r i e , M. 1964. Invloed van carbaryl op de vermeerdering van de v r u i t s p i n t m i j t . J v e r s l Proefstn. F r u i t t e e l t , Wilhelminadorp, 112-113. van de V r i e , M., J.A. McMurtry and C.B. Huffaker. 1972. Ecology of t e t r a -nychid mites and t h e i r natural enemies: a review. I I I . Biology, ecology, and pest status, and host plant relations of tetranychids. Hilgardia 41: 343-432. Watt, K.E.F. 1961. Mathematical models for use i n insect pest control. Canadian Entomologist Supplement 19 (Accompanies Volume XCIII). 62p. Welch, S.M. 1979. The application of simulation models i n mite pest manage ment. Recent Advances i n Acarology 1: 31-40. Wellington, W.G. 1964. Qualitative changes i n populations i n unstable environments. Canadian Entomologist 96: 436-451. Wellington, W.G. 1979a. Dispersal and population change. Keynote address presented to the Second Joint Meeting of the I.U.F.R.O. Working Parties on the Integrated Control of Scolytid Bark Beetles and the Population Dynamics of Forest Insects, Sandpoint, Idaho. 137 W e l l i n g t o n , W.G. 1979b. Insec t d i s p e r s a l : a b i o m e t e o r l o g i c a l p e r s p e c t i v e . I n : Movement o f H i g h l y M o b i l e I n s e c t s : Concepts and Methodology i n Research . R . L . Rabb and G . G . Kennedy (eds . ) Nor th C a r o l i n a S ta t e U n i v e r -s i t y , p .104-108 . Whi te , N . D . G . and J . E . L a i n g . 1977a. F i e l d obse rva t i ons o f Z e t z e l l i a m a l i (Ewing) ( A c a r i n a : St igmaeidae) i n Southern O n t a r i o apple o r cha rds . Proceedings o f the En tomolog i ca l S o c i e t y o f O n t a r i o 108: 23-30 . Whi te , N . D . G . and J . E . L a i n g . 1977b. Some aspects o f the b i o l o g y and a l a b o r a t o r y l i f e t a b l e o f the a c a r i n e p reda to r Z e t z e l l i a m a l i . Canadian Entomolog i s t 109: 1275-1281. Wolfenbarger , D .O. 1975. F a c t o r s a f f e c t i n g d i s p e r s a l d i s t a n c e s o f smal l organisms. E x p o s i t i o n P r e s s , H i c k s v i l l e , N . Y . 230p. W o l l k i n d , D . J . and J . A . Logan. 1978. Temperature-dependent p reda to r -p rey mi te ecosystem on apple t r e e f o l i a g e . J o u r n a l o f Mathemat ica l B i o l o g y 6: 265-283. 138 APPENDIX A Method of degree-day calculations When development rate is a reasonably linear function of temperature, as has been assumed to be the case for many arthropod within the range of normally-encountered f i e l d temperatures, integrated heat, or "degree-days", may be useful in simplifying predictions and simulations. The approach originated with Shelford's 1927 report on the codling moth and has been widely applied. A variety of short-cuts have been applied, some of which are s t i l l in use (e.g. Ito et a l . , 1968; Trottier and Heme, 1979). The simplest i s ° DTI = (T™„v + T . )/2 - T (1) TI max min 1 where T , T . and T, are the day's high temperature, the day's low . max' mm 1 J & r temperature and the lower threshold temperature below which no development is assumed to take place. However, when T m^ n < T^, this method under-estimates the heat summation. Another method, 'DT1 = l (T + T . )/2 - T i f T . ^ T (2) TI I max mm 1 mm ^ 1 (T + T )/2 - T, i f T . < TI ^ max 1-" 1 min tends to overestimate °D T 1 when T m i n < T . An example is shown below. 139 °D 10 daily max daily min method (1) method f2) best estimate (°C) C°C) 19.6 0.5 0.05 4.80 3.06 15.0 3.0 0 2.50 1.44 15.2 4.0 0 2.60 2.60 18.3 4.4 1.35 4.15 2.93 21.7 5.6 3.65 5.85 4.66 26.1 8.9 7.50 8.05 7.62 22.2 11.1 6.65 6.65 6.65 31.1 12.2 11.65 11.65 11.65 The solution, of course, is some simple calculus. Integrations can be performed via electronic computer for each specific data set, or tables may be produced (eg. Brunner .§_t .al., 1982). Integration may be done numerically i f temperatures are collected at several points over the day. More often the data are maximum and minimum temperatures only. Periodic models, usually sine (e.g. Baskerville and Emin, 1969; Frazer and Gilbert, 1976) or cosine (e.g. Lund and Herne, 1980), give good approximations to daily temperature fluctuations (Rosenberg, 1974). The algorithm I used is derived below. If we model daily temperature fluctuations (recalling that one day need not be defined as midnight to midnight) as y = M + Acos (wt) where M = (max + min)/2 A = max - M w = 2 7T (i.e. the period is 1), then the lower threshold, y = T , intersects this curve at t = P, given by ( T l " M \ / P = arccos I — J^ w 140 since Tj =/* + Acos (coP), and the area under the temperature curve is given by 2 1 y dt (see Figure Al) It is straightforward to integrate this simple model: yu. + Acos (w t) dt / I t + - sin (wt) = jy? + i sin (u>p). We must subtract off the area below the threshold, / A ? + ^ sin <W) - Tl P = P (JJL - T p + ^ sin(wPj, and twice this area gives the heat summation for the day. The FORTRAN program used to perform the calculations for lower and upper thresholds is shown in Table Al. 141 - i — 0.6 — I — 0.8 Legend Ihreshold •temperature One Day Figure A l . Cosine model of a 24-hour temperature fluctuation. The curve is used for integration only. The shaded portion represents the degree-days accumulated over the 24-hour period. 142 Table A l . The algorithm used for degree-day calculation (written in FORTRAN). C I n t e g r a t i o n under a d a l l y t e m p e r a t u r e c u r v e . C o s i n e model C Dan L. Johnson, December 30, 1981. C REAL *8 MAX,MIN,THLOW,THUP,P,DDL0W,DDUP.DD.CUM,AV,AMP,W INTEGER YEAR.MONTH.DAY,JUL C W=2*3. 141592654 THL0W=10.0 THUP=35. C-- THLOW and THUP a r e the lower and upper t h r e s h o l d s . LASTMN=1 JUL=0 CUM=0 WRITE(6,8)THL0W 8 F0RMAT(/10X.'Temp Summation ' , 1 5 X , ' T h r e s h o l d = '.F4.1/) WRITE(6,10) 10 F0RMAT(8X,'Date',5X,'Day*',3X,'Max. M1n. DegDays Cumulat 1ve DD', &/6X. ' ' ,3X. ' ' ,3X. &' ') 2 READ(5,5,END=90)YEAR.MONTH,DAY.MAX,MIN 5 F0RMAT(6X.3I2,2F6.1) IF (MAX.LT.MIN) WRITE(6,6)MAX,MIN 6 FORMAT(//' *** ERROR *** max<m1n '.2F8.1) C IF(LASTMN.NE.MONTH) WRITE(6,10) LASTMN=MONTH JUL=JUL+1 AV=(MAX+MIN)/2 AMP=MAX-AV IF (MIN.LT.THLOW) GOTO 30 DDL0W=(MAX+MIN-2*THL0W)/2 GOTO 50 C-- C a l c u l a t e I n t e g r a t i o n above lower t h r e s h o l d . 