OVERDUE pINES ARE 25¢ PER DAY PER ITEM Return to book drop to remove this checkout from your record. DEVELOPMENT OF IMPROVED DENSITY ESTIMATORS FOR LARVAE OF THE CEREAL LEAF BEETLE, OULEMA MELANOPUS (L.) By Patrick A. Logan A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Entomology 1977 ABSTRACT DEVELOPMENT OF IMPROVED DENSITY ESTIMATORS FOR LARVAE OF THE CEREAL LEAF BEETLE, OULEMA MELANOPUS (L.) By Patrick A. Logan Good pest management decisions begin with an accurate estimate of pest density and rely on close predictions of future densities. Sweepnet sampling was evaluated as a low cost method of arriving at an index of density. Attempts at determining instar specific multiplication factors to convert catch per sweep to catch per square foot lacked precision. Incorporation of degree days, windspeed, rainfall and crop height into the conversion factors did little to improve the model. Use of logarith- mic transforms (i.e., a multiplicative model) improved the characteris- tics of the statistical residuals slightly. An additional data set was also evaluated, but this served only to cloud the picture, casting doubt on the generality of the conversion factors. "Two linear feet" or square foot quadrat samples were also evalu- ated. Instability of the variance of egg and larval counts was charac- terized by a double log second order relationship with the mean. Taylor's "power law" was brought to question as a result and was found to be subject to density dependent mortality effects. Negative binomial sta- tistics were calculated for all samples and a common k was derived and found to be independent of mean density and population age structure. The search for variance stabilizing transforms led to selection of a simple log transform over inverse hyperbolic sine, power or more complex log transforms, although the differences were small. 0f several environmental factors, crap height and degree days were found to be of most use in fitting density data. Instar, rainfall, field moisture and sprouting date by themselves were less significant, though some of this may have been due to measurement error. A simple regression of degree days (2nd order), crop height and an interaction term was used to develop a filter for conserving previous measurement information, allowing its incorporation into present measurement esti- notes. Extension of the density rate change equation to prediction of seasonal incidence curves was also attempted and may have general utility in extension work. Extension of sweepnet, quadrat and filtering from CLB larvae to the problem of sampling internal larval parasitoids was impaired by. inaccuracies in the dissection data. to Ron Wilson, Bert Murray, and Jack Mullins for getting it all started ACKNOWLEDGEMENTS I am sincerely grateful to Dr. Fred Stehr for the support and direction provided during my program. I am equally indebted to Dr. Dean Haynes for many hours of discussion, for many insights and contributions to this work. Dr. Ivan Mao has served as an inspiring statistical mentor and his influence pervades my thinking and this thesis. I am grateful also to Dr. Stan Wellso and Dr. William Cooper, the remaining members of my committee, for help in organizing a meaningful graduate program and for providing editorial advice with the dissertation. I am most appreciative to Dr. James Bath and the Department of Entomology for the almost unlimited research opportunities, for facil- ities, finances, and guidance. Many other people have helped me in many ways. Chief among these is Mr. Ken Dimoff, whose genius added immensely to the quality of my program. Drs. Lal Tumalla, Robert Gallun, and William Ruesink, Mr. Al Sawyer, and Mr. Winston Fulton likewise made significant contributions to thesis work. A hundred others, Jan Will, Ginny Powers, Bill Paddock, Steve Koenig, Duane Jokinen among them, also deserve acknowledgement for their many hours of assistance. . Finally, I wish to thank Dr. George Ayers and Dr. Richard Casagrande for their support during the final phases of thesis work. Kingston, R.I. November l5, l977 "...always bear two points in mind: firstly, that to the biologist computation is always a means to an end, the real objective being some greater insight into a biological problem; secondly, that the ease with which computation is now possible places on the biologist an even greater responsibility than before to understand something of the principles on which his numerical or statistical analyses are based." (Davies, l97l) TABLE OF CONTENTS LIST OF TABLES LIST OF FIGURES INTRODUCTION l.l Statement of the thesis problem 1.2 Literature review l.3 Biomonitoring and cereal leaf beetle management A CEREAL LEAF BEETLE LARVAL SWEEPNET MODEL The sweepnet as a sampling tool for cereal leaf beetles Sweepnet and quadrat sampling methods Sampling technique errors Absolute density versus sweep catch: simple corre- lations and regressions Weighted regression analysis Inclusion of additional variables in the sweepnet models An additional data set ANALYSIS OF QUADRAT SAMPLES OF CEREAL LEAF BEETLE EGGS AND LARVAE 3. 3.2. 3.2. 3.2. 3.3 LINEAR 4. b-hA-b-hh l NNNNNN N-J mm-wa—a “Nd Introduction Pubescent wheat study adult sampling methods The relationship between variance and mean for CLB counts Statistical distributions of CLB egg and larval counts Variance stabilizing transforms PREDICTORS OF CEREAL LEAF BEETLE DENSITY Introduction Parameters of the temporal distribution of CLB eggs and larvae Adult influx Crop height Egg parasitism by Anaphes flavipes Age of the larvae Rainfall Insecticides Page vi ix #00“ k0 Filtering General considerations Density models The weighing of the second observation with the predicted density. . Filter operation 4. Further limitations of the filter & h-b-Db «b 00000000 de ESTIMATION OF LARVAL PARASITISM BY TETRASTICHUS JULIS SUMMARY AND CONCLUSIONS 6.l Summary 6.2 Concluding remarks APPENDIX Al - Michigan release sites for Tetrastichus julis (Walker) APPENDIX A2 CLB parasitoid census results A2a - Parasitism rates in samples from MSU Department of Entomology CLB state census, l973-l975 A2b - l974-l975 MSU Cooperative Extension CLB parasitoid recovery program APPENDIX A3 The non-crop larval ecosystem Introduction Literature review Methods Field studies Laboratory studies Results Field studies Laboratory studies Discussion APPENDIX A4 - l975 PWS field maps. Field identification codes. 1975 PWS oat and wheat fields. Supplementary oat fields used in 1975 PWS study. APPENDIX A5 - Identification of l975 PWS cultivated fields by crop and acreage. APPENDIX 6. Pubescent Wheat Study total egg, larval, and adult cereal leaf beetle count data A6a - PWS egg count data. A6b - PWS larval count data. A6c - PWS adult count data. iv Page 85 85 87 93 95 100 104 110 ll) A6-l A6b-l A6c-l APPENDIX 7. Pubescent Wheat Study individual egg, larval, and adult count data. A7a - Adult square foot count data. Adult sweep count data. A7b - Egg square foot count data. Larval square foot count data. 2 A7c - Data_for between + within field log 5 x log x regression analysis. APPENDIX 8. Pubescent Wheat Study weather data. Temperature records for 3 locations near the study area. Niles precipitation, 1975, in inches. APPENDIX 9. Examples of filter operation. A9a - Application of filtering technique to l974 Gull Lake data. A9b - Application of filtering technique to l976 PWS data. References Cited Page A7a-l A7a-7 A7b-l A7b-4 A7c-l A9a-l A9b-l B-l Number 10 ll 12 13 LIST OF TABLES The relationship between catch per square foot and catch per sweep for CLB larvae. Correlation coefficients for count data from oats and wheat. Statistics of combined data set, used in simple (y=a+bx) larval sweepnet models. Zero-intercept (y=bx) larval sweepnet models. Regressions of catch per square foot on catch per sweep, accumulated degree days, crop height, rainfall, and windspeed. Regression of log transforms of catch per square foot on catch per sweep, accumulated degree days, crop height, rainfall, and windspeed. The APHIS (USDA) data set, describing the relationship between catch per square foot and catch per sweep for CLB larvae. The relationship between log within field variance and log mean. Regressions of log 52 on log m, using log 52 = a + b log m, and tests for goodness of fit of joint.count frequency distributions to the Poisson for egg and larval samples (given sample mean less than .5 per square foot). Estimated negative binomial k's for egg and larval sample counts and test for goodness of fit using sample k and common kc’ The relationship between sample negative binomial k's and log mean for egg and larval counts. The relationship between sample negative binomial k and the variance: mean ratio over mean instar. Comparsion of transforms optimized for minimal c. vi Page 20 21 23 24 32 36 38 45 49 SO 55 55 59 Number Page 14 Comparison of transforms optimized for minimal correlation between transformed variance and mean. 61 15 An example of higher order variance dependencies following transform. 61 16 Mean eggs per female CLB per degree day. 66 17 The relationship between log between-plus-within field variance on log mean. 68 18 Analyses of variance for adult CLB's per sweep in wheat and oats and for adults per square foot in wheat. 69 19 Wheat and oat planting in the PWS area. 69 20 Analyses of variance for wheat and oat crop height. 74 21 Percentage of eggs parasitized by Anaphes flavipes, PWS 1975. 74 22 Statistics of average egg density by date and degree day. 76 23 Analyses of variance of larval instar over degree days and crop height. 81 24 Summary of larval instar samples. 82 25 Univariate models of density (Y = log(y+c)) over several environmental factors. 88 26 Representative multivariate models of density (Y = log (y+c)) over several environmental factors. . 91 27 An example of sensitivity of the filter to accumulated degree-days. 98 28 Average parasitism rates of CLB larvae by I, julis, 1975 PWS. 105 A3-1 Distribution of eggs per square foot in cage studies. A3-6 A3-2 Relative frequencies of grass types and eggs laid on them in 4 cage studies. A3-6 A3-3 Results of sweep catches in oats and an adjoining grass (+ alfalfa) field, Kellogg Farm. A3-9 A3-4a Relative frequencies (% of total) of oviposition when presented with equal numbers of a selection of hosts. A3-13 vii Number Page A3-4b Oviposition preferences when given different amounts of each host. A3-13 A3-5 Development times per instar on four hosts. A3-14 viii Number 10 ll l2 l3 LIST OF FIGURES Michigan townships in which I, julis has been released and/or recovered, 1966-1975. Carboy unit used to extract cereal leaf beetles from sweepnet sample bags. Error in estimating average age class from sweepnet data. Distribution of sample sizes for paired data used in sweepnet regressions. Plots of residuals (yi-yi) from sweepnet model for 3rd instar in oats (Table 5) over predicted y and over de- gree days. (continued) Plot of residuals (Yi’yi) from sweepnet model for 3rd instar in oats (Table 5) over crop height (left). Residuals from transformed model for 3rd in- star in oats (Table 6) over predicted (right). The relationship between variance and mean for CLB egg and larval count data. The relationship between MLE sample k and log of the sample mean for CLB egg and larval count data. The relationship between variance and mean for CLB egg and larval counts following transformation. Changes in adult density over accumulated degree days. Location of quarter sections containing oat fields, Pulaski township, Jackson County, Michigan, 1973-75. The relationship between crop height and accumulated degree days for the PWS data. Mean egg density and estimated incidence curve for eggs parasitized by Anaphes flavipes. Relationship between instar and degree days. Plotted points are for square foot samples only. ix Page l5 18 26 33 34 47 54 61a 65 71 73 78 80 *0 .P‘. "EU-y. «4h. Saulomlmgf . .. DJ Number l4 l5 l6 A3-l A3-2 A3-3 A4-l A4-2 A4-3 Relationship between egg and larval densities and degree days and crop height. Examples of filter operation. Use of larval density equations to estimate seasonal incidence. Square foot counts of eggs in a cage study. Average cereal leaf beetle oviposition on 4 hosts. Relative adult mortality on 4 hosts. Field identification codes. 1975 PWS oat and wheat fields. Supplementary oat fields used in 1975 PWS study. Page 90 97 101 A3-6 A3-1l A3-l2 A4-l A4-2 A4-3 INTRODUCTION 1.1 Statement of the thesis problem The absolute density of a phytophagous insect population is assumed to relate predictably to future densities and thus to the potential for insect induced crop damage. Decisions on whether or not to suppress the insect are based on these predictions. The value of a pest management decision depends on information that is only as good as the accuracy of the estimate of absolute density and the closeness of the pest's predicted density to its actual future density. Many pest ecosystems include beneficial predators or parasitoids. Optimization of management decisions over more than one year requires a prediction of interactions between pests and beneficials. The limitations of sampling and prediction are confounded when more than one population is involved. This thesis is an attempt to improve the ability of pest manage- ment specialists to estimate and thence to predict larval densities of a small grain pest, Oulema melanopus (L.) (Coleoptera: Chrysomelidae), the cereal leaf beetle (CLB). Since the density of the larval para- sitoid Tetrastichus julis (Walker) (Hymenoptera: Eulophidae), may also be obtained from CLB larval sampling, this problem will also be dis- cussed. There are at least two reasons for wanting to improve density estimates and predictions. First, a sample taken early in the growing season could be used as a basis for implementing appropriate management 1 strategies during the same season. Second, management strategies must also include consideration of the between generation population dyna- mics of the beetle and its parasitoids. For example, a slightly increased loss during the current growing season resulting from refraining from insecticide use may be more than offset by the bene— ficial effects of parasitoids in subsequent years. Prediction of between generation dynamics rests on the accuracy of initial density estimates and on the corresponding projection of the within generation dynamics of both populations. This thesis, accordingly, deals with two problems. The first is the problem of estimating absolute density of the larval cereal leaf beetle. The second is predicting within generation dynamics of CLB larvae. Before evaluation of the accuracy of the density estimate and prior to analysis of environmental or biological factors affecting the dynamics, fundamental statistical problems must be addressed. Most of the concern of this thesis is thus directed toward solution of these underlying statistical questions. Data for the following studies was gathered under a joint pilot research program involving the Agricultural Research Service (USDA) and Purdue and Michigan State Universities, titled "The Effect of Pubescent Wheat on the Population Dynamics of the Cereal Leaf Beetle“. This program is being conducted (1975-1978) in a 16 square mile area near Galien, Michigan, where the CLB was first discovered in North America. In 1975, the area had a high density of beetles and well established populations of egg and larval parasitoids. Although this thesis is meant to emphasize the goals stated above, additional information from the Pubescent Wheat Study (PWS) will be 3 included in appendices for future reference. The results of 3 years of MSU CLB population dynamics survey work are also relegated to the appendix. Finally, a laboratory and field study of some aspects of noncr0p populations of the CLB and I, juljsyare also included in the appendix. In this way I hope to attain a meaningful balance between conciseness in reaching thesis goals and completeness in presenting reference data for future Use. 1.2 Literature review The cereal leaf beetle is an economically important pest on small grains. Originating in Europe, it was found in southwest Michigan in the 1950's but was not identified until 1962 (Castro, Ruppel, and Gom- ulinski 1965). Information for constructing life table responses to several environmental parameters was obtained by Yun (1967), Castro et al. (1965) and by Shade, Hansen, and Wilson (1970). Helgeson and Haynes (1972) have presented a model for within generation papulation dynamics. Haynes (1973) has reviewed features of CLB life history of most importance in developing pest management models. Several aspects of the ecology of the CLB and its principle larval parasitoid, I, julis, have been summarized in a simulation of the beetle ecosystem by Tummala, Ruesink, and Haynes (1975). ' I, jgli§_was first released at the MSU W. K. Kellogg Biological Station Experimental Farm in 1966 and was established by 1969 (Stehr 1970). This parasitoid is bivoltine and has a facultative diapause. Some of the offspring of the first generation emerge in midseason while the rest enter diapause. Diapausing first generation parasitoids and all of the second generation winter in soil in the pupal cells of the host. There are usually 4 to 6 parasitoids per host larva. 4 North American releases of other CLB parasitoids are documented by Dysart, Maltby, and Brunson (1973). These include the egg parasitoid Anaphes flavipes (Foerster) and 3 larval parasitoids, Lemophagus gurtus Townes, Diaparsis carinifer (Thompson), and Diaparsis new species. It is uncertain whether 2, carinifer has become established to date (Miller 1977). Further subcolonizations and recoveries made by the CLB Parasitoid Rearing Laboratory, Niles, Michigan, are updated annually by the Animal and Plant Health Inspection Services (USDA- APHIS 1972a,b; 1974). I, julis was the subject of an extensive subcolonization program throughout the lower peninsula in 1971-1974. Agents from the Michigan Cooperative Extension Service collected parasitized larvae from the. Kellogg Farm area and released them at preselected nursery sites in their home counties. A follow-up Extension Service recovery program in 1972-1975 revealed widespread establishment and dispersal (Fig. 1; Appendix l). 1.3 Biomonitoring and cereal leaf beetle management In an optimal pest management program the tactics chosen should supplement the effects of natural control agents if possible (Rabb 1970). Following its successful subcolonization, I, julj§_may be regarded as a natural control agent for the cereal leaf beetle. Pest control recommendations should be adjusted accordingly.’ This requires identification and quantification of the relationships between pest density, a critical set of biotic and abiotic factors (eg. parasitoid density and accumulated heat units) and crop damage. To a certain extent, many biotic states can be related to abiotic L JULIE RECOVERIES D TOWNSHIPS m WHICH 1, JUL|§ WAS RELEASED I RECOVERY MADE AT RELEASE SITE DURING uses-1975 7 RECOVERY IN TOWNSHIP WITH NO PREVIOUS RELEASE SITE Fig. 1. Michigan townships in which_I,'ju1is has been released and/or recovered, 1966-1975. LiLLLLS RECOVERIES U TOWNSHIPS IN WHICH _T_._ JUL|§ I wAS RELEASED I RECOVERY MADE AT RELEASE [.1 SITE DURING 1969-1975 RECOVERY IN TOWNSHIP WITH ( NO PREVIOUS RELEASE SITE I II II F1 F1 e—— ’ II [T_J 7%] ' 52 i .- .. é Fig. 1. Michigan townships in which I, julis has been released and/or recovered, 1966-1975. -‘ 6 factors. Plant growth, for example, may be predicted from plant age, rainfall, temperature, and soil nutrients. Since monitoring abiotic factors can Often be done cheaply and in an automated fashion (Haynes, Brandenburg and Fisher 1973), there have been attempts to use abiotic measurements to make regional projections Of biotic states (eg. popu- lation maturity) (Fulton and Haynes 1977). The reliability of popu- lation projections is restricted by the fidelity of the predictive models and the accuracy of the initial measurements on the population. Since projection errors will arise from even the highest resolu- tion models, periodic field measurement are needed for recalibration. The nature of these measurements, ie. the design Of the sampling program, depends on the type of projection being made. An individual farmer will tend to be most interested in the within generation dyna- mics Of a pest. His sampling program would be intensive, marked by somewhat detailed and frequent counts Of pest density, measurements of average larval Size, feeding damage, etc. If a population manage- ment approach is being used, between generation dynamics of the pest and major biological control agents should also be Of interest. Here, the sampling would be over a larger region and would be limited in detail and frequency by the logistics and economics involved. A general requirement Of pest management schemes is that the cost Of information must be lower than the cost of proceeding without it (Headley 1975). For several years, the sweepnet has been used to sample CLB adult and larval populations as part Of a continual moni- toring Of several major small grain producing regions in Michigan, Indiana, Ohio, Pennsylvania and Canada (Fulton, Haynes, unpublished reports). The sweepnet has the advantages Of low cost per sample, 6 factors. Plant growth, for example, may be predicted from plant age, rainfall, temperature, and soil nutrients. Since monitoring abiotic factors can Often be done cheaply and in an automated fashion (Haynes, Brandenburg and Fisher 1973), there have been attempts to use abiotic measurements to make regional projections of biotic states (eg. popu- lation maturity) (Fulton and Haynes 1977). The reliability Of popu- lation projections is restricted by the fidelity of the predictive models and the accuracy Of the initial measurements on the population. Since projection errors will arise from even the highest resolu- tion models, periodic field measurement are needed for recalibration. The nature Of these measurements, ie. the design of the sampling program, depends on the type of projection being made. An individual farmer will tend to be most interested in the within generation dyna- mics Of a pest. His sampling program would be intensive, marked by somewhat detailed and frequent counts of pest density, measurements Of average larval size, feeding damage, etc.' If a population manage- ment approach is being used, between generation dynamics Of the pest and major biological control agents Should also be of interest. Here, the sampling would be over a larger region and would be limited in detail and frequency by the logistics and economics involved. A general requirement of pest management schemes is that the cost of information must be lower than the cost of proceeding without it (Headley 1975). For several years, the sweepnet has been used to sample CLB adult and larval populations as part of a continual moni- toring Of several major small grain producing regions in Michigan, Indiana, Ohio, Pennsylvania and Canada (Fulton, Haynes, unpublished reports). The sweepnet has the advantages Of low cost per sample, 7 minimal time and labor requirements, and minimal damage to the crop sampled. It is also a simple tool, making it possible for relatively untrained personnel to serve as biomonitors. The sweepnet's value as a sampling tool rests on the assumption that the number of insects per sweep can be converted to an estimate Of absolute density. Ruesink and Haynes (1973) developed a multipli- cative conversion factor for adult CLB's that was affected by crop height, wind, temperature, and solar radiation. For larvae, they could find no relationship between the conversion factor and the environmental factors. They settled on a coefficient of 1.02 to convert sweep catch to a square foot density estimate. Recent ques— tions (Gage 1974, Fulton 1976) have prompted a reevaluation of this relationship. This is presented in section 2. In addition to possible biases in sweep catch, the sweepnet is not useful when the crOp is Short. For the intensive early season sampling needed for the within generation forecasts Of most concern to farmers, a more direct population estimate is needed. Square foot quadrats (or equivalently 2 linear feet of crop row) and individual stem counts have been used as sample units. Square fOOt units were used in this study because past experience indicated this provided the best tradeoff between variance and sample cost (Southwood 1966, Helgeson and Haynes 1972). A statistical artifact Of such counts is the dependency Of their variance on the mean. There are two problems caused by this. Since counts which have a higher mean also tend to have a higher variance, a standard error on the estimate is not usable (Sokal and Rohlf 1969, p 381f). This must be replaced by a skewed confidence interval with width dependent on the mean. The second problem caused by the insta- 8 bility of the variance (ie. dependency Of the variance on the mean) is the violation Of normality assumptions necessary for performing hypo- thesis tests Or analyses of variance. These difficulties are discussed in section 3. Density is a function not only Of time but also of accumulated heat, rainfall, windspeed, and probably other environmental factors. Many of the relationships between environmental factors and fecundity, mortality, and development rates have been studied previously (Castro et al. 1965, Yun 1967, Helgeson and Haynes 1972, Gage 1974). The role Of these factors is reexamined in section 4 using new data and a transformation developed in section 3. Estimates Of changes in populations Of I, jgli§_are most directly made by sampling adults. However, the parasitism rates of CLB larvae and pupae also contain usable information on parasitoid density. Larval samples are much easier and less costly to Obtain than pupal samples. For purposes of between generation regional projections of parasitoid densities, larval samples may be Optimal, despite the dif- ficulties of estimating seasonal density. These problems and estima- tion procedures are discussed in section 5. A CEREAL LEAF BEETLE LARVAL SWEEPNET MODEL 2.1 The sweepnet as a sampling tool for cereal leaf beetles Accuracy in biomonitoring is the result Of the mechanical effic- iency of the sampling method in detecting (capturing) the creature being monitored. Although direct counts Of such torpid insects as cereal leaf beetle larvae are generally accurate, the cost of making enough counts to Obtain a minimal acceptable sampling variance (ie. variance within 10% of the mean) may be prohibitive for most survey programs. A less accurate index such as sweepnet count is frequently the only acceptable alternative. The sweepcatch index is related to other factors in addition to density (Morris 1960). Delong (1932), in one of the first reports on factors affecting sweep efficiency, reported that capture of active insects such as leafhopper adults varied with temperature, plant height, and type Of stroke or backstroke. .Hughes (1955) found wind affected catches of a chloropid fly in wheat. Larval size and time of season affected sweepnet efficiency with green cloverworm in soybeans (Shepard, Carver and Turnipseed 1974). Time of day and changes in vertical distribution also affected sweep catch of several small grain insects (Vickerman and Sunderland 1975). Ruesink and Haynes (1973) developed 2 models for sweepnet samp- ling Of the cereal leaf beetle. A conceptual model related the "mult- iplication factor", M, required to convert sweep catch to density, to net diameter, d, length Of the sweep stroke, 1, the proportion of the 9 ., _.... Ann‘lVH‘.‘ _ 10 population in the net's path, b, and the probability of getting caught, p, once in the path. That is, M=1/(d1bp). (l) A regression model for each instar suggested that wind, crop height, temperature or cloud cover did not affect M. There were no conclusive differences in M values between the instars (for the ranges; crop height, 38 to 112 cm; temperature, 24 to 29°C; wind, 6.4 to 16 kph; and solar radiation, .7 to 1.1 cal/cmZ/min). Relating this to the conceptual model, there were no detectable differences in instar specific probabilities of getting into the path of the sweepnet or of getting caught once there. This conclusion was in conflict with an earlier Ruesink data set (Ruesink 1970) which found higher M values for younger larvae (averag- ing 1.41, 0.84, 0.72, and 0.40 for lst through 4th instars). However, the inability to accurately determine instar and small sample size (5 fields in the original data set, 10 in the later) made both sets Of results inconclusive. The instar determination problem has since been solved (Hoxie and Wellso 1974). Head capsule widths have 4 discrete ranges and are now used regularly to determine the 4 instars. The ranges established by Hoxie have been found to apply to Michigan CLB survey data from 1971~72 (Fulton 1975) and 1973-74 (Logan, unpub- lished data). Gage (1974) found that catch per sweep underestimated the total larvae per square foot by at least half, noting that very few lst instar larvae were being caught in the net. Fulton (1975) reported serious discrepanciés between instar proportions in sweepnet samples and the proportions found in square foot samples taken at the same 11 time and place. Fulton concluded that sweepnets are unreliable as a method to assess average age of the population. Reasons for the discrepancies in instar specific catch efficiency were not given by any Of the above. Ruesink (1970) found more younger larvae on the lower 2/3rds Of 38 inch wheat but was unable to estab- lish this as a general rule. Jackman (1976) found the majority of all instars were on the upper 3 blades Of oats throughout the season. Wellso and Cress (1973) Observed a significant difference in oviposi- tion and feeding between top and next lower blades with most eggs laid on the upper, more succulent leaves. Thus it is still indefinite what factors affect the probability of a CLB larva getting into the path of the sweepnet. Also, there have been no studies Of the factors influ- encing the probability Of getting knocked Off the plant and into the net. Because the many advantages Of the sweepnet were placed in jeopardy by questions of its accuracy, a large set of sweep and square foot samples were gathered for comparison as part Of the PWS (section 1.1) program. Methods used are described in section 2.2. Correlations and simple regressions Of sweep and square foot catches are given in section 2.3. Section 2.4 presents an outline of a stat- istical technique appropriate to analysis Of subsample data. Section 2.5 further examines the data to determine whether more detailed statistical models would improve sweepnet conversion factors. Section 2.6 presents an additional data set and its implication (and, alas, complications) for the sweepnet model. 