30 IF (MAX.GT.THLOW) GOTO 40 DDLOW=0 GOTO 50 40 P=(DARCOS((THLOW-AV)/AMP))/W DDL0W=2 * (P*(AV-THLOW)+(AMP/W)*DSIN(W*P)) C C-- Now s u b t r a c t o f f i n t e g r a t i o n above upper t h r e s h o l d . C 50 IF (MAX.GT.THUP) GOTO 52 DDUP=0 GOTO 55 52 P=(DARCOS((THUP-AV)/AMP))/W DDUP=2 * (P*(AV-THUP)+(AMP/W)*DSIN(W*P)) 55 DD=DDLOW-DDUP CUM=CUM+DD WRITE(6,60)MONTH,DAY,YEAR,JUL,MAX,MIN,DD,CUM 60 FORMAT(5X,313,16,2F8.1.F8.2.F12.2.5X.F12.2) GOTO 2 C-- ( r e a d max&mln d a t a from next c a r d ) C 90 STOP END APPENDIX B Summary of recent weather at the Research Station, Summerland, B.C., Canada. (Compiled by the author from Agriculture Canada meteorological data tapes). Date 1 / 2 / 3 4 5 6 7 8 / 9 / 10 / 11 / 12 / 1970 1970 / 1970 / 1970 / 1970 / 1970 / 1970 1970 1970 1970 1970 1970 Mean temp ( O -3.5 1 .7 4.5 7.7 13.4 19.9 22.3 21.3 13.0 7 .9 0.7 -2.5 Mean d l f f (mx-mn) 6.0 6.6 9.5 10.4 13.2 14.0 14.3 14 .6 11.2 9 . 1 5.8 6.0 T o t a l r a i n (mm) 0.0 12.4 7.9 4.7 8.5 14.8 25 .8 7.4 5.4 16.3 17.3 5.9 T o t a l snow (mm) 542 . 3. 0. 0. 0. 0. 0. 0. 0. 0. 219. 300. (31 d ays) (28 d a y s ) (31 d ays) (30 d a y s ) (31 d ays) (30 d a y s ) (31 d ays) (31 d ays) (30 d ays) (31 d a y s ) (30 d a y s ) (31 days) H i g h Low 6. 1 - 14 . 4 9.4 -5. 6 13.9 -10. 0 17 .8 -2 . 8 29.4 0. 6 35.6 5 . 6 36 . 1 8 . 3 33.9 7 . 8 26. 1 1 . 1 21.7 -3 .3 11.1 - 15 .6 8.9 - 16 .7 Mean Mean temp d1 f f Date (C) (mx-mn) 1 / 197 1 -2.5 6.0 2 / 197 1 0 7 6 . 2 3 / 197 1 2 . 1 8 . 4 4 / 197 1 8.6 12.0 5 / 1971 14.5 12.0 6 / 197 1 15.3 11.8 7 / 1971 21.0 14.8 8 / 1971 23 . 3 14.9 9 / 197 1 13.6 10.2 10 / 197 1 7 .9 8.8 1 1 / 197 1 3.3 5.9 12 / 197 1 -5.7 5 . 5 T o t a l T o t a l r a i n snow (mm) (mm) n 6 1 3 5 9 . (31 clays) 7 4 101. (28 days) 1 8 99. (31 days) O . 13.7 0. (30 days) 45.0 0. (31 days) 52.4 0. (30 d a y s ) 12.0 0. (31 d a y s ) 30.2 0. (31 d a y s ) 19.3 0. (30 d a y s ) 7.2 67. (31 days) 11.8 3. (30 days) 0.0 782. (31 days) H i g h Low 10.0 - 17 . 2 11.1 -11. 7 14.4 - 10. 6 22.8 -2 . 2 28.9 2 . 8 28 . 3 5 . 0 36 . 1 5 . 0 37.8 8 . 3 26.7 3 .9 22 .8 -6 . 1 10.0 -7 .8 6.7 -20 .6 Date / 1972 / 1972 / 1972 / 1972 / 1972 / 1972 / 1972 8 / 1972 9 / 1972 10 / 1972 11 / 1972 12 / 1972 Mean temp (C) Mean d i f f (mx-mn) T o t a l r a i n (mm) -6 . 4 6 . 9 0. 0 - 1 . 2 7 . 4 22 . 3 4 . 8 9 . 3 27 . 4 6 . 2 10. 7 34 . 1 13 . 9 1 1 . 4 28 . 7 16 . 5 10. . 2 47 .6 19 .7 13 . 7 20 . 2 2 1 .0 13 . 4 13 . 4 13 .0 1 1 . 2 19 . 9 6 .9 9 .5 5 .0 4 .0 4 .9 1 1 . 4 -3 . 2 5 . 5 21 . 4 T o t a l snow (mm) 4 7 3 . (31 days) 91 . (29 days) 5 3 . (31 days) 0. (30 d ays) 0. (31 d ays) 0. (30 d ays) 0. (31 d ays) 0. (31 days) 0. (30 days) 0. (31 d ays) 0. (30 d ays) 112. ( 3 1 days) H i g h Low 7.8 -22 . 2 11.7 - 18 . 3 17.8 -6 . 7 21.7 -3 . 3 31 . 1 0. 0 28.3 7 . 2 33.9 7 . 8 36 . 1 7 . 2 28 .9 -0 .6 22 . 2 -5 .0 11.1 -5 .6 11.1 - 18 .9 144 Mean Mean T o t a l temp d l f f r a i n D a te (C) (mx-mn) (mm) 1 / 1973 -4 . 0 5 . 8 1 .3 2 / 1973 0 . 0 6 . 1 6 .7 3 / 1973 5. 3 8 . 8 16 . 2 4 / 1973 9 . . 1 11 .7 4 . 6 5 / 1973 13 .9 13. 1 17.9 6 / 1973 16 .9 12 .5 23 .2 7 / 1973 21 .6 14.2 1 .3 8 / 1973 19 .9 13 .8 11.4 9 / 1973 16 . 1 11.1 34 .8 10 / 1973 8 .5 7 . 1 6 0 . 6 1 1 / 1973 0 .8 5 . 0 24 .5 12 / 1973 1 .6 4 .5 12.1 T o t a l snow (mm) 8 5 . 275 . 0 . 0 . 0 . 0 . 0 . 0 . 0 . 0 . 378 . 153 . (31 d a y s ) (28 d a y s ) (31 d a y s ) (30 d a y s ) (31 d a y s ) (30 d a y s ) (31 d a y s ) (31 d a y s ) (30 d a y s ) (31 d a y s ) (30 d a y s ) (3 1 d a y s ) H i g h 10 .0 10 .6 15 .6 2 0 . 6 3 0 . 6 31 .1 34 . 4 3 5 . 6 32 .8 19.4 10 .0 7 .8 Low 18 . 13. -3 . -2 . 0 . 4 . 9 . 6 . 4 . O . -7 . 2 - 9 . 4 . 3 .9 9 .2 .6 .4 .4 1 .4 .6 M e a n M e a n T o t a l t e m p d i f f r a i n D a t e ( C ) ( m x - m n ) (mm) 1 / 1974 -2 . 2 6 . 2 13.8 2 / 1974 2 . 6 6 . 1 8 .8 3 / 1974 4 . 2 7 . 8 24 .7 4 / 1974 9 . 5 10 .3 18 .8 5 / 1974 1 1 . 5 11.6 37 . 5 6 / 1974 18 . 7 14.2 5 . 1 7 / 1974 19 , . 1 13.4 27 . 8 8 / 1974 20 . 7 14.2 20 .4 9 / 1974 16 . 1 13 .3 6 . 2 10 / 1974 9 . 5 10. 5 2 . 3 1 1 / 1974 3 . 7 5 . 9 25 . 2 12 / 1974 0 . 7 5 .4 15.1 T o t a l s n o w (mm) 183. 73 . 18 . 0 . 0 . 0 . 0 . 0 . 0 . 0 . 6 . 118. H i g h (31 (28 (31 (30 (31 (30 (31 (31 (30 (31 (30 (31 d a y s ) d a y s ) d a y s ) d a y s ) d a y s ) d a y s ) d a y s ) d a y s ) d a y s ) d a y s ) d a y s ) d a y s ) .0 .6 . 9 .6 . 2 10. 10. 13 20. 22 . 32 . 8 35 .0 35 . 28 . 20 . 1 1 7 . 0 9 .6 . 7 8 L o w -17 . -6 . -8 . - 1 . 0 . 3 . 7 . 8 3 0 -6 -8 M e a n M e a n T o t a l t e m p d i f f r a i n D a t e ( C ) ( m x - m n ) (mm) 1 / 1975 -3 . 8 6 2 0 0 2 / 1975 -4 . 8 7 .2 0 . 3 3 / 1975 1 . 7 8 . 2 14.1 4 / 1975 6 . 9 11.4 11.0 5 / 1975 12 . 9 12 .9 18 . 9 6 / 1975 16 . , 3 12.2 23 .4 7 / 1975 22 . 3 13 .5 19 9 8 / 1975 17 .5 11.5 35.4 9 / 1975 15 .6 13.2 2 .8 10 / 1975 8 .9 8 .4 14.7 1 1 / 1975 2 . 7 6 . 0 3 . 9 12 / 1975 - 0 .5 5 . 7 9 . 2 T o t a l s n o w (mm) H i g h L o w 407 . (31 d a y s ) 6 . 7 - 18 . 9 535 . ( 28 d a y s ) 7 . 8 - 1 7 . 2 8 45 . (31 d a y s ) 12 . 2 -7 . 5 . (30 d a y s ) 17 . 8 -5 . 0 0 . (31 d a y s ) 26. , 7 2 . 2 O . ( 30 d a y s ) 28 9 4 . . 4 0 . (31 d a y s ) 36 .7 10 .0 0 . (31 d a y s ) 31 . 7 7 . 8 0 . (30 d a y s ) 29 . 4 5 . 0 0 . (31 d a y s ) 25 .6 -2 . 2 464 . ( 30 d a y s ) 18 .9 - 1 2 - 1 1 . 8 203 . (31 d a y s ) 1 1 . 1 . 7 145 Date 1 / 1976 2 / 1976 3 / 1976 4 / 1976 5 / 1976 6 / 1976 7 / 1976 8 / 1976 9 / 1976 10 / 1976 11 / 1976 12 / 1976 Mean Mean Total temp d l f f r a i n (C) (mx-mn) (mm) -1.2 6.0 9.9 0.6 6.3 1.3 2 . 1 8 . 1 6.8 8.2 11.5 15.5 12.6 11.8 25.4 15.0 12.8 40.5 19.3 13.6 49.0 17.4 9.8 80. 1 15.4 11.8 4.3 8.2 9.3 15 . 4 3.4 6.0 10.4 -0. 1 4 .8 1 .3 Total snow (mm) 290. 116. 15. 0. 0. 0. 0. 0. 0. 0. 0. 142 . n High Low (31 days) 10. 0 -9 . 4 (29 days) 10. 6 -11. 7 (31 days) 13 . 9 -16 . 7 (30 days) 20. .0 -2 . 8 (31 days) 25 .6 1 . 7 (30 days) 32 . 8 2. . 2 (31 days) 35 .0 6. , 7 (31 days) 27 .2 6 . 7 (30 days) 30 .6 3 . 9 (31 days) 21 . 1 - 1 . 7 (30 days) 13 .9 -6 . 7 (31 days) 8 .3 -7 . 8 Mean Mean temp dl f f Date (C) (mx-mn) 1 / 1977 -4 . 9 5.3 2 / 1977 1 . 4 7 .4 3 / 1977 4 . 4 8.6 4 / 1977 9 . 9 13.5 5 / 1977 1 1 . . 7 11.3 6 / 1977 18 .9 13.6 7 / 1977 19 .8 13.9 8 / 1977 22 .0 13.5 9 / 1977 13 .6 10.5 10 / 1977 8 . 7 10.6 1 1 / 1977 1 .5 6 . 5 12 / 1977 -2 .6 5 . 3 Tota 1 ra 1 n (mm) . 2 .0 . 4 .9 5 . 1 13 0 8 13 46 10 20. 1 13.5 23 . 4 8.2 18.2 38 . 5 Total snow ( mm) 246 . 53. 69 . 0. 0. 0. 0. 0. O. 0. 188 . 220. (31 days) (28 days) (31 days) (30 days) (31 days) (30 days) (31 days) (31 days) (30 days) (31 days) (30 days) (31 days) High 5.6 12.8 14.4 27 .8 22 . 2 32 34 35 23 2 4 0 9 17 .8 12.2 8.9 Low -13 . -8 . -3 . -2 . O. 6. 5 7 3 -O -16 -13 Date 1 / 2 3 4 5 6 7 8 / 9 / 10 / 11 / 12 / 1978 / 1978 / 1978 / 1978 / 1978 / 1978 1978 1978 1978 1978 1978 1978 / Mean Mean Total temp di f f ra i n (C) (mx-mn) (mm) -2.8 4 . 3 31.6 0.6 5 . 8 6 . 2 5 . 1 9 .O 9 . 1 8 . 2 9.5 57 .7 12.7 11.6 35 .0 18.0 13.4 20.6 2 1.8 13.7 20.3 19.1 11.5 40.6 13.6 9. 1 36 . 7 8.9 10. 2 6.4 -O. 2 7 .0 18.0 -4 . 4 7 .5 0.0 Total snow (mm) 398 . 22 1 . 0. 0. 0. 0. 0. 0. 0. 0. 36 . 150. High (3 1 days) (28 days) (3 1 days) (30 days) (31 days) (30 days) (31 days) (31 days) (30 days) (31 days) (30 days) (3 1 days) .6 .8 .9 5 7 18 22 .8 26.7 33 35 35 26 22 15 8 Low -15 -7 -8 -2 1 5 7 7 3 -4 - 13 -22 0 8 9 2 7 O .0 .5 .0 .5 . 5 .5 Mean Mean T o t a l temp d l f f r a In Date (C) (mx-mn) (mm) 1 / 1979 -9. 3 6.5 0.0 2 / 1979 -1 . 2 7.6 3.4 3 / 1979 5. .5 1 1 .0 4.9 4 / 1979 8 . 2 13.3 13.4 5 / 1979 13 . , 7 13.7 16.6 6 / 1979 18. .3 13.9 21.3 7 / 1979 21 . 7 14.6 19.7 8 / 1979 21 . 0 12.9 53 . 3 9 / 1979 16 . 1 11.8 33.7 10 / 1979 10 .7 8.8 28.3 11 / 1979 0 .9 5.6 9.0 12 / 1979 1 .9 6 . 1 6.5 T o t a l snow (mm) n H i g h Low 201 . (31 days) 3.0 -22 .0 67. (28 d a y s ) 12.0 -18.0 0. (31 days) 17.0 -5.0 0. (30 days) 25.0 -4.0 0. (31 days) 28.0 2.5 0. (30 days) 33 .0 4.0 0. (31 days) 39.5 5.0 0. (31 days) 33.0 1 1 .0 0. (30 days) 28 . 5 6.5 0. (31 days) 22.5 0.5 0. (30 days) 7.5 -7.0 161 . (31 days) 12.0 -13.5 147 APPENDIX C Maps of the spatial arrangement of counts of Typhlodromus spp. (mainly T. caudiglans), active stages of the European red mite, ERM summer eggs and active stages of Zetzellia mali on the 64 trees in the orchard. Each count is the number on a 10-leaf sample. Refer to Figures 3 and 4 for details on the layout of the orchard. Date # Actual sampling date 1 May 13,14 2 May 27,28 3 June 10,11 4 June 24,25 5 July 8,9 6 July 22,23 7 August 5,6 8 August 19,20 10 September 30, October 1 Date #9 was incomplete (see Discussion, Chapter One) and is not included here. 148 N-f-> Phytoseiids by r,c: (Date: 1) 4c Mc Mc Mc Del Del Del Del 1 6 0 1 5 0 20 0 3 2 1 4 3 0 6 2 0 0 1 1 3 1 2 1 1 4 1 3 6 2 4 0 4 3 1 12 1 4 1 0 10 3 1 8 4 3 8 1 3 1 0 13 1 2 2 0 2 4 3 1 2 3 4 11 Act ive ERM by r,c: (Date : 1) Mc Mc Mc Mc Del Del Del Del 1 3 1 39 18 2 35 68 4 0 10 6 4 41 70 63 1 6 45 0 1 8 48 281 62 15 63 16 18 37 293 64 194 42 0 17 2 24 6 67 24 80 0 1 9 44 13 86 35 21 24 6 1 25 57 22 18 5 21 1 1 79 27 45 15 16 0 9 149 ERM eggs by r , c : (Date: 1 ) c Mc Mc Mc Del Del Del Del 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Z. ma1i by r,c: (Date: 1) Ac Mc Mc Mc Del Del Del Del 1 1 1 0 5 2 1 0 0 6 2 2 2 1 8 0 0 3 5 5 13 5 1 4 8 6 2 6 1 0 2 6 0 4 9 8 1 2 1 6 0 1 0 0 4 2 2 1 1 6 6 o 15 0 17 5 0 5 1 0 3 1 1 1 3 1 150 Phytoseiids by r,c: (Date: 2) Mc Mc Mc Mc Del Del Del Del 0 1 0 4 3 10 2 1 1 0 6 6 9 4 1 0 1 0 0 1 1 2 I 1 0 1 4 0 1 1 0 0 6 8 1 0 0 1 3 3 0 3 4 0 7 ) 0 3 5 4 3 . 7 1 j 1 1 0 5 5 5 2 Act ive ERM by r,c: (Date: 2) c Mc Mc Mc Del Del Del Del 1 17 4 2 16 25 18 26 7 0 0 9 58 15 8 49 3 3 1 26 43 10 12 24 5 12 1 1 21 15 43 64 45 2 0 10 8 56 12 24 15 5 5 36 2 1 1 26 10 19 2 2 15 12 31 7 8 26 4 9 15 18 32 34 0 24 .ERM eggs by r,c : (Date: 2) Mc Mc Mc Mc Del Del Del Del 4 198 21 9 74 216 91 238 19 1 5 66 519 112 52 566 4 13 9 177 294 248 80 240 33 48 77 175 96 556 556 389 6 0 99 179 759 87 168 73 47 26 41 1 20 110 171 60 76 9 9 56 64 288 71 85 85 26 27 99 66 140 96 5 124 Z. ma1i by r,c: (Date: 2) Mc Mc Mc Mc Del Del Del Del 0 5 10 1 13 0 0 1 3 4 5 2 4 3 4 0 3 7 1 8 3 13 39 7 1 12 9 13 1 1 24 18 1 3 5 9 1 12 0 1 10 6 1 0 2 1 2 9 1 1 6 0 1 16 4 4 8 2 4 152 Phytoseiids by r f c : (Date: 3) Ac Mc Mc Mc Del Del Del Del 0 1 0 0 0 0 2 0 1 1 0 0 0 0 6 1 0 0 0 0 • 7 0 0 0 1 3 0 0 1 0 2 0 0 0 0 0 0 3 1 1 0 0 0 0 3 2 0 1 0 0 3 1 8 4 1 0 0 0 1 3 2 0 4 0 Active ERM by r,c: (Date: 3) Mc Mc Mc Mc Del Del Del Del 0 1 4 2 0 65 45 29 79 8 0 0 1 1 13 1 4 27 162 3 4 0 13 74 45 31 128 20 2 7 82 31 46 33 1 1 2 0 0 0 41 121 24 7 9 6 1 38 2 43 17 14 28 2 1 1 19 42 1 1 2 2 1 17 24 36 76 36 7 81 ERM eggs by r,c: (Date: - Mc Mc Mc Mc 6 384 37 0 145 2 4 28 20 20 9 124 136 46 35 363 7 0 33 275 22 14 235 24 21 12 4 126 23 96 149 152 Z. mali by r,c: (Date: Mc Mc Mc Mc 0 0 0 0 0 5 1 1 1 2 1 0 0 1 1 0 0 1 0 1 1 0 8 0 1 0 10 3 1 2 0 0 3) Del Del Del Del 796 189 129 216 382 131 84 719 291 224 1 39 285 78 95 93 332 1052 89 49 15 317 31 49 24 359 55 43 9 421 215 26 310 3) Del Del Del Del 6 0 0 0 4 2 5 1 9 3 3 0 2 4 1 4 2 0 2 3 0 0 5 1 3 0 0 1 0 0 2 0 154 Phytoseiids by r,c: (Date : 4) lc Mc Mc Mc Del Del Del Del 0 0 0 0 0 1 1 0 1 0 0 0 0 0 2 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 8 4 0 0 0 2 0 4 0 0 2 0 1 0 6 4 1 4 0 0 13 1 12 4 0 3 0 0 2 0 12 0 6 0 Act ive ERM by r,c: (Date: 4) Mc Mc Mc Mc Del Del Del Del 44 29 73 25 186 141 17 164 62 18 27 69 297 1 14 28 267 5 67 21 199 140 73 44 224 35 22 26 109 137 182 37 1 72 1 5 36 74 37 345 61 240 30 1 84 65 185 74 48 86 72 18 45 7 143 91 37 116 75 89 74 62 28 85 143 62 238 155 ERM eggs by r , c : (Date: 4) Mc Mc Mc Mc Del Del Del Del 139 307 344 187 1631 923 264 1223 302 1 15 274 458 3269 750 291 2221 16 319 74 2260 1019 656 393 946 200 170 142 666 1 1 38 1269 448 843 121 214 322 339 2431 361 1 721 1 33 13 549 385 1338 357 294 37 1 442 1 22 362 65 796 387 360 818 338 235 274 292 1 1 4 441 1022 295 1352 Z. m a l i by r , c : (Date: 4) Mc Mc Mc Mc Del Del Del Del 4 0 6 2 29 2 2 5 8 2 2 8 1 2 8 10 4 3 0 6 6 9 10 6 10 3 2 0 7 10 27 21 23 2 5 0 0 6 8 16 6 1 1 1 0 0 4 9 16 0 5 3 4 0 8 20 4 9 1 2 0 C 4 20 5 5 1 156 Phytoseiids by r,c: (Date: 5) Mc Mc Mc Mc Del Del Del Del 0 1 0 4 0 0 18 0 4 0 0 0 0 0 5 1 0 0 1 0 0 1 0 1 5 2 0 0 0 0 4 3 0 2 0 7 2 1 0 2 4 0 4 1 0 1 0 9 0 1 1 0 3 0 0 4 0 0 5 2 2 0 3 0 Active ERM by r,c: (Date: 5) Mc Mc Mc Mc Del Del Del Del 229 494 207 93 1668 2160 548 2488 586 49 260 763 2597 1240 224 2510 62 490 268 1835 701 301 528 1 604 101 185 565 1033 1933 2058 31 1 975 408 161 865 1 35 720 352 1928 128 1 28 918 433 2515 491 359 1291 445 418 852 77 771 1 40 1081 1788 347 626 360 1037 164 158 2084 303 1975 157 ERM eggs by r,c: (Date: 5) MC Mc Mc Mc Del Del Del Del 134 315 102 109 959 1 138 615 1 193 456 24 92 462 1202 782 188 1853 6 321 74 1 185 318 210 554 955 68 63 364 668 924 1 1 04 237 746 129 99 464 169 371 152 1780 100 1 00 810 364 486 147 1 36 840 397 198 550 77 309 88 487 1550 351 320 298 837 98 73 990 220 2060 Z. mali by r,c : (Date: 5) Mc Mc Mc Mc Del Del Del Del 2 1 1 19 20 4 1 3 19 5 16 5 3 1 1 3 18 9 0 0 3 1 13 8 21 1 1 12 28 0 1 10 4 46 7 2 5 7 1 4 18 13 8 2 13 5 16 4 0 3 32 40 0 5 0 32 20 4 3 14 8 15 2 19 6 0 0 Phytoseiids by r,c: (Date : 6) Ic Mc Mc Mc Del Del Del Del 0 1 0 0 1 0 35 0 2 0 0 5 0 2 10 0 4 0 0 0 10 10 0 0 0 3 0 2 1 0 1 0 0 6 0 6 2 2 2 4 0 0 0 3. 