12 2.2.1 Sweepnet and quadrat sampling methods Wheat and oats were sampled in the PWS area near Galien. Sweeps were made after the host field was at least 8 to 10 inches tall. A sweep is one level pass of a 15 inch diameter net. A sweep covers about a 5 foot swath through the top 8 to 15 inches Of the crop. The sweepnet was lined with a new 30 gallon plastic bag for each field to prevent small larvae from working through the mesh and to keep counts from each field distinct. The sweeper took 150 sweeps, beginning 20 to 30 feet from the edge of the field. The choice Of 150 sweeps was arbitrary. Previous survey work had used a standard of 100 sweeps. The sweeper walked at a Vigorous gait 120 to 180 feet into the field, made a wide u-turn, and exited sweeping until the 150 sweeps were complete. If there appeared to be less than 10 larvae in the bag the sweeper would take 100 more sweeps with the same bag. In practice, workers soon recognized certain fields were sparcely popu- lated and took the 250 sweeps virtually automatically. Having com- pleted the sweeps, the worker washéd down the larvae on the sides Of the bag with FAA, a preservative, and labeled the bag with field number and date before sealing. We used a formalinzacetic acid:alcohol:water mixture (FAA) in a 2:1:50:47 ratio. This preserved larvae suitably for dissections. While the sweeper worked, a second sampler made square foot counts. The sampler walked 20 to 30 feet into the field carrying a lightweight 2 foot wooden stake. He tossed this, letting it fall where it would. This was done to help improve the randomness of the quadrat location. Workers took samples from points far apart in the field. The result was a semi-uniform grid of sample areas with a single count taken 13 randomly in each area. The stick was moved 5 feet down the same row to reduce the disrupting effect of the toss. Care was taken to avoid sampling where the field had just been swept. The sampler picked over the plants in 2 feet of 1 row, recording the number of eggs on a hand counter while mentally counting larvae. The first 50 larvae found were put in a Vial for later study of field age distribution. Vials were labeled with date and field number. They were half filled with water and were returned to the lab and frozen. After collecting the larvae and counting the eggs in a 2 foot section, the sampler moved to another site and again tossed the stick. A minimum Of 10 such sample points per field was about as much as could be taken with PWS resources. However, if 50 larvae had not been col- 1ected within this minimum, an extra 5 points were covered or enough to get 50 larvae, which ever came first. When time permitted, separate records Of each 2 foot egg and larval count were made to provide a measure Of within field variance. The ruled sweepnet handle was used to estimate average standing crop height. Time of day, number of sweeps and square feet, windspeed (taken with a styrofoam ball anemometer), and field wetness were also recorded. Field wetness was classified as either "dry", "wet" (full Of dew or fresh rain), or "damp" (not glistening wet but able to dampen trousers of someone walking through it). When possible, sampling was limited to times when fields were dry. Sweepnet sample bags were taken to the lab and were stored in 9°C refrigerators until processing. Three weeks' samples had accumu- lated before processing began but there was no apparent damage to l4 larvae as a result Of the delay. Larvae were removed from the bag by floatation in alcohol. A sample cleanup unit was constucted from an inverted 5 gallon plastic carboy with the bottom removed (Fig. 2). A large nylon funnel was fit into the carboy to eliminate the shoulder. A 3/4 inch 10 clear tygon tube was attached to the bottom of the funnel. A 2 cork system, one on the bottom of the tube and a second attached to a stick and inserted from above into the funnel, allowed rapid isolation of the larvae. The carboy was partially filled with 95% alcohol. The plastic sweep bag was inverted and washed in the alcohol. Insects quickly sank to the bottom, accumulated in the tube, and were drawn Off into a 2 oz. plastic cup. Plant debris was skimmed from the surface of the alcohol. Since the volume Of alcohol was relatively small, replacement Of all the alcohol was infrequent. All that was necessary was a periodic topping up and a daily flushing of the whole system. Two carboys were used and kept one man busy. Labels were transferred to the new cups and date and field number were repeated on the cup lid. After separa- tion alcohol was poured Off and fresh FAA was added to the cups. Larvae and adults were then counted. The contents Of a cup were poured onto a tray and sorted. Adults were first removed and their number recorded. Larvae were separated from remaining insect and plant debris as they were counted. In very large samples (greater than 1000 larvae) a quartering system was used to speed counts. All adults were first counted and removed. A slurry of larvae was made so that larvae were distributed approximately uniformly on the tray. Larvae in 2 Opposite quarters *were removed. The remaining larvae were again slurried. Counts were STICK FOR REMOVING UPPER STOPPER CARBOY —— —— FLUID LEVEL -——- NYLON FUNNEL P COLLECTING DISH Fig. 2. Carboy unit used to extract cereal leaf beetles from sweepnet sample bags. 16 made Of 2 Opposite corners and the result was multiplied by 4. About 5% of all samples had to be treated this way. Larval instar was determined by measuring head capsules with an ocular micrometer. An aliquot subsample was prepared from the contents Of a sample cup. Up to 50 larvae from each sweep collection were measured. All larvae retained from the 2 foot counts were measured. At the same time,the larvae were dissected to determine species, life stage and number Of parasites. 2.2.2 Sampling technique errors Several counting errors can arise during the above processes. Briefly, these are as listed: A. Precount errors 1) When hand picking, not seeing or losing a larva that is present in the sample quadrat. 2) Sweepnet bias, as discussed in section 2.3. 3) Transfer losses from sweepnet to count dish. 8. Count errors 1) Miscounting. 2) Age group misclassification. 3) Failure to count any members Of an age class. 4) Aliquot biases. C. Analysis errors, as discussed in section 2.4. With the PWS data set I did not measure the efficiency of finding larvae in the field. I assume 100% Of the larvae in a sample unit are. found and counted or returned to the lab, as over optimistic as this may be. The washing down of larvae on the sides of sweepnet bags and later removal from the bag is highly efficient. Once in a particular bag, there is little chance that a larva will not make it into a sample 17 cup, ready for counting. Repetition of counts on a few samples Of 50 to 200 larvae indi- cated precision to be within about 3% of the actual number present. Failure to properly classify the development stage of an insect is unlikely because the overlap Of head capsule widths between instars is quite small (Hoxie and Wellso 1974) and ocular micrometer measure- ment is relatively precise. By limiting the number Of measurements taken from sweep samples to 50, the probability of failing to count any individuals in a parti- cular age class is increased. If there are 4 lst instar larvae in a sweep catch Of 100, there is a probability Of .34 that the lsts will be underestimated and .06 that they will not be counted at all in a sub- sample Of 50, all other biases ignored (Larson 1969). Fulton (1975) suggested that the average age Of a sample would be estimated with reasonable accuracy (5% Of the mean) when 50 larvae were measured. Logistics restricted the measurements done during the PWS to 50 per subsample. There was a major failing with the aliquot method. TOO many large (4th instar) larvae were sampled. This is reflected in Fig. 3 in which 2 groups Of 50 larvae were measured from each Of 48 samples. The age estimate from the first group varied by roughly .5 to .2 from the second estimate. Because of this, a second batch of up to 50 larvae were measured for all of the samples used in the following analysis. This meant measurement and dissecting an additional 2271 larvae from the oat samples (for a total of 5677 measurements in 85 cases) and an extra llOO larvae from the wheat samples (total 4136 in 62 cases). In the oat samples, 45 cases had all the larvae in the net measured. The 40 Fig. 3. 4.0a 3.0‘ 2.0 Average instar - 2nd measurement set LO ' do I So ' lo Average instar - lst measurement set Error in estimating average age class from sweepnet data. 19 cases with an incomplete measurement set averaged 89 measurements per case. In the wheat samples, 31 cases had all the larvae measured. The average subsample size for the remaining 31 cases was 70. Table 1 com- pares the measurement sets further. This includes the slope and intercept coefficients of the square foot and sweep catch regressions. The problem with interpreting the differences of Table 1 is that there is no way to tell whether the higher proportions Of 3rd and 4th instars in the partial measurement samples are the result Of measure- ment bias or are only reflective Of the Older age structure of the higher density samples. In oats, comparison with associated square foot samples suggests the latter is true. In wheat, there is less Of a distinction. The decision to either disgard the incomplete measurement sets or to combine the 2 was resolved in favor Of combina- tion (see Table 3), with the caveat that the result may be an over- statement of the relative proportion of Older larvae in the sweepnet samples and a corresponding bias in the sweepnet conversion factors. 2.3 Absolute density versus sweep catch: simple correlations and regressions For a sweepnet index Of CLB larval population density to be Of use, the catch per sweep of each instar must be correlated to absolute density. Table 2 gives the correlation matrix for the 4 instars in the paired sweep and square foot samples, separated into cases for wheat and oats. Correlation coefficinets significantly different from O at the .05, .1, and .2 levels are indicated (Sokal and Rohlf 1969, p. 516). Note the diagonals of the lower left 4 x 4 submatrices, the correlations between sweep and square foot catch for each instar. In oats, the proportion of 4ths in the 2 sample methods is highly -IIIA it «(to lta..C~ Le-C {cu-T. .2.......a.)£ I .) ..:.>..c— Ego LCC 3....lv Ltc Chafiu tin. “22.; C..r.3?... Lin Equal. Cn...)~.~£ Q‘CUCC‘QnJQL t£k .— O‘C-‘L .. Ii... .H. .1 bfiiupLQ j sum. mmm.~ Laumco cow: com. -~.m Loumco coo: o-._ F_P.— —~c. _mo. omm. new. moo. -~. v mxo.~ oso.— mno. mmm. KN”. omp. nop.p meo.— o mom.m mop.~ «mp. omm. mom. nmm.— ~om.~ m~n.~ m om~.— sop.— mop. mom. ofip. com. com. mm“. m moo. och. NBN. «co. vmm. ~_o.— vo~.P m-.~ ~ Pop. on.. mom. «we. -_. muw. com. omm. ~ nor. mmo. —m~.~ ooo.o m_m. oom. vm~.— ooe.— — opo. coo. ooo.— m~o.- mNo. ouo. mop. no. . _ .o.m ox .o.u on .m.m we .u.m _» va Ecumco .v.w ox .o.m . on .m.m we .o.m _» aov Launc— NoPN omm_ mucDEmcamaoe Pope» momm um“. mucwemcamcoe peach Pm _m «ammo oq ow mmmao Endgame“: usaxqm aeoemaao “ommnmawz usazoo ucaucouu u .o.» .* Loumcw co amwzm can guano mooco>o - vx .e...p u v .w Loumc_ mo uooo ocoacm can :uuou oooco>u a VA .um>La— moo Lou ammxm Loo guano can woo» mucosa Log guano comzuoa nogmcowuupoc on» .F upon» Table 2. SW1 SW2 SW3 SW4 $01 $02 $03 $04 MISW MISQ SW1 SW2 SW3 SW4 501 802 803 $04 MISW MISQ Correlation coefficients for count data from oats and wheat. 21 SW1 = estimated catch of instar l per sweep; MISQ = mean instar of square foot catches, etc. ABS (r) = 1. .072 1. -.026 .698 1. .018 .350 .735 1. .245 .028 -.055 -.082 .262 .052 -.045 -.076 .221 .108 .170+ .150+ .057 .120 .433 .805 -.259 - 218 .047 .313 -.226 -.056 .085 .261 sw1 swz sw3 504 1. .409 1. .337 .636 1. .302 .629 .923 1 .467 .ngg .139 .156 + .176 .447 376 .297 .059 ;£Q§ 121, .464 -.186+ .107 170* .217* -.387 -.025 194+ 332 -.445 -.210+ 180 146 SW1 sw2 SW3 sw4 absolute value Of the correlation coefficient. Oats (cases = 85) 5 Significant at .05 if ABS (r) > .214 r* significant at .10 if ABS (r) > .180 r+ significant at .20 if ABS (r) > .140 1. .962 1 .815 __.__§_8_q 1. -.095 -.068 .140 1. -.435 -.461 -.326 .292 1. -.327 -.331 -.269 .338 ‘_;911 1. SQ1 502 303 504 MISW MISQ Wheat (cases = 62) [_ significant at .05 if ABS (r) > .250 r* significant at .10 if ABS (r) > .211 r+ significant at .20 if ABS (r) > .163 1. .1221 1. .213* 1922 1. -.018 .197+ L342 1. -.203+ -.006 .253 ,439_ 1. -;§§1 -:319. .200+ 4229 .1991 1. $01 502 sq; 504 MISW HISQ 22 correlated. Firsts and 3rds are also significantly different from O correlation at the .05 and .2 levels, respectively. In wheat, the 4 correlations of interest are all different from 0 at a .1 level. The average age as determined by sweepnet has about a 60% correlation with square foot mean instar in both crops. Table 3 combines the 2 data sets from Table 1 and describes the data in more detail. For oats, 95% confidence limits on the lst instar square foot over sweep catch regression overlap the limits for 2nd instar. The confidence limits for 3rd instar surround those for 4ths and in turn fall within those for 2nds. For wheat, distinction between the instars is easier as only 2nd and 3rd instars have overlapping confidence limits. Note that only 2nd instar larvae come close to the slope Of approximately 1 suggested by Ruesink and Haynes (1973). The general trends tend to support the findings Of Gage (1974) and Fulton (1975), reflecting a sweepnet bias against picking up smaller larvae. It Should be clear from the high coefficients Of variability and the low coefficients of determination (R2 values) (Table 3) that these simple models lack the precision desired for accurate predictions. One difficulty with the regressions is that they all provide nonzero esti- mates for square foot density even when no larvae are caught in the sweepnet. While this may be useful when the number Of sweeps is small (100 or less, say), it may be an overestimation. Table 4 provides the slopes and correspOnding confidence intervals for zero-intercept models fit to the same data. These estimators may be more useful when many sweeps are taken at low densities. An attempt to further improve these regressions by taking into account various environmental factors is made in sectiOn 2.5, following 23 mo. oF. «F.o oo o Foo.v om. ow.mw no w mo. a.ooF o.w on. on. F w a mo. w.wm o.me om.oe om.oe F w v FFo.v mo.m om.mmw on u owF. wm.w om.mow no a ow. o.FwF e.Fw mo.ao mo.eo F w n no. m.FFw wo.w . ww.o ww.o F a m Foo.v ww.w «F.omF oo o awe. ow.w oF.oww no w ow. F.ooF o.mF mo.am mo.em F m w oo. m.oFm eww. wo. wo. F m w Foo.v em.F ma.oFF oo Foo Lotta Faauamow cwo. aw. wo.ww no Foo Loccw Fa=u_muw ww. w.mmF o.oF ao.om Fo.om F Fwo coammocoox F oo. m.mFo Fm.m ma.F ma.F F Foo coammocmog F M1 .3 as m: mm .2. 8:8 E .232: w”. .3 Sm mm. W1 ca 838 E .32; a a woz<~w<> go m~m>ggLoF Fxn + a u xv qust gF tum: .amm caoo vwcagsou mo muFamFaaam .m oFgoF 24 Table 4. Zero-intercept (y = bx) larval sweepnet models. OATS WHEAT Instar (i) bi 95% C.L. bi 95% C.L. 1 11.838 3.708,19.967 12.792 8.294,17.289 2 2.116 O. , 4.552 1.549 1.043, 2.055 3 .542 .233, .852 .598 .403, .793 4 .581 .507, .655 .099 .037, .160 a brief statistical argument, presented for future reference in section 2.4. 2.4 Weighted regression analysis It is important to note that the instar specific variables "catch per sweep" and "catch per square foot" are themselves estimates. They are based on subsamples Of larvae. For example, catch per sweep Of 2nd instar larvae in sample j is x2j = (IZj/"j)(Lj/Sj) (2) where Izj is the number Of 2nd instar larvae in the subsample nj; Lj is the total number of larvae from which the subsample was drawn; and Sj is the number of sweeps taken. Sampling errors for these estimates are related to subsample size. Since the standard error Of the sample is s.e. = SQRT ( £(xij - “ij)/"j2) (3) where pij is the unknown population mean (estimable only in repeated data sets) and SQRT() is the square root Operator, sampling error is inversely proportional to square root Of subsample size. In this sence larger samples may be said to be more reliable than smaller ones. There are no standard criteria for evaluating the relative relia- bility Of a sample estimate. By comparing values of l/n for different 25 n's, Searle (1971, p. 365-73) suggested ratios of such values were possibly useful. Ratios of 4 : 1 or higher were indicative that sample error should be taken into account. It seems clear that there is little loss in reliability in going from a sample size of 100 to 50. This is equivalent to a ratio of 1.41 : l, which would rule out an adjustment under Searle's criteria. However, comparing samples Of 6 and 50 would result in greater than a 4 : 1 ratio. As Searle noted, the 4 : 1 ratio is an arbitrary norm. If one were to use a more con- servative 2 : 1 ratio. adjustment of sampling error would be considered appropriate when less than 12 larvae were present in a sample. This assumes that 50 is the Optimal sample size (Fulton 1975). (If 30 were considered optimal, based on some other analysis, 8 larvae would be the lower limit, using the 2 : I ratio.) > ' Comparing plots Of sample size for the sweepnet samples with their paired square foot samples in Fig. 4, 21 Of the cases in oats (24.7%) have at least 1 Of the samples smaller than 12. Only 3 (4.9%) of the samples in wheat are less than 12. This suggests that the age specific density estimates in oats ought to be subjected to a weighted regress- ion analysis to adjust for the large percentage Of small, "less relia- ble", subsamples. Here again, the criteria are arbitrary.v As a sug- gestion for future use, perhaps a maximum of 10% "small" samples should be considered as a basis for deciding the appropriateness Of such analysis. TO perform a weighted regression analysis, the total sum Of squares and the reductions in sums Of squares used in the regression must be weighted. 'This is done by weighting each term in the sum of squares in inverse proportion to the variance (Searle 1971; Steel and Torrie 1960, 26 .mcomemEomg amcommZm :F omm: mamo omgFmo Low mmNFm mFoEmm mo coFasoFLamFo szmmmzwIImmow mgmzqm on“ o: ONF P p p > p p .r I I I I I I I I l I I I I - I II I I I qurz v I I I r T T 001 0. OS 0' OZ 1 on I ”I .F 1003 BUUDDS-TBZIS 31dNUS szmmmz ”II .c .mFm mmom wgmzaw OOFF .5 owu . Om— I L I PINE—.08 I I . I 1. ”U I Hw II II I I ITMIG I I II fil I I I I 33 I I I :08 I Tl. _. .__. _. .«Z n _. . 3 I '— 10— I I I I II W I I IIIfi I I n! ”Hymn III? I I II I a ”U .. .aa fl: I T I Yul: flU quo . 0 El 27 . 180), p -) (4) = z ' w. sx/var(x1 1 where wi is the weighting coefficient, 5; is the variance Of all the sample means for variable x, and var(§) is the variance associated with the Specific subsample mean. The usual simplification of (4), keeping in mind that the variance is inversely proportional to sample size, is to use subsample size over the total number in all subsamples as the weight wi = ni/n. (5) where the "n." notation indicates summation over all n (Searle 1966). Steel and Torrie (1960, p. 181) point out that it is the relative rather than the actual weights that are important. Since samples do not really add new information to the estimates in direct proportion to their sample size, it is proposed that a modification of (5) be used, wi = SQRT (hi/n.) .(6) Using a weight proportional to subsample size, a sampling of 100 would count as twice as reliable as a sample of 50. Using weight pro- portional to square root Of subsample size, 100 would only be judged 1.41 times as reliable. Samples less than 12 receive a rapidly decreas- ing relative weight. This seems a reasonable compromise between using no weights and using subsample size as a weight factor. Regression based on weighted means uses a weighted variance covari- ance matrix, each element of which has the general form cov(xj,xk) = Ewij(xij'ij where summation is over all i, wi's are the weights from (6), Xi and )wik(xik-;k) (7) xk are any variables in the regression and i's are weighted means, x = 2(wixi)/Zwi 28 Weights are based on (6) for means (number per age class per sample unit) but are set equal to l for all other variables. Covariances (and variances by an analogous process) are computed from the expansion of (7), cov(xj,xk) = zwijwik(xijxik"xikik'xikij+§jik) (9) which for parameters with weights equal 1 reduces the usual form, kl-(injzxikl/mJIIm-I) (10) where m is the number of cases. A correlation matrix based on the cov(xj,xk) = [£(xijxi above with elements rjk = c0v(xj,xk)/SQRT(var(xj)var(xk) (11) can be used to compute least squares estimates of transformed correla- tion coefficients, ai. The normal equations are . " r l 1 r12 rIn a11 r1y r21 ‘ r2n a2 r2y 'rn] rn2 1 an rny (12) with solution 4 L . . . A = -1 Estimates Of the original coefficients are obtained frOm bi = aiSQRT(var(y)/Var(xi)), (14) for i = 1...n, and from DD = y-blxl-bzx2-. . .-bnxn, (15) where ai's are the transformed regression coefficients, var(y) and var(xi) are weighted (by "i or by unity) variances, and y and ii are the weighted means. It should be noted that the same logic applies to models contain- ing classification terms. For example, if a measurement were assigned 29 a coded (0,1) value depending on whether it was taken from a particular group, mean, variance, and correlation terms could be calculated as above. The normal equations from the transformed correlation coeffi- cients would be 32'x 32' 9 56y where the BX'X submatrix is an incidence correlation matrix, 31.2 is the regression matrix as in (12). The diagonal Of BX'X would have all elements equal 1 and all Off diagonal elements would be negative with absolute value less than 1. Having computed the weighted means, variances, and the correlation matrix, standard statistical packages are available to complete (13) through (15)(Nie, Hull, Jenkins, Steinbrenner and Bent 1975). Carrying out the above on the data used for the sweepnet regressions involved weighting only the sweepnet and square foot instar specific densities by subsample Size, assigning a weight of l to all other measurements. The resulting correlations and regression statistics were not appre- ciably different from the results presented here, which are based on an unweighted analysis. Since the amount of programming time required to adhere to the weighted regression analysis was not thought to be justified by the refinement in results, the technique was dropped. Apparently there were sufficient cases that disparities caused by small samples were averaged out. With smaller data sets or with sets involving a large number of small sample versus large sample pairs, this may not prove to be the case. Application Of the above to a correlation analysis by Fulton (1975) using data with these latter characteristics reversed an original conclusion of no significance 30 between instar specific square foot and sweepnet catches. 2.5 Inclusion of additional variables in the sweepnet models The simple regression Of catch per square foot on catch per sweep (section 2.3) did not provide the precise conversion factors desired for the sweepnet model. It was hypothesized that environmental factors were influencing sweep catch, although this hypothesis was rejected by Ruesink and Haynes (1973). Several environmental measurements had been made for each PWS sweep and square foot density estimate (section 2.2.1). These included crop height and windspeed, which had been taken in each field. Temperature and rainfall were estimated from a single station in the center Of the study area, within 3 miles of each field. Because there was some doubt as to the accuracy Of the central site thermograph, data from a New Carlisle, Indiana (5 miles south), and a South Bend, Indiana (12 miles southeast), weather station were averaged with the thermograph when computing degree day accumulations. These data are listed in appendix 8. This decision to average may not have been justifiable but was not reconsidered until later (see section 4.3). The full regression model that was evaluated was yi = a + blxi + bzdd + b ch +_b r24 + bSW + e (17) 3 4 instar specific catch of CLB larvae per square foot; where yi a = a constant common to all Observations; xi = instar specific catch Of CLB larvae per sweep; dd = accumulated degree days (base 48F) on sample date; ch = standing crop height on sample date (inches); r24 = rainfall during previous 24 hours; W = windspeed (mph) at time of sample; e = residual error, assumed ~N(Q,IOZ). 31 The regressions were carried out in a stepwise inclusion following the initial fitting of sweep catch (see Draper and Smith 1966, p. l7lf). Results are listed in Table 5. Most Of the parameters in the model (17) are included in the regressions because the regression program used was set for very loose inclusion criteria. For practical purposes, 2 values could variables that add less than (arbitrarily) 5% to the R have been excluded. The differences between predicted and Observed y's are called residuals. For each regression in Table 5, plots Of residuals over predicted y and over the independent variables (x,dd,ch,r24, and W) were made. This technique is advocated and well discussed by Draper and Smith (1966, p. 86-99, 132) as a way to check model accuracy. Briefly, these authors recommend examining such plots for depEndencies Of the variance on the prediction or on the independent variables, a linear Slope to the residuals, or curves in the plots Of the resid- uals. These conditions are indicative respectively of variance instab- ility (see section 3), incorrect fit or omission of the intercept, or model inadequacy and a need to include higher order terms or interactions in the model. The first type of abnormality (variance dependent on the predicted y) is especially crucial as a visual test Of the critical assumption that the error term is normally distributed, an assumption necessary for valid testing. A few examples Of such plots are given in Fig. 5. These indicate, respectively, a) overall model inadequate as there is a clear dependency Of the variance on the prediction, b) ade- quacy over degree days and c) crop height as there are no apparent additional curvatures that would demand higher order terms. TO adjust for the variance instability, a multiplicative model of the form 132 com. mom. mwm. oFN. Fvo. mmm. mmm. an. Own. mow. mmm. mww. www. cow. «mm. vmm. omm. wow. mFN. (I ("I m.mmF u .>.u ccoov.o n u no. mv.F1 FacoamEOOV FFoo. mFoo. m go oFo. «mo. 0 2 mFoo. mmoo. m an ow. Fv. ~ «we owe. wwo. F a. FacuocoooocFv «a om.m u um o.omF - .>.u ooomn.m n u om.v mmF. FucoamEOUV moo. muc.o m vu vF.~ mm. v «N; «F. mmo. m 3 mo. FF. N :u Foo. Fae. F ma FacoucwawocFv ma Fm.w . HF F.00F . .>.u ..e_.a . a Fo.w mF.F Facoamcouv .F.c nu FF. mFo.- v 1 Fmo. NFo.- m gu ov.F no.F m ”we aw. 8F F w. FacoocoooocFv w» Fw.F . am o.FmF . .>.o ..mq.a . a cm.m Fm.F Facoamcouv moo. FFooo.I m vv ov.F nu._- Q v~L oF. nF. m ) mmo. Fvo.- ~ gu oo.w «F.o F F. Facoocoqochv F» .Idmqumd any coFmaFucF oFoemem LoLLw acoFoFaaooo co Louco oceocaam coFawoLoum quxz achuFFFoou mF .>.u .amLFF comeoLmog wga cacF GIULOF we) Quota Lon guauu moo. Fmo. omo. mmo. mvo. mom. cmw. mmw. omF. owe. mmw. mmw. moF. moF. moo. now. new. oNN. mom. ooo. (I Fm. - mm m.ma . .>.u occmm.0m . a FacoachUV «NE 3 cu cu "I Facoucwoochv «x .>.u ooFm.v n a Facaawcoow 3 owL to go MI Faeaocoacac_o FF .>.u comm.v . u Facaamcouv 3 QNL go on w. FacmncoooocFv NA mNN. n um m.oFm n .>.u ochF.m n u Fm. Fw.- om. wn. m oFo. FFo. o moooo. woooo. n moo. oFo.. w mo. wm. F oo.w . a” o.FoF . Fm.F an.¢ ovo. mwo. m Fm.w oo.F- a oFoo. wnoo.. n owo. moo.- w FF. oF. F ow.“ . aw ~.mmn - oo.F oo.m Favo. oooo.- m c.~ Fo.n. o wo. wo.- n FFoo. mmoo.. w Fm.F on.- F om. mm.F FmFo. Fmoo.. m wooo. mooo.- v FF. mm.F- m mooo. oFoo.- w wo.” oq.F F ~5.o.wu «no co_msFucF Loccw acoFoFaaooo co LooLo vconcoam coaamutoow mFao .Lotco FoacFmoL ooaoeFamo «F no .coFaoFLo> co .Fgo oooaaoc_z aca .quco FFaacFac .Fguv agOFo: ooLu .Fuoo «Foo oocoov uwacFaeauuo .F.F oooza Loo goaou co FFV aocw «canon Loo guaou co acoFawuLooa Facaamcouv 3 go F. Facwocwoochv F» Faa_caF .m azaa 33 . mamo mmgFom gm>o Fm mFgMFV maco :F gmamcF new goo Fmooe amgammzm soak F I < Fdwv qumo wFQQ mmmomo F.owaooomma oF gm>o Foam x omauFomS 51.xv mszomeg we maoFo .m .oFm 0:: 000 000 80 0N5 00. DOB 00v 2; a.“ Q.N O.“ vow Duo 9.0 «.00 0.0! —._I. V v . .p I 12 I I I I I I - I . I ill \I II 5 . I I III I ._ .e In :3. I In. I I II 10 . I - I I ‘ ‘ - I I I - - v I I I I I II I I I v I 8 HO I z I I T _ _ _ I I I v S T I I. G I. . flH F I I H I .I r- —.| .l v v I I . w I F I 1 TM IN I." 5" 18001838 .Auzmwgv umuu_umga Lm>o Am m_nmpv mumo c_ Lmumcw uLm Low Fmvoe umELowmcmLu Eogw mpmzuwmmz .AummFV psu_m: aoLu Lm>o Am m_amhv mumo :_ Loumcw ugm Low Pmuoe umcammzm 50L» wauwxv mszuwmmg mo uo—a A.u:ouv .m .mwm Jag 31 Q mm FIDHJI mug) fl u ,L v x H ¥/ r_.u.bflu H Um ,L .u uflm.L m C. n N . o- T m. 9 mm on S on m; 2 m, , P P b F F l..| —> b b r brr+LLP rLri b r++ \— Lr r P ?¥ bLLL I» ‘t[ Y, .0 I V. I I I u I ' VI I I I I IMU I II I m: u I I I " III I .. I m I II In-III w I I I I I I n I I an I I r ‘ ‘ I , I I}. I II HI I I I M I I I I I I I I I I 0 I a 34 WUHUISEE 141m WHHUIQ 35 y. = axblddeChb3r24b4wb5e was tried, where the variables are defined as on page 30 (eqn. l7). This model was evaluated by first taking logs of both sides. Some vari- ables were recorded by adding 1 to avoid taking logs of 0's (Sokal and Rohlf l969, p. 384; Draper and Smith 1966, p. 132). The new model was log(yi+l) = log a + bllog(xi+l) + bzlog dd + balog Ch + b4log(r24+l) + bslog(w+l) + log e ..(l9) Recoding and transforming the variables did seem to produce a better fit to the data (Table 6). As expected, the effects of outliers are greatly reduced as is indicated by a general increase in R2 values. The question of whether this makes for a better model. however, is largely a matter of personal statistical preferenCe. Certainly the' transformation did reduce the variance instability, as can be seen in the residual plot (Fig. 5d). For most practical purposes, though, the increases in R2 values caused by adding environmental effects to the sweepnet model do not seem to justify the extra effort needed to take and use the measurements. The predictions still lack precision. Regretfully, the simple regression models of Tables 3 and 4 are recommended as "best". Instar specific sweep catch does not appear to bear a l: 1 relationship to square foot catch. The effort to establish the expected true relationships has led to equations that may be useful yet lack precision. 2.6 An additional data set Previous data sets comparing sweep catch with absolute density were thought to be either too small or to have unreliable instar identi- fication methods and thus could not serve as a check for the sweepnet 3o onv. any. omv. mum. .m~. oqv. mmv. one. won. NwN. msw. MNN. mmN. -m. o—n. mom. 0mm. OmN. no.— Koo. ~mo. ov. o~. mo. o~.~ wo.— mv. ow. o—. can. Pm. “N. m». w”. o~.n —o. -.— m_. oo. Nmno.wm Lace“ uuuvcaum ~.oop a .>.u oooo~.m - u m, .>.u .uucv. co.aaocoug or» cuc— ooULOe «on nook, Loo cuuou .noov uxqu oucoou vouq_:sauua nn.~- flucaumcoUv Nvm. ONO. m to co. ~mo. n .p.zvoo_ mm. m an ac. “no. o~. .c. ~ a...mcvoop ~mo. moo. __. _ A..n.voo_ oNo. co. Aucoucooooc.. I..~»voo_ ”no. . am o.om . .>.u ...m_.__ . a mm..- ”acouw:0uu vo. ._.: A..nvoo, a... No. o.. a no 90— v... .K. mm. m 5,.qntwoo_ wo.. ~n. an. ~ to no, . . _o. _ ,_.m.voo_ ”MM. M”. Aucoocoooocpv A—omxvoo. vvo. - um 0.0m a .>.u o-omm.m - u ova. flucauwc0uv .m. ._.c on oo. o~.. co. o~. . A_..choo_ o~¢. co. a... n to oo_ mom. ”N. ~_. ~ “.H‘Joo, m-. wo. a“. _ N,...Vom_ ~_o. «n. Aucoocooooc_v ”p.»xvoo— #10. I um v.Vo I .>.U ooomm.m - m mo.~ Aucouwcouv me. nv.. m “pavucuoo— oov. no. an.- . cu oo_ Non. .o. .o.. m no oo_ qma. an. -. ~ A_.1cmo. OON. m_. ”0.. _ fi_._.wmw, 000. .m._ Aucoucooovc.v Apo.xvoo_ 3V :2 S _ o5 Macaw.» ml“ Ifimwmfl “co,u.»e00u co canto Locum co.«mo.noa agave-um bawz: I .>.U oocnv.nv I u hnv.- AucquwcouV ._.c ap.vwcvoo— ._.c -¢3v oo— -. n uv oo— moo.. N go we. om. _ ._...voo_ Aucovcoauuc_v A_onxvoo. owe. . N» m.mo . .>.u coo—v.—— - u ~o.~ Auccumcouv mo. m fl—.:voo_ wo.. v ._.v~cvoo— ~m.. n no oo— Nv.- ~ cu oo— mn. n A_.n-voon Aucoccoaovc.v A_.mxvooF a_o. - Nm m.mv— - .>.u cca-._~ - m mv.n nucouwcouV mo. m A_.zvoo— ~o._- . A—.v~cvoo. _c._- m an ac. mm.- ~ cu oo— n—. — “pow-Voo— Aucoucooaoc.v Ap.~xvoop wmoo. - no ~.M—_ - .>.u oocmm.op a u ~_.~ Aucuumcouv o—o. m A_.3voo_ ~o—.- 9 so oo— Nm.- n A_.«~.Voo, ~o.- N no mo. mo.~ _ A_._~Voo_ Aucwucwaouc_v A—ovaoo_ New co—nguc_ ugtwrco» uco_u_s~oOu .o Lougo co—mmocoom wp ac u:~.u.ecoou .LOLLU ~03U'm0L gag—umo 9.— mm. .Axv vooomoc.n van .Aq~cv _—oec,oc .Acuv “co—on nocu .AIV ooo:« Loo sou-u co Axv uooe atone“ con gouau yo mscoaucogu no. yo co,mmogooa .o o_no» 37 models developed here. A new set was obtained from the APHIS (USDA) parasitoid rearing laboratory at Niles, Michigan. This included paired sweep and square foot samples from l5 wheat and l3 oats fields. Each field was divided into l4 equal size units and a single square foot was sampled in each unit. All larvae were retained for later age classifi- cation. A large portion of each sweep sample was measured and there is no reason to suspect any biases were introduced by the counting or measuring processes. Table 7 describes the APHIS data set. In comparison with the PWS data (Table 3), all slopes from the square foot over sweep regressions are steeper and all intercepts but for 3rds in wheat are not signifi- cantly different from 0. The density per square foot had a higher average for all instars in both crops than the densities found in the fields described in Table 3. To a certain extent, the high R2 values for lst and 2nd instars were aided by the presence of many 0's. There were 53 and 42 pairs of 0's for lsts in oats and wheat, respectively (5l% and 45% of the samples) and l6% of both crops had paired 0 counts for 2nd instars. However, the good fits and high slopes of the regression coefficients in general cannot be explained away. These results were not altered by including windspeed, crop height, or accumulated degree days in the regressions. It must be concluded that the relationship between sweep and square foot catches of CLB larvae is not well understood. Although differ- ences between sweepers has been thought to be negligible (Ruesink and Haynes l973), perhaps this should be reinvestigated. Until then, use of sweepnets for intensive population studies of the CLB has been gen- erally curtailed (A. Sawyer, E. Lampert, pers. comm.). When used in 38 Nwm. woe. New. Pvm. an. no. mm. ov. a: v—m. n—m. .m—. Nmo. Foam mpum no mwmzm mmocm>u a .00.. mm.~c o.vmmq .a w ~.oo_ m~.oP~ -- q.oomo_ _ An.ov a e so.v m.oe~ m.~on~F _m w n.0vfi o.NmP -- m.mmem~ F An.ev a m Foo.v m.m_ o.m~m_ Fm m N.mop ~.NFF -- o.moo_ _ ”n.0v x N _oo.v N~._P o._no_ Fm m N.mmm «.mn -- e.mwa _ an.av a _ nywm maw wm mm umum mmummw ceumcm a am._ om.- emm. o_o._- hoe.~_ MNm.e q co.m mm.~o vow.” mo_.m- omv.o~ mam.“ m Nm.F mo.o_ mew. v¢~. ¢-.m mem.~ N mq.F ¢_.m~ men. own. Nmo.¢ ~N¢._ — .m.m we .m.m _o .n.m _x A_v ccumcH mmmom mFaEmm :_ mo>cm_ .mpOP Oman mucmsmcamowe Pogo» mo mmmou uoou ocuzmm Hdmxz .Loccw + xa + m a x m? pmvoz .guuou do co_uo_>mu ucuucoum . .u.m ,» .wo>co— mgu Lo» amozm Lea guano uca co. m—. FF. mucmwce> um m_mx~oc< ppm. omm. mo—. woo. vm—. ONN. mno. omo. mo—m mmom cop mmmzm .v Loumc_ mo aowzm Log cuuou moocm>a . .x .mucm_uw»uoou co Loccm ncmucoum moo. m.vom m.mmmom No— u v.~oq no.5 -- m.mmep P ”o.mv a e So.v ~.Nmm o.o~mmm mop u m.m~m mo.m_ -- “.mwmm P An.ov a m So.V em.m~ Rm.o~m~ ~o_ m o.NNm wq.~. -- m¢.om~ F AD.QV a N So.v No.~ m~.o_m me. u ~.¢_m am.~w -- em._mF P An.ov a _ 4mqw mflm mm mm qulw mmummw LgmeH 1M. mo.o _m.n_ -~._ mmo. oo~.q~ Nmm.m v up.m Kw.vm ~mm.~ Non.~- vnm.m_ Nmm.v m .m.¢ mm.m— onm. m_~. Nmo.m mum.p N m_.~ ~—.o~ «mp. Pm_.- Nmm.— ch. _ .w.m we .m.m _m .u.m w» Rwy Loumc~ mm~v_ m_QEom c, mc>co_ _cuo~ - Nmmm mucoEmcsmcoe Pope» vo~ mmmou mmmw mummmw wmmm u .m.m .maopm u we .uomucmucv u _o .e...— n P .m Laumcw do poo» wcoacm con guano goo» ocozcm con cuuou cmmzuoo a_zmcovumpmc mcu mc_n_cummu .uom «you Acou .umm\mmpa5om mm cu m .mfiascm\mqoo3m cm on — .xm_coa .muoo .uomcz «aao.mm iiom.vN aaam.mw caav.NN «aao.om atom.om .m.cm. ¢a¢N.Nm «cam.—¢ cram.mp .Locco + xn + o u » mw ~muo: Cum. owm. mmm. mvm. Nam. mom. NRO. wa. vwm. o—m. (Z A-m_ mommy .uom\»ooc acmsam _ x on .mumo . o .uomgz u z A.Eeoo .mcma .mezomv .umm\uoom wgoacm — x on .oupawpo + mumo omo.— mpm. mom. com. one. nmo. new. moo. mo“. omm.- mom. NNF. mmm. mmF.u New. mpo.Pu mom. mom.n cos. Nno.u .u.m x .x .come mop van a .mucowcm> v~m_m cwcu_z mo. cmozumn a_;mcowum~og one mmo. mmm.~ mmo. om~.~ mew. mv~._ mF—. Pmm.~ mmo. vm~.P mmo. mam.— vmv. «mm. mmo. moc.p mmo. coo.p mop. Fo~.F .m.m a mmo. m—m. mmo. Pym. umo. mmm. omo. amp. omo. omm. Ono. mnp. mco. mmm.u mpo. Nmm.u NNo. .va.u ego. ch.u .w.m a .mxog F—ao .ONmF v .wxmg ——:u .mmmp m .xuaum gums: ucmumoaza .mnmp N Aon, xcwmasav .—Fv.P was.” mpm.~ mmm.p mow. mmm. cos. m—m. mom. mpo.a Nmm. mmm. Npm. mpc.pn ovm. mom.~u Nam. moo.Pu mum. Npm.n .U.m A em mm mm mm ow mm woo» mcmamm .mxog ppao .mmmp _ .0 «mac .3 comm v v gmxzom mza m meow mezom mza m m uoom mcoamm m mmm xcwmmaa umcmmmzm w<>m<4 mmzm poo» mcmamm NNN ~— mmzm xcwmoaa a mhgso< oucaom .m apno» ! szfl,“ w. ..~.ll. 4D ..l min-.4 1| Al . a 46 also gave non-different intercepts but a significantly greater slope for the catch per square foot regression. This latter result is espe- cially curious since it seems more logical to expect the relatively greater area covered in the sweepnet samples (50 to 125 square feet per sample) to produce less between sample variance than the approximately 5 square feet covered for each of the absolute counts. The ratio of catch per square foot to catch per sweep, a crude "M" value (as per section 2), for the PWS adult data was 1.96, compared to an average ratio of .66 for 15 sample sets by Ruesink and Haynes (1973). These 2 discrepancies suggest that the adult sweepnet model (Ruesink and Haynes 1973) bears reexamination. There are no significant differences between the regressions of larvae per square foot from the PWS and the 1975 Gull Lake data (Sawyer 1976)(Tab1e 8).- There is also no significance to the difference be- tween PWS and Sawyer data for the egg regressions. It is interesting that the regressions for eggs per square foot fit most closely to the larvae per square foot regressions for the same locations, but, again these differences are not statistically significant. It is unlikely that there is any difference between the variance x mean relationships for eggs and larvae for wheat, oats, and barley. Ruesink's adult results (Table 8) were from 3 crops and there were no detectable differences. Gage's (1970) data likewise had no difference between wheat and oats (Table 8). 3.2.3 Statistical distributions of CLB egg and larval counts Although the PWS regressions of Table 8 adequately fit the data (R2 = .848 and .885 for eggs are larvae), the fitting of 2nd order regressions leads to interesting results (Fig. 6). The 2nd order VRRIRNCE LOG [BRSE 10] VRRIRNCE (BRSE 10] 1.00 Fig. 6. 47 -1.2 —0.0 -0.4 -0.0 0.4 0.8 x.2 1-6 LOG [BRSE 10) MERN 3.2 l LARVAE . LOG [BRSE 10) MERN The relationship between variance and mean for CLB egg and larval count data. 48 regressions provide only a slight increase in R2 (.035 and .028), but they do suggest that the variance x mean relationship is not a simple log-linear one. It is useful to establish the significance of this in terms of the underlying distributions. From Fig. 6 it appears that samples from low densities (less than .5 per square foot) have a poisson variance x mean relation, which is closely approximated by the 2nd order regression. This had been sug- gested by Ruesink and Haynes (1973). Regressions through those points with mean less than .5 confirms that the variancezmean relationship is approximately 1 : l. Tests for goodness of fit of the pooled samples' frequency distribution also suggest a poisson (Table 9). The Poisson distribution is a limiting case of the Negative Bino- mial. As the negative binomial statistic, . k = mZ/(sz-m) - (25) ((21) rewritten) goesto infinity, the distribution approaches the Poisson. The variance x mean relationship at higher densities is that of (25) but, from Fig. 6, may be different from that of lower densities. It is therefore reasonable to ask if the distributions of count data are in fact negative binomial and whether they share a common k. Although several computational schemes exist for estimating k (Anscombe 1949, Beall 1942, Bliss and Fisher 1953, Bliss and Owen 1958) most estimates have biases. An unbiased, not necessarily minimum var- iance, estimate of k can be obtained from the maximum likelihood est- imates (MLE's) of the k's for individual samples, as in Table 10. In Table 10, column 1 refers to the samples listed in appendix 7b. Columns 2 and 3 are the individual sample's mean and variance. The k for the sample, column 4, is computed using Fisher's MLE (Bliss and Fisher 1953, 49 NN.N A:mo.Nx ”44.? u Nx NON. u Amvmo. m\¥ m__o. u m\¥ N.m N V+N o.me .4 P N.Nmp NON o mm>cm4 «N.N Avao.Nx ”Nm.fi u Nx «No. 1 Aevmo.m\¥ mmNo. u m\¥ N._ e v+N m.o_ m N ¢.PN Ne _ m.P¢N omN o momm .MM >ocgwEm Nucmammgm Nucwzmwgm uczou >ogomospox umwumaxm um>gmmno ANN do mmmzooou 44*.F.m ooN. mmN. Nao.- amp. mNm. Nee. moo. NON. mmm.- NP aa>cm4 4*«NO.N _oN. ooN. Nom.- meo. NON._ NNF. eN_. mNN. Nem.- NN mumm .m .mm .u.m x .w.m a .m.m m .u.m x wmmmm muHHmHHwmv mm—qum Fm>gmp ncm mom com cemmwoa we“ op mcowusnwgpmwu zucmzcwgm pcsou ucwon No a?» mo mmmcnoom com mummy can .5 mop n+m u m mop mcwm: .E mop co m mop mo mcowmmmgmmm .m m—amp N N 50 Estimated negative binomical k's for egg and larval sample counts 1.1111 (3 1 I). C and test for goodness of fit using sample k and connmn k . EGGS r rxr----- COMMON K-' LARE(OF1 Orv SA J -----rc --usx~ CHI-SQU K) (8.5. MEAN VA? NC. I C) . 7 O I O O O A O A U 0 U A 0 AA. 0 O O O O o o O O N O N O O O N O NN o O O O O O O O O o I \I””\l’1’,,’\I”””\l””),\I\I"1’)”’))”|””"I'”)|I)”\I\I”’|l)‘0’)”Il,\l’1,11’5 85859685261121311‘2223735673“?1.922110690212621111“25669505113001“2335161“232E 1 l. ‘ll‘l‘l‘l‘(“l|l|‘l|‘(l\l|l\l\l|l\l\l\((1““‘((l\“l\l\(l\‘l\(‘l‘l\‘l‘((“(l\lll\((‘(l‘l‘l“l\‘l\l\‘((‘A‘A‘p r (595029637676557“96.101.903.03153576372283761173§652056Q750306172727131636767 6A 822276689960875093768“k0.3730836797902925“15530681055905966238095562h03659505 2(531360711837516353057327 179311183150793“18.1.9131h2259537155“9303:.39197285327 cocooncoco-00.00.000.000.oooooooooooooooooooooooooooooocoo-cocoooooaooooooopr 91.“.59595283 716 125h7kk 1157 1 2 “1 21 51 5&13329 55 2 372611153E11QC 3 1 4w 1 . 2 1 11 1 F r... 8 A o A c A A AAA M. N N N N NNN U h ,’”)’,’)),,’,)’,”,’,’)),)),”),,,,,),,”’,,’,,,,’,”’,”,,",,,",,”’,,, 0.985qeehzs11521212222287 “5733213211106“8212821111hzs7khh031020c.lv3233“1e 13132 ‘llJIl|((‘(l\(((l\l\‘l\“l\“l|“l"(l‘l‘((((“((l\l\“(‘(l\l‘““((l\‘l‘(11(‘((““((((((l\l\l\, 5 7.6765582533538310 925b715760“31h697.1332.95.»?3013537B827711h1.795331932510185???b 1.216532669809515h5637“53020792332“10672999692015.49900760~20k50J56507638“2.72)“ 9932106172B07833555062030902.4233.4150173036.96201“0066333035022002h513.96“2&979 00000000000000000000.0000000.000.00.00OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOCOOOOFJ 23“.)9615161 2 1 1GQ323 1231 2 b 9 3 121551“ 12 1 113 5 “1 o 7.. 3 : ’,,,)’,),,”,’,,,’,”\l),,,,)))‘I\I’””)”)”,,”’,,,”””’I’”)””‘0)”,”,’k 136.59“ 13977321663721399352h363986575753225823?-““25019790. ~97 “14767166389375ECQ. 65..“231531229“ “027525611030211093056830535““67937573337510.8139“21300255126226618 o 00000000000000000000.0000000000000000000000000000000000...on.ooooooooooooooE 6..) 11 75 5773““8129 731 6 15 2 81171 35111 1 5 13 1 1 7 227. o 1 31 11 7.. 1 e 1 S 1 2 ( (((l‘l‘lIll“(l\‘lI((‘((“(l|l\l\l\((((l\l\“(((‘l\l\‘((“(‘(lI‘AlI‘l‘o-‘(I‘l‘1“‘((l\(((l\l\l\l\All“ C) 5613536163755721fl1997676502396.U55616201669732061’95“37 13566‘25273c177217 165’“ B112.U:H “53531976189955339LQ 21166—58620 77 h6966617226‘55h231“ 226057375“69ficr)66362582 .0000.0.0.0000...000.0000.0000.0.0.000.000.I.OOOOOOOOOOOOOOOOOOIOO0.00..0003 17123 170 3““75ahh961 602 5125 B 23253 26h12 11 .b 2 25 3 1 6 7147.2...» 1 5 a . . nu. . 11 .2 o a o o 3 1 c . g . . .1 1 . . - 1.3 731C211010 11c...—17Q.fl4.5nwn1 Q11.Lc.ns711h 120:0)??qu §9L7%Q%.bhq 30.2.L7 aura/777 17 h1gaLH15q 15.71», .W 90(1er no 915111383h907986968“58123“62133792536997536688616255691306h353h5~c8217h99903K .00...OCOOOOOOOOOOOOOOOOOOOOOOOOOOO...OOOOOIOOOOOOOOOOOOOO000.00.000.00...O 5196032517 11 1 1111163357 53 3“ 1 927?. 171 1 10383722 31 1 «(92392 Q. 1 3 3.52 3E3. 1 9.1 1 1 1 1 2?. 1 1 2 3 OHPCN 0733000002!330003773377707700000777377 0707 070007033330007373337737300333700C 06330...»003.0303000566336660 b60030066636b0606060006533330506363336620 600c033600 63956h282h583750526576688h21301506h30060666006““2175368209k733h836688h51C8BO oOOOOOOOOOOOOOIOOOOOOOOOO000000000000000000.0000...0000000OOIOOOOOOCOOOOOIOE “JG/2769.313 1 21151111173215 3111611 “7.51 1511 21251122 .4 1 11119. d 13111.1 11 1 1 A H I 'l 1?!“967890123567892123h56780C23hc1t749023“,.)67131457890121.678.12.. h5789f.1?1.k67r.1c 1111111112233???27.7.11 33! 331 31.1» AA A b h“§55555c.:her56$.b€. 77777777 R! A.“ RR R Q ”a F. .LILBVAE K we MEAN Table 10 (centinued) NC. 0 .b . 7 O O O O O O A O O O A O A O O 0 AAAA . o. u N o o N o N NNNN O O O O 111,111,1111771111a!”77,11,1717117,7,77,71,1777771’6'”7’771711111111777117115 379397Q69223Q65~kn935k191fi32111679777721032121371877721021111312221k27ZloauOE 1 L (I‘l‘l‘llltt‘ltl|l\‘l\l\‘lll\l\l| ‘(l‘lI(1111110‘1‘1l\l\l\‘l\‘l\(““l\((‘l\‘l\‘(1((l\(((“(‘n\(l\(ll(l\t|l\lll|rr P 0001925032161185801936036563637866727115A21k67151156kh856856h610€09112760781A 80075028A703611092699820A675199“35169079638231751713871123619597360178“008705 27“C.C.C‘1C;1323(~7C.€1~8973501-91Ca23c.0212¢.16C43Canflca226971‘9““JEZZ)&.C7233001'.61~5€““.6017 o00000000000000.0000.00000000000000000o0.000000000000ooooooooooooooooooooooor 690650761 A2106 55961222A63 798366922 11 1&6159671 11 3 2 1A1115651 C 1 21 1 1 12 11 2 .1 s R E 8 o A O A o A AAAAAH N N N NNhNNU K ))))”\I7)\I)’”’)” )7, A)-7,7,77,17,777’7,’77,771.”,77,7)777777777177,77,771), 39308522822265 9 0* h “955 .V 161332121889656521022121381958621121111312221627 200000 1 . ((((l\(‘l\“((“‘(‘l"(((“(‘(((‘(((((l\‘(‘(((l\(‘l\“(“““(‘(‘l\l\““((‘(“‘(‘((, J 730159396365527.55153675019339925257 “39616052566935279363973 3.1135133521903216“ 53033972675all”h6?23232525655622569926067A79k72b58752550089069938“36653103337 5020A80186OAZQHQ7“70%35222203001101339“070671602122996308A53231A780327700000“ o00000000000000.0.000000000000000000000000000000000000000000.0000000000000001 1863632 2 A 233 18A56121 2 3 7QBQ7Q9 12613966 2 1 135 o 1 5 1 : 77”)7”,’,’)’,)’7’),,”),”””)’7)””)”)’,7))777”),”)’),’77”,””)’,)K V.“9753972C69107636757A50.137629515316610533“32A017672372371523C9h7588319860A6 3A19220038320620582165229631h“6957717223098229621217726029“A904E““2375560631o OOOOOIOOOOOOOOOOI.0..0....0......OOOOOOOOOOOO.OOOOOOOOOOOOOOOOOOOOOOIOOOOOOOE :12 1 2 2371135 1“... 321313 3 9 9 Z 60: 23983 S 1 3 3113 262210 12 11 o 1 25 5 7 1“ 22 1 S ( (((l‘l‘(((l‘l‘l‘l‘l‘l‘ll‘(((‘l‘((l‘ll(‘((((((l\(l|l\l\‘l\(‘((l|((((l"ll(‘(l|l\l|l\l\l\l\(ll((l\l|l\l|l|‘11-‘11 1 8631k5629195k966c71““791905898616321161517AA67A631A6859A1269155161&2357627569 37396k1hfi1930506)“28.2503817136380114“635“53131“1915239965657526.‘ 91C8 “33715577211 OOOOOOOOOOOOOOOO.0...0.0.00......0.0.0.0000...OOOOOIOOOOOOOOOOOOO000.00...001 GA 12 6 113562351 122 33912102520 7 6156123266 9 31 2 313211393“ 231213 B . o 1 21. . c . 2 3 g a 1 . 11 c a . . . . o . o 3 o o c o o 2 a. 17.? .HA 1:31 “A 7.0?11P.6261«11~C§1GL1A1Q.“27489.976133627673“147.616.41.31... «9.9.?!Rct.2131! 1! C 1C0. 789577288362~93#72155A87803753685073960“7236907966129“5672755290181202863335K o0.0000000000000000000000000000000.00000000000000000000000000000.00000000000 19%,.0666113279633356875 621 1 0.6657651 23 111223700“ 11 2 1111510 11.13621 7 13 7.3.2 6 2 1 395). 2 A 2 1 OHPCN 77700730700003033377777000000003737330700003730007030300330070307037030337°7C 666306306000330330666660000050036363306C0303630006330300330060?0603653333606 C6002096637A89615A8A0A652390686169037627636729866005216217066A7PAA5229A532?20 o00000000000000...no.0000000000000.0000300000000000...00000000000oooOOOOOOooE 192Q3b173117352232163? 1371 11375b¢1 21 1 15 3Q151 1 1 1 111 21o1 7 I 'l 17.2..«5678C12Arj760.0123A5670123A5E7n.0123A5790123A56780.0123A57931?.3Ar...678c.01h5697:: 11111111??2?????331313133AhhbuuhA95555555556686668577777777778894flan 52 p. l82). For some of the samples, a k was not computable. This was so whenever l) the mean and variance were equal, giving an infinite k, 2) the mean was close enough to the variance to cause a failure to con- verge on an estimate of k within a reasonable series of calculations, or 3) the ratio of mean : lst trial estimate of k was less than -l (see Davies l97l, p. 368-374). Although the negative binomial para- meter k is usually considered to be a positive number, negative esti- mates were also calculated (cf. Bliss and Owen l958). As a check on whether the individual samples were in fact distri- buted negative binomially, the theoretical sample distributions were generated from sample mean and MLE k. Observed and theoretical distri- butions were compared by a x2 test for goodness of it (Table 10, columns 6 and 7). 0f the 68 testable samples of eggs and larvae, only 2 egg and 3 larval count distributions were significantly different from expected at a .l a level. Due to the small sample size, l0 to 15 counts per sample, the minimum expected cell size for the X2 test was set at l, rather than the usual 5 (see Steel and Torrie l960, p. 350; Sokal and Rohlf 1969, p. 565). Such test results should be evaluated cautiously, but it is safe to conclude that the samples can be repre- sented by the negative binomial distribution. From the individual estimates of k and their corresponding vari- ances, a weighted average was computed as the estimated common k, kc = Zwiki/n ' (26) where wi, the weight, is the inverse of the variance of ki and n is the number of sample sets (Steel and Torrie l960, p. 180). The pooled variance estimate is var(kc) = (nl-l)var(k])+(n2-l)var(k2)+...+(nm-l)var(km) (27) n +n +...+ - l 2 "mm 53 These estimates are listed in Table 10 (cf. Bliss and Fisher 1953, p. l94). Two tests were used to determine the commonness of the estimated kc's. The individual sample distributions were compared to theoretical distributions generated from'kC and the respective sample mean. Goodness of fit was tested by a X2 statistic (Table 10, columns 8 and 9). For eggs, l7 of 70, and for larvae, 13 of 69 testable cases (24 and l9%) were significantly different at a = .l, causing some question as to the commonness of the k's. It should be remembered, however, that setting minimum expected cell size to 1 makes the X2 test more susceptable to fluctuations in smaller cells. A 2nd test, illustrated in Fig. 7, was suggested by Bliss and Owen (l958). The individual sample k's are plotted against the sample mean (here transformed to logs to give clearer separation). For neither eggs nor larvae are the points substantially different from the mean common k, and this is confirmed by regression through the points (Table ll). Although there is considerable variance about the kc's, the esti- mates are not correlated with the mean. A check was made to determine if the aggregation of larvae was related to age. Although the individual larval square foot counts were not subdivided into counts of each instar, an overall estimate was made for each field. The individual field MLE k and the variance : mean ratio from Table l0 were each compared to the mean instar for that field (Table l2). There were no correlations and plots of the data suggest no higher order relationships. It is assumed, therefore, that there is no significant change in the distribution of larvae attribu- table to instar. The common k may be accepted as common for all age 54 _m>gm_ cam mmm mbu go» some m_aEmm mgu mo mop new x mFaEmm 042 cmmzpmn qwcmcowpm—mc mch zqmz flea mmqmw 0.~ 0.“ a." 0.0 v.0 _ F F * b h x. y 5 004 0.0: P > b. P F 5 by P P L y 0.0: p F b b F r L b h > L m¢>m¢4 0.0: N. «I 09‘ V v 1 M UBiUNIiSB .oumn pcsou g .3... quz 000 mmqmg 004 om. rllm....flb...ne m5-.-na-..wa-.-m¢x r I TL... I 0 I I— VI I . I I II I flI IulnIII I ID II Idfllr I3... II I .I I I v I v 1% m M UBLUNILSB 55 .m.:~mm.~s Fmo. em. Fm.m .m.:npn. ~00. «m. Fm.m .m m .u.m x N .Lmumcw came m? x van .5 :mmE Lw>o prmL cme " .m.=oom. Foo. we. mmm. .m.c¢¢_. coo. me. emu. .m .mm .a.m 2 map .mgczou Pm>cmp ucm mam cow cams mop vcm m.x _wweocwn m>wummwc mpasmm :mmzpma nwgmcowpmpmg map mm. mm.~ wo.- om.~ .m.m a Ne. no.0 No.m om.~ mm 5 u > cm. mm.~- mo.w P~._ mm x .m.m m .u.m N mmmou Lm> Lo vx m? x mews: .xn + m u a mw choz .memcv .m_ mpnmh NN.P en._ Fm. mm. mm.m No.p on mm>gm4 m¢.¢ mm. mn.~ N~.~ mm.¢P mm.N mm mmmm .m.m a .m.m m .u.m u mammu mucmmcm> mzu can x mesocwa m>wummmc mFaEmm :mwzuma awzmcowwmpmg ask .15 mop n+m n wx m? quoz .FF mFQMF 56 groups. There seems to be no reason for not taking the estimate of kc for eggs to be the same as that for larvae, using the above tests. A single estimate of common k, kC = 2.75, is the arithmetic mean of the egg and larval kc's. It is adopted for the development of the related transforms, below. 3.3 Variance stabilizing transforms The negative binomial is a skewed distribution, especially at lower densities. Transforms that reduce variance instability (depen- dency on the mean) may also reduce the skewness of the frequency dis- tributions. Results from different transforms are often similar and choice of transform may be up to user preference, with little apparent difference in the variance : mean relationships after transformation. The choice of transform depends on the perceived relationship between variance and mean. For contagious negative binomial type dis- tributions, Beall (l942) used an inverse hyperbolic sine transform, 2 = sinh'](SQRT(kx))/SQRT(k) ~ (29) with k, the negative binomial statistic, z, the transformed count, and x, the original count. Kleczkowski (l949) used the simpler z = log (x+c) (30) with c, a constant. Noting that the transform was less effective at low densities, Kleczkowski (1955) proposed the curvilinear modification 2 = log ((x+c+SQRT(x2+2cx))/2) ' .(31) Choice of c was dependent on the intercept of a standard error x mean regression for small, untransformed counts. To compensate for the apparently common difficulty of transform failure at low density, Hayman and Lowe (l96l) used the log variance x 57 log mean relationship (22) of Wayman (1959). This is equivalent to s2 = loamb (32) which was simultaneously used by Taylor (l961a) to describe the vari- ance x mean relationship of count data from 24 biological surveys on various animals, from viral lesions to ocean fish. From this relation- ship, these authors used Kendall's (1973 (lst ed. 1948)) transform, 2 = fx(l/SQRT(§))dx. (33) which leads to z = x(]'b/2) ~(34) The "power law“, equation 32 (Taylor l965),and its transform (34) minimized the range of variances of several aphid count sets (Taylor 1970) and was found to be slightly better at this than the commonly used log (x+l). Note that (34) is not useful when b = 2, in which. case Taylor used the log transform (Taylor 1961a). Table 8 suggests that a common regression of 2 = .25+l.33 log m (35) log 5 is a reasonable expression of the within field log variance x log mean relationship for both the egg and larval square foot counts under normal farming conditions. Using Bliss's (l967, p. 128) correction for under- estimation of the arithmetic mean from log data, ma = mg+l.1513ss (35) with ma the arithmetic mean, mg the geometric mean of the log data, and 52 the error mean square (here the squared average s.e. for eggs and larvae in Table 8), leads to a variance x mean relationship of 52 = 2.llm]'33 (37) for eggs and larvae. Using (34), an appropriate transform is z = x'33 .(38) 58 Note that this transform depends on a lst order relationship between log variance and log mean. Fig. 6 suggests that this may not be valid. Ruesink and Haynes (1973) observed that "b" was closer to unity when the sample means were less than .5 per sample. Perhaps this is why Taylor (1970) found the transform less effective at stabilizing vari- ance at low density. Development of second order regressions, however, leads to expressions which are beyond the scope of this thesis to integrate. The above transforms are compared in Table 13. Two criteria are used here to evaluate transform effectiveness. The simplest test is to compare the ratio of maximum to minimum variance with a ratio established under normality assumptions (Hayman and Lowe 1961, Taylor 1970). The test statistic is c = (1n p)/2. (39) a normally distributed variable with variance 32 = l/df, where df is the degrees of freedom of the sample variance, and 1n p is the natural logarithm of the ratio of maximum to minimum variance. Here, the average df is 12.1. The expectation of a range in a normal distribution is approximately oSQRT(n), where n is the size of the sample supplying the range (Masayuma 1957). Thus, under normality assumptions, for both eggs and larvae the expected minimum ; = SQRT(92/12.1)/2 = 1.38. Initially, all transforms were evaluated using a search technique to minimize p, the ratio of maximum to minimum variance. That is, the transforms were optimized for ;. Constants of transformation were selected and evaluated iteratively until ; changed by less than .l%. Subsequently, the correlations of the variances and means of the trans- ‘Formed sample counts were computed (Table 13). 59 mm.p oop. No.1 ok_. om._ ooo. Po.F oop. F~._ okp. . oo._ ooo. om.” _oo. oo.o oo.P moo. No.o oo_. oo.P oPN. oo._ ooF. oo.~ oo.. oo.. moo. oo._ Foo. oN.o .u .w 0m.m_ mw.P¢ mm.mm ¢~.m~ m¢.om N0.mm FF.¢~ um.wmvm on.mF ¢—.Pm mm.~p 00.