3 4 0 7 2 0 1 0 9 1 6 4 0 1 1 1 13 7 0 2 1 Active ERM by r,c: Mc Mc Mc Mc 250 205 95 134 204 36 231 333 46 156 239 273 152 96 352 325 198 151 468 127 123 574 332 382 149 436 136 196 308 477 140 151 e: 6) Del Del Del Del 746 934 312 486 732 753 374 1 102 188 145 356 340 771 308 262 106 632 341 712 84 273 256 758 170 302 344 1014 21 1 189 662 377 1 003 159 ERM eggs by r,c : (Date: 6) Mc Mc Mc Mc Del Del Del Del 734 633 225 407 727 2486 2635 2153 598 255 41 1 474 694 2124 2264 2806 380 237 896 602 263 630 1 498 966 377 276 912 330 1368 . 674 2317 565 876 293 985 431 1244 2069 407 378 865 897 932 489 1598 949 1834 336 739 669 342 : 314 681 1459 3050 707 581 882 255 521 986 1381 2303 577 Z. mali by r,c': (Date: 6) Mc Mc Mc Mc Del Del Del Del 5 0 2 17 5 4 22 5 7 28 0 4 10 18 2 1 12 6 1 2 1 12 8 7 1 27 1 1 5 7 29 6 1 4 0 21 32 32 31 6 16 16 26 41 15 0 9 17 23 34 0 2 3 1 1 20 0 8 7 0 23 3 1 1 3 0 2 Phytoseiids by r,c: (Date: 7) Mc Mc Mc Mc Del Del Del Del 1 0 0 3 2 0 17 0 3 2 0 1 2 0 15 1 1 1 4 0 10 14 0 0 2 16 2 0 2 0 7 0 0 1 4 0 12 1 1 1 13 19 2 4 2 2 7 0 6 1 4 3 5 22 16 4 3 2 0 7 16 1 1 0 28 1 Act ive ERM by r,c: (Date : 7) Mc Mc Mc Mc Del Del Del Del 1 42 206 127 67 899 879 268 90 97 49 48 1 13 145 327 1320 1830 148 438 331 65 326 543 746 242 198 244 678 124 879 162 606 135 619 352 48 38 554 1333 158 125 307 746 273 136 934 685 847 173 192 734 12 166 676 955 2260 664 570 188 158 259 31 1 221 1 151 1 33 ERM eggs by r,c: (Date: Mc Mc Mc Mc 21 1 354 143 141 1 49 97 100 228 335 427 369 125 492 308 398 282 930 440 67 67 226 748 338 107 474 621 29 95 714 183 191 229 Z. mali by r,c: (Date: Mc Mc Mc Mc 2 3 13 3 18 17 0 1 6 25 44 33 4 43 25 5 4 5 42 3 8 2 9 1 4 6 15 18 19 22 16 1 9 23 30 161 7) Del Del Del Del 431 557 393 162 113 398 1 158 1439 550 1027 623 414 750 289 519 187 514 977 275 250 746 889 933 327 669 61 1 1363 1015 462 382 822 172 7) Del Del Del Del 40 8 4 7 1 2 1 3 23 8 29 48 27 7 17 4 21 30 5 6 25 74 2 1 2 32 63 1 1 9 13 33 29 2 7 0 162 Phytoseiids by r,c: (Date: 8) MC Mc Mc Mc Del Del Del Del 0 1 7 3 1 0 4 0 0 0 0 1 0 0 34 1 0 0 2 0 9 8 0 0 1 6 0 0 0 0 18 1 0 12 0 1 0 7 0 6 20 3 2 1 3 9 2 15 2 0 2 0 13 1 3 6 2 0 12 1 2 0 5 0 Active ERM by r,c: (Date : 8) Mc Mc Mc Mc Del Del Del Del 24 42 42 10 143 150 59 18 4 21 7 24 32 16 236 1 37 9 141 80 1 1 82 191 117 21 47 75 68 63 49 49 85 37 62 74 52 1 3 98 370 94 83 68 131 91 23 242 216 403 145 102 1 16 1 1 24 179 109 323 173 370 43 25 52 224 44 208 50 163 ERM eggs by r,c : (Date: 8) Mc Mc Mc Mc Del Del Del Del 225 98 28 30 166 332 250 82 43 54 44 41 130 32 304 172 24 246 110 15 144 371 201 99 184 98 81 86 135 1 12 234 75 306 215 1 1 5 46 200 412 91 1 90 103 176 1 13 58 420 324 338 254 146 144 1 4 1 1 3 310 216 636 185 535 168 66 103 518 1 46 315 97 Z. mali by r,c : (Date: 8) Mc Mc Mc Mc Del Del Del Del 2 0 1 3 1 1 29 7 1 3 0 4 3 13 7 6 34 0 12 17 10 2 6 16 16 9 13 1 6 9 8 6 2 1 1 27 6 26 0 2 4 1 8 27 6 1 4 1 4 6 3 20 1 1 27 7 7 1 8 3 2 4 22 1 3 7 10 6 0 0 C Phytoseiids by r,c: (Date: 10) Mc Mc Mc Mc Del Del Del Del 0 3 1 1 4 2 2 0 1 6 4 0 0 0 2 5 3 0 0 6 0 21 1 1 1 0 2 2 1 0 0 1 24 5 1 1 1 0 0 0 3 1 6 10 4 1 0 2 6 0 1 3 1 0 0 5 0 1 0 0 2 6 0 2 0 4 0 Active ERM by r,c: (Date : 10) Mc Mc Mc Mc Del Del Del Del 17 16 0 0 34 15 17 9 38 4 3 1 10 21 14 10 0 5 1 9 26 34 17 10 5 6 6 1 1 58 34 2 18 2 5 41 12 16 26 5 0 4 5 2 16 34 23 12 3 3 8 0 10 32 15 15 18 20 10 0 7 15 34 16 4 ERM eggs by r,c: (Date: Mc Mc Mc Mc 2 10 0 0 42 7 0 0 0 6 0 3 0 4 0 3 0 1 16 2 1 1 4 1 4 1 0 0 2 16 0 0 0 Z. mali by r,c: (Date: Mc Mc Mc Mc 9 1 1 1 5 6 41 14 7 20 1 2 56 16 1 1 1 41 29 9 35 52 1 1 4 1 7 23 1 3 1 4 30 18 8 15 54 2 8 7 165 10) Del Del Del Del 12 24 10 21 2 4 4 21 1 15 5 8 8 6 0 1 18 16 0 0 9 15 1 0 10 25 4 1 6 19 4 2 10) Del Del Del Del 20 7 0 4 12 8 21 8 13 30 12 1 3 29 25 20 9 2 5 6 16 9 1 18 1 2 12 0 5 6 4 0 2 0 166 Appendix D Treatment totals of predators and prey in the orchard experiment at the Summerland Research Station, 1982. Treatment totals: n = 80 leaves each. SPRAY BAND: WEEDS: Date 1: May 13,14 Typhlodromus adults Active ERM ERM eggs Zetzellia mali  Campvloma verbasei Date 2: May 27,28 Typhlodromus adults Active ERM ERM eggs Zetzellia mali  Campyloma verbasei Date 3: June 10,11 Typhlodromus adults Active ERM ERM eggs Zetzellia mali  Campyloma verbasei Date 4: June 24,25 Typhlodromus adults Active ERM ERM eggs ZetzelUa mali Campyloma verbasei Y Y Y Y N N N N Y Y N N Y Y N N Y N Y N Y Y Y N 23 20. 39 12 33 34 16 23 181 . 419 242 630 212 547 111 146 0 0 0 0 0 0 0 0 17 18 31 23 27 45 28 38 0 0 0 0 0 0 0 0 26 12 22 15 17 20 22 13 122 172 97 194 91 180 113 104 979 1831 563 1764 482 1492 553 1064 27 53 73 37 16 59 55 55 0 0 0 0 0 0 0 0 3 0 1 1 13 17 17 12 237 292 218 415 154 233 135 215 1138 1657 1558 1538 843 1416 621 1028 5 15 15 15 7 12 20 15 0 0 0 1 0 2 0 0 1 1 1 2 14 21 31 24. 1033 1095 850 1210 329 476 250 577 7317 8224 5734 8507 1823 3509 1597 3210 27 67 47 72 38 58 56 66 0 0 0 0 0 0 0 0 Treatment totals: n = 80 leaves each. SPRAY: Y Y BAND: Y Y WEEDS: Y N Date 5: July 8,9 Typhlodromus adults 2 2 Active ERM 12414 9430 ERM eggs 6722 5281 Zetzellia mali 60 87 Campyloma yerbasei 0 0 Date 6: July 22,23 Typhlodromus adults 10 5 Active ERM 4805 3314 ERM eggs 11050 8539 Zetzellia mali 31 85 Campyloma verbasei 3 2 Date 7: August 5,6 Typhlodromus adults 11 2 Active ERM 5954 2146 ERM eggs 4746 2780 Zetzellia mali 95 86 Campyloma verbasei 4 1 Date 8: August 19,20 Typhlodromus adults 5 4 Active ERM 793 731 ERM eggs 1733 1105 Zetzellia mali 31 74 Campyloma verbasei 1 0 Date 10: September 30, October 1 Typhlodromus adults 4 6 Active ERM 196 96 ERM eggs 90 50 Zetzellia mali 132 89 Campyloma verbasei 0 0 Y Y N N N N N N Y Y N N Y N Y N Y N 0 8073 5967 71 0 6 11233 6977 108 0 31 2320 1529 51 0 30 4790 3216 86 0 17 1286 752 94 0 16 2978 3027 58 0 4 4178 7925 92 4 4 4119 7065 95 3 57 1715 9058 86 5 39 1640 6292 124 6 33 1315 5982 86 0 32 1632 6036 83 0 12 4314 3795 121 1 13 3125 3314 104 1 47 4036 4318 168 7 61 2559 2768 174 1 69 2912 3572 210 0 98 3404 3742 169 2 4 611 1172 81 0 4 937 1530 69 0 24 1044 1718 49 0 57 505 1101 40 2 62 