~N m0.NN m0.w~ em.mp m¢.mmm¢ .m. Poacoz ono ok.o Aooooom\hflxoocoomvp-ooom ooo. . onpooo\fiflxovooomo_-oooo ooo. ox oo.x oN.F AN\AAxoo+oxocooo+o+xoo oo_ ooo. Ao+xo oop uw>cmmno PoEcoz ono ok.o Aooooom\fifixovooooo_-ooom Nko. Aooooom\fifixovooomoF-oo?m moo. ox oo.x ook. AN\AAXUN+Nxvooom+o+xvv oop pom. Au+xv mop um>cmmao u .ucopmcou ecowmcoch .N\Ao :PV u u .NN. cogu memmcm mm mo. u a no ucouwmwcmwm .pcmwuwmmmoo cowgopmccou u c .mucmwco> Eoewcws op Easwxoe mo ovum; .u Foeocos Loy umNTEouoo macommcmcu $0 comwcoosou mo>co4 mmmm .mp mFQMF 60 Another criteria for optimizing transforms is to select constants so that the correlation between variances and means of the transformed counts is minimal (Table 14). This also tends to reduce the c statis- tic. A graphic check was made to be sure the lack of correlation was not due to an introduced curvilinearity. There is evidence of residual correlation at low density with all transforms (Fig. 8, for example), but only 1/3 of the overall variance after transform could be accounted for by fitting up to 4th order polynomials (Table 15). The effect of transforms on the sums of squares and the analysis of variance has been discussed by Beall (1941). In general, gross changes in significance of test statistics will not occur as a result of transformation, but some differences abOut the relative importance of less correlated variables may occur. The log transform of Table 6, for example, changed few of the conclusions from Table 5, except for the cases of lsts and 2nds in wheat, in which the order of inclusion in the regression was altered. For all of the transforms evaluated, chosing constants for minimal correlation between variance and mean helped reduce the skewness and kurtosis of the counts, although all remained leptokurtic and somewhat . positively skewed. From Tables 13 and 14, all transforms are better than none for reducing variance instability and approximately normal- izing the distributions. The simple log transform, with constants .132 and .170 for eggs and larvae, will be used here as it is the simplest and has marginally better c's.l Table 14. Eggs Larvae Table 15. Eggs Larvae 61 Comparison of transforms optimized for minimal correlation between transformed variance and mean. Transform Constant, c .3 Observed 4.23 log (x+c) .132 1.64 log ((x+c+SQRT(x2+ZXC))/2) .337 1.60 x2 .280 1.60 sinh-](SQRT(cx)/SQRT(C) 1.485 1.60 Observed 4.39 log (x+c) .170 1.72 log ((x+c+SQRT(x2+2xc))/2) .450 1.71 xc .283 1.81 sinh-](SQRT(cx)/SQRT(C) 1.112 1.71 An example of higher order variance dependencies following transform. Model is y = big, y is transformed variance, g_a vector of orders of the transformed mean, using 2 = log (x+c). Variable* bi s.e. 33_ £33 Overall 22 -.525 .172 .290 .290 36.83 24 .111 .222 .322 .032 21.12 2 -.133 .071 .327 .005 14.23 23 .256 .153 .348 .021 11.60 constant .386 .024 . 22 -.559 .134 .240 .240 28.39 23 -.043 .193 .331 .091 22.00 24 .247 .198 .334 .003 14.68 2 .095 .077 .345 .011 11.45 constant .340 .021 * Listed in order of inclusion (Nie et a1. 1975) ** All F ratios significant at P <.OOOl ** 61a VQRIQNCE Z fl. LFlRVQE ." Z:LOG(X+C) - :Il-III‘l ‘ 'I I ’ I 'I ll. 0 n... VRRIRNCE Z D '00‘w’1—r—r v v v I 1 'fi ‘ v v v 1 r v v u v v °-i.2 ~0.a —o.4 —o.o 0.4 0.8 1.2 1-6 2.0 MEQN Fig. 8. The relationship between variance and mean for CLB egg and larval counts following transformation. LINEAR PREDICTORS OF CEREAL LEAF BEETLE DENSITY 4.1 Introduction The goals of this section are to find a way to use previous mea- surement information to reduce the error term of estimates of CLB density and to predict seasonal density using early season measurements. This requires a search for relevant environmental factors and linear coefficients relating these to historical changes in density. Rates of population change based on these regressions are then used to estimate present population levels, according to recorded weather and / or biol- ogical information. The most recent measurement, if any, may be aver- aged with the projection for an improved estimate. Models of expected future weather or biological events may be used for a density forecast over an entire season. The PWS environmental factors measured were temperature (described as degree days, base 48F), windspeed, wetness, rainfall, and crop height (see section 2). Average larval age was estimated from subsample head- capsule measurements. Egg parasitism by Anaphes flavipes was determined from laboratory rearing of weekly collected field eggs. The applicability of statistical description of change in CLB larval numbers in response to the environmental factors may depend on the relationships among the independent variables. Several of these bivariate relationships are discussed in section 4.3. The development of improved estimators for larval density follows in section 4.3. A multivariate regression is fit to the PWS data, using transformed 62 63 counts as the dependent variable. Partial derivatives of this estimator are then used as “filters", in the sense of Hildebrand (1976) and Sakawa (1971). 4.2 Parameters of the temporal distribution of CLB eggs and larvae. Multiple regressions tend to mask relationships among variables. For the CLB, these relationships have intrinsic properties of interest which have been studied in an ongoing research program for several years. The PWS data set will be used here as a vehicle for review and corroboration of several of these underlying bivariate relationships. .The distribution of error terms on the independent variables will not be discussed here because most of the field measurements (ie. crop height, windspeed, etc.) were made only once per count period. In future studies, these error terms ought to be given the same attention given to the error term of the dependent variable (section 3)(see Box and Tidwell 1962). Out of necessity, these errors will be considered normally distributed with stable variance. 4.2.1 Adult influx The most important variable initially affecting egg (and hence larval) density is the spring influx of adults. An estimate of the number of adults available for oviposition may come from regional sur- vey data taken early in the season or from samples near the field of interest. Casagrande (1976) found the rates of adult emergence from wintering sites was quite similar for 4 years of Kellogg Farm data. Regional incidence curves for 3 years of sweepnet data also showed a consistant pattern with peak regional density at roughly 200 dd's and a gradual decline over the next 600 to 800 dd's. This is essentially 64 the pattern for PWS adults (Fig. 9). The PWS peak adult density in wheat was at 206 and 238 dd's for the square foot and sweep sample units. Buildup to this point is quite rapid as the bulk of adult emer- gence takes place over a relatively short span. The adult population then decreases in an exponential fashion, -A(t-200)’ (40) N = N e t where Np is the peak poputhion and t is "time" in dd's. The mortality rate, A, has been estimated to be .0071, .0071, .0037, and .0045 per dd for the 4 years of Kellogg Farm survey. The factors that determine this rate are unknown. Casagrande (1976, p. 29) reported A was independent of host crop type. The seasonal total of "adult degree days" (the area under the adult incidence curve) is sensitive to this parameter. Mortality rates of .007, .005, and .003 applied to the PWS regional density in wheat at 200 dd's resulted in 55.9, 70.9, and 100.4 adult dd's. The number of adult dd's may be used to estimate total egg pro- duction. Table 16 reports results of several studies. Disregarding the lowest Teofilovic estimate, mean eggs per dd from 6 studies is .35 (s.d. = .044). Wellso (1976) reports average total oviposition per female in laboratory studies is similar to that of field beetles. Multiplying .35 times l/2 the total adult dd's gives an estimate of total eggs produced. The rate of adult movement into oats depends on factors which have not yet been analyzed. Oats is a preferred host for CLB's and acts as a sink, with beetles leaving oats more slowly than they arrive (Casa- grande 1976, p. 48). Although this should eventually establish an equi- librium with other hosts, the level at which this occurs and the factors RDULTS/SHMPLE 65 L31 * SYMBOL CROP UNIT J . A". ..-'~_ 9.- 7 :i: r»: r ‘ - >K HERT so .FT. . C) ORTS SHEEP . fit 1 30 = SRMPLE SIZE I l 400 1200 DEGREE DRYS (BRSE 48F] Fig. 9. Changes in adult density over accumulated degree days. Symbol size indicates number of field means included in sample point. 66 Table 16. Mean eggs per female CLB per degree day (48F). Source Eggs/dd Conditions Teofilovic (1969; ref. in .325 30°C (86°F) Wellso, Connin and Hoxie .055 16°C (61°F) l973) Wellso, Connin and Hoxie (l973) .272 26.7°c (81°F)a .362 26.70C (81°F)b .38l 26.7°c (81°F)C Wellso, Ruesink and Gage (1975) .388 26.700 (81°F)d Tummala, Ruesink and Haynes .370 unspecified (1972) a laboratory cages, equal numbers of both sexes b laboratory cages, sexes separated before oviposition trial c field collected, oviposition measured in laboratory d mated once weekly determining this level are unknown. In the lst simulation of the CLB ecosystem, Ruesink (T972) assumed 90% of the beetles in wheat went to oats. Casagrande (1976, p. l44) plotted adult densities in strips of oats. Density increased until it was 10 to 20 times as great as in wheat. The proportions of land area in small grains and in spring grains are important in determining relative densities (Ruesink 1972, p. 28). The spatial arrangement of the host fields is also undoubtedly important in this respect. In the l975 PWS area, oats were about l2.7% of the small grains and were about evenly interspersed with wheat (map, appendix 4). The adult incidence curves vary considerably from field to field. Between plus within field variance is more than within field variance 67 for all life stages (Table l7). The Ruesink regressions for adults and larvae in Table l7 are the same because they came from pooled data. There was no visible difference between the regressions for the 2 life stages (Ruesink, pers. comm.). For the PWS data, fields were grouped according to sample date (appendix 7c). Analysis of variance for the 3 incidence curves in Fig. 9 reveals that much unexplained variance remains after inclusion of many envir- onmental variables and after adjusting for between field differences. Table l8 includes dd's, crop height, windspeed and wetness. Between field differences were handled as classification effects, as were the field wetness ratings. Each field or wetness classification was given either a "0" or a "l", depending on whether a sample fit that classifi- cation. Singularities in the §f§_(or the correlation) matrix were. avoided by eliminating l wetness classification and l field, as per Draper and Smith (1966, p. l37-46). Less than 5% of the cases had missing crop height, windspeed, or wetness variables. All cases were used and the correlation matrix was constructed with pairwise deletion of missing variables (Nie et al. l975). * Apparently, either sample technique or selection of variables were inappropriate for the adult density model. 0f the environmental factors, dd's accounted for most of the variance. Between field differences accounted for about l/3 of the variance (Table l8). Two final observations on adult densities are appropriate here. The amount of host crop planted is highly susceptable to economic factors that are not necessarily related to pest density. ‘For example, Table 19 shows the change in total acreage and numbers of fields in the PWS area from l973 to 1975, when the relative value of wheat was 68 .mmmp .xvsum poms: ucmummnza m .Foo.v a pm acmu_wwcmwm mummy d m .AMNm_v mmcxmx can xcwmmsm 206» muwumwpmpm .mawcmczOu m_ we comm c? m_asmm mcmz meLmn can :606 .mumo .pamzz co m6~m_c mar 0“ 6: .mPQEMm 666 56666 m? “we: .xm>a:m msu m>vpatmaoou .mom_ .N®m_ _ 55N.~ “mm. «mm. mmm. mNN. Nem.~ Koo. moo. . mmc. mNF._ m mza poo» mcmzcm m00m 885m.mm mom. Nam. Nmm. moo. _¢o.P omm. “we. NNQ. m_m.— o mza poo» mcmmvm -- Ffim. -- -- mmo. cmm.~ mmo. moo. -- _ -- mu xcwmmaa N F um: mmzm wq>m<4 55ao.mp mam. owe. omm.- mmF. 0mm._ moo. moo. omw. Nmo.F- N_ mmza poo» mcmswm aao.m 0mm. com. mm_.- mom. mqo.m «mo. arm. mac. mpm. o mmza -- me. -- -- omo. «mm.F mwo. moo. -- l. mu xcwmmzm N F um: mmzm mp42o< u mm .u.m x .m.m n .m.m a .u.m x : wucaom .Loccm + xn + m u x m_ pmuoz .x .cmme oo_ :0 .x .mucmwcm> U_mww c_;uw3 mafia cmmzump mop cmmZumn awcmcowumpmc mzk .NP m_nm» 69 Table 18. Analyses of variance for adult CLB's per sweep in wheat and oats and for adults per square foot in wheat.$ Model is 2 y = p + afx_+ error, where bf = (dd dd2 ch ch w mf), dd = degree days, ch = crop height, w = windspeed, mf = wetness classifications, and f‘ = field effects (0,1). _a_ SSR (d.f.) SSE (d.f.) _F_ fi Wheat: catch per sweep b_ 44.9 ( 7) 204.1 (394) 12.38*** .180 .f 75.7 (60) 167.3 (364) 2.75*** .379 (9_ f)‘ 127.3 (67) 121.7 (334) 5.22*** .511 Wheat: catch per square foot b_ 4.1 ( 7) 27.1 (396) 8.53*** .131 .f 10.0 (63) 21.2 (341) 2.55*** .412 (b_ f)‘ 17.2 (70) 14.0 (333) 5.85*** .552 Oats: catch per sweep .b .24 ( 7) 1.45 (125) 2.97** .143 .f .59 (23) 1.10 (122) 1.33n.s. .293 (b_ j)' .67 (30) 1.02 (102) 2.24*** .397 $ = Reduced models are not to be confused as being parts of compound model. Differences in degrees of freedom are attributable to missing data elements. Table 19. wheat and oat planting in the PWS area. Figures in paren- theses are for Michigan only. Michigan data, 1973—74, from M. Holmes (APHIS-USDA, pers. comm.). 1974 PwS figures from R. Gallun (USDA, Purdue University, pers. comm.). 1973 1974 1975 Fields Acres Fields Acres Fields Acres Oats (12) (115) 14 (11) n.a.(94) 17 (15) 181 (162) Wheat (11) (111) 46 (45) n.a.(519) 73 (58) 1234 (933) 7O responding to high international export volume. Oats acreage was com- paratively stable during this time. The presence of a constant CLB population in the area from 1973 to 1975 would have resulted in an apparent decrease in average field density in wheat. The distribution pattern of agricultural fields may remain stable from year to year. Fig. 10, for example, maps the distribution of 1/4 sections containing oats in the Jackson County, Pulaski township, Mi- chigan, CLB survey site for 1973 to 1975. Although the number of oat fields dropped slightly in this area, there were few sections in 1974 or 1975 which did not contain or were not adjacent to a section with an oat field the previous year. Similar maps of all the CLB survey sites listed in Appendix 2 suggest that this continuity of pattern was gen- eral. If the degree of aggregation of a population is intrinsic to a species (0. Kendall 1948, Taylor 1965, 1970), it must nevertheless be susceptable to the spacial dynamics of crap planting pattern. Assuming that the adult spatial wintering pattern is closely related to crop pattern in the previous summer and that the spring host selection flights of adults can cover at least a mile diameter area, present crop planting habits in the Michigan CLB survey areas will do little to alter the between field CLB adult count variance x mean relationship from year to year. The remaining error in the adult density models is large, but is understandable. Sweepnet and stick toss techniques have long been suspect as adult sampling methods, but have been retained for lack of suitable alternatives. Larger sample size would reduce error in density estimates. Further discussion of the nature and causes of between field variance of adults will be deferred to a later study (A. Sawyer, 71 .xpcsou cemxumw .awcmczou mempza .mnfimwm 3.1 . III. $1930» .xmdé thGO zomxos. .mN-MNm_ .cmm_:6,z “no mcwcwmycou mcowuumm cmucmsc mo mcowumqu .o_ .626 1: $19 MN? 72 PhD thesis, Michigan State University, ca. 1978). 4.2.2 Crop height Crop height (ch) is an indicator of several "states" of the plant, including its age. It is an indirect summary of all the factors that have been experienced by the plant, such as fertilizers, water and warmth. Each state in turn may be important for its influence on CLB density. Those spring crops which are planted late will be exposed to a decreasing regional adult population and will receive fewer eggs, even though younger crops are preferred for oviposition (Shade and Wilson 1964a). There is a tendency for spring grains to have higher densities than the fall planted winter wheat for the same reason. The relationship between wheat and oat ch and accumulated dd's has been graphically illustrated by Gage (1972) and Sawyer (1976). The relationship is variety specific and consistant, and, like crop ma- turity, may normally be subject to little modification by other environ- mental effects (Wiggins 1956). It is essentially that of the PWS data (Fig. 11, but note that a base of 48°F was used, not the more common base 42°F). Further analysis of this relationship suggests that growth of wheat may be much more uniform across the region than the growth of oats (Table 20). Degree days account for over 80% of the seasonal variance for wheat, but less than 40% for oats. A hypothetical explanation for this lies in the difference in planting times for the 2 crops. Although the pattern will vary from year to year, there was a 12 day period of cool, dry weather in October 1974 (59 dd accumulated) when it is assumed most wheat was planted. Wet 30 35 25 20 15 CROP HEIGHT 10 5 73 GP 4 . NHEFlT J ‘ ORTS J 1 .1 .4 .J I 7 I ‘ l ' l ' 1 ' l 20 40 60 80 100 120 DEGREE DRYS (BRSE 48F) x10 The relationship between crop height and accumulated degree days for the PWS data. Regressions are: oats... y = -19.5 + .069x - .000022x2, R2 = .388; wheat... y = .47 + .061x - .000026x2, R2 = .837. Table 20. |m Oats dd dd. dd, Wheat dd dd, dd. Table 21. Date Oats 6/2 6/9 6/16 6/20 6/25 Wheat 5/14 5/21 5/28 6/2(3) 74 Analyses of variance for wheat and oat crop height. dd = degree days, f_= field effects (0,l). Model is y = u + alx + error. SSR (d.f.) SSE (d.f.) §_ 33_ 5412.5 ( 1) 8652.6 (172) 107.6*** .385 dd2 5462.4 ( 2) 8603.1 (171) 54.3*** .388 dd2,_f 12666.0 (25) 1399.4 (148) 53.6*** .901 46709.0 ( 1) 10057.0 (840) 3901.0*** .823 dd2 47515.0 ( 2) 9251.0 (839) 2154.0*** .837 ddz, i 52200.0 (65) 4566.0 (776) 136.5**? .919 Percentage of eggs parasitized by Anaphes flavipes, PWS 1975 (data in Appendix 6). Degree Number of Number Day (48°F) Fields of Eggs "9°" Perceft Sampled Parasitized (-s.d.) 565 18 413 11.0 (10.5) 663 16 435 12.6 (9.4) 804 16 449 86.6 (14.9) 924 19 444 94.6 (8.2) 1067 13 193 98.8 (3.2) 220 17 960 0.1 (.46) 345 17 1002 0.0 (0) 492 17 390 7.3 (6.4) 565 17 360 21.0 (20.3) 75 spring weather drew out oat planting times over roughly a 3 week period of 260 dd's in May. The result is a greater spread of oat planting times on a physiological scale than was experienced by wheat, hence the larger between field difference. Inclusion of between field differences 2 in the oat crop height regression thus brings the R value to .90. 4.2.3 Egg parasitism by Anaphes flavipes Anaphes flavipes is an established CLB egg parasitoid in the Galien area (T. Burger, pers. comm.). In the PWS survey, parasitism was moni- tored weekly from May 14 to June 2 in wheat and from June 2 to June 25 in oats. Leaves bearing CLB eggs were clipped, returned to the lab, and transferred to a plastic petri dish. Emerged larvae, or eggs with clearly identifiable contents were noted and removed daily. All egg development was usually completed in 7 to 9 days. These data are in- cluded in the individual field summaries of Appendix 6 and are summed up in Table 21. Table 22 lists the mean and standard deviation of total egg counts made in wheat and oats. In addition to the separate egg counts dis- cussed above (section 3), many fields were simply given a total count per 10 or 15 square feet. All egg samples are thus included in Table 22 and the statistics given are the statistics of field means. The individual field density means were used to determine the least squares fit of density over polynomials in degree days (up to 4th order). The percentages of parasitism in Table 21 were fit with 2nd order curves and the estimated number of parasitized eggs per square foot was calcu- lated (Fig. 12). The areas under the total egg and parasitized egg incidence curves were compared for an estimate of seasonal egg parasi- 76 Table 22. Statistics of average egg density by date and degree day. De ree Number of Density (per square foot) Date Day ?48°F) Fields Mean S.D. Oats May 27 478 8 21.133 12.687 30 533 7 4.343 5.111 June 2 565 13 3.195 4.276 3 579 1 .200 0.000 5 621 4 10.500 5.696 6 637 11 3.636 3.261 7 643 1 .500 0.000 9 663 17 2.388 2.202 12 725 19 3.002 3.048 16 804 17 2.543 2.823 17 829 1 4.067 0.000 18 858 21 1.248 1.679 20 924 21 1.457 1.577 25 1015 21 1.111 1.218 Wheat May 7 142 27 8.788 13.077 8 154 8 17.600 12.094 9 170 18 2.878 1.783 13 210 33 20.582 21.033 14 220 16 6.850 5.698 16 231 42 22.674 23.985 19 290 4 16.025 13.805 20 320 39 15.105 13.920 21 345 17 18.508. 20.854 23 386 5 22.471 16.064 27 478 19 8.684 6.004 28 492 27 3.332 2.587 29 512 6 5.394 4.715 30 533 38 3.266 3.072 June 1 544 5 2.840 2.970 2 565 5 2.220 3.190 5 621 4 1.250 .994 6 637 12 1.281 .729 7 643 5 .347 .345 77 tism. For wheat, this estimate was 2% and for oats 29%, although the early part of the season was apparently missed in many oat fields (Fig. 12). The effect of egg parasitism was to prevent the later part of the larval incidence turve from developing. A, flavipes thus competes with larval parasitoids, as has been discussed by Haynes and Gage (1972). The seasonal rate of egg parasitism depends on many as yet unknown factors and varies by region. The extent of egg parasitism has been higher and has occurred sooner in several eastern U.S. survey areas (T. Burger, pers. comm.). Although the biology of A, flavipes has been studied (Anderson and Paschke 1968, 1969, 1970), little is known about native alternate hosts and there is no knowledge of the wintering , habits. It is unknown how the A, flavipes density pattern of the Galien area in 1975 related to other areas or times. 4.2.4 Age of the larvae The average age of larval insects is of interest if it relates to changes in the efficiency of a sampling technique. This problem was discussed above for the CLB. The average age also has value if it is relatable to changes in insect density. If differences in crop planting time, for example, make it impossible for insects to lay eggs in field 8 until much later than A, the subsequent average age of larvae in B would be expected to be less than that in A at any single time. The incidence curve in B will likewise be shifted to a later part of the time axis. The question then is: If mean instar is substituted for degree days (or another such "time axis") can the predictability of the incidence curve be enhanced? 78 “: 0818 EGGS / SDURRE FDDT 600 480 560 840 720 000 080 980 1040 DEGREE DRYSlBRSE 48F] ” - . WHERT EGGS / SDURRE F001 .1. . o V V T—rfif Y Y I v v v 1 r Y Y I v v v I v v v " v V T—riv v Yj '0 180 240 320 ‘00 4.0 530 .40 720 DEGREE DRYS (BRSE 48F] Fig. 12. Mean egg density (upper curve) and estimated incidence curve for eggs parasitized by Anaphes flavipes. 79 Following the notation of Lee, Barr, Gage and Kharkar (1976, pg. 40-43), the physiological age of an individual CLB can be repre- sented by 2, an indicator with a value of z = O for a newly laid egg and z = 2 for the oldest possible age of an individual. Then at a time, t, the number of individuals in a particular age group between (21’ z. + dz) is x(zi,t)dz. For CLB larvae, 2 takes on the 4 discrete 1 integer values for the 4 instars, as no continuous indicator of age is available. The average for many individuals, however, is a continuous variable, with values from 1 to 4. Lee et al. developed a partial differential equation to describe the maturity distribution of the population as a function of time. For now, the change of the average age of larvae over time, 2 x (zi,t)/i, is all that will be considered. Fig. 13 shows the relationship between average instar and dd's for wheat and oats. The regression based upon the corrected sweepnet samples is included for comparison. The individual field instar means, as determined by square foot sampling, are plotted but not the means based on sweepnet. The second order dd equation accounts for approximately 1/3 and 1/2 of the instar variance for square foot counts in oats and wheat (Table 23). Crop height has relatively less relation to instar and inclusion of ch and dd's in one model does not improve fit (Table 23). For wheat, the regressions of Fig. 13 and Table 23 may not repre- sent the true regional pattern as only a small part of the total number of fields were included in any day's samples (Table 24). Also, the square foot samples in wheat represent only a short part of the total incidence curve. The sweepnet samples are, of course, subject to the 80 INSTQR MEDN V I Y Y V Y I Y Y V f I’ +1 1' V V r r T V T I Y Y T r T r V r 400 480 560 640 720 800 880 960 1040 DEGREE DRYS [898E 48E) 1 NHEDT MEGN INSTRR v TY’I Y f Yfi—f r V 1 f I ff 1' r Virfi' T Y Yfi ' 1 . . . 1 360 400 440 480 520 560 600 640 680 DEGREE DGYS [898E 48E] Fig. 13. Relationship between instar and degree days. Plotted points are for square foot samples only. 81 Table 23. Analyses of variance of larval instar. dd = degree days; ch = crop height. Model is y = a + bx, 5, SSR (df) SSE(df) .5 33 Oats-Square Foot dd 27.32 (1) 22.68 (116) 139.8*** .55 ch, ch2 10.50 (2) 39.50 (115) 15.3*** .21 dd, dd2, ch, ch2 30.70 (4) 19.30 (113) 44.9*** .61 Oats-Sweepnet dd, ddz 3.20 (2) 26.80 (137) 8.2*** .11 ch2 .18 (1) 29.82 (138) .8 n.s..01 dd, ddz, ch, ch2 3.93 (4) 26.08 (135) 5.1** .13 Wheat-Square Foot dd, dd2 7.96 (2) 14.00 (92) 26.2*** .36‘ ch 2.22 (1) 19.74 (93) 10.4** .10 dd, dd2, ch 8.37 (3) 13.59 (91) 18.7*** .38 Wheat-Sweepnet 2 dd, dd 35.41 (2) 21.00 (185) 156.0*** .63 2 ch, ch 10.93 (2) 45.47 (185) 22.2*** .19 dd, ddz, ch, ch2 35.97 (4) 20.44 (183) 80.5*** .64 82 Table 24. Summary of larval instar samples. ------- Square Foot------- ------Sweepnet---------- Date dd Number of ---Instar---- Number of -—-Instar---- Fields Mean S.D. Fields Mean 5.0. Oats May 27 478 3 1.816 .369 3 2.394 .345 29 512 1 1.875 0.000 30 533 6 2.464 .500 2 3.238 .081 June 2 565 12 2.451 .617 3 579 6 3.118 . .189 4 599 7 3.520 .319 5 621 1 3.520 0.000 6 637 5 2.542 .789 10 3.390 .264 7 643 1 2.500 0.000 9 663 12 3.396 .265 12 725 19 2.576 .532 18 3.109 .319 16 804 16 2.953 .288 17 3.196 .563 17 829 1 3.153 0.000 1 3.420 0.000 18 858 20 3.273 .391 17 3.383 .339 19 891 4 3.594 .081 20 924 20 3.601 .307 20 3.617 .215 25 1015 16 3.675 .372 20 3.467 .665 Wheat . May 23 386 2 1.526 .339 6 1.728 .094 27 478 13 1.970 .414 19 2.705 .277 28 492 26 2.377 .378 27 2.768 .323 29 512 6 2.421 .454 6 2.708 .781 30 533 36 2.706 .312 15 3.207 .279 31 544 5 2.991 .574 June 1 556 5 2.691 .428 8 3.113 .548 2 565 3 579 2 3.690 .099 4 599 48 3.525 .343 5 621 6 3.570 .122 6 637 1 2.809 0.000 13 3.575 .274 7 643 1 3.357 0.000 4 3.423 .275 9 663 17 3.478 .232 18 858 17 3.663 .240 83 sampling biases discussed above. The samples in oats were accumulated over a longer period. A relatively complete representation of the area's fields was obtainable within one day (Table 24). 4.2.5 Rainfall Shade, Hansen and Wilson (1970) recorded the mortality of the 4 CLB instars during 2 rainstorms in 1966. Third instar larvae experienced a significantly greater mortality than either lst's or 2nd's. The rate of precipitation in the storms was .31 and .36 inches per hour, falling over periods of 45 and 25 minutes. Total mortality was 14.5 and 13.5 percent for the two storms. Little else has been done to look at the dynamics of populations in storm conditions. Although a rain gage kept track of the precipitation at the center of the PWS area, this is not. presumed to be an accurate estimate of the precipitation in all fields. Rainfall is entered in the following regressions only as total inches in the approximate 24 hour period prior to sampling (based on the single rain gage). 4.2.6 Insecticides One of the reasons that the goodness of fit of instar over degree days (Table 23) is not higher than .55 may be the occurrence of insec- ticide applications on some of the farms. These roughly coincided with "normal" peak larval density, during a period from 600 to 800 degree days. This is reflected in a decline in mean instar. The mean for the oats' square foot sample on dd 725 (Table 24) was 2.58, which is close to the 533-565 dd measurement (mean instar = 2.46) and is less than the value predicted from the overall regional regression (Fig. 13) (predicted mean instar = 2.85). 84 Insecticide use on CLB's is extremely common and is a major factor affecting population levels and population dynamics. Although they are usually ignored in studies on the ecology of the CLB, insecticides are in fact a part of the normal agricultural ecology of most regions. Any attempt to predict future regional beetle or parasitoid densities, or to evaluate spatial or temporal distributions, must be preceded by an under- standing of the dynamic effect of insecticide usage. For example, it should be realized that the perception of between- field variance (52) x mean (m) relationships may be affected by insecti- cides. Two cases will illustrate. Case 1: log 52 = a + b 109 m. The use of sprays at higher density will still produce a lst order estimator with expected coefficients ”a" and "b". Case 2: log 52 = a + b1 log m + b2(log m)2. If a lst order equa- tion is fit to the data, the coefficients "a" and "b" will depend on the degree of spraying. Removing those sub-populations with higher means will result in a lower estimate of the lst order slope. A few moments relfection on Fig. 6 will make this clear (see Pielou 1969, pg. 91-97). ' Note that although use of 2nd order equations to describe variance x mean relationships may produce little increase in goodness of fit (section 3.2.4), the resulting estimator may be preferable, depending on the likelihood of insecticide usage. Although it is not always possible to witness insecticide application directly, the incidence curves for eggs and larvae and the instar curve may contain sufficient information for inference. A precipitous drop in egg and larval density and concurrent retardation of the rate of instar 85 advance may be taken as indicators of insecticide use. Since actual knowledge of whether a field had been sprayed existed for only 5 of the 21 oat fields studied (maps, Appendix 4), §_posteriori inference was necessary. Graphs of square foot density and instar changes in each field plus sweepnet and crop height information were all reviewed. On this basis, it was determined that 10 of the fields were not sprayed and 11 probably were. In the following derivation of a filter for reducing measurement error, only the unsprayed 1975 PWS fields were used. 4.3 Filtering 4.3.1 General considerations Any estimate of beetle density based on count data has an error term associated with it, as discussed in section 3. This section is an evaluation of a process called filtering, which is an attempt to improve current density estimates by using previous measurement in- formation. Extension of this same process to a complete seasonal incidence estimate is also discussed. Filters are also called "linear minimum variance sequential state estimation algorithms" (Sage and Melsa 1971). Their basic idea is not new (Weiner 1949, Kalman 1960) but application to entomological problems is uncommon (Hildebrand 1976). The parametric approach used here is a very simple "first pass" at using the technique. This section is thus meant to be an evaluation of the potential for develop- ing a complement to the more sophisticated discrete component simula- tion methods (Gutierriz, et a1. 1974; Tummala, Ruesink and Haynes 1975) which will ultimately serve as the basis for an on-line pest management system for CLB (Haynes 1973). In the interim, the predictor may also provide a useful, low-cost estimator of seasonal density. 86 Gage (1974, p. 79) reported the initial, peak and last CLB larval populations in oats (:_s.e.) to occur at 375 (:12), 723 (:25), and 1218 (:46) dd, respectively, for 6 years of Gull Lake Kellogg Farm studies. In what follows, it is assumed that the shape of the larval incidence curve is determined by the environmental factors discussed above plus the available ovipositing adult population. The incidence curve clearly cannot begin until after the crop sprouts. The last eggs will be determined by some unknown combination of adult mortality rate and plant senescence. For normally planted oats it is assumed that the curves should approximate Gage's (1974) characteristics. The area under the curve also will determine its shape. This area is related to the potential egg input and thus to the adult influx. Either of these could be measured early in the season to be used as an indicator of the larval curve's area, but not without complicating the measuring and modeling errors. The algorithm used here, which is meant to apply to CLB's but which is generalizable to other pest/crop systems, requires only an initial estimate of larval density and initial environmental condi- tions. Current environmental measurements plus the derivatives of a multiple regression of historical density and environmental variables are then used to project a current or future point. Since it is beyond the scope of this thesis to evaluate the environmental and sociological factors that govern density patterns in sprayed fields, these were not included in the analysis. The only suitable large data set available for parameter estimation was the 1975 PWS unsprayed oat fields. This provided 80 density measurements in 10 fields. The 1976 PWS data and 87 three fields from the 1974 Kellogg Farm work were available for evaluating the filter. 4.3.2 Density models Table 25 lists several univariate models of density (transformed as per section 3) over environmental factors for the 10 unsprayed PWS fields. Transformation used the z = log (y + c) relationship developed in section 3 to improve variance stability prior to analysis. The relationships between log density and dd's or ch (Table 25) are clearly curvilinear for larvae. Eggs show less reliance on higher order terms probably due to failure to sample during the lst part of the incidence curve. This is because most oat fields were not located until very close to or after peak egg density. The resulting density pattern has a negative decay rate somewhat like the adult mortality pattern (Fig. 9). The factors, field wetness (m), rainfall in the previous 24 hours (r24), average larval age (1), seem to have little relative influence by themselves on density. This, however, may be due to the way the rainfall was measured (section 4.2.5) and to inaccuracies in instar determinations (section 2.2.2). Certainly, further work on the impact of rainfall is needed. Sprouting date (5) was also included as a variable. These data were determined by applying the quadratic formula to the ch regression of Fig. 12, adjusting the y-intercept according to a field effect vari- able. That is, another regression was run using ch = a + bij_+ bzdd + b3dd2 as a model, with ch = height and 3:, a vector of field effects, coded (0,1). Combining the constant "a" and the bl: product into a single term produced a ch equation of the form, ch = a° + bldd + 88 Table 25. Univariate models of density [Y = log (y+c), section 3] over several environmental factors. Eggs Mode1a Cases R2 c.v.b F. sig.C dd 80 .327 84.5 .0001 dd. dd2 80 .330 84.8 .0001 ch 80 .334 84.0 .0001 ch.ch2 80 .358 82.5 .0001 ch. ch2.ch3 80 .359 83.5 .0001 .f (10) 80 .354 87.4 .0001 m2 79 .009 102.5 .411 m3 79 .014 102.3 .305 m2,m3 79 .031 102.0 .304 r24 80 .039 100.9 .079 i 61 .312 85.6 .0001 1.12 61 .318 86.0 .0001 s 80 .072 99.2 .016 Larvae 2 c.v.° F sig.c .009 141.4 .393 .323 117.6 .0001 .012 141.2 .331 .303 119.3 .0001 .376 113.7 .0001 .347 119.5 .0001 .006 141.6 .512 .012 141.2 .328 .014 141.9 .574 .021 ' 140.8 .268 .104 135.8 ‘ .041 .037 139.4 .089 aModel is of form Y=a + b x, where x includes variables listed: dd=degree-days base 48°E5—ch=crop height (in.); £?(0,1) field effect (10 fields); m , m =damp or wet field conditions; r24= rainfall in previous 24 hours; i=instar; s=sprouting date bPercent CProbability regression is N.S. 89 bzdd. The quadratic formula was used to find the roots of this equation, one of which was easily accepted as the degree days accumulated by the sprouting date. The results (Table 25) indicate that sprouting time by itself contributes relatively little to the total variance. 0f the variance in larval density explainable simply as between field effects (f), only about 10% could be attributable to sprouting time. Figure 14 shows the log density regressions over dd's and ch. The predicted density was transformed (y = 102 - c) before plotting. The poor fit to either variable by itself is evident, but not unexpected. In Table 26, combinations of several of the environmental factors are evaluated. Although inclusion of additional variables will in 2 values, up to R2's of about .78 for full models general improve the R for both eggs and larvae (Table 26, bottom), these larger models are not very useful. The inclusion of all measured variables plus "field effects" in the full model yields an "adjusted R2" (Cohen et al. 1976, p. 16) of only .71. The combination of dd's (2nd order), ch, and the dd x ch interaction term provides an adjusted R2 of .454 for larvae. This combination is simple in terms of numbers of variables to be measured but it provides significance and reasonable fit. The effect of sprouting time could be accounted for in the interaction term. Also, the fit of ch over dd's (Fig. 11) makes substitution possible, facili- tating differentiation. This regression is, therefore, chosen for use in the rate calculations. Supplying the appropriate coefficients, the larval predictor is 2 Idd + bzdd + b3ch + b4dc (41) 2 60 + bldd + bzdd + b3(p]dds + pzdds) + b4dd(p]dds + pzdds) log (y + .17) b0 + b FT. EGGS/SO. FT. EGGS/SO. 90 I. A O ” 4 FT 0. 1 LRRVRE/SO. . . -.I . I “ u I I ( . I ' ' . . _ ' - II I. . V ' v ' f j v V v ‘ v ‘ v ‘ 4‘. Jo 030 no no no no um am m 000 000 m can no no I000 01' DEGREE DRYS (BRSE 48F) LRRVRE/SO. FT. 0 I W I! 20 II N ' ‘0 CROP HEIGH (INCHES) DEGREE DRYS (BRSE 48F) CROP HEIGHT (INCHES) Fig. 14. Relationship between egg and larval densities and degree days and crop height. Table 26. 91 Representative multivariate models of density [Y = log (y+c), section 3] over several environmental factors. Mode1a dd,ch,dc dd,ch,dd ,dc dd,ch,ch2,dc dd,ch,dd2,ch2,dc dd,s dd,s,dds ch,s ch,s,chs dd,s,dd2 ch,s,ch2 dd,ch,s,dd2 dd,ch,s,dd2,ch 2 2 dd,ch,s,m,r24,i,dd dd,ch,s,m,r24,i,dd ,ch 2 2 ----- Eggs----------—-- ----Larvae-------- §a§§§_ BE_ CLULE F Sig. RE_ C.V.b F sig.c 80 .446 77.6 .000 .417 109.9 .000 80 .473 76.2 .000 .481 104.3 .000 80 .474 76.1 .000 .427 109.6 .000 80 .438 75.6 .000 .518 101.2 .000 80 .404 80.0 .000 .045 139.7 .167 80 .459 77.3 .000 .168 128.7 .001 80 .391 80.9 .000 .039 140.1 .217 80 .418 79.5 .000 .269 123.1 .000 80 .407 80.3 .000 .362 115.0 .000 80 .410 80.1 .000 .343 116.6 ‘.000 80 -- s not entered-- .390 113.2 .000 80 .501 74.6 .000 .525 100.5 .000 ,ch2,i2,dds,chs,dd25,chzs I 79 .573 72.2 .000 .594 98.5 .000 2,i2,dds,chs,dd s,ch2s,: (full model) 79 .578 57.0 .000 .781 75.9 .000 aModels are of form Y = 95, where elements of_K are degree-days base 48°F (dd); crop height (ch); sprouting date (5); plant wetness (m) (see text); rainfall in previous 24 hours (r24); instar (i), and field effects (f), plus intercept (b0). bCoefficient of variability, percent. cProbability b_= O. 92 where: bo = -5.516 :_1.07l* b] = 1.345(10'2) :_3.12(10‘3), b2 = -6.814(10'°) :_2.224(10'°), b3 = 0.105 :_3.302(10'2) and b4 = -1.515(10'4) :_4.041(10’5). Also, p1 = 6.838(10'2)‘:_1.567(10'2), and p2 = -2.257(10'5) :_1.016(10'5), from Fig. 11. Then, since ; = 10K - .17, (42) where K is the right hand side of (41), 91 = (y + .17)(c1 + c dd dd + C3dd2) 1nlO, (43) 2 ('5 II 2 1 b1 + b3(p1 - 2p2dds) + b4(p]dds + p2ddS ), 2 2(b2 + b3p2 + b4(p1 - 2p2dds)), and C3 = 3b4p2. with dds, the estimated dd's at time of sprouting, as discussed above. where: ('5 II This equation (43) describes the rate of change in the average 1975 PWS unsprayed oat field larval density in response to dd's and ch. To adjust the rate for other densities, the initial measurement in the field is compared to the predicted density from (42). The ratio (measured + .17) / (predicted + .17) is multiplied by the rate (from (43)) to produce a corrected rate. The addition of the .17 was neces- sary to allow the predicted density to asymptotically approach zero, instead of being restricted to a finite range. * :_standard error of the coefficient 93 4.3.3 The weighting of the second observation with the predicted density. An initial field measurement of larval density is needed to start the filter. The rate of change in density (43) can then be applied (section 4.3.4) for an estimate of current density. The question of interest in this section is how to best average this estimate with a corresponding current measurement. The choice is between equal weight- ing (a simple average) and weighting somehow based on the relative reli- ability of the 2 numbers. I'll evaluate the latter choice first. The initial measurement (the one that starts the filter) has a variance associated with it which can either be measured or estimated from the equations of Fig. 6 or from (21) using the common k of Table 10. One approach to weighting is to use a rate of change in this error term similar to the rate of change of the density estimate. A simple version of this was used by Hildegrand (1976) who simply multiplied the variance of earlier measurements by an apparently arbitrary 1.05 per day, "thus weighting the past measurements less." This is intuitively appealing in that older measurements somehow seem less reliable than more current ones. . A similar result is obtainable from evaluating the rate of change of the confidence limits about the prediction equation (42). In general, the symmetric confidence interval around any estimate of E (yo) corresponding to the set 50 is 1" A I I ‘1 20b 1 0t 50 (l )9 30 (46) (Searle 1971, p. 108, eqn. 95), where b_is the vector of estimated regression coefficients, t is a value of Student's t statistic, o is an estimate of residual error variance (Searle 1971, p. 93), (111) is from 94 the original data set, and x6 is a specific set of x's. For retransform- ing from the logarithmic relationship of (43), the confidence limits are 10 '°" ° __ '° -.17 (47) The derivative of (47) involves the solution of a quadratic of the form d x'Ax / dt, which at the time of this writing prohibits further progress. Neither the divergent trend of Hildebrand's multiplier nor the rate of change of the prediction equation confidence interval is entirely satisfactory. Continuous divergence of the confidence limits is a very conservative approach, eventually negating the worth of early sea- son measurements. Even altering the value of the constant is objection- able until an empirical basis for this change is obtained. 0n the other hand, the confidence limits of the prediction converge as they move toward the peak density, and diverge away from the peak, ignoring the loss in reliability due to the age of the previous measurement informa- tion. A third approach to the variance estimate is simply to use the estimate from Fig. 6 (or (21)) applied to the predicted current density. This again ignors the age of the previous measurement information. How- ever, it does make the larger densities, which inherently have greater variance, account for less when averaged. This is a problem since the weighted average will be used as the starting point for the next itera- tion of the filter. The net result will be bias toward low seasonal incidence estimates. 95 Simple arithmetric averaging of the predicted and the measured den- sity produces an unbiased estimate which incorporates past information, weighting most recent measurements more than previous, though it still ignors the relative ages of the measurements. Despite the latter draw- back, it is the averaging technique that will be used in the following. 4.3 Filter operation The filter algorithm is quite simple. The filter is initiated by _ the first non-zero field density measurement. CH and dd accumulation are also needed. Ch is incremented in 1 dd steps according to the relationship of Fig. 11. Ch and dd are then entered into the rate equa- tion. The calculated increase is adjusted by a factor that is the ratio of the density measurement to the predicted estimate (from (42)). This same factor is retained for each dd's rate calculation. The amount of increase is added to the density measurement and the process is repeated up until the date of the next measurement. The resulting prediction is then averaged with the observed density, as discussed above. After ad- justing for ch error by using the new ch measurement (see below), a new correction factor and a new rate is calculated and the process is repeated. In evaluating the dd and ch components of the filter, it was noticed that the regional (PWS unsprayed fields) average larval density seemed to peak later than the averages for 6 years of Gull Lake data (Gage 1974, p. 79). Re-examining the dd data used (Appendix 8), I decided that the decision to average the PWS site center data with the New Carlisle and South Bend stations (see 2.5) was a mistake. Rather than convert the dd accumulation, the rate equation was merely shifted to 30 96 dd's earlier. This was the difference in accumulated dd's between the PWS site center and the average, for the site center 700-750 dd period (the expected peak period). Filter operation is illustrated in Figs. 15a and b, using 1 field from the 1976 PWS data (Sawyer, pers. comm.) and 1 from the 1974 Gull Lake data (Sawyer 1976). Other fields in these studies were also used to evaluate the filter and these results are listed in Appendix 9. The filter behaves roughly as expected, with incidence curves similar to Gage's in respect to peak and last observed incidence. De- pending on ch, larval density tends to peak at around 750 dd's, and to go to zero near 1150 dd's as in Fig. 14. The filter also noticeably conserves previous information, which is, of course, its purpose. This makes initial measurement accuracy quite important. A low measurement initially contributes to a low estimate at the next and subsequent observation periods. The filter is sensitive to the accuracy of the dd day accumulation. A shift of the rate equation a few dd's either way can be quite impor- tant. Table 27 shows the effect of the 30 degree day shift mentioned above. The filtered peak density in this example increased 19% follow- ing the shift. There is also a problem with the plant growth equation as used in the filter. Measurement of standing crop height is subject to errors. To allow for changes in crap height (and thus the larval density rate change equation) over time, the crop height over dd equation of Fig. 11 was used to predict the path of crop height between measurement periods. The next new measurement and the prediction were averaged arithmetically and the filtered crop height term was used as the beginning for the next crop height path prediction. .mquEmm poo» mcmzcm om co ummmn cowgm>cmmno comm .Amump smxzmmv 8686 axes P_=u aem_ .Aeem_tv he .Avm;m__n:ac: .mezmmv open mza mNmF ou cowumuw_aq< .Aummpv Am .cowumsmao cmupwm we mmpanxm .m— .m_u .mmpasmm poem cmmcwpum om op mp :0 women :owum>cmmno some Hmwv mmcmv w>¢o mmmomo Amwv mmcmw myco mmmomo 97 0mm" own: owo omm omc own 82 owe. owe owm omv own 71f»..tsp..tpst._ O a.1._.Mt..».L..t_s..:..v....r0 ”I. .- f . : e I . uu .. 5 no . .4 . An TQU 17¢ .11— .- . // . .. 68 m. E i nu i v n. ”U T . no : l.3 _. 9 i2 .. = 1 . .0: . LI 1 0 im1; .. :3 .u.m3mmmfi_E e . @2855: a .. 88285 a . $533: a .. . 02,580 a. 1.... 05,580 a T. 1003 388009 / BHABUW 98 Table 27. An example of sensitivity of the filter to accumulated degree-days, Gull Lake field 916 (see Fig. 188). O(Y)= observed larval density; E(Y)=predicted from rate equation; F(Y)=filtered estimate. Original Shifted by 30 dd Degree-Day 0(Y) E(Y) F(Y) E(Y) F(Y) 313 0.00 - - - - 378 0.00 0.00 0.00 0.00 0:00 440 .03 0.00 .03 0.00 .03 488 .06 .35 .20 .39 9 .22 567 1.40 1.02 1.21 1.28 1.34 663 3.29 2.99 3.14 3.98 3.63 760 4.56 3.99 4.27 5.64 5.10 835 3.47 3.46 3.47 4.87 4.17 926 1.61 1.80 1.71 2.71 2.16 985 1.03 .81 .92 1.21 1.12 1043 1.06 .30 .68 .47 .76 1157 .11 0.00 .06 0.00 .06 1249 .06 0.00 .03 0.00 .03 99 The ch equation of Fig. 11 needs to be adjusted for the sprouting time of the particular field. One way of doing this is to use the ch curve and the first measurement to project backwards to the zero height (sprouting) dd, as discussed above. Subsequent applications of the ch curve could then be based on this figure. Without knowing the accuracy of the initial measurement, however, there is no reason to accept this sprouting time as correct. This is perhaps more clearly a problem when the field is not located immediately after sprouting but instead is found when it is already, say, 8-12 inches tall, as is most commonly the case in survey work. A better estimate of sprouting time might be the filtered estimate obtained after the second measure- ment time. But using this procedure a logical question is, why stop here, accepting only one estimate of the sprouting time? In the filter as used here, the approach was taken that each new measurement of ch, filtered using previous measurements, provided a valid estimate of sprouting time. Crop growth rate and density rate equations were thus altered for subsequent prediction periods. There are obvious drawbacks to this approach and refinements could be made. But perhaps the most limiting is the observation that the estimated sprouting date seemed to be later as later measurement information was added. Also, estimates of crop height were most frequently higher than observed, in- dicating plants were growing more slowly than predicted. For ch to be used in the filter, a better model is needed. In most plant studies, _however, yield is emphasized and data on height is often secondary or ignored (eg. Wiggins 1956; Holt, Bula, Miles, Schreiber and Peart 1975; Jackman l976). 100 Re-estimation of sprouting time and filtering of ch changes the density rate equation slightly at each time of observation. This change is in addition to the change from the adjustment factor, sec- tion 4.3.2. Thus, for observation near peak density, for example, it is possible for a negative predicted trajectory to become positive at the time of a new observation. An additional application of the density rate equations is to pre- dict seasonal incidence curves. This can be done from a single measure- ment, projecting both forward and backward in time. Measurement error, however, is extremely critical. Fig. 16 illustrates the use of the rate equations for 3 different observation points, treating each obser- vation as though it were unique. 4.4 Further limitations of the filter Because of the simplistic nature of the density rate change equa- tion, the effect of many potentially important factors is ignored. Perhaps the most critical is the effect of egg parasitism, which may vary greatly between regions. The net effect of increasing egg parasi- tism is to accelerate the rate of decline of the larval density in late season. On a dd basis, this may be compensated for by reducing the absolute value of the coefficients of either the 2nd order dd or the interaction term in the larval predictor equation (41). The magnitude of correction necessary for a given level of egg parasitism is unknown 'but experimentally determinable. The alternative, not explored here because of numerous missing elements in the data set, would be to include egg and parasitized egg densities at previous measurement points as part of the larval density regression. The spacing (separation in dd's) of 101 .Amump cmxzmmv open mxmq szo dump ”pew; .m>c:o ccwpuwucen uWPOm An woumowmcw mw . a .mpcwog m>wumucmmmcamc mzq mum? Eocm coepmpoqmcpxm .mucmuwucm chomswm mpmewumm op mco_um:vm zfiwmcmu Pm>cmp wo mm: Hume mwcmw w>¢o mmmomo omw— owe“ owo 0mm owv . p ; b F L Pl» F I? PLL PL b h F 4 4 d 4 omhumommm I om>mmmmo d u.\ omN I v T V 1 fi fi' I V Y I 1’ ‘U 1 fi Y 91 ‘ I 0 E 9 100:1 ‘ 380008 / BUAHU'I 6 21 81 .Avmcmw_asacz .cmxzmmv mumu oww~ .O p .um |l va mmmm: mEo mmmomo omo— omo omm owe . €F>¢.L_>»_.>.bet3\: d . 4 8:885 .. 83:80 a. 0mm #10 2 I8 1003 388005 / 38A861 V 102 sample periods and great difficulties in analysis made this approach virtually useless as a practical tool. Difficulties in estimation and prediction may not be entirely the fault of the filter. Sawyer's (1976) data uniformly showed peak densi— ties of larvae at the 640 dd observation. This is somewhat earlier than expected if the Gage characteristics (Gage 1974, p. 79) are held to be generalizable. This might be attributable to an error in measuring dd accumulation, as it was warm in late March. For several years, it has been customary to begin marking the accumulation of dd’s on April 1. There is no reason for this other than the generalization that there usually are no dd's before this in many parts of Michigan. A better standard might be to begin accumulating after the end of the last full 2 week period in which no dd's were accumulated, or to use some simi- lar standard marking the cessation of cold weather. In 1976, there were a few warm days at the end of March which could have contributed to the accumulation of dd's experienced by the population of beetles. The 1976 PWS adult population peaked near the 200 degree-day mark, however (Sawyer, pers. comm.). removing some of the doubt about dd accumulations. Another cause could have been a more general use of insecticides than was observable. Average density on dd 640 was over 12 larvae per 2 linear feet, with some fields reaching 20 and even 43 larvae per sample. It is not known how farmers perceive these densities, nor what factors go into a decision to spray, but it should be noted that the pattern of field 1024 (Appendix 9), a field known to have been sprayed (Sawyer, pers. comm.) had a below-average density and displayed a less precipitous decline than other fields not labeled as sprayed. Investiga- tion of egg and adult density curves and larval age structure for the 103 1976 PWS data may shed further light on this problem, but that work is yet to be completed (Sawyer, Ph.D. thesis, Mich. State Univ., ca. 1978). ESTIMATION OF LARVAL PARASITISM BY TETRASTICHUS JULIS To estimate the fraction of the CLB larval population that is parasitized by I, julis, or to obtain density estimates, requires taking a larval sample. The statistical difficulties in this have been dis- cussed above. Several additional technical problems with larval dis- sections also require investigation before addressing the problem of improved parasitoid density estimators. Sweepnet sampling to obtain an estimate of larval parasitism is affected by sweepnet bias. The percentage parasitism of each instar must be adjusted by the instar specific catch correction factors. Overall apparent (i.e., at the time of sampling) parasitism rate is thus dependent on the accuracy of the correction factors. For the 85 cases of paired sweep and square foot samples used in section 2, proportions of parasitized and total CLB larvae (Table 28a and b) were corrected by use of the simple multipliers of Table 4. The averages for the cor- rected samples (Table 28c, column 5) generally reflect the effect of larger numbers of younger instars. Younger instars have lower rates of parasitism (Table 28d). Thus, application of sweepnet bias correc- tion factors lowers the estimated rate of parasitism (Table 28c, column 3). Use of the correction factor doesn't account for all the differences in parasitism rate between sweepnet and square foot, though. The re- maining difference is most probably caused by preservation and/or microscope technique (section 2.2.1). Square foot sample larvae were preserved by freezing and sweepnet samples used FAA. Based on past 104 105 Am.a V Aa.a V Am._ V Ac. V 8.8 V m.mmV e.m V ©.mNV o. V Ao.NmV Ak.o_V Am.m V a._FV m.m_V m.mm m.me o. AAA/NA VVV A.U.m m.N m.N o._ m.N n.m N.m u.o~ o. m.om v.m_ m.m F. o._~ N.m_ m.mm ¢.mo ,o. wkfi mw.MN om.mN Pm.mN o.mp mp.op v0.0 m.w m.— o. m~.Nm No.9m mw.Nm o.mN cm.mN mN.mF m.mv m.om o. w. IIIIIIIIIVIIIIIIIII .umpummmeu LmnE:: cams A_.o.V Am.m V Am._ V Ac. V Aa.m V “4.5 V Ao. V AN.mNV Ac. V Am._aV Am.m~V Ae.e V Ao. V Am.o_V A_.e V AV.mNV Am.mmV ho. V .U.m ¢.m ~.m m. o. m.p m.m o. N.mp o. m.~v o.mN m.m _. 0.x _.¢ F.—N m.N¢ a R mo.m om.NF mm.op o.Nm mm.op NN.—F m.n o.o o.N— -.FF no.mp mo.mN o.mN pw.mm mm.NN mN.Nm m.mv o.m .m IIIIIIIIIIMIIIIIIII .mza m~m_ . 2V3“ 4H s6 dastaV mDQ co ho. Ac. Ao. Ao. Am.m V AV._ V Ao.oVV AF.“ V Ac. V Aa.mmV Am.6mV Am.m V Ac. V Aw.m V A6.P V Am.m V Ac. V Ao. V .b.m GHN mm. m¢.N mo.m o.o mm.m mm.op o.m o.m o._N o.N nm.N om.v o.N mm.m mo.m mN.m_ o.m o.m— U IIIIIIIIIINIIIIIIII N.“ P OOOOMOOO Ao. V 0. Ac. V 0. Ac. V 0. Ac. V o. A_.mNV m.“ Ac. V o. Ao. V O. AD. V o. “O. V o. A.e.mV nee IIIIIIIFIIIIIII mmpmc Emwpwmmcma mmwcm>< mo. mp mFOP wp. «F «No mm. o_ mmm o. — mNm wm._ mp vow oo.m mp mNN mN.m c nmo o.~ N mmm o.N P mud m_aEmm poo; msmzcm .m o. m— mFOP om. «F «Na mo. op mmw o. P me mF.F m— vow 5e. w_ mNN mN._ e Nmm o. N mmm o.m F NN¢ m_qum pwcammzm .< m. mameu aweaa "Lmumcfi .mN mpamh Table 28 (continued) 106 C. Average percent parasitism (I, julis) for all larval samples.* DD48F Cases 478 1 533 2 637 4 725 18 804 16 829 1 858 16 924 14 1015 13 Corrected Sweepnet Square Foot Sweepnet 22.22 .0 10.09 33.51 12.14 29.71 29.71 1.47 20.54 10.05 1.41 7.68 4.48 .69 4.13 8.0 .0 7.21 4.08 .72 3.8 9.63 1.26 9.98 62.08 6.32 60.38 D. Parasitism rate by instar, all samples. Instar 1 2 ‘ Average percent = 100( Z ( 2 TJ / E CLB)) / n ... ---------- Sweepnet------------ No. CLB's 42 461 2184 2949 No. Parasitized 2 35 308 581 j 1 1 --------- Square Foot------- N0. CLB's 121 456 1092 1447 No. Parasitized 2 4 28 42 106 Table 28 (continued) C. Average percent parasitism (I, julis) for all larval samples.