864 1502 116 2 67 823 1458 109 1 13 99 26 142 0 5 81 57 132 0 19 112 43 68 0 71 112 80 109 0 22 58 29 109 0 47 110 32 157 0 APPENDIX E Counts of miscellaneous species and stages. Rust mites and McDaniel spider mite, the alternate prey of Typhlodromus, were not present in the orchard during the 1982 experiment until late summer. Treatment totals: n = SPRAY: BAND: WEEDS: Date 1: May 13,14 Typhlodromus eggs Typhlodromus nymphs Apple rust mites McDaniel spider mites Thrips 80 leaves each. Y Y Y Y Y N Y N Y 1 1 4 0 0 0 0 0 0 adults 0 0 0 nymphs 0 0 0 0 0 0 Y N N N N N Y Y N N N Y N Y N 0 3 2 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Date 2: May 27,28 Typhlodromus eggs Typhlodromus nymphs Apple rust mites McDaniel spider mites adults nymphs Thrips 0 1 1 0 0 1 0 0 1 2 0 0 0 2 3 6 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Date 3: June 10,11 Typhlodromus eggs Typhlodromus nymphs Apple rust mites McDaniel spider mites adults nymphs Thrips 0 0 0 0 0 0 0 0 0 0 0 0 8 11 11 11 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Treatment totals: n = SPRAY: BAND: WEEDS: Date 4: June 24,25 Typhlodromus eggs Typhlodromus nymphs Apple rust mites McDaniel spider mites Thrips 80 leaves each Y Y Y Y Y N Y N Y 0 0 0 0 0 0 0 0 0 adults 0 0 1 nymphs 0 0 2 0 0 0 Y N N N N N Y . Y N N N Y N Y N 0 0 0 0 0 0 12 15 19 8 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 Date 5: July 8,9 Typhlodromus eggs Typhlodromus nymphs Apple rust mites McDaniel spider mites adults nymphs Thrips Date 6: July 22,23 Typhlodromus eggs Typhlodromus nymphs Apple rust mites McDaniel spider mites adults nymphs Thrips 0 0 0 0 0 0 0 0 0 1 0 0 6 8 5 3 0 0 0 0 0 20* 0 0 0 0 1 0 0 0 0 0 0 0 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 4 0 0 0 22 8 9 22 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 1 6 1 3 0 2 2 Date 7: August 5,6 Typhlodromus eggs 0 Typhlodromus nymphs 0 Apple rust mites 0 McDaniel spider mites adults 0 nymphs 0 Thrips 3 0 0 0 0 2 0 0 0 0 1 6 7 6 15 0 0 0 52* 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 15 3 5 9 *on one leaf Treatment totals: n = 80 leaves each Date 8: August 19,20 Typhlodromus eggs 0 0 Typhlodromus nymphs 0 0 Apple rust mites 0 20 McDaniel spider mites adults 0 0 nymphs 0 0 Thrips 3 1 Date 10: September 30, October 1 Typhlodromus eggs 0 0 Typhlodromus nymphs 0 0 Apple rust mites 39 0 McDaniel spider mites adults 0 12 nymphs 0 9 Thrips 2 2 APPENDIX F Listing of the simulation model described i n Chapter Four. 1 c 2 C-- SIMULATION OF EUROPEAN RED MITE / TYPHLODROMUS POPULATIONS. 3 C-- WITH PARTICULAR REFERENCE TO 4 C-- DEVELOPMENT. AGE STRUCTURE. PREDATION. WEATHER AND DISPERSAL. 5 C 6 C-- DAN L. JOHNSON. INSTITUTE OF ANIMAL RESOURCE ECOLOGY, 7 C-- AND THE DEPARTMENT OF PLANT SCIENCE. UNIVERSITY OF BRITISH COLUMBIA. VANCOUVER. BRITISH COLUMBIA. 8  —9 C 10 C 11 C 12 C-- PROGRAM STRUCTURE: 13 C-- MAIN PROGRAM WITHIN-STAGE DEVELOPMENT; SURVIVAL; REPRODUCTION. 14 C-- SUBROUTINE PRED PREDATION: FUNCTIONAL.NUMERICAL RESPONSE. 15 C-- SUBROUTINE MERGE TRANSFER BETWEEN LIFE STAGES. ADJUSTING FOR RATES. 16 C-- SUBROUTINE OUTPUT... CONSTRUCTS SUMMARY. TOTALS. PLOT FILE. 17 C 18 C 19 C 20 COMPILE: RUN *FTN SCARDS*=ERMPRED 21 C-- (OR USE OTHER SUITABLE FORTRAN COMPILER.) 22 C 23 C-- RUN -LOAD 1-WINIT 7-BOLA:TCHARTS(90.) 9--PL0T 24 C 25 C-- THIS WILL PROVIDE A RUN BASED ON THE MAX.MIN 26 C-- TEMPERATURES OURING THE PREDATION & DISPERSAL EXPERIMENT. SUMMERLAND. 1982. 27 C-28 29 C 30 INTEGER DISPSW.PREDSW 31 C SWITCHES FOR DISPERSAL. PREDATION 32 C 33 C 34 COMMON EEGG(30).EJUV(3O).EAD(3O).WEGG(30).TIM(3O).TA0(30) 35 C EEGG...DEVELOPMENTAL VECTOR OF ERM SUMMER EGGS 36 C WEGG. . .DEVELOPMENTAL VECTOR OF ERM WINTER EGGS 37 C EJUV...DEVELOPMENTAL VECTOR OF ERM LARVAE & NYMPHS 38 C EAD ...DEVELOPMENTAL VECTOR OF ERM ADULTS. M & F 39 C 40 C TIM ...DEVELOPMENTAL VECTOR OF TYPHLODROMUS E.L.N 41 C TAD ...DEVELOPMENTAL VECTOR OF TYPHLODROMUS ADULTS 42 C 43 C POSITIONS IN THE DEV VECTORS ARE INCREMENTS IN PROPORTION OF 44 C DEVELOPMENT COMPLETED. AND VALUES ARE THE NUMBER OF MITES IN 45 C A PARTICULAR DEVELOPMENTAL STATE (I.E. 0-5%. 5-10% C 90-95*/. OF DEVELOPMENT COMPLETEO. IF N - 1/.05). 46 C 47 C 48 C 49 C 50 C I N I T I A L I Z E AND S E T 51 DO 6 I • 1.30 52 WEGG(I) " O. 53 EEGG(I) • O. 54 TIM(I) - O. 55 TAD(I) • 0. 56 EJUV(I) - 0. 57 6 EAD(I) • O. 58 C C O N S T A N T S 172 5 9 C SET PARAMETERS FOR TVPH IMMATURE DEVELOPMENT (LOGAN,1977) e o P - .0603 R • .0774 TM - 18.9 TB - 1B.3 61 62 63 64 DT - 1.5 65 C 66 C-- READ IN WINTER EGGS TO INITIALIZE POPULATION: 67 READ (1.7) NWCS. (WEGG(I).I - 1.NWCS) 68 7 FORMAT (I3.30F3.0) 69 C 70 WRITE (6.8) 71 8 FORMAT (///'SET DISPERSAL SWITCH:'/' 1 - ON.'/' O - OFF.') 72 READ (6.9)DISPSW 73 9 FORMAT (II) 74 C 75 C-- WHEN DISPSW « 1. A STOCHASTIC DISPERSAL COMPONENT IS ADDED 76 C-- EACH DAY TO TKfttff RANDOMLY CHOSEN AGE CLASSES OF ERM FEMALES. 77 C-- THE ADDITION FOR EACH CLASS IS RANDOMLY CHOSEN BETWEEN 78 C — -1.0 AND 1.0 FROM A REGULAR DISTRIBUTION. 79 C 80 C 81 WRITE (6.12) 82 12 FORMAT (//'SET PREDATION SWITCH:'/' 1 - ON.'/' O • OFF.') 83 READ (6.13)PREDSW 84 13 FORMAT (II) 85 C-- IF PREDSW » 1, THEN PREDATORS ARE PRESENT AND 86 C — MAY EXERT AN EFFECT ON THE ERM. 87 C 88 READ (6,14) TEST 89 14 FORMAT(F2.0) 90 WRITE(6.16) TEST 91 16 FORMAT('ECHO. TEST IS '.F3.0) 92 C 93 ITER • 304 C T U I I I »TF 214 DAYS. APRIL 1 TO OCT 31 94 C-- SIMULA E UtI J I . 95 C-- STARTS ON APRIL 1 » UULIAN *90 96 C-- TIMESTEP IS ONE DAY. BUT DEVELOPMENT IS HOURLY. 97 C 98 0 • .95 99 C-- 95% OF AN AGE CLASS MOVES INTO THE NEXT IF CONDITIONS ALLOW. 100 C 101 EATPER - 10. 102 EATYST - 10. 103 C-- INITIALIZE NUMERICAL & STARVATION RESPONSES. 104 C-- EATPER IS CALCULATED OAILY FROM HERE ON. 105 C 106 WEX » 6. 107 C-- EXPONENT IN WINTER/SUMMER EGG DETERMINATION. 108 C 109 N - 20 110 C-- TWENTY DIVISIONS IN DEVELOPMENT VECTORS. 111 C 112 PI • 3. 1415926 113 C 1 14 C 115 C-- ZERO THE DEVELOPMENT RATE OVERRUNS. 116 C THESE ARE SAVED SO TRUNCATION ERRORS ARE AVOIDED. 173 117 OVERW • 0. 118 OVERE - O. 119 OVERJ - O. 120 OVREA - O. 121 OVRETI - 0. 122 OVRETA • O. 123 C 124 C 125 C-- ZERO THIS YEAR'S WINTER EGG CROP. 126 WCROP - 0. 127 C 128 C 129 C 130 C 131 C 132 C 133 C BEGIN TIME ITERATION 134 DO 1000 IDAY - 90,ITER 135 C 136 C 137 C 138 IF (IDAY.GT.95) GOTO 20 139 C-- WAKE UP SOME DIAPAUSING PREDATORS. 140 DO 18 I-1.N 141 18 TAD(I)«TAD(I)*0.5 142 C 143 20 CONTINUE 144 C 145 C 146 C 147 IF (IDAY.LT.140) GOTO 30 148 IF (DISPSW.EO.O) GOTO 30 149 C-- DISPERSAL SWITCH: 1-YES, 0=NO 150 C 151 C ERM IMMIGRANTS... 152 DO 28 ID=1,3 153 IR=IFIX(FRAND(0.0)»20.) 154 DISP=FRAND(1.O) 155 EAD(IR)«EAD(IR)+DISP 156 28 CONTINUE 157 C 158 30 CONTINUE 159 C 160 C 161 C-- READ DATE AND TEMPERATURE FIELD DATA. 162 READ (7.32) NY,ND.NM,TMAX.TMIN 163 TMAX"=TMAX+TEST 164 TMIN=TMIN*TEST 165 32 FORMAT (6X.31 2.2F6.1) 166 TBAR « (TMAX+TMINJ/2 167 C APPROXIMATE A 24-HOUR TEMPERATURE FLUCTATION. 168 C (A VARIETY OF PERIODIC OR FOURIER SERIES WILL DO.) 169 W • 2*PI/24 170 AMP » (TMAX-TMIN)/2 171 C 172 C--CALCULATE PROPORTION DEVELOPED TODAY. 173 RW - O. 174 RE • O. 174 175 RJ - O. 176 RAO • O. 177 RTI - 0. 178 RTA • O. 179 C 180 DO 100 IHR > 1.24 181 C CALCULATE TEMPERATURE. DEVELOPMENT HOUR BY HOUR. 182 T • TBAR*AMP*SIN(W,FLOAT(IHR)) 183 C 184 C ERM 185 ADDW » 1/(24*EXP(5.0326-.10225*T)) 186 ADDE - 1/(24»EXP(5.459-.1548»T)) 187 ADDJ • 1/(24»EXP(4.982-.141*T)) 188 ADDA • 1/(24*EXP(3.962-.117*T)) 189 C 190 IF (A0DW.LT.0.0O05) ADDW - O. 191 IF (ADDW.GT.0.01) ADDW • .01 192 C 193 IF (ADDE.LT.0.0004) ADDE - O. 194 IF (ADDE.GT.0.04) ADDE • .04 195 C 196 IF (ADDJ.LT.0.0006) ADDJ - O. 197 IF (ADDJ.GT.0.04) ADDJ « .04 198 C 199 IF (A0DA.LT.0.OO1) ADDA - O. 200 IF (ADDA.GT.0.048) ADDA • .048 201 C 202 C TYPHLODROMUS 203 TT • T-TB 204 TAU » (TM-TT)/DT 205 RATE » AMAX1( P»(EXP(R*T) - EXP(R*TM-TAU)). O.) 206 ADDTI • RATE/24 207 ADDTA - l/(24*(25.60 -.462*T)) 208 C 209 IF (ADDTA.GT.0.01) ADDTA".Ol 210 IF (ADDTA.LT.0.001) ADDTA-O. 2 11 C 2 12 C ADULTS ARE FAIRLY LONG-LIVED. AND DO NOT HAVE AS 213 C WIDE A RANGE OF AGING RATES AS THE PREY. 2 14 C 215 C 216 C ACCUMULATE A TOTAL FOR THE DAY. 217 RTI - RTI+ADDTI 218 RTA - RTA+ADDTA 219 RW « RW+ADDW 220 RE • RE+AODE 221 RJ « RJ+ADDJ 222 lOO RAD ' RAD+ADDA 223 C 224 C-- UPDATE DEVELOPMENTAL/AGE CLASS VECTORS. 225 C 226 PTI « FLOAT(N)*RTI+OVERTI 227 PTA « FLOAT(N)*RTA*OVERTA til PW - FLOAT(N)»RW+OVERW it% PE - FLOAT(N)»RE*OVERE l i l Pd • FLOAT(N)»RJ*OVERJ 23, PA - FLOAT(N)'RAD+OVREA 232 175 233 C THE INCREMENT OF DEVELOPMENT (POSTIONS MOVED TOOAY): 234 INCTI - INT(PTI) 235 INCTA - INT(PTA) 236 INCW • INT(PW) 237 INCE « INT(PE) 238 INCJ - INT(PJ) 239 INCA « INT(PA) 240 C-- SAVE OVERRUN FOR NEXT DAY. 241 C-- OVERRUN IS DIFF BETWEEN DEVELOPMENT AND TRUNCATION. 242 OVERTI - PTI-INCTI 243 OVERTA « PTA-INCTA 244 OVERW « PW-INCW 245 OVERE » PE-INCE 246 OVERJ • PJ-INCJ 247 OVREA • PA-INCA 248 C 249 C 250 C-- NOW UPDATE AGE CLASS VECTORS. 'INC IS THE INCREMENT 251 C-- OF ADVANCE IN THE AGE CLASS VECTOR FOR A PARTICULAR DAY. 252 C 253 NMAX • N+INCTI 254 DO 250 I » 1.N 255 K » NMAX+1-I 256 250 TIM(K) - (1-Q)*T1M(K) + Q*TIM(K-INCTI) 257 C 258 NMAX « N+INCTA 259 DO 260 I - 1.N 300 EEGG(K) • (l-Q)'EEGG(K) • Q*EEGG(K-INCE) 260 K - NMAX+1-I 261 260 TAD(K) - (1-Q)*TAD(K) + 0*TAD(K-INCTA) 262 C 263 C-- WINTER EGGS DO NOT DEVELOP TO HATCHING 264 C-- LATER THAN EARLY SUMMER, SO DON'T BOTHER UPDATING: 265 IF (IDAY.GT.170) GOTO 290 266 NMAX «= N+INCW 267 DO 280 I • 1,N 268 K = NMAX+1-I 269 280 WEGG(K) • (1-Q)*WEGG(K) + 0*WEGG(K-INCW) 270 C 271 290 NMAX = N+INCE 272 DO 300 I « I.N 273 K «= NMAX+1-I 274 ' 275 C 276 NMAX = N+INCJ 277 DO 310 I • 1.N 278 K «• NMAX+1-I 279 310 EJUV(K) • (1-Q)*EJUV(K) + Q*EJUV(K-INCJ) 280 C 281 C 282 NMAX • N+INCA 283 DO 320 I » I.N 284 K » NMAX+1-I 285 320 EAO(K) « (1-Q)*EAD(K) + Q*EAD(K-INCA) 286 C-- (AOULTS WHICH GRADUATE OUT OF THEIR AGE 287 C-- CLASS VECTOR ARE DEAD.) 288 C 289 C 290 C 176 291 C 292 C-- ZERO THE CLASSES (VECTOR POSITIONS) THAT HAVE 293 C-- OPENED UP AS THE VECTORS MOVE FORWARD. 294 C-- (NOTE THAT THERE IS NOTHING TO FILL SOME 295 C-- VECTORS. FOR EXAMPLE. THE WINTER EGGS.) 296 C 297 DO 332 I • 1.INCW 296 332 WEGG(I) - 0. 299 C 300 DO 334 I » 1 . INCE —' » * « o 301 304 334 EEGG(I) • 0. l ° 0 l C DO 336 I • 1.INCJ 336 EJUV(I) « 0. 305 306 DO 338 I « 1.INCA 307 338 EAD(I) • O. 308 C 309 DO 340 I « 1.INCTI 310 340 TIM(I) - O. 31 1 C 312 DO 342 I » 1.INCTA 313 342 TAD(I ) = 0. 314 C 315 C-- NOW TRANSFER THOSE WHO HAVE COMPLETED DEVELOPMENT 316 C-- TO THE NEXT CLASS. DIFFERENCES IN REALIZED DEVELOPMENT 317 C-- PER DAY ARE ACCOUNTED FOR BY SQUEEZING OR STRETCHING 318 C-- THE DEV STRUCTURE OF THE GRADUATES TO MATCH THE 319 C-- DEV RATE OF THE NEXT AGE CLASS. I.E. N POSITIONS INTO M. 320 C 321 C 322 IF (IDAY.GT.170) GOTO 350 323 C — ...SO AS NOT TO WASTE EFFORT UPDATING AN 324 C-- EMPTY VECTOR OF WINTER EGGS. 325 C 326 C--TRANSFER ERM WINTER EGGS HATCHING THIS DAY. 327 C-- PRESERVING HATCHING SEQUENCE... 328 CALL MERGE(INCW.INCJ,N,WEGG,EJUV) 329 C 330 C--TRANSFER ERM SUMMER EGGS HATCHING THIS DAY. 331 C-- PRESERVING HATCHING SEQUENCE... 332 350 CALL MERGE(INCE,INCJ.N,EEGG,EJUV) 333 C 334 C--TRANSFER ERM NYMPHS MOLTING THIS DAY. 335 C-- PRESERVING MOLTING SEQUENCE... 336 CALL MERGE(INCJ,INCA.N.EJUV,EAD) 337 C 338 C--TRANSFER IMMATURE PHYTOSEIIDS WHO HAVE COMPLETED 100% 339 C-- OF THEIR DEVELOPMENT TO THE ADULT ARRAY. 340 C — PRESERVING MOLTING SEQUENCE... 341 CALL MERGE(INCTI,INCTA.N.TIM,TAD) 342 C 343 C 344 C-- NOW KILL ADULTS WHO HAVE COMPLETED 100% OF THEIR 345 C-- LIFESPAN TODAY, AND ZERO TRANSFER POSITIONS OF OTHERS. 346 C 347 348 N1 - N+1 177 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 360 370 375 380 384 NMAX « N+INCW 00 360 I • N1.NMAX WEGG(I) - 0. NMAX « N+INCE DO 370 I • N1,NMAX EEGG(I) > O. NMAX " N+INCJ DO 375 I • N1,NMAX EJUV(I) - 0. NMAX -DO 380 EAD(I ) NMAX • DO 384 TIM(I) N+INCA I « N1,NMAX - O. N+INCTI I « N1.NMAX • 0. NMAX » N+INCTA DO 388 I - N1.NMAX O. 388 TAD(I) C-- IF LATER THAN DAY 170 (MID-JUNE), C-- HAVE NOT YET HATCHED MUST DIE. IF (IDAY.NE.170) GOTO 385 DO 386 I - 1.N 386 WEGG(I) • O. 385 CONTINUE WINTER EGGS C--C--C--c--C-NOW TRANSFER FECUNDITY FROM OVIPOSITING ERM ADULTS... FEC - O. FECUNDITY IS A FUNCTION OF FEMALE AGE AND TEMPERATURE. I USE A FUNCTION FIT TO DATA OF RABBINGE, 1976, CONFIRMED AND ADJUSTED TO MY LAB DATA. DO 390 I » 1.N X « TBAR**4 XCURV « 20**4 MAXIMUM POTENTIAL EGGS LAID IN ONE DAY « 6. EMAX « (6*X)/(XCURV*X) A = FLOAT(I) AGE CLASS (1 N » 20) EGPER » EMAX*(A-3)*EXP(-.35*(A-3)) LAG 3 SINCE THE FIRST 3 AGE CLASSES ARE PRE-OVIPOSITIONAL. IF (EGPER.LT.1.E-10) EGPER • 0. FECAD " EAD(I)*0.7 SEX RATIO: 70*/. FEMALES. 390 FEC • FEC+EGPER»FECAD WRITE (13.391) IDAY,EGPER 391 FORMAT ('IDAY: '.13.', EGPER: '.F6.2) -- LEAVES ARE VERY NUTRITIOUS IN EARLY SPRING. -- AND LESS SO IN SUMMER. TAPERING OFF UNTIL :-- AUTUMN. LEAF BIRTH AND GROWTH RATES FOLLOW C--1 7 8 407 C-- THE SAME TREND. 