* Corrected DD48F Cases Sweepnet Square Foot Sweepnet 478 1 22.22 .0 10.09 533 2 33.51 12.14 29.71 637 4 29.71 1.47 20.54 725 18 10.05 1.41 7.68 804 16 4.48 .69 4.13 829 1 8.0 .O 7.21 858 16 4.08 .72 3.8 924 14 9.63 1.26 9.98 1015 13 62.08 6.32 60.38 D. Parasitism rate by instar, all samples. ---------- Sweepnet------------ ------—--Square Foot---—--- Instar No. CLB's No. Parasitized No. CLB's No. Parasitized 1 42 2 121 ' 2 2 461 35 456 4 3 2184 308 1092 28 4 2949 581 1447 42 Average percent = 100( E ( 2 TJ / Z CLB)) / n ... i = l..4; j = l..n j 1 i 107 experience with FAA preserved specimens, above stage microscope light- ing was used for dissections. The ivory colored I, juli§_eggs and larvae were teased loose from the beetle body and were easily discern- able. This was not the case with the frozen samples, however. When frozen, 1, julis eggs and larvae tend to rupture. Detection of the membranous chorion is possible and reportedly there is little confusion with other membranes; but this requires substage lighting. Freezing and dissection with substage lighting has been reported (T. Burger, pers. comm.) to be superior to use of FAA or alcohol preservation and above stage lighting (i.e., the parasitism rates were closer to cor- responding live sample results). However, when using above stage lighting, as was the case in the PWS dissections, I, juli§_eggs and‘ larvae are much more difficult to find. Parasitism rates for square foot samples (Table 28c, column 4) are thus underestimated by an un- known magnitude. These shortcomings in the PWS data set, and in all similar existing sets, make it impossible to meaningfully proceed with development of filters or incidence curve predictors for parasitoids at this time. Nevertheless, it is possible to provide some guidelines for future efforts and to suggest some of the variables that may be most useful in model building, based on previous research on I, juli§_biology (Gage 1974). Accumulated degree days play an important role in the emergence of adult wasps from wintering Sites. The soil temperatures experienced by the wasps will rely on soil type, ground cover and mois- ture variables, but Gage (1974, p. 39) concluded that at least a field average relationship between soil and air temperatures justified use of air temperatures in predictive equations. 103 Degree days also can be used to predict the rate of larval de- velopment (Gage 1974, p. 32). Unfortunately, the relationship between mortality rate and temperature is less clearly understood. Gage (1974, p. 70) used accumulated dd base 85°F as an index of mortality in the soil but had no data on mortality rate in CLB larvae on foliage. Thus, there is substantial but incomplete evidence, sufficient to suggest the potential utility of heat units as a significant variable for deriving rate equations for filtering. Just as the CLB larval incidence pattern is affected by the time of host plant sprouting, so I, julj§_density patterns would be deter- mined by the time of availability of the host. It is not unreasonable to evaluate crop height as an indirect but possibly useful predictive variable (See Gage 1974, p. 92). The attack behavior of I, julis is apparently greatly affected by wind and rain (Gage 1974, p. 58). Just as with the CLB larval density estimators, a more thorough investigation of the effects of rain dura- tion and intensity is badly needed. Also, a concept of "wind days" above a threshold may have value in predicting rate of parasitism. I, julis is apparently inhibited by winds, so that a higher than average accumulation of wind days would mean a lower than average attack rate. The deve10pment of filtering equations for I, jujj§_density in CLB larvae should consider at least the above set of variables. Devel- opment of the rate equations for lst generation I, julj§_follows analo- gously the CLB density change rate equations in section 4. Second generation curves need to include an addition set of factors to account for the diapause rate of T. julis. Considerable work has been done to 109 determine causes of diapause rates (Gage 1974; S. Koul, pers. comm.; Logan, unpublished notes), but neither temperature, soil moisture, age of female, or constant light (e.g., l6 hrP, 8 hrS) regimes defi- nately cause changes in diapause incidence. Predicting the path of the bimodal parasitized larval incidence curve is difficult due to the number of potentially significant vari- ables affecting rates. Any meaningful attempt at the multiple variate analysis and regression needed to fit data requires reasonably acturate measurement of parasitism rate throughout the season. With the current problems in the PWS and all existing data sets, any effort to develop filters along the guidelines laid down above is presently premature. SUMMARY AND CONCLUSIONS 6.1 Summary Sweepnet sampling for cereal leaf beetles is inaccurate. Conver- sion to square foot density equivalents may be accomplished by use of the multipliers of Table 3 or 4. Inclusion of some additional environ- mental parameters provides little improvement in the estimator. Sweep- nets may be used as tools of detection but statements about the degree of confidence in the detection statement (Ruesink and Haynes 1973) ought to be subject to larval bias corrections. Analysis of factors affecting changes in low density populations, where some samples are disproportionately more "reliable“ than others due to sample size (section 2.4.) may require a weighted regression analysis for correct interpretation. The variance:mean relationships of CLB egg and larval counts (sections 2.2.1 and 3.2.2) suggests increasing aggregation at higher densities; i.e., variance is unstable. Use of 2nd order log/log transforms produces a nearly Poisson variance:mean relationship at low densities and accounts for increased variance at higher densities. Taylor's power law, which relies on an assumption of a first order linear relation between log variance and log mean and hypothesizes that the slope for that relationship is a biological constant, is cast in doubt by the second order fitting. Density dependent mortality thus would produce a changing first order slope, according to, for example, the regional tendencies to apply insecticide. Several transforms are more or less equivalent in reducing variance instability and at normalizing the distribution of egg and 110 111 larval square foot count data. A log transform was selected from the alternatives due to its simplicity. Accumulated degree days and crop height are useful in filtering larval density data, accounting for half the variance in the PWS data set. Derivatives of a log transform of density over these terms is useful for predicting future (including total season) densities, and for conserving previous measurement information to improve current density estimates. Inaccuracies in previous measurements will be retained in the filtered estimate. The use of larval maturity as an improved scale for time (a sub- stitute for degree or calandar days) has limited utility. The numbers of larvae to be collected and measured for precise estimates of propor- tions is logistically limiting at the initial research stage and would be prohibitive at an implementation phase without low cost automated age group classification techniques. The problem of estimating I, julj§_densities within CLB larvae is compounded with the difficulties of larval density estimation. The complexity of the bimodal parasitism curve demands accurate measurement data before parameterization is possible and current data sets are limited in this respect. Although there is sufficient understanding of .I-.lEli§ biology to allow speculative guidelines for development of fil- tering algorithms, accurate estimation of parameters will require more precise quantification. 6.2 Concluding Remarks The review of larval CLB sweepnet models confirmed doubts about the net's sampling accuracy. The 1976 PWS program incorporated this and 112 greatly reduced the use of the sweepnet for sampling. The work on distribution of count data relative to a standardized sampling scheme has significant applicability in the development of sequential sampling algorithms. Corroboration of the negative bino- mial relation with a common k along with development of a suitable dynamic damage threshold model for small grains would allow farmers to minimize the expense of detecting currently threatening population densities. The procedure for doing this is laid out in "The Rationale of Sequential Sampling, with Emphasis on its Use in Pest Management" (Tech. Bul. 1526, USDA-ARS, 1976, 19 p.) The use of filtering equations could further reduce the required number of samples for determining an updated density estimate. Finally, the use of density rate change equations, despite their simplistic assumptions, should provide growers with an improved ability to interpret current density information. In these regards, continuation of the evaluation of and further refinements of the sweepnet and quadrat sampling and the filtering models discussed in this thesis seems certainly merited. Al-l mN m mEWeVU .Lm Nmm :owuumm _chuc:;u mgemz ecomxo couxeo mamepcoz eccm>em c8.6.5 m_V_>cwmwem eeH ecomemm mmecam FPOmcmmcH emcee: copsm>wm eweeoc< eeoELUWe oomoH Eecuza e_aeemme_e commex e_m_csez exogea mmecu weecopoozum ecoocou epeeceaae meze» ocma :eEngm m_;mczoh zesmmo ecemoo ecepxeo omXezmz commxmzz EPeuucoz mocmoz Ucepvwz came: mmumwcez nEouez :oummcw>w4 mmzecmm :ece_mm4 meeem mxem ooeeEeFeg comxuen e__mnemH : oomoH Neceou cowuumm eFecom copmom segmcH upmwmxoocm mcweemm e_m_m>ez comw>eo m_pe>pcoEcm> meF_eo ee>o m>wpo czouempez ceewcmcm copcmm com_w3 xmmcu cm>me ewcw_o> esomxuu smuezepou mFQoz umwp>smpe3 mxem mpqaeccoch eceppam cocheu excem com_w3 ommmpo m_F_>mwsLe: mezmczoh ewcoH EecmcH coca: mpeem__w: mmem>esp ecesw mmmmcmo couem coucPPu mcepu cemzoomgu xwo>mpcecu mmeu czocFeu cocmcm :mwcumm xguem Ewcpc< echF< :emm__< ecoop< xuceoo PNQP Acmx_e2V mwmmw mzcuwumecpm» com mmuwm mmemhmc cemecowz 1 F xwecmga< A1-2 m wN :o_uumm czoumczocm couceu cmsam ce> mfi_w>;ucoz LmEF< eanc< mcsam xgsnmpevwz coquNe: muum_eez cc»; coux_oa ecomxo ce>_—F:m mp__>cwmwem eeH eweeoc< ep__ee:: peesufioozum FPmnmEeu EVE... Eezmcn smVecesu ecomumcb mesmczoh maxez epoumah mmmmezeflcm oefiwcem cwepu .um ezepuo eceero commxmsz .- moccoz mmum_ce2 coummcw>wm oo~eEeFe¥ e_:oH Eezmcm coca: mmmmcmw Npcaou cowuumm pmmcou e_>o Eecmcwm czoucmuez xocmm xu:_:o mxem ecepmooz coumpumeu m—aaecsosp smem_;3 cpsoezFa m>oco mere couPPEe: epmne< ememswem coccm> muuoeem cc»; mmep< necx—mm ;u_3eou m:_Eoo ecosco_m eeeVemez mwzmczoh mmmmcmw cepee_m neon—em cucecm :mwecmm ecuem uecms< NNmF meme: :mczm ue> epoumzk mmmmezewsm uepmcem eweFU .um zecwmem m_mH mzcmmsa ommmuo eeoumo epommmo Neczou A1-3 NF eoozcmmsm m xoofiwz ~N.N Ammmem_ms NV ceEcmnm m_.e :eewcmcm ON gowzcoz coeuumm menmczoh ecemoo omxenmz Npcsou om.mN.¢N.MN.oN.oP.¢ cowpomm ucecw Pszeoow upmwmceu amchm cm>cma couxeo coummewcm mwceca mwm ecepcm< epecom ecepmcoa mac»; maceco .- mcmmx eHcoH eCVLam czommmsee :we_aao couxe—u spouse; mezmczoh ewcoa vmmp ezeemo someWVU uecms< Nuczou mmmp .mnmmxm ooF\m.mVu gems mueeFummLmuc: ou crocx menu.» FucmeFuue aeF cF umoF mmFQEemcc .ucaoF oe>seF Ace mum: atone cm>Fm .F u \FF uFmFF coF muec smFaneceoV u .ucmeFuue oeF cF unoF mqueemc .memmzm ooF con coax" N .coFFeum meuemz <ez xmz - Fucaou mmmmexerm Fm FF F New FxF on mm m «Na FN\o Fe FF F FoF mN\o an com c vFo oF\o oN oF m «om mF\o FF oFm m mom «F\o ochczoF mchou “mm: 1 Fucaou eeoumo m m oF FFo m\F 0N MN oF me N\F FF eoF FF eoF FN\o o woF FF CNN NN\o FF ea 4F eoe eF\e n ma FF Fee ~_\e qugmczoF egocFQmeegu . Fucaou use; oF Fm FN New FN\o Fe m ee owe ~_\e a. on se wee _N\e we MN MN ”MW mm“w 11 om o mN mFm FM\m me o MN mFm m\m . anmczoF FFmeFae - Fucaou cemxuee 0L nn mN mN FN FmF m\F mN FF mF eFm oN\o oF cmF mF oFo «Nxo mF Non oF mmv mF\o c.2238 fommm .. 3:38 “mg-5 FF me .n oee .-\e m. mFN oN owe e~\e WW m". ”w mm”. mm“w ON Fm F~ ooe eF\e mF Foe em oFe o~\e NF owe Fe OFe .F\e me «FF mF com m\o mN mFN on ooF NF\o anmcon mFooeccogF . Fucaou Futon Fm mm Fe oee NN\e F oFm em ome F~\e Mm mm m” mmmF “M“w om mm oN ooF FloF\o m NMN mN omF ON\o NF Fm NF mmF NFxo as as we con e\e Fe meF ON wee oF\e anmcon csoo . Fucaou ceomFF< acmFmFmeLea mm.mmu meFmFu Fo-o mueo smFuFmecee mm.mVu meFmFu o-o mueo NEmFuFmecem mm.mmu muFmFu Fa-o memo e » cemx cemz Fo .oz N n cemz cemz Fo .oz F a cemx gem: mo .02 mnm— vFoF muo— .mFoF-mFoF .mamcmu mueum emu aaoFoEoucm Fo ucmEuceamo am: soc» mmFQEeu cF «cue; smFuFmecea .N xFucoaq¢ Appendix 2b. 1974-76 MSU Cooperative Extension Cereal Leaf Beetle Parsitoid Recovery Program Key to footnote symbols used in total larvae column: Symbol none \JCWU'l-wa-fl: Sample Unit lOO sweeps hand picked, no unit 25 sweeps 50 sweeps 125 sweeps 150 sweeps 200 sweeps 250 sweeps 300 sweeps A2b-1 OO Om OOO FO OFFOFOF FNFV mOFFatO< EezmcF OF NO NO OFFOFFN O.OO ON ON OFFOFON O.NN ON Fe O OFFOFOFFOFV eFaFFOOOOO coca: F.FN ON ON OO OFFOFNF O O O O OFFOFFN FONV OeFeeaO dFeemFFFI ON m m OFFFFN N.OF mm mm OFFOFON O.OO ON ON O OFFOFNF FFV OFsO F.NO OF OF OFFFFN N.NO Fm Fm OFFOFON N.OO NO NO O OFFOFFF FFNV EeeOeFO OOFOFFO m.Fe O O OFFFFNF F.FO F F OFFFFO m.OF OF OF O OFFOFFN FNNV eoeeaO cemxonmzu Om Om OON OFFFFF Om Om OO O OFFOFON FOV Ee< OOOOO xwo>mFmeu O OO NOO mO OFFOFNF O OO OON OO OFFOFOF O OO ONF O OFFOFON FFNV eFeFFOF mmeu Lee & mFo OOF umo: mFea F.ummV czoF iuiaume>se41111 \xpczou .Eecmoce >cm>oomm eFouFmeceO memmm Fem; Femsmu muF>cmm cochmFxm m>FFecmOooo 3m: ouivaF O O O 3 OFFOFNF O.NO FF FF O3 OFFOFN eeFmFeaOmeO ON OOF ONF OFFOFON FF NFF NFF OFFOFOF N.OF ON ON O OFFOFN FOFV F6263 caocqu O F F m: OFFOFFF O O O 2 OFFOFFN eaFeFdeOmeO cucmgm O Om OON F3 OFFOFON N Om FF F3 OFFOFOF FOV eFaFFFOO mchmm ON m m O3 OFFOFON OOFFFOOOOOO O OO ONF OFFOFFN O OO OO OFFOFNN Ne Om OOF O OFFOFFF FONV aFaOeeeaOF F.N Fe Fe OFFOFFF ON Om Fe 3 OFFOFOF FOFV 44 OO NNF O OFFOFFN FOV OeeFOOO O.FN OF OF :3 OFFOFFF O ON ON 3 mFeeFO FONV OOFFEOO Fecem O F O 3 OFFOFNN O OO FOF O OFFOFNN FNNV O.F OF OF OFFFFN O OO OFF OFFOFON F.F ON ON 3 OFFOFOF FFV O N N OFFFFN ON Om OOO OFFOFON Om Om OOF O OFFOFOF FFV ”seem eFeFee Lea N mFo FOF “mo: mueo F.ummV czoF 1-11-1me>ce4111 \Fucaou .ON xFOeaOOO A2b-2 O.N OO OO OFFOFON ON OO OO OFFOFOF O F F O OFFOFOO FOFV ON OO OOF 3 OFFOFNF FOV OFOOOO O ON ON O OFFOFF F.OO OO OO O OFFOFF FFV OOOFOO O FF FF 3 OFFOFF O OO ONF O OFFOFOF FOOV OF OO OO OFFOFFN OF OO OFO O OFFOFOF FNOV OOOOOO O OO FON OFFOFNF O.OO OO OO O OFFOFFN O.OF OO OO O OFFOFO OF ON ON O OFFOFOF FFV OFOOO O O O OFFOFNF ON O O 3 OFFOFON OO N N 3 OFFOFO FFFV OFOEOFOOOOO F.FO NF NF O3 OFFOFFF OF O O O OFFOFOO FOOV O.OF OF OF O3 OFFOFO FNV OOOOO N OO OFO O OFFOFF FNOV OOOOOEOO F.OO O O OFFOFOF O N N OFFOFFN O.OO OO OO OFFOFO OF OOF OFN OFFOFNF OO OO OOF O OFFOFN FOOV O.OO OO OO O OFFOFO FNNV FOOLO OF OO FOO OFFOFON OONOEOFOO NN OO OOF OFFOFOF O.OO NO NO O OFFOFNF FOFV O.OO OO OO OFFFFF O.OO ON ON OFFOFFN OO OO OF OFFOFON O.OO NN NN OFFOFOF F.ON OF OF O OFFOFOF FOFV ezeOOoz OOF FO FO OFFOFO OO OO OF OFFFFF OF OO OOO OFFOFON NO OO ONN OFFOFON OF OO OFO OFFOFFN O.FO FO FO O OFFOFOF FOFV OFOFFEOOO NN OO NON O OFFOFOF FFV OFFEOO eFanemH ewcoH O.OF NO NO O OFFOFFN O OO OOF OFFOFON OO OO OO OFFOFOF O O O OFFOFFN F.OO FO FO OFFOFF 0.0 FF OF OO OFFOFOF FNV eOOOOF O.OF FN FN OFFOFOF OF OO OFO OFFOFFN F.OO OO OO OFFOFO OF OO NNOF OFFOFON OO OO FO 3 OFFOFN FFOV OFOOOO OF OO OFN O OFFOFOF FNNV OOFFOLOO choF EecmcF Lee 3 OFO OOF “mo: mueo F.omOV czoF Fee & OFO OOF “mo: mOeO F.OmOV 33°F nuiiime>ce41111 Opcaou iuulume>ee41111 \xucaou Ab2-3 O.OO OO OO O OFFOFON O O O OFFFFF NN OO ONN OO OFFOFOF O NF NF OFFOFON NO OO OFO OO OFFOFO FFFV O FF FF 3 OFFOFOF FFNV OOOOOO O OO NOO OFFOFON OOOOFOOO OF OO OOOF OFFOFOF OO OO OO O OFFOFOF FNV EOOFOO O O O. OFFOFON OOF FF FF O OFFOFO FFNV OO ON ON 3 OFFOFOF FFOV OFOFFOez OO OO OO FO OFFOFON OOF N N OFFOFNN O.OO OO OO FO OFFOFOF OF O O O OFFOFOF FFNV OFOFFOOO OO O O OO OFFOFO O.OO FO FO OFFOFFN OF OO NO OFFOFON OOF O O OFFOFOF OF OO FNF OFFOFOF OOF O O O OFFOFF FONV OFOFFLOOO N.NN OO OO O OFFOFOF FONV OOF OF OF OFFOFFN OF ON ON OFFOFOF ON O O OFFOFON OOF OF OF O OFFOFO FNNV ON O O O OFFOFFF FONV O.OO OF OF O OFFOFOF FOV OOOOF OOF F N O OFFOFFN FNNV F.OO NO NO OFFOFON OOF F F OFFOFON O.NO OF OF OFFOFOF OO NF NF O OFFOFFF FFNV OFOOOOO OOF OF OF O OFFOFO FOV OOOOO LOOOOO F.FO ON ON O OFFOFO FONV OOOOOFOO OOFOOOFsFO 0.00 O O OFFOFOF OO OF OF 3 OFFOFO FOOV OLOOFN O OO FNO FOVO OFFFFF O O O OFFFFN O OO FOF FOVO OFFOFON O OO OFN OFFOFON N OO OOO FOVO OFFOFOF O FO FO O OFFOFOF O OO FNO FOVO OFFFFF O F .F OFFFFN N OO OOO FOVO OFFOFON O F F OFFOFON OF OO OONFFOVO OFFOFOF FONV OOOOOO OOF F F 3 OFFOFOF FNV OOOOO :mcmmeJ 0x0..— LeO & OFo Fob FOOI muea F.omOV :30» Leg N OFQ FoF «mo: muea F.OmOV OZOF -lnuume>ee31111 OFcOou unilume>se41111 \OFOOOU A2b-4 O.OO FF FF OFFOFON O O O OFFOFOF N.ON OO OO O OFFOFO FOOV OOOOOO O.O ON ON OFFOFON eateoz O OO OO OFFOFOF O.FN NN NN OFFOFOF OOF O O OFFOFOF O OO OOO OFFOFON OO N N OFFOFFF OF OO OFO O OFFOFON FOFV O F F OFFFFF OO OO OF OFFOFON OF OO OOF OFFOFON O.FF NN NN OFFOFFN ON OO NFN O OFFOFFF FOFV OOFOtesFO OOF F F O OFFOFOF FFV FOOFOFO OOF N N OFFOFOF OOOOO OOF O O O OFFOFFF FOV OOOFOOFO OOOOOOOFz N OO FO OO OFFOFOF FOOV N OO FO OO OFFOFFF FOFV OOOO eeOO O.OO FN FN O OFFOFFN N OO OOFN O OFFOFFN FONV OOOOOOO O O O FO OFFOFOF O OO OFO O OFFOFFN FOFV OzotO OO ON ON OO OFFOFOF FFNV OOOOOOz OO OO NON OFFFFOF OO ON ON OFFOFON O OO NOOF OFFFFN O.FO OO OO O OFFOFOF FNV OOOOFO O OO ONFN O OFFOFFN FNFV OFOOOOO mumoumz mmumwcmz OF OO OO OFFOFON OO OO FO O OFFOFON OOFFFOOOOOO N OO OON OFFOFOF OEOOOO F.NF OO OO O OFFOFNF FOFV OOOOOO O F F OFFOFON OF OO OF O OFFOFON O OF OF OFFOFOF N.OO FO FO FO OFFOFOF OO NF NF 3 OFFOFFF FFFV eOeeOOO O.FO FN FN OO OFFOFO FOFV O OO OOF O OFFOFOF FOFV ON OO OO OFFOFON O OO ONN OFFOFFN NN OO OOO OFFOFOF O OO FOO O OFFOFOF FOV OOOeesFO OO OO OOF O OFFOFOF FOFV OFFFOOOO come: OOFOOOF>FO sea & OFo FOF uOo: muea F.OmOV OBOF sea a OFo FOF Fmoz mpeo F.OmmV OzoF -----me>ce41111 \Oucsou -11--me>ge31-11 FOFOOOO A2b-5 F.O OF OF OFFFFO OO OO OOF OFFOFON OF OO OOO O OFFOFFF FFNV NF OO NOF OFFOFFF O O O OO OFFOFOF FOFV NO OO OF OFFOFOF OOF O O OFFOFON O OO OOO OFFOFFN O.OO OO OO OFFOFFF N OO OOO OFFOFFN OOF OF OF O OFFOFFF FOV OF OO FON O OFFOFOF FOOV mOFOOOO 0.00 NF NF OFFOFON F.FO F F OFFOFON F.OF FO FO OFFOFFF O.OO NO NO OFFOFFF OO OO OO OFFOFFF OO ON ON O OFFOFOF FONV OOO OO O O OFFOFON 0.00 FF FF OFFOFOF NF OO OF OFFOFFF O.OO OF OF O OFFOFO FOFV FFOOOOOO NO OO NFF O OFFOFFF FOV eeeeseO OO O O OFFOFOF N OO OFO OFFOFNN OOF O O O OFFOFO FFFV LOEOOFO N OO OOF O OFFOFOF FOV OOOOOOOZ EFOOFOOO Oommxmaz F.OF OF OF OFFOFOF OF OO OO OFFOFNN O ON ON OFFOFOF N.OO OF OF OFFOFOF O.ON ON ON O OFFOFO FOOV O.FO OO OO O OFFOFO FNV OOFO O.NO ON ON OFFOFON OF OO OOOF OFFOFFN OOF O O OFFOFFF OO OO OOO O OFFOFOF FONV OF O O O OFFOFO FOOV O.OF ON ON OFFOFON O.O FF FF OFFOFNN OF OO OO OFFOFFF F.OF OO OO OFFOFOF OO OO OF O OFFOFOF FONV EFOOOeOz F.OF O O O OFFOFO FFNV O.OO OO OO OFFOFON O.FF O O OFFOFON F.FO NF NF OFFOFOF OO ON ON OFFOFFF O.FO OO OO O OFFOFO FOFV OOFOFOLFOO OOF OF OF O OFFOFOF OFFVOFFFseFOFeO OO OO OO OFFOFON F.OO ON ON OFFOFOF OO OO OO OFFOFFF F.F ON ON OFFOFOF N.NN O O O OFFOFOF FOFV eeetOeese O OO FO O OFFOFO FOV OOFFz EFeOFcoz moses: Lee R OFo FOF FOO: mpeo F.ummV czoF Fee & mFa FOF “mo: mueo F.umOV c3OF luuume>ee411111 Ouczou -lluilme33e41111 \Oucsou A2b-6 O OO OOF FO OFFOFOF NF OO OFF OFFOFNN N OO OOF FO OFFOFO FFFV OOOFOOFO O.NF ON ON OFFOFOF OOOOOO N.OO FO FO OFFOFF NO OO FOF OFFOFOF O.OO OF OF OFFOFNN NO OO OO OFFOFO OF OO OO OFFOFOF ON OO FOO OFFOFOF O.OO NF NF O OFFOFO FNFV OOOLO ON OO OOF O OFFOFOF FOOV OF OO OFO OFFOFON OO OO ONF O OFFOFN FFFV OOOEOOFO O OO OFO OFFOFON OO OO OON OFFOFNN ON OO OOF O OFFOFOF FNNV O.OO OO OO OFFOFOF F.NF FF FF OFFOFFN O.FO FO FO O OFFOFOF FFFV OO OO OO OFFOFOF OO OO ONN OFFOFOF OOF ON ON OFFOFO OF OO OO OFFOFOF O F F OFFOFOO NO OO OO OFFOFO N OO OON OFFOFFN ON O O OFFFFO OF OO FOF OFFOFOF OO OO OON OFFOFON O.FN FO FO OFFOFON NO OO FON O OFFOFOF FOFV OOOFOOFO O.ON OO OO OFFOFOF O.OF FO OO O OFFOFOF FOV OOFFOO O.OO OO OO O OFFOFFN FOOV FOOOOO O.OF FN FN OFFOFON OFOOOOO O OO OOF OFFOFOF ON OO OOF O OFFOFOF FFOV OOF F O OFFFFO O OO ONF OFFOFON F.NO ON OO OFFOFON OF OO OOO OFFOFFN O.OF ON ON OFFOFFN OF OO OO O OFFOFOF FNFV OOOFOOO F.OO ON ON OFFOFON OOOOOZ OO OF OF OFFOFON F.OO O O O OFFOFOF OOFFFOOOOOO OF OO NOO OO OFFOFNF FOFV eesFFFOO zeEmmo commxmaz Fee a OFo OOF uOo: mueo F.OmOV czOF Lea F mFo “OF uOo: mueo F.OmmV czoF -----me>ge41111 \Ouczou -----me>se411111 \Ouczou A2b-7 ON OO FOF OO OFFOFON O OO OO OO OFFOFON ON OO OOF OO OFFOFOF FOOV O FN FN OO OFFOFOF O.OO NF NF OFFFFF O NO NO OFFOFO N.NF OF OF O OFFOFON FONV FLOOOFOOFO OF OF OF OFFOFOF OO OO OOF O3 OFFFFF F.OF O O OFFOFOF NO OO OFF O3 OFFOFON O.OO OF OF O OFFOFO FFV .OOFO OOOO NN OO OOF O3 OFFOFOF FOFV OOFOFNOO . OOOOOO .OO O.FO NO NO OFFFFF OF OO OFF O OFFOFON O O O OFFOFON O.OO OO OO 3O OFFFFF O O F OFFOFOF ON OO OFF 3O OFFOFON O O O O OFFOFF OOOO OO OO OFF 3O OFFOFOF FOFV OFOFFLFOO EFOFO .FO OO OF OF OFFFFO OO OO OFF OFFOFFN OO OO FOF OFFOFFN OO OO ONF OFFFFF OO OO OF O OFFOFON FFNV OO OO FFF OFFOFON O O O OFFOFFN OO OO OOF O OFFOFOF FOV OEOO O O O 3 OFFOFON FONV $on 2%: OOOOO3OFOO OO NO NO OFFFFO OO OO FOF OFFOFFN O.OO ON ON O OFFOFOF FOOV OFFOFOOO ON OO OFF O OFFOFON FNNV FeOeOOtO ueFFcem 3eOFmem F.OO NO NO O OFFOFO FOV OFOLOOO OOF F F O OFFOFON O O O OFFOFOF O O O OFFFFO O O O OFFOFOF O O O OFFOFON N.O NO NO 3 OFFOFO FONV OLOOOOOO O F F 3 OFFOFOF FONV O.FO OF OF OFFOFO OF OF OF OFFOFOO O OO OO OFFOFOF N.OO OF OF O OFFOFON FOFV NF OO OO OFFOFOF OOF N N OFFOFOO O OO OOF O OFFOFO FOFV 856: O.FO O O O OFFOFON FFFV OOFEOO OOmOOO .Fm eeouOo FeO.N OFQ FOF “mo: mpeo F.OmOV czoh Fem F OFa FOF FOO: muea F.OmmV czoF iiilme>ce311111 \Oucaou 11-1-me3Fe411111 \Opcaou A2b-8 F.F OF OF NO OFFOFOF O OO OO NO OFFOFFF FONV 3Oa 3OO OOF F F NO OFFOFFN O O O NO OFFOFOF FFV OOFFFEOO O.OF ON ON NO OFFOFOF O OO OO NO OFFOFOF FONV OOFOOFFLO O O O OFFOFON O O O OFFOFON O O O O OFFOFON OOFFFUOOOOO c925 cm> OO OO FO OO OFFOFON ON OF OF FO OFFOFNF FOV OOOOOO OOF OO OF O OFFOFON FFNV O OO OOF O OFFOFOF NO OO OO OFFOFON OO OF OF FO OFFOFOF ON OO OON O OFFOFFN FOFV OO OO FOF O OFFOFO OOF OO OO O OFFOFON FFFV OFOOOO O F F O3 OFFOFOF OFOOOOF FO NO NO OFFOFOF OO OO NO OFFOFO OO OO OO OFFFFN O O O OFFOFFF OO OO OO O OFFOFON FOFV OO>OO 3O: 0.0F NO NO OFFOFOF 0.00 OO OO OFFFFN O O O 3 OFFOFFO FFOV OOOOOO3 OF OO OFF O OFFOFFN OF O FOF OFFOFFF OO OO OFF 3O OFFFFF OO OO OO OFFOFOF OO OO ONF 3O OFFOFON O N N O OFFOFFO FOOV LOFXOO ON OO FO OO OFFOFOF FFV eOeeO> Emcmunmmz wwmmmZmEm Lem N m5 “FOF. pmoz wueo A603 :32. Fan— & m5 90H “mo: wumo Tummy :32. -----me>ce4111- \Opcaou -----me>se411111 \Oucaou Appendix 3: The Non-crop Larval Ecosystem Introduction Not all cereal leaf beetles or I, juli§_are found in fields of cereal grains. Fence rows, pastures, fallow fields, scrub woods and similar uncultivated areas are also capable of supporting beetles. This appendix examines some of the factors that affect the number of CLB's and para- sitoids that can be found outside of crops. These include oviposition, survival and deve10pment rates, and incidence of parasitism. Literature Review The literature on the relationship between the cereal leaf beetle and various hosts is extensive. Hodson (1929) assigned a descending order of feeding preference to barley, oats and wheat, respectively. He also noted that orchard grass (Dactylis glomerata) became a staple food after cereal grains had been harvested. Venturi (1942) claimed that the CLB was polyphagous within the Gramineae. Balachowsky (1963) found the CLB in Europe would feed on quack, timothy, rye, orchard, oats and canary grasses, among others. _ Wilson and Shade (1964a) described the feeding and oviposition pre- ferences on several grasses and related these to the survival rates of the larvae. They f0und oats to be preferred over barley and wheat of the same age for oviposition, and the latter 2 were preferred over corn. The age of the plant was a critical factor, however, and wheat or barley were more preferable to oats that was 10 days older. A3-1 A3-2 Wilson and Shade (1964b) also studied differences in post-diapause adult weight gain and survival on different grasses. Any Small grain was accepted before sorghum or corn. Foxtails (Setaria sp.) were antibiotic to CLB's. They hypothesized that Q. glomerata was used as a spring adult food if small grains were unavailable. Larval survival and development rate also varied according to host (Wilson and Shade 1965). The small grains plus orchard, timothy, brome, quack and rye grasses all had larval laboratory survival rates above 50%. Corn, which was an acceptable food for prediapause adults, was less suitable for larvae. Leaf-vein Spacing was one factor that affects larval survival (Wil- ' son and Shade 1966). The mouthparts of first-instar larvae were too large for insertion between veins, and the veins themselves were too tough to chew through. Leaf pubescence was also a deterrent to ovi- position and feeding (Schillinger and Gallun 1968, Wellso 1973, Webster, et a1. 1973). A general theory of host suitability was developed by Caswell, Reed, Stephenson and Werner (1973). The plants described above as unfa- vorable or antibiotic to the CLB were found to have a common metabolic pathway, designated C-4. Plants suitable for CLB's had a different pathway, labeled C-3. There was a good correlation between metabolic pathways and leaf-vein spacing. There are at least 3 reasons to expect to find cereal leaf beetles in wild grasses. Adult beetles may be drawn to non-crops when crop age (Wilson and Shade 1964a). Feeding damage to crops may reduce moisture content and succulence, prematurely aging the crop and causing adults to look fOr new hosts. Random flight activity may carry adults out of the A3-3 field, perchance to an acceptable grass host. Surveys of adult beetles at the Kellogg Farm during 1973-1974 (Casagrande 1975) showed that during the spring and early summer grasses in fence rows, pastures, and stubble fields held a sizeable reservoir of adults. Methods 1. Field studies: Three types of experiments were carried on in the field during the summer of 1973 using the MSU Kellogg Farm as a re- search area. In one set, large (18 x 6 x 6 cu. ft.) cages were set up half over grass and half over adjoining winter wheat, and others for grass and spring wheat, and grass and oats (as in Figs. 3-1). The cages on wheat were moved later, and the experiment was repeated to look for changes due to the age of the wheat crops. The cages on oats were set up for only one period. For each experimental treatment, 2 cages were used at a time. About 250 adult CLB's were introduced to each cage, and 100 were added 2 days later to make sure there would be a large number of eggs laid. After 1 week, the crop and the grasses were sampled to determine egg counts, densities of various grasses, and distribution of eggs within the cages. - A second field experiment involved taking a sweepnet catch of larvae in both an oat field and an adjoining grass field. Fifty sweeps were taken in oats and 100 sweeps were taken in the grasses, where the density was somewhat lower. This was done as a check on the distributions found in the cage studies. Finally, beetles that had been reared in the laboratory, and which could not have been previously exposed to parasitoids, were set out on pots in the same oat field and at varying distances into the grass field. A3-4 This was to see whether parasitoids could find the larvae in the grasses or whether the parasitoids would avoid this area. After 2 days, the larvae were brought back into the lab and were dissected for parasitoids. 2. Laboratory studies: Four different series of experiments were conducted in the lab. All 4 series were done in the CLB rearing room in the MSU Plant Science greenhouse. The temperature throughout the ex- periments was a constant 24°C with 60-80% RH and a 16 hour daily photo- phase. Seeds for 4 grasses were collected near the Kellogg Farm. These included orchard grass (Dactylus glomerata), 2 types 0f brom (Brgmu§_ inermis and B, tectorum) and barley. Timothy (Phleum pratens) and quackgrass (Agropyron repens) seeds failed to germinate, even when held for 3 months in a refrigerator. It was suggested later that a better way to get quackgrass would be to dig up roots in the autumn. When used in these experiments, all grasses were 4-6' tall. A test was made to check fOr oviposition rates on each grass by itself. A minimum of 24 pairs of adult CLB's were set up on individual caged pots for each of the 3 grasses and for barley. The cages were made of "lumite" screening, which fit snugly into a 2-inch flower pot. The cages were checked daily for eggs, the numbers were recorded, and the fresh eggs were removed. Plants were replaced every 3 days. Counts were made daily for 3 weeks. A 2nd series of experiments was conducted on oviposition preference. For this test, five 2 x 2 x 2 cu. ft. cages were used. In one set of experiments, the 5 cages each held barley, orchard grass and B, tectorum ("downy brome"). A second set also included B, inermis ("smooth brome") and the timothy that would germinate. A third set added "smooth brome," but had no timothy. A3-5 In each cage, 10 pairs of sexually mature adults were established. After 2 days, the plants were removed and counted for eggs. A 3rd test was run to check egg-laying preference on unequal amounts of orchard grass, "red brome" and barley. Five 2 x 2 x 2 cu. ft. cages were set up with a total of 6 potted plants per cage. The eggs laid were counted at the end of a 3-day period. A final test was undertaken to study the survival and development rates of larvae on the 4 grasses. A minimum of 30 individually caged pots (like those of the lst series of lab experiments) were set up for each of the grasses. 2 eggs were placed on the upper side of the lower leaves of the plant. Twice daily checks were made to note the larval instar and when pupation occurred. Survivorship was also noted. Results 1. Field studies: Cage studies to evaluate the relative oviposi- tion in grasses as crops are extremely susceptible to artifacts. That is, the presence of the cage alters adult behavior. Fig. A3-l shows the results in 1 of the cages. The beetles tend to pile up along the edges of the cage, especially in corners. Square foot foliage samples taken 1 foot away from the edge have 1/3 the density of eggs as samples taken against the cage wall. In analyzing egg counts, a paired-t test was used, grouping square-foot counts from opposite sides of the cage (eg. 45-20 in Fig. A3-l). Results of the cage counts are given in Table A3-1, with mean egg density for both sides of the cage. In addition to the counts taken in the 4 early season wheat/grasses cages, 3 square-foot foliage samples were taken from each cage. Grasses were divided into 3 distinguish- able types: bluegrass (Poa canadense), quackgrass and a conglomerate "wild oats," which later proved to be a mixture of timothy, oats (Avena sp.) and A3-6 «OOO.O O.NF+ OO.FNOO.N O.FNOO.OF OF OOOO I. .1 Femcz 4ON.N O.N+ F.F+OF.F O.O+OO.O OF OOFOOO . mmepm 1. 