408 C 409 DAY » FLOAT(IDAY) 410 OUAL « ((DAY-100)*EXP(-.02*(DAY-100)))-.5 411 C-- NOTHING FOR ERM TO EAT BEFORE DAY 100 AT EARLIEST. 412 C 413 IF (OUAL.LT.O.) OUAL • O. 414 C-- BEST APPROXIMATION OF PRESENCE. GROWTH AND NUTRITIVE 415 C-- VALUE OF APPLE LEAVES. 416 F • FECQUAL/10 417 C 418 C-- ONLY SUMMER EGGS ARE LAID UNTIL AUGUST. AFTER AUG 1. 419 C-- (213) A PROPORTION OF THE FEMALES LAY SUMMER EGGS AND 420 C-- THE REST LAY DIAPAUSE EGGS. THE PROPORTION LAYING 421 C-- WINTER EGGS APPROACHES 1.0 IN OCTOBER. 422 C 423 C-- 0% WINTER EGGS LAID AUG 1. 424 D » AMAX1(FL0AT(IDAY)-213.0.) 425 C-- D IS THE * DAYS PAST THE BEGINNING OF AUG. 426 C-- 213 + 45 IS THE DATE BY WHICH ABOUT 50'/. OF 427 C-- THE EGGS BEING PRODUCED ARE DIAPAUSE EGGS. 428 PROW » (D**WEX) / (45*»WEX + D**WEX) 429 C-- PROW IS THE PROPORTION OF THE EGGS LAID THAT 430 C-- DAY WHICH ARE WINTER EGGS. THE PROPORTION 431 C-- OF THE EGGS WHICH ARE DIAPAUSE EGGS INCREASES IN A C-- TYPICAL LOGIT FASHION TO A LATE SEASON MAXIMUM OF 100% (BASED ON LITERATURE AND MY OWN OBSNS). FW • F*PROW 4 » FS • F-FW 436 C-- WRITE (2.399)PROW.F.FS.FW 437 399 FORMAT ('PROP WINTER. F. FS, FW: '.4F8.3) 438 C 439 C-- LAY TODAY'S SUMMER EGGS. 440 IE • INCE 441 IF (IE.LT.2) IE • 2 432 433 C — 434 435 DO 392 I " 1.IE 392 EEGG(I) - EEGG(I)*FS/IE 442 443 444 C 445 C-- LAY TODAY'S WINTER EGGS. IF ANY. 446 C-- USE WCROP AS A BASKET TO STORE THIS YEAR'S WEGGS. 447 WCROP « WCROP+FW 448 C 449 C 450 IF (PREDSW.EO.1) CALL PRED(IDAY,TBAR.N.EATPER) 451 C 452 C-- FECUNDITY BASED ON RECENT AVERAGE FOOD. 453 EATPER*(EATPER+EATYST)/2 454 C 455 TFEC »(EATPER/(5+EATPER)) 456 C LITTLE OR NO EGG PRODUCTION FROM STARVING FEMALES... 457 C HOWEVER. SOME TYPH EMERGING IN MAY HAVE 458 C EGGS OR THE ENERGY TO PRODUCE THEM FROM 459 C THE PREVIOUS YEAR'S FEEDING. 460 IF (IDAY.LT.140.AND.TFEC.EO.O.) TFEC-0.2 461 C 462 C-- TYPHLODROMUS LAYS UP TO ABOUT ONE EGG/DAY ON AVERAGE. 463 C 464 TF • 0. 179 465 C 466 DO 450 I • 1.N 467 450 TF • TF+TAD(I)*TFEC 468 C 469 IF (PREDSW.EO.O) TF-O. 470 C-- NO GROWTH OF PREDATORS IF NO PREDATION. 471 C-- A SWITCH FOR RUNS WITHOUT PREDATORS. 472 IT - INCTI 473 IF (IT.LT.2) IT - 2 474 DO 460 I • 1,IT 475 460 TIM(I) » TIM(I)*TF/IT 476 C 478 c ~ WRITE (6,351)INCW.INCE.INCJ.INCA.INCTI INCTA 1?5 351 FORMAT CINC W.E.J.AD:'.416.'. ' 2 1 6 ) 480 C 481 C 482 C-- APPLY SURVIVORSHIP/DAY MULTIPLIERS. 483 SURVTA • AMAX1((.9-.01*TBAR), EATPER/(0.2+EATPER)) 484 IF (SURVT.GT.99) SURVT-.99 485 C SURVIVAL PER DAY IS ABOUT 80% FOR STARVING ADULTS, 486 C DEPENDING ON TEMPERATURE. 487 IFUDAY.GT.273) SURVTA-.9 488 C BEGIN DIAPAUSE PREP IN OCT... FUNCTION UNKNOWN. 489 SURVTI • 0.995 490 SURVE "0.99 491 SURVJ « 0.98 492 IF (TBAR.GT.30.) SURVE -0.7 493 IF (TBAR.GT.30.0R.TBAR.LT.10) SURVJ • 0.8 494 DO 400 I « 1,N 495 TIM(I) « TIM(I)*SURVTI 496 TAD(I ) « TAD(I) *SURVTA 497 WEGG(I) » WEGG(I)*SURVE 498 EEGG(I) • EEGG(I)*SURVE 499 EJUV(I) = EJUV(I)»SURVd 500 C-- APPROXIMATE ADULT MORTALITY AS F( AGE. TIME. TEMPERATURE). 501 C-- (FROM LITERATURE. MAINLY). 55§ 2 3 SURVA F.T{(ioi852.AGE.AGE+o00061.T.T-.3053) l°l IF (SURVA.GT.0.99) SURVA - 0.99 505 400 EAD(I) • EAD(I)*SURVA 506 C 507 C 508 C 509 C TRUNCATE DENSITIES. . . c . n DO 500 1=1.N t\°, IF (EEGG(I).LT.1.E-4) EEGG -O. 11 IF WEGG(I).LT.1.E-4) WEGG -O. 5a IF EJUV(I).LT.1.E-4) EJUV(I)-0. HI IF (EAD(I).LT. 1.E-4) EAD(I)-0. 5 1 4 IF TIM(I).LT.1.E-4) TIM(I -0. 500 IF (TAD(I).LT.1.E-4) TAD(I)-0. 515 516 517 C 518 C 519 C \ 180 562 563 523 C-- FUNCTION IS UNKNOWN, BUT IS CONSIDERED HERE AS BEING 1%/DAY. 524 CALL OUTPUT(IDAY,TMAX,TMIN.RW.RE,Rd.RAD,F,N.WCROP,ND,NM,NY) 525 C 526 C 527 EATYST-EATPER 528 C 529 C-- READ NEXT DAY'S TEMPERATURE AND 530 1000 CONTINUE 531 STOP 532 END 533 C 534 C MAIN PROGRAM ENDS 535 C 536 C 537 C 538 C 539 C 540 C 541 SUBROUTINE MERGE(NFROM,NTO.N.FROM,TO) 542 C-- MERGES HATCHERS OR MOLTERS INTO NEXT STAGE. 543 C-- PRESERVING SEOUENCE AND AGE STRUCTURE DUE 544 C-- TO DIFFERENT AMOUNTS OF DEVELOPMENT ATTAINED 545 C-- BY DIFFERENT STAGES ON A GIVEN DAY (F(TEMP)). 546 C 547 C-- THANKS TO MIKE PATTERSON FOR PROGRAMMING 548 C-- ADVICE ON THIS SUBROUTINE. 549 C 550 REAL FR0M(3O).T0(30) 551 IF (NFROM.LE.O) RETURN 552 IF (NTO.LE.O) NTO • 1 553 NOVER • N+NFROM 554 NOFIRS • NOVER-NFROM-M 555 C 556 C-- WRITE (6.12) (FROM(I). I - NOFIRS. NOVER) 557 12 FORMAT ('FROM:'.10(1X. F6.1)) 558 C-- AEND IS THE END OF THE GROUPS MOVED AT THE PREVIOUS STEP. 559 C-- AMOVE IS THE NUMBER OF GROUPS TO MOVE INTO EACH 'TO' GROUP: 560 GEND « NOFIRS 561 GMOVE • FLOAT(NFROM)/FLOAT(NTO) 562 C-- THE AMOUNT MOVED TO THE PREVIOUS 'TO' GROUP: PREV « O. 564 C 565 C-- MOVE GMOVE FROM 'FROM' TO EACH 'TO': 566 DO 160 I • 1, NTO 567 GEND • GEND+GMOVE 568 C-- BREAK THIS INTO A WHOLE GROUP AND ANY FRACTIONAL GROUP: 569 I END « GEND 570 GPART - GEND-FLOAT(IENO) 571 C-- DON'T MOVE A FRACTION AND THE WHOLE OF THE LAST GROUP: 572 I END » IEND-1 573 C-- COUNT ALL FROM GROUPS UP TO GPART: 574 AMOVE • O. 575 IF (IEND.LT.NOFIRS) GOTO 140 576 DO 120 d - NOFIRS. I END 577 120 AMOVE • AMOVE+FROM(U) 578 C-- ADD IN THE FRACTIONAL PART IF ANY. 579 140 IF (GPART.GT.1.E-10) AMOVE « AMOVE*GPART«FROM(I END*1) 580 C-- THIS 'TO' GROUP HAS THE CURRENT VALUE LESS THE VALUE OF THE 181 5B1 C-- P R E V I O U S 'TO' GROUP. 5 8 2 T O ( I ) « AMOVE-PREV 5 8 3 C-- REMEMBER THE CURRENT VALUE FOR THE NEXT 'TO' GROUP. 584 160 P R E V - AMOVE C — WRITE ( 6 . 13) ( T O ( I ) . I • 1. NTO) 13 FORMAT ('TO : ' . 1 0 ( 1 X . F6.D) 5 8 5 C 5 8 6 5 8 7 5 8 8 RETURN 5 8 9 END 5 9 0 C 591 C 5 9 2 C 5 9 3 C 5 9 4 C 5 9 5 C 5 9 6 C 597 C 5 9 8 SUBROUTINE P R E D ( I D A Y , T B A R , N , E A T P E R ) 5 9 9 COMMON E E G G ( 3 0 ) . E J U V ( 3 0 ) . E A D ( 3 0 ) . W E G G ( 3 0 ) . T I M ( 3 0 ) . T A D ( 3 0 ) 6 0 0 C 601 C-- LARVAE DO NOT FEED. PREY ARE ERM J U V E N I L E S . 6 0 2 C-- PUTMAN AND HERNE, 1 9 6 4 . CAN. ENT. 9 6 : 9 2 5 - 9 4 3 . 6 0 3 C 6 0 4 C 6 0 5 C-- T. C A U D I G L A N S EATS A MAX OF ABOUT 10.0 JUV/DAY. 6 0 6 I F ( T B A R . G T . 0 . ) GOTO 8 6 0 7 E A T « 0 . 6 0 8 E A T P E R « 0 . 6 0 9 RETURN 6 1 0 C 6 1 1 C 6 1 2 8 DEM « 20*TBAR / ( 1 5 . + T B A R ) 6 1 3 S R A T E - ,05*TBAR / ( 1 5 . + T B A R ) 6 1 4 C 6 1 5 S U M J » O. 6 1 6 SUMT = O. 6 1 7 R E M A I N - 0. 6 1 8 C 6 1 9 DO 10 I ' 1.N 6 2 0 SUMJ ' S U M J * E J U V ( I ) 621 10 SUMT • S U M T + T A D ( I ) 6 2 2 C 6 2 3 C 6 2 4 C I F NO PRED OR NO PREY. RETURN 6 2 5 I F (SUMJ.GT.O.AND.SUMT.GT.0.) GOTO 15 6 2 6 E A T P E R » 0 . 6 2 7 EAT=0. 6 2 8 R E M A I N ' S U M J 6 2 9 RETURN 6 3 0 C 631 C 6 3 2 C 6 3 3 C 6 3 4 6 3 5 15 NR - I F I X ( . 6 * N ) * 1 .rr-r-r C-- ONLY THE LAST OF THE IMMATURE C L A S S E S 6 3 6 C-- ARE PREDATOR NYMPHS. LARVAE DO NOT FEED 6 3 7 DO 2 0 I - NR.N 6 3 8 2 0 SUMT - SUMT«-TIM( I ) 182 €39 C 640 C-- PREDATION BY JUVENILE T. CAUDIGLANS IS NOT SUBSTANTIAL. 641 C-- BUT COULD BE UNDER DIFFERENT SCENARIOS... 642 C 643 EE - (DEM*SUMT/SUMJ)«( 1-EXP(-SRATE*SUMJ/DEM)) 644 EAT » SUMJ*(I-EXP(-EE)) 645 C 646 EATPER - EAT/SUMT 647 C 648 DO 30 I • I.N 649 EJUV(I) « EJUV(I)-EAT*EJUV(I)/SUMJ 650 30 REMAIN - REMAIN+EJUV(I) 651 C 652 C100 WRITE (4.50) IDAY.SUMJ.EATPER.EAT.REMAIN 653 100 CONTINUE 654 50 FORMAT ('DAY*'.13. ' J : '.F8.1.'. '.F5.2.' EACH; '.F8.1.' EATEN, , 655 &F9. 1 , ' REMAIN.') 656 C-- WRITE (11.999) IDAY.SUMJ,SUMT.EATPER 657 999 FORMAT ('DAY*'.14.'SUMJ.T • '.2F8.4.', EATPER • '.F8.4) 658 RETURN 659 END 660 C 661 C 662 C 663 SUBROUTINE OUTPUT(IDAY.TMAX,TMIN,RW.RE.RJ.RAD,F,N.WCROP,ND.NM,NY) 664 COMMON EEGG(30).EJUV(30).EAD(30),WEGG(30).TIM(30),TAD(30) 665 C — SUM STAGES AND WRITE OUR SUMMARY. 666 SUMW-O. 667 SUME - O. 668 SUMJ = 0. 669 SUMA - 0. 670 SUMTI • O. 671 SUMTA « O. 672 DO 5 I » I.N 673 SUMTI « SUMTI+TIMU) 674 SUMTA » SUMTA+TAD(I) 675 SUMW=SUMW+WEGG(I) 676 SUME « SUME + EEGGO) 677 SUMJ « SUMJ+EJUV(I) 678 5 SUMA « SUMA+EAD(I) 679 C 680 C-- ADD IN THIS YEAR'S WINTER EGG CROP (LAID AFTER AUG) 681 SUMW • SUMW+WCROP 682 C 683 FEM * SUMA*0.7 684 C-- ERM POPULATION IS 70% FEMALES. 685 C 686 C-- LOG TRANSFORM FOR PLOTTING. 687 SUMW-ALOG(SUMW+1) 688 SUME=ALOG(SUME+1) 689 SUMJ=AL0G(SUMJ+1) 690 SUMA=AL0G(SUMA+1) 691 SUMTI«ALOG(SUMTI*1) 692 SUMTA=AL0G(SUMTA+1) 693 C 694 C 695 C-- WRITE (6.10)ND,NM,NY,IDAY.TMAX.TMIN 696 10 FORMAT (//I4.'/'.12.'/'.12.' (DAY* '.13.'). T MAX. TMIN: '.2F5.1) 183 697 C--698 15 699 C 700 C--701 C — 702 c — 703 704 c--705 706 16 707 C--708 C 16 709 C — 710 C--711 C--712 C--713 C--714 C--715 20 716 717 718 719 C--720 23 721 C 722 c 723 25 724 C 725 726 WRITE (9.15) WCROP,SUME,SUMJ,SUMA FORMAT (4F8.2) WRITE (3.16)N0,NM.NY.SUMW WRITE (4. 16)N0.NM.NY.SUME WRITE (5.16)N0.NM,NY.SUMJ WRITE (8.16)ND,NM.NY.SUMA WRITE ( 14. 16)ND.NM,NY.SUMTI WRITE (15.16)ND.NM,NY.SUMTI FORMAT (I2.'/'.I2.'/'.I2.'.'.F12.9) WRITE (9,16)SUMW,SUME,SUMJ,SUMA,SUMTI,SUMTA FORMAT (6F10.6) WRITE (6.20) (WEGG(I).I -1.N) WRITE (6.20) (EEGG(I).I - 1.N) WRITE (6.20) (EJUV(I).I • 1.N) WRITE (6.20) (EAD(I).I « 1.N) WRITE (6.20) (TIM(I).I • 1.N) WRITE (6,20) (TAD(I).I • 1.N) FORMAT (20F7.1) IF (FEM.EO.O.) EPER • O. IF (FEM.EO.O.) GOTO 23 EPER » F/FEM WRITE (6.25) IDAY.RW.RE.RJ.RAD.EPER CONTINUE FORMAT ('DAY* '.14,'. RW.RE.RJ.RAD, EGGS PER FEM: '.5F6.2) RETURN END Author Anderson,N.H. 54 Arrand,J.C. 2,3,10,11,88,44, 70,72 Asquith,D. 55,93 Avery,D.J. 92 Barfield.C.S. 13 Baskerville,G.L. 139 Baumgaertner,J. 59 Beirne,B.P. 21 Berryman,A.A. 98 Blatt,D.W.E. 109 Bliss,C.I. 55 Bradner,L. 9 Brickhill.C.D. 72 Briggs,J.B. 92 Brown,A.W.A. 10 Brown,L.G. 106 Browne,R.W. 92,100,106 Bruinsma,J. 94,95 Brunner.J.F. 139 Cagle,L.R. 103 Caltagirone,L.E. 9,21,44 Cameron,P.J. 13 Chant,D.A. 2,4,32,54,107 Chapman,R.B. 34,39 Clark,W.C. 110 Collyer,E. 39 Comins,H.N. 109 Cone,W.W. 18 Cranham,J.E. 8,9 Croft,B.A. 7,10,14,18,23,32, 37,38,39,55,93,98 Crowley,P.H. 109 Curry,G.L. 106 Davis,D.W. 106 Demichele,D.W. 106 Dohse,L. 13 Dover,M.J. 55,98 Downing,R.S. 3,4,8,9,10,11,18, 28,30,38,44,70,72,76,112 Easterbrook,M.A. 8,9 Emin,P. 137 Eveleigh,E.S. 107 Index Fisher,R.A. 18,55 Flaherty,D.L. 18,34,39 Flint,W.P. 3 Fransz,H.G. 98 Frazer,B.D. 59,97,98,100,106,139 Gadsby,M.C. 14 Gerson,U. 3 Getz.W.M. 98 Gilbert,N. 59,97,98,100,106,139 Glass,E.H. 93 Groves,J.R. 3 Gurney,W.S.C. 109 Gutierrez,A.P. 59,106,108 Hall.F.R. 9 Hantsbarger.W.H. 9 Harries,F.H. 21 Hassell,M.P. 109 Hastings,A. 109 Helle,W. 9 Herbert,H.J. 7,41,101,103 Herman,S.G. 14 Herne,D.C. 8,9,18,33,39,51,99,100, 101,102,104,107,138,139 Hesper.B. 109 Hilborn,R. 109,114 Hogeweg.P. 109 Holling,C.S. 98,110 Horn,H.S. 13 Howard,L.O. 3 Hoy,B. 6 Hoy,M.A. 18,63,98 Hoying,S.A. 23,38 Hoyt,S.C. 9,18,21,44,70,93,100,106, 112,139 Hudson,W.B. 21 Hueck,H.J. 18 Huffaker,C.B. 2,4,7,8,9,11,14,18,21, 34,39,54,109 Hull,L.A. 55 Hussey,N.W. 7,54,81,82 Huxley,T.H. 40 Ito.Y. 136 Jaeger-Draafsel,E. 18 Jeppson,L.R. 18 Johnson,C.G. 13 Johnson,D.L. 61 Johnson,D.T. 14,23,37,38 Johnston,D.E. 81 Jones,R.E. 97,98 Krebs,C.J. 13 Kuenen,D.J. 18 Laing,J.E. 106 Lee,M.S. 106 LeRoux,E.S. 7,9 Levin,S.A. 109 Logan,J.A. 98,100,103,106,107, 110 Lord,F.T. 7,9,10,70 Lund,C.T. 99,100,101,139 MacArthur.R.H. 13 MacPhee,A.W. 7 Madsen,H.F. 2 Marle,G. 80,81,82 Marshall,J. 1,4,6,7,8,9 McGroarty,D.L. 39 McMurtry,J.A. 2,4,7,8,9,11,14, 18,21,39,54 McRae,K.B. 101 Mead,R. 61 Metcalf,C.L. 3 Meyer,R.H. 39 Miraura,M. 109 Mitchell,R. 81 Miyashita,K. 13 8 Moilliet,T.K. 4,8,28,30,38,44, 76,112 Morgan,C.V.G. 54 Morris,R.F. 13 Morrison,G. 55 Mowery,P.D. 55 Murai,M. 82 Murdoch,W.W. 109 Murray,J.D. 109 Myers,J.H. 13 Nachman,G. 54 Nisbet,R.M. 109 Nishino,M. 114 Oaten,A. 109 Oatman.E.R. 54 O'Neil.W.J. 9 Orchard,0. 80 Parent,B. 10,70 Parr,W.J. 81,82 Patterson,N.A. 9 Penman,D.R. 34,39 Perry,J.N. 61 Petkau,A.J. 61 Pickett,A.D. 7,9 Pielou,D.P. 8,21,55 Pienaar.L.V. 98 Platnick,N.I. 80 Potter,D.A. 81 Pritchard,A.E. 39 Putman,W.L. 7,9,18,33,39,51,80, 104,106,107 Rabbinge,R. 1,81,98,103,104,105 Readshaw,J,L. 4 Regniere,J. 100 Renshaw,E* 109 Rice,R.E. 93 , Roff,D.A. 109 Rosenberg,N.J. 13 Roush,R.T. 18 Ruesink,W.G. 98,100 Sabelis,M.W. 14,54,98,106,107 Sanford,K.H. 18 Santos,M.A. 70 Schuster,R.O. 39 Shea,K.P. 14 Shelford.V.E. 138 Skeith,R. 106 Skellam,J.G. 109 Smilanik,J.M. 63 Smith,J.M. 109 Southwood,T.R.E. 13 Stark,R.W. 98 Stearns,S.C. 13 Steer,W. 80 Stewart,P.G. 18 Stimac,J.L. 13 Stinner,R.E. 13 Strong,D.R. 55 Stulte,N.T. 9 Summers,C.G. 59 Takafuji,A. 14,114 Tanigoshi,L.K. 92,100,106 Taylor,L.R. 55 Thomas,W.P. 39 Thompson,W.A. 13 Treherne.R.C. 1,6 Trottier,R. 101,102,138 Tummala,R.L. 98 186 van den Boer,P.J. 18 Vandermeer,J.H. 109 van de Vrie.M. 2,3,4,7,8,9,11, 14,18,21,39,54 Vertinsky.I.B. 13 Wang,Y. 106,108 Watt,K.E.F. 98 Wearing,C.H. 39 Welch,S.M. 55,98 Wellington,W.G. 13,38 White,N.D.G. 70 Woiwood,I. 55 Wol£enbarger,D,0, 81 Wollkind,D.J. 98,100 Wrensch,D.L. 81 Wright,M.A. 139 Yamada,H. 138 Yates,F. 18 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0095862/manifest

Comment

Related Items