1. pemcz OFOOOO OF OO.F O.F+ m.F+om.F NF.N+OO.m oF LchF: NemzzimueO OO. ON.O+ NO.OmOO.OF OO.FFmFO.ON O FOOO3 OO.F N.O+ NO.O+OO.OF ON.O+OF.OF O OOFEOO 1. .I chFm; Oocu OONN. NN.O1. O.NFHON.OF O.mHFO.F m Femgz OOF cu om. o.N+ o.O+oo.mF O.oF+oo.ON m smFOFz eoFuxFLem F mmcmsmFFFo Omecu Oocu 1:. Oosu :oOemm Fo mEFF O.1 O+Oemz .FO .OOFOOOO .mmeu mg» Fo meFO Ooem co emFOEeO meF meeOOO Fo smOEOc u OOO .OmFOOFO mmeu OF NOOF meezcm emu Ommm Fo coFFOOFFFOFa .F-m< mFOeF .OOOFO mmeu e OF Ommm mo Oucaou poem mFeOOm .F-m< mcszO mF ON Omega F ON NF OF m FN m z + m Femgz NF oF 3 OO MOI NOF A3-7 downy brome, with others possible. The number of stems in each set of 3 sq. ft. and the total eggs on those stems are given in Table A3-2. The "oats" made up only 6 percent of the stems, but contained 41 percent of the eggs. Forty~eight percent of the stems were bluegrass, but these contained only 12 percent of the eggs. A separate count a week later of eggs on 30 stems gathered in each of 3 cages had a mean eggs per stem of .823, 1.80, and 0.07 for quack, "wild oats" and bluegrass, respectively. A later repetition of the 30 stem egg count was made in 4 cages, just prior to the "boot" stage in the wheat. By this time timothy had been separated from the other grasses. Egg density was quite low. The average eggs per stem on wheat, quack, timothy and brome (8:993; sp.) grasses was .067, .017, .167, and 0.0, respectively. ‘ Sweepnet catches in an oat field and an adjoining grass field (an alfalfa field over-run by quackgrass), taken at the Kellogg Farm in 1973 (June 13), reflect the same relative preference for oats over grasses. This is summarized in Table A3-3, which also lists results of dissections to estimate parasitism. There were 7 cases of superparasitism in the 332 larvae dissected for Table A3-3, all of them in the grasses. No hypothesis to explain this relatively high level can be offered at this time. It is known that adult Disparsis n.sp. feeds frequently on a variety of common hosts, including daisy fleabane (Erigeron annus), yarrow (achillea millefolium), chickweed (Stellaria media), and mustards (Lepidium sp., Barbarea vulgaris, and others), (D.C. Miller, pers. comm.). Although these may have been attrac- tants to the grass field, the superparasitism in the face of so many unparasitized larvae remains unexplained. A3-8 Table A3-2. Relative frequencies of grass types and eggs laid on them in 4 cage studies (early season - paired with wheat). "Oats” include wild oats, downy brome, and timothy. - Stems in Relative Eggs in Relative Cage Grass 3 Sq. Ft. Frequency(%) 3 Sq. Ft. Frequency(%) 1 Quack 497 91 100 77 "Oats” 42 8 3O 23 Other 7 l 0 2 Quack 108 23 17 41 Blue 276 67 6 15 "Oats" 27 7 18 44 3 Quack 48 11 11 10 B1ue 343 75 13 12 "Oats" 47 10 86 78 Clover 19 4 0 O 4 Quack 161 39 22 52 B1ue 253 61 20 48 A3-9 .EOFFFOeceO mFOFuFOE Fo OmucmeFucF meOFOOF OFFOOmm «a .OOecm :F ooF OOFeo OF OOmmZO omF O 44O.F OON.OO OO OO. OO OOF.OF OOO.ON OO OO. OO 44O.OF 44O.ON OO OO. OO O ON ON ON. OO OF.FF OOO.ON OO OO. ON OON 0.0F OOO.OF OO OO. O OOOLO F O OO OOF NN.O - OOOO OOOOOO e4 .OOO OFOOOOOFO OFFOO.NH OOOOOOOFO OOOOO FOOV OFOFO FOO FOOO EOFOFOeseO F smeEOz Ome>FeO Eocm muceFOFa .MFmF .mF mean .Ecem mmoFme .eFmFF FeFFeFFe +V Omega mchFowue :e ece Oueo OF Omsupeu Omm2O mo OFFOOmm .m-m< mFOeF A3-10 Dissection of larvae that were set out on pots showed essentially the same results as Table A3-3. At a distance of 10 feet into the field, there were no larvae parasitized. Perhaps I, julj§_is able to respond to the presence of oats at that distance so that it spends less time in the grass searching for hosts. 2. Laboratory studies: When caged on a single host Species, ovi- positing CLB females laid more eggs on barley than on orchard grass or brome grasses (Fig. A3-2). There was considerable variation between in- dividual oviposition rates on any host. There was no apparent difference in survival of adults on the 4 hosts over the 2-week period (Fig. A3-3). When given a choice of plants, barley is clearly preferred for ovi- position, over orchard grass, smooth or downy brome, and timothy (Table A3-4a). This is true even when given a surplus of the less favored host (Table A3-4b). Development time per instar is given in Table A3-5. Larval mortality is not listed in the table because it is not certain which larvae died and which were able to escape the cage. The numbers of larvae in the table reflect lost larvae as well as those found dead. Discussion Native grasses are capable of sustaining cereal leaf beetle adults and larvae in different degrees. Some are better as oviposition sites than others. Bluegrass, for example, does not seem to serve well as a host for egg laying. The role of wild grasses in maintaining a non-crop papulation of the cereal leaf beetle is a real one. The grasses serve as food during times when the crop is unpalatable. During this time, eggs are laid on the grasses. The grasses are then capable of supporting larval development. The numbers A3-11 .OLOOO OF .OOOFF OFFOO m.FOOOO ..O.O FOoON-ON .OOOFOOOOEOO NOOOFFFOOOU .chEchOxm mg» Fe ecm ece mchcmeO mcu Fe OmFeEmF mOFFFOOO 1F>o Fo LmOEOc mcu mFe Ocmeazz .OFOOO O co coFFFOOOF>o mFmeO FemF Femcmo Feem>< .Ntm< .OFO m> an NN.O / \\ / \\ / 1V 3 / / 9 .. 9 n3 Fl. lmw nu Ozomm Eoozm i... .m mmdmo 016.1081... V. wZOmm >2300 11 >mqm230O ... .oNd OO1_>w4mce4 cmO waeoiuliunu 111111111 1 1111111111 .Oczo; OF .mOeOOouoza FOF.FeFO mNomiom .OFFOFEOO m...VomNiFON .mFOFeLmOEmF ”:oFuFecou acouecoaeF .OFOOO NOOF co seFOOF LmO OmEFp ucmeOoFm>mO .m-m< mFOeF A3-15 caught in sweepnet indicate a potentially large number of beetles could be found on other than crop hosts. Non-crop populations of cereal leaf beetle larvae are in turn capable of supporting larval parasitoid populations. The level of these populations will depend on the suitability of the grasses present and the availability of parasitoid adult food sources. The field studied at the Kellogg Farm could have served well as a reservoir of beetles and parasitoids, acting as a buffer to the effects of insecticide usage in the host crop. A more complete study of the wild grasses would require an esti- mate of their relative abundance over large areas. Also, the effect of age and the synchronies of grass types should be evaluated further.' As least, this study indicates that significant numbers of beetles and parasitoids may reside outside of host crops. Appendix 4. 1975 PWS Field Maps A4-l Field identification codes (see Appendix 5). A4-2 1975 PWS Oat and Wheat fields. A4-3 Supplementary oat fields used in 1975 PWS study (located approximately 1 mile due east of main study area). 114-1 1 ,_ l uncuIIAll A4-l. Field identification codes (see Appendix 5). A4—2 ..A.1i~ V?“ \ S ”1 DJ 15 1 'f‘ 73 Q \ \ 1915 ._ fl SS.\\ WHEAT 53033301113 A4-2. 1975 PMS Oat and Wheat fields. A4-3 n. DAVVON lll‘ll llll / ! V*%?? $97 55; C , r I! llllil MItuIGAN I~DIANI ’5 .a. I." h—l m 0m A4-3. Supplementary oat fields used in 1975 PHS study. Appendix 5. Identification of 1975 PWS Cultivated Fields (Map. Appendix 4) by Crop and Acreage. (Survey taken 28 July through 1 August, 1975). Field # l-l-l l-l-2 l-l-3 l-l-4 l-l-5 l-2-l l-2-2 l-2-3 l-2-4 l-3-l l-3-2 l-3-3 l-3-4 l-3-5 l-3-6 l-4-l l-4-2 l-4-3 l-4-4 l-4-5 l-4-6 l-4-7 l-4-8 2-l-l 2-l-2 2-l-3 2-l-4 2-l-5 2-l-6 2-l-7 2-2-l 2-2-2 Crop Corn Soybean Corn Corn Corn Wheat Soybean Corn Soybean Wheat Wheat Oat Alfalfa Soybean Corn Wheat Wheat Oats Corn Pasture Corn Hay Corn Wheat Wheat Oat Plowed Hay Soybean Hay Wheat Soybean AS-l Field Identification Acres 37 20 45 l5.5 l9 40. 26. l4. 46. 23 6 l3. 25. l8. 68. 20 l9 5 33 l4.5 9.5 l2 ll 01010101 01010101 l3 7 9 28 ll l6 5.5 l2 lO.5 Field # 2-2-3 2-2-4 2-2-5 2-3-l 2-3-2 2-3-3 2-3-4 2-3-5 2-3-6 2-3-7 2-4-l 2-4-2 3-l-l 3-l-2 3-l-3 3-l-4 3-l-5 3-l-6 3-l-7 3-2-l 3-2-2 3-2-3 3-2-4 3-3-l 3-3—2 3-3-3 3—3-4 3-4-l 3-4-2 3-4-3 3-4-4 3-4-5 Crop Alfalfa Corn Hay Wheat Soybean Corn Corn Hay Corn Hay Oats Wheat Wheat Wheat Soybean Hay Corn Soybean Hay Wheat Corn Soybean Hay Wheat Wheat Corn Soybean Wheat Wheat Oat Oat May Acres 29.5 79 35 24 37.5 l6 l4.5 20 25.5 l2 3l 10.5 22 l5 Field # 3-4-6 3-4-7 3-4-8 3-4-9 3-4-10 . 3-4-11 4-1-1 4-1-2 4-1-3 4-1-4 4-1-5 4-2-1 4-2-2 4-2-3 4-2-4 4-2-5 4-2-6 4-3-1 4-3-2 4-3-3 4-4-1 5-l-l 5~l~2 5-l-3 5-2-l 5-2-2 5-2-3 5-2-4 5-2-5 5-3-l 5-3-2 5-3-3 5-3-4 5-3-5 5-4-l Crop Corn Soybean Corn Hay Corn Hay Wheat Corn Soybean Corn Soybean Wheat Wheat Hay Soybean Pasture Oat Wheat Wheat Pasture Corn Soybean Pasture Plowed Wheat Corn Pasture Pasture Corn Barley Wheat Wheat Alfalfa Corn Wheat Acres 67 20.5 37 l4 l2 2l.5 l7 l6 ll l4 ll.5 l9.5 32 l0.5 57 22 l4 l4 l8.5 l8 3O l0 l5 l8 l0 49 74 25 AS-Z Field # 5-4-2 5-4-3 5-4-4 5-4-5 6-l-l 6-l-2 6-l-3 6-l-4 6-l-5 6-l-6 6-l-7 6-2-1 6-2-2 5-2-4 6-2-5 6-3-l 6-3-2 6-3-3 6-3-4 6-3-5 6-3-6 6-3-7 6-4-l 6-4-3 6-4-4 6-4-5 7-l-l 7-l-2 7-l-3 7-l-4 7-l-5 7-l-6 7-2-l 7-2-2 7-2-3 Crop Corn Hay Corn Oat Corn Hay Pasture Alfalfa Pasture Corn Wheat Corn Corn Corn Wheat Soybean Hay Corn Hay Corn Plowed Hay. Hay Hay Pasture Wheat Wheat Soybean Pasture Wheat Wheat Wheat Wheat Oat Acres l9 34 l7 9O 2l 32 lO l9.5 ll.5 46 l7 26.5 l2.5 l2 l4.5 l7.5 2l.5 l2 37.5 l9.5 l8.5 l2 77 l2 2l l7 25 22 ll l5 Field_fi 7-2-4 7-2-5 7-2-6 7-2-7 7-2-8 7-2-9 7-2-10 7-3-1 7-3-2 7-3-3 7-3-4 7-3-5 7-3-6 7-3-7 7-3-8 7-3-9 7-3-10 7-3-11 7-4-1 7-4-2 7-4-3 7-4-4 7-4-5 7-4-6 7-4-7 7-4-8 8-l-l 8-l-2 8-l-3 8-l-4 8-l-5 8-l-6 8-l-7 Crop Corn Hay Plowed Oat Hay Soybean Corn Rye Wheat Corn Hay Corn Hay Hay Hay I Hay Hay Hay Wheat Wheat Soybean Plowed Corn Hay Corn Alfalfa Wheat Wheat Wheat Wheat Corn Corn Corn Acres ...a—I.—a.—a_a oomom—‘w UT d ...J—l DONNC‘LDNO‘U‘IVKOGDUW U'l l0.5 l0 ll.5 2.5 l4 l5 l2.5 l5 l4.5 l4 l0.5 J- Field # 8-2-l 8-2-2 8-2-3 8-2-4 8-2-5 8-3-l 8-3-2 8-3-3 8-3-4 8-4-l 8—4-2 8-4-3 9-1-1 9-1-2 9-2-1 9-2-2 9-2-3 9-2-4 9-3-1 9-3-2 9-3-3 9-4-1 9-4-2 lO-l-l lO-l-2 lO-l-3 lO-l-4 lO-l-5 lO-2-l lO-2-2 l0-2-3 lO-2-4 l0-2-5 Crop Wheat Wheat Corn Wheat Soybean Wheat Corn Wheat Hay Wheat Plowed Wheat Corn Soybean Corn Hay Plowed Plowed Plowed Corn Wheat Corn Wheat Hay Plowed Wheat Corn Wheat Wheat Oat Corn Soybean Acres 49 5.5 29.5 4.5 l3 l0 ll l6 27 73 24.5 28 l9 l7 56 44 l2 l6 l6 l9 l2 l8 9.5 Field # lO-3-l lO-3-2 l0-3-3 l0-3-4 lO-4-l lO-4-2 lO-4-3 lO-4-4 lO-4-5 ll-l-l ll-l-2 ll-l-3 ll-l-4 ll-l-S ll-Z-l ll-2-2 ll-2-3 ll-2-4 ll-2-5 ll-2-6 ll-3-l ll-3-2 ll-3-3 ll-3-4 ll-3-5 ll-3-6 ll-3-7 ll-3-8 ll-4-l ll-4-2 ll-4-3 ll-4-4 ll-4-5 ll-4-6 Crop Hay Corn Corn Hay Wheat Corn Hay Plowed Hay Wheat Wheat Hay Hay Corn Wheat Plowed Wheat Soybean Pasture Hay Corn Corn Hay Pasture Hay Corn Plowed Hay Hay Corn Hay Corn Acres l9 4O 36 25 58.5 32 ll.5 20 33 56 23 50 29 25 l8 58 25 37 l3.5 l2 28 33 37.5 A5-4 Field # l2-l—l lZ-l-2 l2-l-3 l2-l-4 l2-l-5 l2-2-l l2-2-3 l2-2-4 l2-2-5 l2-2-6 12-2—7 l2-2-8 l3-l-l l3-l-2 l3-l-3 l3-2-l l3-4—l l3-4-2 l3-4-3 l3-4-4 l3-4-5 l3-4-6 14-1-1 14-1-2 14-2-1 14-2-2 14-2-3 14-3-1 14-3-2 14-4-1 14-4-2 14-4-3 14-4-4 14-4-5 Crop Wheat Wheat Hay Corn Wheat Corn Soybean Oat Oat Oat Soybean Wheat Pasture Hay Corn Wheat Wheat Wheat Corn Corn Soybean Rye‘ Hay Corn Corn Hay Pasture Corn Wheat Wheat Pasture Pasture Oat Acres 27 33 35 l8 27 l3 27 l7 l0 28 60 26 i 44.5 33 ll 20 46 37 12 23.5 30 l4 70 23 29 47 bSDNOSN 01 Field # l4-4-6 l5-l-l l5-l-2 l5-l-3 l5-l-4 l5-l-5 l5-l-6 l5-2-l l5-2-2 l5-2-3 lS-3-l l5-3-2 l5-3-3 l5-4-l l5-4-2 l5-4-3 l5-4-4 l5-4-5 l5-4—6 l5-4-7 l6-l-l l6-l-2 l6-l-3 l6-2-l l6-2-2 l6-2-3 l6-2-4 l6-2-5 l6-2-6 l6-3-l l6-3-2 16-3-3 Crop Pasture Corn Hay Corn Corn Pasture Plowed Corn Hay Wheat Pasture Pasture Wheat Wheat Alfalfa Wheat Corn Soybean Soybean Oat Corn Alfalfa Wheat Hay Wheat Corn Soybean Corn Wheat Wheat Corn Acres 7.5 78 l5 43.5 9.5 28 l2 20 20 25 l6 l9 25 ll l2 l4.5 3l 34 32 l4.5 l7 22 l5 l7 J- Field # l6-3-4 l6-3-5 l6-3-6 l6-4-l Cr 9 Alfalfa Alfalfa Corn Corn Acres 22 37 28 23 Appendix 6. Pubescent wheat study total 699» larval, and adult cereal leaf beetle count data. Appendix 6a. DD CH M WI SQFT TOTE ED EP PWS egg count data. degree days base 48°F crop height (inches) field wetness (l=dry; 2=damp; 3=wet) wind speed (mph) number of linear feet of row sampled total eggs counted in SQFT subsample of eggs dissected (if any) number of eggs in ED parasitized by Anaphes flavipes a-l NM sm mm '0 mN PM sN mN Nm om Nm mm on sP 0N 0N Nm cm am moP oN mo N 0N on PS n PomP “oooPm MP on mo P 3P oN so P Pm oN on P mm on oo P PoNP ”oomPo o on no m P om mo P o om No m m om :o N mN om :P P oP oN oo N Pm om mo P o om No m moPmoPoomPu o on no P oP om mo P s on no m s om so N mN om MP P mP 0N mo N sP om mo P m om no N NoPmouaomPu P on :o m m om o P o om so P oP om mo N SN om PP P oP oN oo N Nm om mo P o om mo m PoPmofioquu 0 on so 5 m on No P o on no m oP om mo P PoPaoPongm kah Hmom P3 : mP ONN NP os— mm No: NN on oP oNN o osP sooP mm mPoP mm aNo sN omm mm com mN mNs mN moo oN smo oP mom sooP on mPoP mm aNo om mom 0m :0m NN mNs oN moo oP smo oN mom sooP o: mPoP mm =No om omo om sow u oN oN moo mP smo PN mom on mPoP om =No 0N mmm mN com :0 no on mm 0N mm oP om sP mm mm mP 0P 0P 0m mm mm NN om mm PN oP oP an em Nm NN am on om so P No oN so P PoPaoPooum m on No m m om No m moPmoPoqum N on No m NoPmoPoooPm m on No m PoPmoPoqum PN om mo N oN cm 00 N oP oN co m mP om so N oP 0N mP P sN oN no N snP cm no P ooP cm 00 m Van Ca ac w NoPNouoqum sN cm mo N s on no P oP oN co m mm 0N so N mN oN mP P o oN oo N NmP om so P amP om oo m mm on mo m PoPNonoqum s om mo m c om oo m N oN oo m oP oN so P s 0N no N 3P oN so P on on oo P om om so m aoPPonoomPu who» know P: : mP mNs oN moo 0P asm oP mom NP mom mP mom sooP oN mPoP sN aNo sN omo sN zoo mm mNs mP moo mP swc NP mom «P «so PM mPoP sN =No sN omo sN zoo mm mNs mP moo NN smo NP mom NP mmm sooP mm mPoP 0m aNo mm :om 0N moo PN smo mom mP mmm :u on PP s NP m PN P 0P oP oP sN sN on on m cm NN NN oN Pm NN Nm m sN P oN am om a om mo N m om oo N P oN oo m N oN so P N oN NP P a oN oo N NP om mo P s on mo m moPPouounu oP on no P nP on no m m cm oo m mP om so P MP 0N oP P =P cu so N a: on co P 0: on ao m NoPPouaqum o: on No m Po om oo m NN oN co m Pm oN oP P o oN oP P PP oN oo N 2N om so P ma om mo m PoPPmuoqum c on no N N om mo N o Om No P sNP ON 00 P PONNPPQJmHh P om mo P 0 nm mo N 0 0m No P =mN en so P oommPnagmHu mHOH Hmom u) x mm am no mN 2N cc mu sooP mPoP =No omo sow mNs moo smc «(a mmm sooP mPoP :No wmm sow mmb moo smo mom mmm mPoP =No mmo as: mPDP aNo mmw vs: a: N m NN NN OCB PN on Nm PN mN P~ mm 0P Pm mP PN mm 0m Pm PN mm mm aw o om mo P m om no N c on No P mo: oN so P moNNPPoomPu 0 on mo m m om oo m oP om oo N a on mo P Q on oo P a oN co m c on mo m o om mo m smm oN oo P mONoPPooum co on mo m N: on No P No om oo m as on mo P Nm om oP P NP oN co m n oN a: m osP on so P Nmm on :o P moNs PoSNPu NP on so P mm om «2 u «N on so P sm on so P oP on so P o oN so N no on mo P NP om No m PoPo Poquu N: on no P so om oo m o om mo P om on no P moP om oP P moan "oomPa NHOH know HI I mm em mm cP mPoP amm mmn ms: mPoP :om mmm ms: mPoP :No mmo com mms moo mqo PNo mom No: ms: sooP mPoP an mmm sow mNs moo smo mom sooP mPoP aNo own now mNs on CD .— mm Pm mN om mP am mm mm oP Pm PN nm em aw om nP om 3P am 0P om om an om sm om me P 3N ON no N on on mo P moan ”ooqu 3N om mo P ss om mo m (P om mo m o: om mo P no on mP P Posm Poquu mN om mo P P: on mo P =m rm mc m msP om mo P =PP Cm «P P mm 0N oo N ooN om so P smP om =o m ssN cm mo P moPN uoqum oN on so P sP on No P m on no m oN om mo P om om mP P Nm oN oP P so om mo P m: om o: om oc P moaP PoowPu oN om mo P mm om so P Pm om oo P Po om mo P om om oP P so cN NP N uNP cm :0 P as Ca n: n oaP on no P momP Pooqu mHOH know «3 z no mo 3P mP NP =P mP mP PP :P ac o: co m: PP PP mP ac :0 - Nu mNs mco smo ncn L sooP mPoP amo mmm sow mNs sooP mPoP oNo emu son unb moo PNo mom ms: sooP mPcP =No omm so: mNs moo PNo mom ms: sooP mPoP aNo cmm oNo mNs mwc PNo «(a ms: no A6a-2 NP ON mm mm mm Om so P OOPm Poowuu omP Om OO m ss ON OO P NO ON 00 P m Om mo P mOPm POONPN NOPN Poawmu POPO "ogmmm O ON NO P OP om mo P N on no P NN OM mo P NN Om co m nP ON OP P PP ON :0 P sP on co P NOOs "OANPm P ON oo P OP Om so N mm Om so P mP Om so P ON om oo m cm ON OO P mNN ON mo P mO Om mo P Poss POAme N ON oo P s Om no N Po ON OO P as ON oo P POP Om co m NOms unaqu msos soon P: x OP NOO so ONm mo osP Om mOO Nm PNO mom mN mmm NOO OP Os. mP mOm OP PmN ONN NP OPN OO NOP Pm mOO ON PNO mom NN mmm NOO PN msO NP mOm NP PmN ONN OP oPN so NOP Om mOO ON PNO mom ON mmm NOO mN OsO OP mOm ONN :0 GO ON Os mo OOO mP sO mm 000 OP sN 0m 0- 00 mm om mOP ON mo P OPP Om mo P NOms noqmmu mP Om NO P OP ON so P ON ON oo P POP Om OO P mm ON Nm Om no P NONs "comma s ON no N OO Om OO N OO Om co P omP ON so P OOO Om OO P OOO ON mo P mON Om co P PONs Poqmum mN ON mo P on Om co P mmP Om so P OsN Om oo m sOP ON Oo P om Om Oo P NOPs "OOOPO rP CO (C P sP Om no P sN Om so P oN Om 00 P s ON mo P oP On :0 P POPs "comma wsos smom P) x NP oPN oo NOP Om smo Om mmm NOO sN osO ON mam PmN can OP NOP sN mOo mN PNO OP mmm NOO OP csO mP mOm NP PmN ONN oo NOP Om smo mom ON mmm NOO mN msO OP mOm OP PmN ONN OO NOP «N swc mom Om mmm NOO sN msO PN mOm ONN PP oPN PP NOP :u on no Ou PP Om mo P m ON Oo N mN ON mo N am ON OO P OOP Om mo m ca (N Oc P OPP ON co P NOP ON mO N ONP Om mo P POmo “OJNPm mN ON mO P o ON mo N OP ON OO N OOP ON Oo P ss Om mo N OPP ON Oo P OPP ON mo P ssP ON No N mOP Om mo m PONo uoquu sP ON mo N :0 ON co P so ON :0 P ss ON mo P so Om so N POOm Poqum «N on sc N Om ON mo P «On cm vc P Pmm ON mO P ONm ON mo P mOm Om mo P mOmm Poquu o ON OO N sN ON NO P mOP ON mo P mmN ON mo P mmm ON NO P NOmm POONPO msos snow P: x mm Om mN mN sP ON mP PP OP ON cP so OO OO PN mN sP NP :0 mOo mmm NOO msO mOm (or PmN enn OPN NOP smo mom mmm NOO msO mOm ONm PmN ONN oPN OmP mmm osO ONm PmN OmP mmm NOO CNN PmN OPN NOP mmm NOO PmN OPN on omm Om OO P NOmm POONPO os ON NO m sNN ON mo P moP Om mo P Pon uoqum mo ON OO P POPm POONPO mP ON Oo N om ON No P osm ON mo P Omo OP OO P OOO ON mo P Omo Om Oo P NOmO "quPu OO ON mo N OO ON mo P OmO ON mo P 0mm ON OO P OOO ON Oo P wmm Om NO P POmO "OOOPO Om ON mO m sON ON OO P mmN ON mo P ONm ON OO m sow ON mo P Pm: Cm Cm m NONO "OJmPu so ON mo m NOP ON OO P NPN ON mo P omO ON OO m 0mm ON mo P OOO Om mo P PONO noqum ss ON No N sm ON NO P OPm ON OO P mmm ON mO N POPO Poqmum wsos snow P: I 00 NP P7 oo 00 mm NN sP mP OP NsP mOm oPN NOP oPN mmm NOO ONm PmN OPN NOP mmm NOO ONm PmN oPN NOP mmm oom ONm PmN oPN OmP mmm OOm ONm PmN OPN OmP mmm NOO ONm PMN on OOO ON ONm Om PoPr KP cn Nm ON OO ON sO ON mo ON OPP Om NOOm m ON mN ON OoP ON OOP ON PmN ON omP Om POOm Pm ON OOP ON smN ON POO ON OPO ON ss Om Pomm 2 ON O ON NsP ON CON cn OmP ON OON Om NOPm NP ON Om ON ONP ON ssN ON PoN ON NmP Om PoPm OO om Om ON NO ON NOON msos so No P co P "OOOPO NC A No P so P so P so P 00 P "oomP; mo N NO P mo P mo P Oo P mo P "OONPO M O v-O-o—v-r-N POJqu no N P mo P a P P "oowmm m 0 Fo-N’PN "nome mo P 0o P oP P ”oomPu om P: 2 OP an Om sP sP NP OP ON OP mP OO OP mo OPN OmP Pow NOO ONm PmN oPN NOP mmm NOO ONm PmN OPN OmP mmm osO ONm PmN oPN NOP mmm NOO ONm Pwn oPN OmP mmm NOO ONm OmP NPm ONm PmN co OO ON mo P sP Cu «C P NOON POONPO OOO ON Oo P mOO ON so P OON ON mo P sPP ON Oo N mN Om OO P POmN "OONPN oo ON mo P OOO ON mo P Oom ON mo P NNN ON mo P NO Om so P PONN ”oquu mO ON mO P mo ON 00 P OmP ON OO P NO ON mo N NO Om sO P NOPN POONPO Po Om mo P COP CN :0 N os ON mo P PO CO «C n PN Om 00 P POPN POONPO NO Om mO P ON ON OO P sm ON mO N O Om mo N NOOP "OONPO O . Om OO P OP ON mo P OP ON mo N mo Om PO m NOmP POONPO so ON OO P mmP Om OO P mo ON OP P POmP "OONHO msos soon H) I NP ac sP mP OO OO so sN mP mP OP OO mm ON mP OP OP mm «P NP PP so PN mP PP so ON OP mP :4 OP mm mm ON Io ONN «OP Omm ONm PmN omm NOP NPm ONm PmN ONN NsP NPw ONm PmN ONN NOP NOO err PmN can OsP NOO ONm ONN OsP A6a-3 PO Om :0 P O ON OO P m ON NO N OP Om OO N POOP "comma mOOs POONPO sP ON NO P mP ON OO P vP cm sc P NO ON so P OO Om OP P POmOPPOOOPm PO ON NO P sNP ON mo P OON ON OO P POP ON OO P OOOmPuoomHm Pm ON mo P ONP Om mo P OOO ON OP P OOm ON Oo P OOP ON OP P on Om OP P NOOmPPOOmHm OP ON mo m OP om OO P Os ON mo P sm ON Oo P Om Om OP P POOmPPOOmHm msos sham H: 1 OP NOO mP ONm OP ONN Oo osP sN OOO Pm NPm sP OON OO NOP mm OOO ON NPm OP OON mP OPN NN mmm sP ONm 2P PmN PP oPN OO NOP ON 32m sN NPm sP OON mP OPN OP NnP :0 oo OP ON oo P Om om OO P mON ON mO P OON ON so P smN ON OsP om OP P PommPPoquu mO Om co m ON on 00 P MN ON mo P OP ON Oo P am ON so N P O Om mP NOOOPHOOOPO Om Om mo m mo Om No P POP ON oo P mO ON OO P OO ON mo N NN om OP P POOOPnoomPO NP om mo m NN Om OO P Om ON OO P mN ON mo P OP ON mo m m Om ON P PoPmPHOOum m om no N m ON NO P mm ON OO m O ON mo N s ON mo P OONNPOOONPO sP Om mo N OP ON NO P sNP ON oo P OPP ON mo m ON ON mo P OOP Om mo P NONNPHOOOPO msos sham H: : NN 23m PN Non mP PmN OPN OO NOP PN mmm OP NOO PP ONm OO PmN OP oPN Oo NsP ON mmm ON NOO mP PmN PP oPN oo NsP ON sum ON NO: OP ONm sP PmN mP oPN OO NnP PN mmm on No: mP mam sP PmN OP oPN ON mmm mN NOO mP ONm NP oPN OP osP mu on no om O ON mo N O ON so P om ON OO P OP ON Oo N om ON mo P Os om oo P PONNPPOONPO PmP ON mo P OPN ON OO P OOP ON Oo P NmP ON Oo P Os Om OO P NOPNPPOOOPO NP Om so N NP ON mo P sP ON mo P POPNPPoqum PP ON me P Pm ON co m Pm ON mo P ON ON so P O OO OP m MONPPPoqum NONPPuoquu OP ON 00 P Os ON so P MOP ON mo P Pmm ON so P PONPPuoqum O ON Oo P NoPPPuoouHm msos show H: : mN OP NP mP OP PP mP so NP 00 mo PN mP sP ON NN ON sP 2P ON NN ON mP mN :u mmm osO ONm PmN OPN osP osz ONm PmN oPN osP mmm PmN OPN smo mom mmm NOO osO mOm PmN ONN NOP smo mom mmm NOO osO mOm PmN ONN smO mom on sP OP ON on OOON mm on Om ON sO P OO ON mo P smP om mo m OOP ON OO P NOPPPuoomHh ON ON me P mO ON OO P :PN NN. 00 P sPa om mo m OOP ON mo P PcPPPPOAMHm OP ON «O P Os ON OO m OOP ON so P mNN ON OO P OOO Om co m OON ON MP P PoaoPqumHh 2 ON so P s ON OO P NO ON mo P POP on No m OO ON mo P sP on O0 P NONOPuOONPu O ON so P O ON OO m mP ON OO P OO ON OO P Nmm Om mo m mmN cN me P PONOPNOOOPO Nsos snow P: 2 ON OP OP mP sN OP OP OP NP Pm PN PN mP OP Pm ON ON sP mP PP ON mN ON ON OP OP mmm NOO oom mOm PmN ONN smO mom mmm NOO com mOm PmN ONN smc mom mmm NOO osO mOm OON ONN smO mom mmm NOO osO mOm PmN ONN OsP smO mom mmm NOO osO mOm PmN ONN oo PP 2N Om CDOC'O'il~ :2 mm om msos show H: : mO om mo P PONOP"OOOPO m ON so P MP ON co m m ON so P ONP ON mo P OOO om NO m ON ON mo P sm Om Oo P PoPoPPooum NN om mo N mm ON so P OO ON OO N so om OO P momo POONPO OP ON mo P ON ON OO N ON om mo P Nomo ”OONPO O om Oo N OP ON so P PP ON OO N mP om mo P Pomo "OOOPO mm Om mO N ON ON so P OO ON mo m om om OO P NONO "OOOPN Ps Om No P Om ON oo P OO ON OO N O: Om mO N PONo "OOOPO mN om NO N NN ON OO P Nm ON OO N OOPO Poqwnu mP ON ON ON OP sP PP ON sP OP ON OP 2P NP ON OP PP osP smo mom mom mmm NOO OsO mOm PmN ONN osP NOO ONm ONN osP ONm ONN osP NOO ONm ONN osP NOO ONm ONN osP NOO ONm ONN csP Nos ONm ONN oo Appendix 6b. DD CH M WI UNIT NSU TOTC PNS larval count data. degree days base 48°F crop height (inches) field wetness (l=dry; 2=damp; 3=wet) wind speed (mph) either linear feet (SQFT) or sweeps number of UNITS sampled total count of larvae in NSU UNITS numbers of larvae in a subsample grouped by instar. First number of pair is the number of that instar in the subsample. Second number of pair is the number that were parasitized by Tetrastichus julis. b-l O IO 0 Io P INm P ION N ImN P nOP ONIOO O IOO s an oPnNm O nsP 0 IO IO ( I IO IO IO tJC)('D O I '. NOCOFFDO "0 000 3 0 000 O O 0 IP IP Im IP nmN IN: nsP Is no nP nm IPN IzN IsN IO IO Io Io IP 0 OD 000 800000.30 O NOPOOOPOOO 0 000 O O 0 IO IO IO IP IO IO IPN nmP IO Io IO IO IP IOP no no no nN no IO nsP lmn. nOm In IO 0— IN IP IP IP IO 000000 30 O 0° 00° 0 0300000000 0 000 O O 0 ID IO IO IO IO Is Im IO IO IO IP IO IO IO IO IO IO IO Io IO IN Io IO IO IO IO Im IO Io Io NNNPIOMOOOOO usos Om soon mo P OOP noon mo P Om soon mo N OOP noon mo N Om soon No P OOP nozn NO P ON soon so P OOP noon so P OONNPHOOOHo Om soon mo P OOP noxn mo P Om soon mo N OOP noon mo N Om soon NO P OOP nozn NO P ON soon so P OOP nozn so P OONNPHOOOHo Om soon OO m OON noun no m Om soon OO m OOP nozn OO m Om soon OO N OOP noon OO N Om soon OO P OOP nozn no P Om soon Oo P OOP nozn OO P ON soon OO m Om soon no m Om soon Oo m OON noun Oo m ON soon OO P mONOPHOOOPo Om soon mo m OON noun mo m Om soon No m OON nozn NO m Om soon co m OON noon OO m Om soon OO P OON noon mo P Om soon OP P OON noun OP P ON soon co m OON noun OO m mONs "OOOPo on: stO H: x OPOP mPOP ONO ONO OOO OOO msO OsO mPoP OPOP OP ms: OPoP OPOP OP ONO OP ONO OP OOO OP OOO OOO OOO mNs ONs mOO mOO :O OO O IO O an PmnOO o nom an nOP nON In ION nO nnm ImO IN Imm IP OQOIO NONONOM m IPO POIPO O ION O ImO INN IOO INP ION IN IsP IO 0 ONOOOOO IO O nOP OOIOO o an OPnos P nOO nNm IOP nON Im 00*00 O IP P INP OPION O ION O nOm P nsN N nPN o nsN o nsP o nPP P IOP Im INP Is Im 0000 Is IOP IPN IOm ImN Ism Imm IPN IsN ION IOP .- O ONOOOPOIDORN IO IO IPN ION nsP an nNs ImN IOP OOOOOONOWO O 0000 OOOOOOMONO 0 00000000300 OOOOOOOOOO IO IPP nmP IN nP nsP no nON no IO IN IO Im IP IO ImP Im IO Im In IO IOP Imm IO Io IN IO IP IP Is IN IsN Im 0000 0000000000 0 0000000030" 0000000000 IO IO Io IO IO nm IO nm Io IO IO IO IN Io IP IP IO I: IN IP IO IN I0 I. I0 ION IOP IO IO lo 00 IN IO IP IO IN IO Ns F- on OP OOP mO OOO OOP OPPP sO sOO OmP mN usos Om: sax: H: : ON soon OO m OP OOP noun OO m OP Om soon Oo P OP O noun Om soon OO P so mONs "OOOPo Om soon OO P ON OOP noun OO P ON Om soon mo N ON OOP noon mo N ON Om soon Oo P OP OOP nozn Oo P OP Om soon OO P OP OOP noon OO P OP Om soon OO P OP OOP noon OO P OP ON soon so N OP OOP nozn so N OP Om soon mo P OO OOP noun mo P Oo Om soon NO m OO POPO "oooso Om soon OO m OP OON noon OO m OP Om soon OO m mP OOP noun co m mP Om soon mo P mP OOP no3n mO P mP Om soon mo P OP OOP noun mo P OP Om soon OP P so OOP nozn OP P so Om soon no P OO ON soon OO N OO Om soon OO P OO mOOm "Ooumo Om soon Oo P OP OOP noon no P OP Om soon mo m mP OOP noon mO m mP Om soon mO m NP OOP noon mo m NP Om soon no P OP OOP noon OO P OP om soon OP P so OOP noun OP P so POON "oaunh MOO Os: mPOP mPoP ONO ONO OOO OOO OOO OOO mNs mNs mOO smO OPOP OPOP ONO ONO OOO OOO OOO mNs mNs OO nmN IOO an nms nsm nsm nmP an no IO 8 OONOOOOONO ION IO IO 000 N I I N OONONOOOOO I N lb O INP nmN IsP ONnOm O no O nO O nO nO nPO nNP nON nOP nmm an IOm IOP 3 OONOOOOONN O ION nm IO IP IOP IN IOP ION IOO OPOOOOFOQO IzP P Is OPION O IO P ION O ION O nmP O INm N ImN o IsP m IPm PPIOP P IrP O I: O IP 0000000000 0 0C OOOOOONONO OPOOOOOO‘OO IN IP Io IN I: I0 Is IO IsP IP IO IP IP IP Im IO IN IO IP Im IuP IOP nm IO IO IN Im IO IO IN IPN IN no nO nmP IO OOO 0000000000 0000000000 0000000000 0000 Io IO IO IO Io IO IN IO nm IO IO Is IP IO IO IO IO IO IN nm In Io IO IO IO IO IP Io IO IO IO IO IO Im Is I: P'FFP—Pmm—Pp— PMFFNFpr'mm—FP— o-Nm—v-o-v-MMr-v—v-o— Om Om soon Oo mmm OOP noun Oo OOP Om soon mO ONmP OOP noun mo OOP om soon mo PsO OOP noun mo osP Om soon mo PsOP OOP noun no OOm Om soon OP ON OOP noon OP PPm ON soon OO PN Om soon so OOP OOP noun so O Om soon OO N om soon mo mOPN "cowso O Om soon Oo Om OON noun OO O Om soon No OmP OOP noun NO OP Om soon mo os OOP noun mo Om Om soon OO POP OOP noun Oo Pm om soon OP sP OON noun OP OP ON soon OP O Om soon no O Om soon Oo mOOP Poouso PN Om soon mO PO OmP noun mO OO Om soon OO OOO OOP noon OO OOP Om soon OO NOO OOP noun oo sOP Om soon no OOO OOP noon no OOP Om soon OP PO OOP noun OP On ON socn NP mPN OOP nozn NP Pm Om soon OO Om OOP noun Pm Om soon NO ON OOm nozn PP Om soon OO momP Poqumo usos on: sat: H: I OPOP OPOP ONO ONO OOO OOO OOO OOO ONs ONs mOO PNO PNO OOO OsO OPOP OPOP ONO ONO OOO OOO OOO OOO ONs ONs mOO PNO OsO OPOP OPOP ONO ONO OOO OOO ONO ONO ONs mNs OOO mOO PNO OOO OOO NPO OsO A6b-2 IOm ION IOP nOm nsP IOm InN 'pV 0 MmmOU‘ONIfi an IO IN 08 NPIOP o Is m nom o In NPIOn O IOm o In IOP IOP IOP ION IOP IOO v-Nr-OC-OOP ION m IOP mPINN o IP NPIPm O IO sPIOP O IO P I. O Im OPIOm O IOP P IOP P IOP O O Imm O IPP O IMN PPION 30000000 0 N000 0 0000 00.00 00000 Is Is IO IN IO Im IO IP IO IN IO Is Im IO IN IO IO IPP INP IO IN IN IO 0 00000000 00 00000 0000 0 000" 00000 IO IO IP IO IP IO IO IO IO IO IP IO Is I. IP IO IO IO IO IP IO IO IN IO IO IO IO Is IO IO IO IP IO IN so OP PmP OP soon noxn soon noxn soon nan soon noun soon OOP noun POPnOnoquHo In 0 o-o-o-v-v-u—MMO-u- Om soon No m onP noon No m Om soon No m moPmouoooPo OOP norn NO m Om soon NO m NOPmOPOOOHo OOP nojn NO m Om soon NO m POPMOPoomHo Om soon no N onP noun no N Om soon OO N onP noun mo P onP noun so N ON soon co m onP noun co m ON soon so N onP noun so N ON soon OP P onP noun nP P ON soon no N OOP noun no N Om soon no P onP noon OO P OOP noun Om soon co m Om soon Oo m NOPNOfloooPo Om soon no N onP noon no N Om soon mo P OOP noon mo P onP noun so N ON soon co m POPNOPOquuo usos On: sun: H: : OP OP OP :9 ONO ONO onO onO OOO OOO nNs mOO mOO OsO OsO OOO OsO OOO nPOP nPoP ONO ONO POO one OO O Im P nOm o Io NPInn PPIsP s IOP OPIOP O IP PPImN N IOm O IN 000 0 00"00U‘P00 IO IO IN IO IP IN Im IO IO IO ION IOP IP "000 0000 0 0"00 0 00N0P0P000 000 0 000000P00 IO IO IO ION IOP IPP INP Im IOP IP I— IO IN IO IP IN IN IP I. IO IO IP 0000 0 0000 0000 0 00"0000000 000 0 000000000 Is Im IO Im IO IO IN IO Im IO IO Im IO IO IN IO IO IO IO IO IO IO IO IO IO IO IO IO IO IO I— IP IO IO IO PN OMNFNM'NQW‘NNO '- '00 0' F ON OOP ON OOP ON OOP om OOP OOP soon noxn soon OP norn OP soon noun soon morn noun Om soon on soon POPNOuoqumo om OOP om OOP ON OOP ON OOP ON OOP ON OOP om OOP OOP Q 0 o-I-NNC-O-NN 0 0 MM soon noon soon noun soon noun soon nozn soon nozn soon nozn soon noon nozn om soon OO onP noun OO OOPPOPOOOPo om OON om OOP OOP ON ON OOP ON OOP ON OOP om OOP 0 O PI-v-FIVGIPv-flfiflu-flwwfn MM soon no3n soon moon noun soon soon noxn soon NP nan NP soon noun so0n no3n OOP noln Om soon OO mOPPONOmeo g. 0 Pv-NNo-o-v-O-MPo-NNN M usos an: sun: a) t :0 P Im o IP o IPP o IN o ns O IO s ImP OPIsN o IN o no o INP o IOP O IO o IN P nnm P nNm o IN o In O INn o nmm OPION n InP mPnON n INN NPnNm m nsP P IN O Im o IP o IP o IO O IO o Im o Im O nmm O IPN O ION P nmP o IN o no 0 no o IN P ION N Imn o Im o IN o IPm P nsO OPIPO P IOm s IOm O IO oPnOm N IOP P Im o Is NPIPm oPnNn m In. P IO O IOO N IO P ION P nnm P IO O IOP o no n InP O m 0 000 0 000000N00 0000 0 00P000P000 0000 In IP IO IO In IO It I. IO In IPN IO IO IO IN IO Is IO IO IN C00 0 0000000000 0000 0 0000000000 Io O IO OP IO NP no mo Io m Io sm IN OP In OO IO sP Io OmP sN IO NO ON Io OOP IO PON Im sP Io O IO sP IP OP IO OO Io mo IP s Io O IP OO IP mP IP OOP sP IO OMO O IO N. Io OO IN OP Io OO P O Io O IO as IO P IO OP PON no m. P usos On: sax: u) I Om soon mo m onP noon no N Om soon co m onP noon mo P ON soon co m on. noon co m ON soon so P OOP noun so P ON soon OP P onP noon OP P ON soon so N onP noon so N Om soon Oo P onP noon Oo P onP noon cm soon OO m NoPPOuoooPo om soon No m onP noon no N Om soon co m onP noon mo P onP noun so P ON soon co m ON soon oP P onP noxn oP P ON soon OP P onP noon OP P ON soon Oo N OOm noon Oo N Om soon so P OOP nozn so P onP noon Om soon no m onP noon no m POPPOuoouuo Om soon mo N onP no3n mo N Om soon mo N onP noon mo N Om soon NO P OOP noun NO P ON soon Oo P onP noun OO P sONNPPOJOHo nPoP nPoP ONO ONO POO Ono OOO OOO nNs nNs mOO smO smO OOO mmO nPOP OPoP ONO ONO Ono on: ns- ns- OO A6b-3 Ims IO, IN IO 0000 Im— In In IO— ”COM ION um 00 I— ImN “‘0 IO: Im IN I.. OOFN nmN Im IN Im— MOO.- NPI—N 0 IO m Im— O ImN Imm o o Iwm — a Iwa I5 INF Ir? Imm NCOO F I:N now '— o Im 3 Ion Io ION INN INO HOMO INN I'm ION Imm 000° MPION 0 Im. e INm N ImN O IO O Im O INP O ION IN IFN Ipr IO 0006 O Io O IN» O IPN O I0— I— Ip— I—p ImN 0000 Im Io IPN Imm NPOO O Io— O O O 0000 CO CO OC‘OO 0000 I0 I. In Io— IO I— Im— IO IO IN ION I» IO Im IN Ip IO In Ip Io In Io Im— IO ON ON, OFM mmm MON aON mm son OO :0"— OHOH Om? mmnm 00 m ON know no N ON whom NO w om. mmxm NO — ON whom 2O — Noam uoqmnm oo N no N N: — om, nmzu ON know co when om— mmzm NO — ON Hmom mo p poam "oqmam om— wmzm ON know om— mmzm ON know Omp mmzm ON Hmom «Omm om— mmxm m— N ON hmom 3O N ON hmom so — om, mmzm no — ON buom MO u Norm "OOmHm mO N mO N mo — mo . mo . "OOmHm om, mmzm 00 m ON Hmom mO ON whom no omr mmzm aO ON whom NO popm "aquuh P’PN Or— warm on Huom OO — omp mmzw OO p ON hmaw 0O — NO2N "OJmHm om, warm ON hmom OO — Omp mmzm OO p ON hmom so — FOMN "Ogmum om— mmzm FONN "OOHHL Om: 9H2: H) x cm aw on on up mN 5N (N 0N "N am n~ m, mp m, cm a~ 2N 2N .— mN 2N o— 0. hp pN FN o. p, p, m. :u o¢m mmm ~03 mo: o~m mom mmm man No: own oom mmm mmm mp: m»: ONm mom mmm No: No: ONm can No: No. own .mc N—m N—m own FNO com com ONm FNO OO O I— ImN I— Iw 000 O INN 000 I a "ONF-N I O m I“? O ION O—I—o o Inn 0 Ion O ImN O Ihw a Ian 3 IN— IO Io— Iom NOO In, I.m Imm Ia~ Inc COMOO m IOm o Im~ ..Iom o I., , Iom OONOO N00003 _I—~ Imm Iom Io Imp Imp IN ION Io IN Im IP I: IO— IO Im IaN InN I'm IN I.— IO IO IN Ip Io Io— 000 00000 000000 IO Io IO Im IO IO I: IO IO IF IO IO IN I. In IO IO IO cm 2, ON ON whom no F om, mmzm MO — omp mmzm mo . ON know MO — —ONN "Oawnu Omp mmxm ON know no p om? mmzn MO p ON know OO — NOFN "oqwnm Om Hmom MO — omp mmzm MO , ON know so N FOFN “aqunm om, mmxn cm whom no _ mmam mo . o~ buom no . whom me ~ was. "aqmum om? mme Om know OO — om. mmam OO p ON Hkom MO — NOmp “qunu om— mme ON Huom mmzm on whom mman know O— .Omp uoqmum mo — OO — no p to p — omp mmlm Om know mO — om. warm mO p ON know IO — pON. uoqmnm on Hmam co m cm. warm so m cm whom mo . mmam no , on whom we m cm. mmzm ~o m mo.mouoqmuu Om: kHz: H) 1 FN 5N ON mp mm mm mN mm mm m— mN mN @— mm mm NN N—m N—m ONm .wo ~pm ~,m o~m NO: NO: ONm omm wo. wo. o~m ONN oom NO. NO. own mom mmm mmm Non Non ONm .No woe mo. own m—o— m,ow awo awo one on POPO OFNPO mogcoo—PO’ EN Ian I.m Imm In Im~ Imw I—s IaN Is Ic— Imo Io IN Imw In: —IFN Im— OOFOO OP"C OOPOOPF'OO" I p o In. o I.— apImm o~Io~ o I» GO OFFMOOMOOO I g. In 0‘ ONOOO OOOO OOO’O OOOOOOPOOO OOO'OOPOOO OPNO" In IO— IO I. I0 Im— IO? I— IO It Io IOp IOP Im l— I." I» Iow In Io Im IF OOOOO OOOC COOOO OOFOOOOOOO 00000 0000000000 IN IO IO I— IO IO I— IO IN IO IO IN IO IO IO IF I— IO IO IO In IO IO In IO IO IO IO In IO I— IO IN IN IN IO Ia IO I, IO IO Om wucm om— mmzm Om know 8— Om— mmzm '— ON whom om— mme Om Hmom on. mmzm om— mmzm Om know mO—mouoqmuu Io O mm—wNN——NN on puom om_ mmxm cm hmom mo om, mmam cm Hmom om, mmzm on whom om, mmzm cm Egon m. cm. mmzm m. o~ whom mmzm on whom mo mmzm mo mmzm cm wmom wopmouoquam on know no om, mmzm so om anon o nmam c on whom mmzm cm whom mmzm om know .— mmam _. know mmzm on whom we mmnm mo n.3m om Puom Fopmouoauuu Om know co m Om. warm to m —Ornouaaunh a O WMPPNNF—wwmmq—c—mm m O MMFPNNFPNNMMFFMM oko» an: 9H2: H: : =Ca aom mNh mNh moo moo hmo pmo Ohm mom mpop mFO— mp0— m—Op OO A6b-4 800 O "OOONO O'- O N MN NO 00 INN INn IN— Imn ImN INn INn ImN Imp IO IO IO— I». ION I.n INN M Mu-o 5 IN IOF Im IN I.N Imo Ion INN MOMOOO 00 IO IO, —IN: IO: Ino Io O N n O I:— O N O O IN: IN IO. In. ION ION I—n M N CO OO O I—— OPNOOO 0° 000 O FOOOO O OO O O OO 00 IO I— IN I. IOP IO In IO IN ION Inn Inn IO IO 000000 00 0 000 0 000°C O OO O O 0° 00 IO IO IN IO l— IO IO IO IO In In INN IO IO IO IO IO IO— IO, IO IO IO IO IO IO I— IO IO nn oom oo— moo. no, n~m_ ~n~ FON: ONO. n: Huom mmzm know mmzm Hmom mmzm Hmom warm know on NOns "aqmnm cm— on. on own Om, ON Om— ON Omp On warm mmzm know mmlm Oman know mmzm know mmnm whom NONF "oqwmu OO— om— ON can on own On om. ON omp On mme Oman know warm whom mmum Huom mmxm know mmlm know FONF ”O4 on, own ON ”Nam mmzm khan mmzm mme know Oman Hmom mmxm know OO was Nops "OJmHh om— mmlm :O M'FP' p omN Oman 00 p pow» ”Gamma UHOH On: wax: u) t On Nn :9 OOO noo n:O n:O FNO oom mmm Oh: Oh: mnn :mO nOO on 00°C 0000 0000 OOO.- NOOO U‘OOO IO— Io In In InN IO— IO Io, Ion InN IO IO ION IO IO ION I— IO I—N 300° 000° OONO OPOO OOOO ION IPN In, IO: ImN ImN IbN IO— Ion IO: ImN Imn ION I—N Is IO Ion I:N I— IO Io. In. om: "00° 0°00 "000 .000 I0 Ian INN Io IN IO, I:- IN— I— It Imp I» I: IO— ImN IsN In IN— IN— Ipn In INn In» 00°C 000° 000° 000° 000° 000° IO I» IO IO IO I:— In IO IN In IO IO Io— Ia IO Ion In. IO Io Imn In. O—N Opp :O ONN r: so: nON mop 50: on 0909 On: Hug: a) : om, moan no ~ o~ know no ~ o~ whom we . om. nozm ~o . o~ whom no , ~omm "oouan o~ anon ~o n powm "oouHu .3mNQGC om, no)» no ~ om whom no ~ o~ whom No . om? no)» «o . o~ whom mo . ~omn "gonna om, moan no ~ o~ whom no N o~ yuan no . om. mnzm no . o~ whom no _ .onn "oouuu om. moan co m o~ whom no m o~ whom no , om. moan no . o~ «now mo . ~o~n "cough om. moan o. ~ ow anon mo m o~ whom no _ om, moan no . o~ anon no . .o~n "oomuu on. moan o. ~ o~ anon ~o ~ o~ when mo . om. moxm ~o _ o~ whom oo — porn "oounu on FN ON ON 5. N, on mm mm mm NN On mm on On ON on ON on 0— mp on ON ON ON mp ON ON nN nN :F :0 oom nnm No: No: ONn mnn oom nnm NO: NO: ONm «on nnm No: No: ONn oom nnm OOn OOn ONn oom nnm NO: No: ONn OO POWFMO ONII‘ OOPOU‘F OOOOOOOON In— Ion In, I. In IN: I—N IwN Imn Inn IO— Ip— IO I—n InN I—p IO: IO: ImN I—w IO I— Ian I» IO IO: I'— In FIaN IN INp In ION IN— IO: 000000 I» ION Imp I». Imn IO— InN INw INN 00" 000001! IO Im— IN— I—: I0— INP OOOONOOOO I Q IO— O I—— O ION O IO— O Ion sOInO 000000 "00000 OOO OPOOOOOOO NOOO IO IN um I— In In IO In It Im IN IOP I:— Ion InN IO Imp I—N IO IN— Inn 000000 000000 000 000000000 0000 IO IO IN IO IO IO IO IO IO IO IO In In In IO I— I— IO IO In I». I:— IO IO IN I— IO IN IO IO Oo— Np, no— :N« Opp no Op UHOH am: has: a: t know mmzm mmzm know wmzm hmom Oman know .O—h "nguu om— om. on om. ov— om. mmxm mm)” know mmzm mast mmxm buom know hmom Oman know know I O I-Mo-v-NNNPo-Pu—u- so none ”aqunm om— om, ON om— Om— Mann warm know Oman warm whom khan khan «Nam khan know M O o-NPPNCUNo-c-u-o- CO FONO ”aqunu no n mo N OO » 0O p :O p Oman know khan Oman when Foam “OAOHL om— ON ON Omp ON mmzm :O m know so N whom no — mm)” no p know no p nOnm "OJOHO ’— Op ON O— can mmm mun own own oom nnm NO: No: ONn OO A6b-5 PNO OO O PONOOOO OO O U‘OOO 00 005000 00 INn INN IOn Inn IN IO In: I:N I:N INN In IOO IO 0 000 OO O OPOO OO PONOOOO 00 In INN IO. IO. ION IO IO IO. InN Inm INN Inn IO I. IN. IN. IN. INN IO. INN ION O "000000 00 It 000° 00 0 000 0 ”000°C 00 IO I. IO IO IO I.n IO IN IN I. IO. IO I.n IN I. IN Io. Inn INN In. IO. IO IO IO I.. In ION IN. 0 000 OD OOOOOOO OO O 000° 00 OOOOOO 00 IO IO IO IO IO In. IO IO IO IO IN. IO IO. IO IO IO I. I.. In. IOM IN IO IO IO IO IN IO In. I. nn. NON ON: OO. :ON NON N.. OOn N.. :O O:. nON .N. OO. .O. .N. NN. O. OO. mmxm MO . OO. mmxm O. . ON Nucm rc . om. warm OO . OO. mmzm =0 . ON know NO . ON Hmom Mo . OON mmzm MO . OM Nmom OO M NO..."OOMHO OO. MNJm MO . OO. mmxm O. . ON Nuom OO . OO. mmzm OO . om. mmzm no . ON know OO . OO. mmnm OO . NN Hmom OO . OON mmzm OO . OM Hmom OO M .O....O4m.u OO. mm3m MO . OO. warm O. . ON Nmom OO . OO.Amm3m OO . OO. mm)“ NO . ON hmom OO M ON chm Nc . OO. mmzm NO . ON know no . om. mmzm OO . OM know OO M .OOO..OAmHh OON mmzm OO . OO. mmxm O. . ON Nmom NO . OO. mmlw NO . OO. warm OO . ON know OO . OO. mmjm OO . ON know OO . OO. mmjm OO . OM Nmom NO M NONO..OOth Om: Nun: H: z OOO mOO rm: NmO nnm OOn con mnn OOO Nno .no oom mom nnm nnm ONn ONn mnn ”CO 0 MOOO‘O CO OOOO’F" “‘0 000 00" IO Inn IOn Inn IO IO IO. IO I. IO. IO IO IO. IO In In: IO I. NOOPOFO 00 N NOOOO 00 In In. IO. IO. IO. IN. ION IOn IN In. In In~ I.n In~ InN I.n I- I~m IO I.. IO ION "OOOOOO CO 00°00 CO I. I8 IN IN IO. IO. IO Ia. IN. IO IO I. IN IO. In ION IO IO. In. IO 0000000 00 0 00°00 0° 00° IO IO IO In. IO. I.n I~ IO IO IO IO In IO IO IO. IO I. IO IO IO IO IO IN O. OO NN. OO. OO. NO. NO OO. OO N.. .. oNo. om. NN. ~on .o. ~n. on: n~n on .N c. 0.0N am: has: n) x OON mmzm OO. Omzm ON Nmom OO. mmxn warm Nmom Nuom mmzm Nmom mm3m . Nmon .ONO..OJth 3 O Mo-v-o-v-mO-o-PP— OO. OO. ON OO. OO. ON ON OO. ON mmnm mm3O Nmom mmzw mmmm Nuom Nuom warm Nmom OO. mmzm OM Nmom .0.0."OJqu N . . . . 00 n . . . . n OO. mmzm OM Nmom MO N OO. mmjw MO N ON Nmom NO . MOMO "OOOHO ON Nmom OO . NOMO "OJOHN OO. mmzm OM Nmom :O N OO. main :0 N ON Nmom NO . .OMO "OOOHN OO. Oman OM whom no N OO. Oman MO N ON Nmom NO . NONO .qunh ON ON N. O. NN NN N. O. 3. OOO NnO NnO OOO OOO nnO ON: ON: O:M oom ~on NO: own own .NO Non Non o~n OOO No: NO: own on P NP 0 OOO N *O 00° 00" 000 IOWPN ON ION I. IO I:N I: IN. IO. IO IN IO INN In. IO. IO IN IN ION INn Inn In. ION I. I0. Ion In. . O N 80° MONOO 8 IN. I:n IO IOO IN. I: IO: IO IN. IO. IN In. INN Ia. InN IO. IO. 0 COD 0 00 00000 O 00 In Ion IO In. I. IO. IO: IN I. IN I. It In IO IO IO IO IO I. IO IO IN I: O O 0 000 O DO 000 00000 O DO IN. O IO IO In IO IN IO Io In In IO Io IO IO I. IO I. IO IO IO In IO I. IO IO IN nn. non ONON Om: was: a: : OO. warm OM Nuom NO . OO. ONJO NO . :N know OO . .ONO "Ogunu OO. Oman OM Nmon NO N OO. Oman NO N ON Nmom OO . :0.0 .quum om. moan on .uom no n on. moan no n ON know OO . no.O .oqmnm No.O .oquuu .o.O “ogunm OO. mmzm :O . OO. warm NO . ON whom NO . OON Omxm NO . OM know me . OO. main OM Nmom MO . OO. Omzm Mo . OM Nuom OO . OO. warm OO . OM Nhom OO M No:N "ogumu OO. Oman OO . OO. ONJO OO . ON Nmom OO . OO. Omxm OO . OM Nmom NO N OO. warm on whom no . OO. warm :0 . OM know NO . OO. mmzm NO . OM Nhom M OO .O:N uoqwuh OO. Oman OO . OO. Oman OO . NOMN "aqunh ON ON O. ON ON O. O. O. NO N: on :0 oom ~on Non o~n oom won Non own non Non Non own A6b-6 For“ OPOOO MMO”? ONN IN: ION IO: IOn IOn Io. INn ION Io. I.. In IO Ian IN. IO ONOOO NOCOO IO IO. IN IN I.. INO In I.N InN I.n Inn I.n In. Io. Ion OFOOO OOOOO IO I. IO IO IO IN I. IO I: IO IO. I.. IO I:. ION 00°00 00°00 00° IO IN IO IO IO IO IN IO IO IO IN OO NO NON nNN NO OO. ONJO OM Nmom :O . OO. warm :0 . ON Nuom OO . .On. "oqwuu MO:N unqunu OO. mmzm O. . OO. mmzm NO . ON Nmom NO . ON know :0 . OO. warm :0 . .OMO..OOOHO OO. Oman O. . OO. mmxm NO . ON Nucm NO . ON Nmom nO . OO. mmzm Mo . :O:O..qunu OO. Oman O. . OO. warm OO . ON Nmom OO . OO. Omzm OO . NooO.uO4unu o. n. Nn nn nn :N :N .n ON NN NN oom NO: Non ONn mom OOO nnm N.O N.O OOO OOO nnO N.O N.O ONO» Om: NHzO H: x :0 on 0000 000m .0 000000 ODHOO‘O ON InN o InN I.. o Io. IN o I». In. N I.n I.n o In. In o Inn I. o In. IN o Ion Inn o IO. Io o InN IN o INN Io. o In Ion o I: ION o I. INn o IN InN o ION Io. o I.. Io o Io. Io n Ian I.. n Io. Inn o In. ION o InN InN o Io. IN 0 In. Io. N.INO I: o In IN o In. n n 000° 0000 000000 00 I: IN ION IO IO IO I:. IN. I. IO. IO IO IN IO IO IN I:. INN IO. I. In IN In ION IO: IN. I.n 0°00 000° 0 000000 00 000000 00 IO I: IN IO IO IO IN I. IO IN IN IO IO IO IO IO IN IN IO IO IO IO IO IO Ion In :NO OO. .ON nn P- OM OMN nNn NOO OO: M.n O.. NO. O.. mmnm no n .uom mo N .uom No . on. noun No . oN .uom oo . NoNN..onu.u OO. On ON Omzm MO N whom MO N Nmom NO . OO. Oman NO . ON Nmom :O . .ONN..O4NHO OO. ON ON OO. ONJO MO N ON Nhom OO . OO. Oman OO . ON Hmwm OO . NO.N.quuHh OM whom NO N .O.N..O4wnh OON OO. ON OO. OO. OO. Oman nmxm Nmom warm ONJO mmzm ON Nuom ON Nmom OO. warm MON...O4mHm O mm3m OO . NON..uaquu OO. warm OO . OO. mmzm OO . ON Nmom OO . OO. Oman OO . OO. mmzm OO . ON Nmom NO . OO. warm NO . ON Nuam MO . OO. Oman MO . .ON...quum UNON Om: kHz: .3 2 ON :N MN MN M. ON MN O. O. N. ON O. O. No :0 OOO nnO NO: NO: ONM OOO nnO ON: ONM OOO ON: ON: ONM nnO NOOOO OOCOO OOMO" FO OOOID" OO NOPF IN IO I. IO IO IO IO IO: INN I:N IN. IOn I:. IO. IO I.. I. IO IO r. I:. IO IO IM INN ION IN. NOOOO COOOO OOPOO 0000-- M000 IO. INn IN In I. IO. IO IN I.N INN I.n ION IN. I:N IO. IN. IOO ION IO IO I. IO. I: IO IOn 00°00 00°00 00°00 00°00 0°00 ION IO. In IN Io IN. Ir IO IN IN IO. IN IN ION IO IN IO: IN. IO. I.. IN. I: I. IN 000°C 00 CO 00000 00000 00000 0000 IO In I: I. IO I.. Ic. IO IO IO I. IO IO IN IO IN IO. IN IN IO. In I. IO IO I: IN Io In In I. NN. NON O. On on :.O. .:O :O. OM Nmom OO . OO. ONJO OO . ON Nuom O. . NO:O..OOmHm OON Oman O. . OON Omzm NO . ON know OO M OM Nmom :O . OON norm :c . .O:O.“Ogmnm OO. ONJO O. . OO. Oman OO . ON Nuom OO . OM Nuom OO . OO. Om)” OO . ON Nhom MO . .OMO..OOmHm OO. main NO . OM Nmom OO M OO. Oman OO M OM Nmom OO . OO. Oman OO . ON Nmom MO . NO::..OAmnh OO. mm)” NO . OON Omzm OO . OM Nmom OO M OM Nuom NO . OO. Oman NO . ON Nmom OO . .O::..Oqunh OON OOJO MO . OON Oman OO . OM Nmom MO M OM know :0 . OO. mmzm :O . ON Nmom OO . .O.M..qu~h OO. ONJO OO N OM Nmom OO N ON know NO . OO. warm NO . ON Nuom :O M :ONN..OAN~O UNON Om: N.OO u: t NN NN N. .n ON ON NN NN ON :N :N ON O. ON .N On On O. NO: NO: ONM OOO OOO ::O N.O N.u mom OOO nnO Non Nan oNn OOO MMO NO: NO: ONM OOO nnm Non Non ONM OOO OOO ::O NO: NO: ONM OOO nnm NO: NO: Onn OO Appendix 6c. DD CH M NI SWP ASWP 50 A50 PNS adult count data. degree days base 48°F crop height (inches) field wetness (l=dry; 2=damp; 3=wet) wind speed (mph) number of sweeps number of adults in SWP number of linear feet number of adults in SQ .. n: O: ON 0. PC) «I: tom N: OMDMO OO OO OO OO OO OO OO OO OO OO OO OO om< om NN oO. ON. 0O. oo . 0O. 0O. o. . O . N N o. . N N O . o. . .onN “nONHN . oO. N oO. no . .nN oO. no . .N oO. O . on. oO. N . NO. Oo. N . OON OO. O . n . O . O . .ONN "aqunm O OO. O oO. no . n. OO. N . ON oO. o. . NO OO. N . O n OO. OO. O . NO. OO. : n O N O . N . No.N "OONHN no oO. no . 0.. Oo. O . :N OO. O . nO OO. O N c n O . c . n . N . .o.N "adunm mum: arm .3 x ...— MMU‘OO ONN NN O. M. O. Mn O. O. O. O. .. O. mu .NO OOM OON .nN .ON :O. .N. .O. .NO N.O Own onN ONN O.N :O. .N. .o. .O .NO N.O OON ONN O.N .ON :O. NM. .N. .O. .O NO: ooN ONN O.N .ON :O. .N. .o. .o OO '- OOONMRM OPOOOOO nn .. O. N. F 00° MPFN '- OO. OO. OO. OO. OO. OO. OO OO OO OO. OO. OO. am: am no :3 No:. FOUNMOIV' N NOM. .On. n O. OO OO .O OO ON .ON. "ngmm Or. OO. OO. OO. OO. OO. .OOuHh OO. OO. OO. OO. OO. OO. OO. OO. .OJNHN OO. OO. ON. :O. OO. OO. OO. unqwuu FO mNMOOQNO PMPMMFPP OO :O N. O. II‘FIDFIOM O \OlnIntNInOIDInID o-o-I-Mu-MMv-O- v-Nv-MI-Mr-o-o-P mmm: mum H) I so OOO NO: OON ONN .ON :O. .N. .O. .O aOO No: OON ONN O.N :O. .N. .O. .0 OOO MMO NO: OON ONN O.N :O. Nn. .N. .O. .O .NO No: OON .nN O.N :O. Nn. .N. .o. .O OO OO. OO. OO. OO. OO. :. OO. OO. OO. M0.00.04mHu Oo nc FMNOFMO' O F a O MPNPNMPM OO. or. OO. MO OO. :O OO. M. OO. OO OO. OO . OO. MO NO.OONOJNHN MONMNNCO MPNFNMPM OO. OO. O OO. OO. OO. .. OO. N. OO. MN OO. .0.00noqu.u QONPPO m O MPNPNMv-M OO. :O OO. OO. OO. OO. n OO. .O.:Ouoqunm IAIANMMO M O PFPMPM N OO. NO M MO.MO“O4m.m O. OO. NO M NO.MONOOOHO ON OO. NO M .O.M0.04m.u O OO. OO N O OO. MO . O OO. NO N NO.N0.00w.h m3m< arm .3 z 0.0. :NO OOO zoo ONN nOO NnO ONO 0.0. :NO OOO :OO ONN nOO ONO 0.0. :NO OOO :OO ONN MOO ONO 0.0. :NO OOO :OO ONN MOO ONO ONO ONO 0.0. :NO .OO OO OO. OO. OO. OO. OO. OO. 0”".th NO.N0.04mHu OO. OO. OO. OO. OO. OO. N OO. OO. OPOOMOOO .O.NONOOmHm OO. OO. OO. OO. OO. OO. 0O. . OO. n OO. .OhlflOlflIflOO 00 NO O. OO OO OO no NO NO O. OO NO O0 O0 Mo :0 M N . N . v-NPNNv-N v-v-NPMv-N M :O..Onoqmnu OON OO. OO. OO. OO. OO. OO. OO. OONONOOO PNO-Pv-o-N MO..O"qunm OO. OO. OO. OO. OO. OO. OO. OO. OMIQPNOO man: NIO .3 : OO FNPPMP~ . NO..0.04uHL NN NN nn O. O. On NN NN nn ON NN OOO OOO . OO. Oo n oO. no .. OO. NO OO. O. oO. o. oon Oo O. OO. NO 3. OO. n. OO. Oo n .o..n"oqu.m FNF—Pv—N O OO. MO N O OO. nO N M OO. NO . O: OO. OO . NONN..O4uHh OO. MO . OO. MO N OO. NO . OO. NO . OONN..O4N.& MMO.- .- OO. MO . OO. MO N OO. NO . OO. NO . OONN..O4u.h OO M OO M OO N OO . OO. OO . OON OO M MONO."O4uHh O N O O: OON OO. OO. OO. ONMNIQ OO NO OON OON OON OO OON OO OON O. OON OO OO. :O OON MONN .quum OOONOCOO MMFPMMM mmm: mxm H: x 0.0. :NO OOO ON: 0.0. :NO OOO ON: 0.0. OOO ON: 0.0. :NO OOO :OO nnO 0.0. :NO OOO :OO ONN nOO nao OOO OO O cw. :c . OO OO. MO N OO OO. :O . . OO. :O . .N OO. :O . O: OO. NO N ON OO. MO . .0.0 "OOme O OON OO M O OO. OO M .. OO. MO . :O OO. MO . : OO. O. . Mo:M "oqmau O. OO. OO . N. OO. MO M :N OO. no M M. OO. OO . : OO. O. . .O:N "Danna .. OO. OO . O. OO. MO . .M OO. MO M OM OO. OO . M or. O. . O OO. NO . MO.N "aquuh N. OON :O . N. OO. NO . n OO. nO M M OO. OO . N OON O. . Mo:. NOOmHm N OO. MO . :. OO. :O . O: OO. OO M O. OO. OO . NN OO. O. . MO OO. N. N MN OO. On OOn Mon. .quuu man: Law a) z :. n. o. No .. n. .. n. OO O.o. :NO OOO :OO ONN nOO NnO 0.0. :NO OOO :OO ONN 0.0. :NO OOO :OO ONN 0.0. 0.0. ONN A6c-2 N. OO N. OO OO OO PMO‘ “‘0 OO. OO. OO. OO. OO OO OO InNN mOOO on: on N OO. Oo . Nn OOO N OO. Oo . .n noo On oon no N NN n:o N OO. OOO .o oO. oo . o. nnO O or. No . O. «N: .O oO. o . n. oON .N. oO. n . N. .nN oO. oO. o N .. .oN O . N :O. o N N .N. : . : .o. o. . : NO .oNN "o4”.m o oO. :o . On OOO : oO. oo . On noo n. OO. Oo . On Nno nN oO. OOO .n oO. oo . :N nnO on OO. No . ON ON: oo. oO. O . o. oON n: oO. o. . O. ONN .: oO. o n O. o.N N n .. .oN ON oO. N . .. :O. Nn oO. O n o. Nn. : N o .N. o N n .o. o . O No No.N uoqmnu o OO. :o . On OOO M OON OO . Nn nOO N oO. oo . ON Nno . oO. OOO O oO. no . on nnO N oO. No . NN ON: .. oO. O . .N oON : oO. O . o. ONN o oO. o n .. o.N :: oO. o . o. :O. .N oO. n n o. Nn. : . o .N. o N O .o. N . O No .o.. ":OOHN n oO. no . Nn OOO O oO. no . Nn noo .ono ”ogunm man: man a: x :O OO MNFFQMMN N RPOII‘U‘N ' O 0 ID ‘0 o oO. no . : oO. no . o. oO. :o N n oO. :o . OO OO. O . on. oO. o . :: oO. O n O:. OO. O . On OO. O . no oO. : n O N O N .ono ”aqu.m N oO. no . O. oO. Oo . n oO. no . NN OO. no N O. oO. :o . OoN oO. o. . :o. oO. O . oo oO. : n no oO. o . O n : N a N .ONO "OOOHO n. oO. no n n. oO. oo . nN oO. O . No oO. O . :N oO. N n O . O n n N .o:O ”oOONN NN oO. :o n oNN oO. no . ON oO. O . ON. oO. n . ON. oO. n n N . N n O N N : MOMO .OOwuh O30< mum .3 I On n:o On Nno ONO ON ON: oN oON O. .nN O. .oN .. :O. 0 Nn. o .N. N .o. o NO On OOO NN nOO On Nno on ONO nN ON: N. oON oN .nN :. .oN o. :O. N .N. O .o. : NO O. OOO O. ON: o. oON o .nN N .oN O :O. o .N. n No .. OOO oN NO: o. OON N .nN n .oN : :O. O .N. n .o. N NO mo no N OO O oo n. oo nN oo . oo N on :. OO O. oo O. oo ON OO o oo. .. oo O. OO N on Nn NO no N: OO o: no :: Nn :O .N oo O. OO .: :O :n oO No o: :: o: N. on nO oO N. oo o. oo N oo .. oo on: on n oO. No N n. oO. No . on. oO. O . no. oO. O . On. oO. n n OO. OO. O . N n O N o N NonO "oqm.m NN. oO. N . OO. OO. : N o:. oO. : . n n : N : N .oNO "oqm.u O N .o.O “o4“.u oN oO. No N OO oO. No . os: on o . 0.0 oO. : . noN oO. n n ..N oO. O . O.. oO. N . ON oO. n n N o O . Non: "OOOHN . oO. oo N ON oO. no . oON oO. o . :OO OO. o . NOO OO. : n :OO OO. O . NO OO. : . nO OO. N n : N : . .on: "aqunu O oO. no n no oO. :o . OON OO. O . NoN: "ogunu Nam: mxm H) x OM OM O. ON :0 OOO NO: OON .nN .ON :O. .N. .O. NO OON .oN :O. .N. .O. NO .O. OOO NO: OON .nN .ON :O. Nn. .N. .O. NO OOO NO: .MN .ON :O. Nn. .N. .O. NO OOO OOM OON OO NN OO :O OO O: O: .. OO O OO O OO N. OO NN OO ON OO ON OO M. OO : OO O OO Mn OO ON OO MM OO NM OO : OO O. OO O OO ONUOQU‘IIB 0 ‘0 O. UPOFN p am: am onN oO. n n NNN oO. : n oOn oO. o . O n o . : . NoN: "gamma m oO. o. N NO oO. :o . .N. OO. O . nON oO. n n No: oO. O N OO. OO. O . O n O . : . .oN: "n.a.. O. oO. o. N NN oO. No . oon oO. O . O:N oO. n N no. oO. : N OO. OO. O . o n N . n . .o.: “oOOHN O OO. Oo n .N oO. No . :: oO. N . n: oO. : . NO oO. : N on OO. O. . o: oO. o . .N oO. o n N N o N No:n "o4”.m N oO. no N N. oO. No . .n. oO. O . NON OO. O . OON OO. O n No. oO. O . N n o n .o:n “ogunu min: arm H) x .nN .ON :O. .N. NO OOO NO: OON .MN .ON :O. .N. .O. NO OOO NO: OON .nN .ON :O. NM. .N. .O. NO OOO No: oON .nN .ON :O. .N. .o. 00 O. OO :. OO N. o-Nflpm ON '- O O n N n .O:M .Oqunm O OO. O OO. OO N :N O OO. OO . O. OOM OO. : . :. nON OO. n . O N.. OO. M M O O . O O M O O . M N N N .OMn ”aqunu . OO. n. N On O: OO. :O . :N OO. OO. O . :. OO. OO. M N .. N.. OO. O N O. .O. OO. O . O. : M N O N n O n M NO.n "OOOHO N OO. 0O M ON OO OO. :O . O. N:N OO. N . O. Mn. OO. O N O OO. OO. O N O NO OO. N . O N n O M . O O N M .O.M "agunm N OO. O OO. OO . .N OO OOr N . O. NN OO. O. . O. O: OO. O N n. O M :. :. OO. N . O. O N O O . O n . O NO:N "OOOHO NO OOO nnO ON: oON .nN .ON :O. .N. .o. NO OOO NO: OON .nN .ON :O. .N. .O. OOO NO: OON .nN .ON :O. .N. .O. NO .NO N.O OON ONN O.N .ON :O. .N. .O. .O man: man .3 x :0 OO A6C-3 N. 00 o. OO .. OO MNFMV‘F NM 00 00 ON OO NN OO O. OO N. OO om< om Nnn OO. O. . NON OO. N. . .O. OO. O . O.N OO. O. n ON OO. O . o n O . O . .ON..HOOO.O O OO. no . Na OO. o. . O. OO. Oo . nN OO. no . OOO OON no . NO. OO. N. . an. OO. o. . on OO. o. n ON OO. O . a . O . N N No...“OO0.0 o OO. no . N OO. O. . O OO. Oo . N OO. so . NN OO. OO . On: OON oo . OON OO. N. . OO OO. o. . On. OO. O n nN OO. O . a . O . N n .o....aqnnm . OO. no . N OO. o. . ON OO. Oo . .n OO. No . nO OO. No . NN. OO. no . O.. OO O. . OOO OO. O . .O OO. O. . NO. OO. M . O N .o:o..aamuu N3m< Nam a: z .. n. O. .. o. OON OON ONN .ON am. NM. .O. NO OOO NnO nnO Onn OON ONN .ON :O. .N. .o. NO OOO NnO nnO nnO OOn onN ONN .ON .O. .N. .o. NO OOO MOO NMO OOO MMm ON: OON .ON .ON ON. NM. OO O OO . OO OFNO‘O‘R O \O co co co OO 00 MFMF‘DM '- NN OO OO oo OO P'NON O‘ OO Jmc OO O . N N .OOO..OOOHO o OON OO N OO. o. N OO. NO O. OO. OO O OO. N oO. nN OO. O. .O OO. NO OO. OO. NPNO-MPPr-o-O-Pv-o- NONO.”O4th OON OO. om. NO OO. mo om. om. OO. O. OO. O. OO. O. OO. O OSOO-ON—P '2 mg mm PPNPMPI—o—o—PPPF .ONo.quuHm OO. . OO. Om. NO OO. mo om. NO om. mo OO. M. .O. OO. O MON OO. O OM on. N O M O '— MQMON '- '- .M om. .0.0."O4mum O. OO. MOMO .OJOHN N3n< N13 an t OtN :0 0.. OOO MOO NMO MMO ONO OON ONN .ON :O. .N. .O. NO OOO MOO NmO OOO ON: OON ONN .ON :O. .N. .O. NO OOO .0 0000'- '- OFNM‘O .3 '- OOOMO ON :- o. NN Om< oo. oo oo. oO OO. oo. oo. 00 OO 00 OO. 03. OO. OO OO OO OO OO. om .. OO. no N n OO. O. . OO OO. n n ON OO. N N O . a . nonO "OOO.N o OO. O. . O OO. O M O OO. O n n n On OO. O . N . O N O N NonO "OOOHO o OO. O OO. so N o oO. a. . O. OO. O n n OO. O n O n n . n . n . n . .onn "OOOOO O. OO. O OO. no N N OO. N. . On OO. O O OO OO. O N O n O . n . n . O . NONO "OOOHO N. OO. No. OO. No . NO oO. N. . On OO. o. . O.. OO. O N n n NO OO. O . O . .ONO "OOOHO OJOO NOO H: x ON .. a. NO: OON .MN .ON OO. .O. onN .nN O.N no. .N. .O. .O .NO NO. OON .MN O.N .ON :O. .N. .O. .O OOm NO: OON .MN O.N .ON :O. .N. .O. .O OOm No: OON ONN O.N .ON cm. .N. no NONOOmO PO FPOOO O OO. OO NMNM. NM 0 0 OO. OO. OO. OO. OO o 00. OO. OO. :O. OO. 00. OO OOO on a . N . .ONO "OOOHN M OO. .: Om. NO N O. OO. N. . NO OO. O. . O: OO. O N N M O: OO. O . M . O . O . O0.0 "Gaunt N Om. NM OO. :O M N. . O. . N N N . M . O . m . Mo.O "oqwnu O . N0.0 "aqunm O . .0.0 "oqmnm N om. so . O Om. No . a. OON NO . m OO. O. OO. MO . O. OO. OO . O. OO. O . :. OO. O M : OO. O M O M N . N . O . o. M NOON "oqmmm N3m< Nxm a: 1 ON O. :. WMOFMM v-u-r- P MMMU‘OFFO OOm NO: OON ONN O.N .ON :O. .N. .o. .O OOO NOO OON ONN .ON :O. .N. .O. .O .O .O OO0 MOO MOO OOO MMO ON: OON .MN O.N .ON :O. .N. .O. .O OMOMPOM RQN OMMN nomcnm FN O O 00 OO OO OO co 00. NM 0 O .- O Q O cu. «O . . OO. OO . . OO. Oo . O. OO. .N OO. so . O OO. No . N OO. N . O OO. O n n oO. O n O n N N O N N . N . .OON ”OOOHN o OO. Oo . O OO. Oo . o OO. no . O OO. O OO. no . O. OO. no . ON OO. O . Nn. OO. N n O. OO. N N n . O N O . a n NOMN "OOOHN o OO. no . O OO. no . N OO. No . N OO. O oO. No . N OO. OO . O. OO. N . nn OO. n n O. OO. O N nn OO. O . nN OO. O N O n a . O . NONN "OO0.0 NnOO Nam .3 z NN OOO ON MOO .n nnO MMO .N ON: N. OON N. .mN O. O.N O. .ON 0 :O. N .N. N. .O. O .O OOO on nOO On nnO OOO ON nnO nN ONO O. OON M. .mN N. .ON O :O. N .N. N .c. O NO OOO Nn nOO on NnO OOO on nnO NN ON: ON OON N. .nN O. .ON O. OO. N. NM. O .N. O .o. O NO zu no c-4 ooP ooP ooP on om ooP ooP OMPfiOOO oo om 00‘ om on co co co om om Pv-U\=::r-— r—v- oo oo oo on oo co co NMMU‘OVOM N'- om< om o omP om onP :o P o omP NP P P omP oP P o N o P N P = P m P PoaP PQPNPN o P oP N moaP PQJNPP o omP oP P a omP No P m omP so P a omP P P PP omP P P moP omP mo P moP omP mo P o N m N N P PomoPuongu m omP oP P :P omP N: P NP omP mo P PNN omP o P Pm omP P P No omP a P woP omP P P m N m N a N zo=mPPquHm m3m< mam H) z mom Nm: omN mNN PON nmP PNP PoP Pm O‘OU‘QFO‘MO '- r-v- POP Po mm cm oom PN omm Pm NPm PP ooN mP oPN PP oPP PP OPP oP NmP m oPP w Pm NM mom mm omm 2N NPm 0P ooN m PmN mP oPN NP sz NP NmP m oPP :u on Na PP Pm mP :P OFF"...- PP N'- "VOQMVO OOOPNO oo oo oo ow GDP 00 co co 00 on ooP 0m 00 ooP ooP ooP om< cm 0 omP oP P P omP mo P o omP mo P NP omP mo P NP: omP P P oPP omP m P NPN omP oP P Po omP oP P a m P N m m Noszuoaum P omN oP P o omN No P PP omN so P «P omP m P mN omP a P ooP omP m P o N m N P N PoamPPoquu m omP mP P NP omP mo P c cmP mo P Pm moP mP P ooN omP m P moP omP oP P a m o N o N PommPPounN m omP Po P m omP co m mm omP oo P 0N omP oP P P omP a P m omP P N oP P m m a N N P NozaPnoqum m3m< mmm H: x Pm 0N PN PP mP PP oP :0 com 0mm mmm No: ooN PmN oPN sz NmP oPP Pm mom 0mm NPm ooN oPN OPP NmP oPP Pm mmm 0mm aw: ooN oPN sz NmP OPP Pm mmm mmm No: com mNN oPN amP NmP oPP Pm no on on Do ooP om ooP own-coma ow ow ooP oo oo NOF—OFN co co oo oo om NQOONN oo co 00 cm on om v-v-NMv-m 33 N ow om< om mP omP Po P mP omN mo P Po omP No P on omP m P Pm omP o P Nm omP P N as omP m P a m o P mP P PozaPuaqum PP omN mo P m omN mo P N omP so P mm omP NP P m omP o N m omP oP P m omP oP P m m o P N P PoPmPunqum o omP co N N omP No P m omP o P m ouP m m m omP a N oN mP m P P P m P :ONNPPQPNPN o omP so m mP omP No P mm omP oP P mo omP m P moP omP o m mo omP m N so mP o P o P m P NONNPPQJNHN : omP no N m omP Po P oP omP NP P NP omP oP P PONNPPoqum m3m< nsm H) : amo NNM QPOQPU‘O v-II-N 2N mN 2P mP 3P 0P cow 0mm No: ooN mNN oPN nmP NmP oPP Po mom omm No: ooN mNN PON nmP NmP PoP Pm 0mm Nan mNN Pom amP PNP PoP Pm mmm No: ooN mNN PON sz PNP PoP Pm mmm sz ooN mNN an m on P 00 :P 00 3 cm 0 co 0P co oo 00 GOP om PFQMR cm oo 00 00 or 00 ONNOMP N 00 P 00 oP 00 NP cc 2 on ow< om m3w< arm H3 2 om omP m m on omP m P mm omP Q P w P s N PONNPPoqum Om omP mo N 0P omP mo P PmP omP oP P No omP OP P mP omP h m Nm omP m P m P o P N m NoPNPuaqum h omP oP P m omP o m cm omP m P o P w P N .P PcPNPucquu o omN 00 P o omP NO P N omP mo P o omP mo P a omP so P NP omP mo P PP omP PP P 0N omP o P mP omP 0P N PM omP oP P a m o P MONPPuoqum o omP mo P P P NONPPPogmHm omP 00 P omP mo P omP co P omP mo P P omP Po P omP me P PONPPPoquHm FFO‘MQO PP oP Nmmfifi mm mm 2N 0N NN 0N PON nmP PNP PoP Po mom mP: ooN mNN PON amP PNP PoP Pm mNN PON amP PNP POP Pm mmm moo Pmo omm cmm mPa ooN mNN Pom sz NmP PoP mom Pm mmm moo Pmo mmm mmm mPn oo Appendix 7. Pubescent wheat study individual egg, larval, and aduit count data. Appendix 7a. Adult count data. There are 6 sample repititions for sweepnet and 6-l0 for linear foot counts. A "30-0" in a linear foot count is a combination of 3, 10 foot samples with no insects found. DD = degree days base 48°F CH = crop height (inches) M = field wetness (l=dry; 2=damp; 3=wet) WI = wind speed (mph) SQ = number of linear feet SW = number of sweeps KNT count per SQ or SN A7a-l PouoP PonoP oonoP POIOP POIoP oonoP OOIOP :OICP mPno ooaoP oonoP oonoP ooaoP zonoP PoaoP POIoP oonoP oonoP couoP oonoP monoP monoP monoP oouoP oonoP oouoP oonoP POIOP motoP NOIoP mPcoP NouoP NOIoP oonoP oouoP oonoP NouoP oOIoP Hzxuom NOIoP NOIOP monoP oonoP NOIoP oonoP NosoP oouoP cowoP zouoP oonoP POIoP oonoP oonoP oonoP oouoP oonoP oonoP oonoP oonoP nonoP MOIOP mouoP oonoP PouoP oonoP oonoP DOIDP nono— PouoP mOIoP oonoP POIoP POIoP oonoP oouoP NOIoP mouoP oonoP oonoP :ouoP oonoP oonoP oonom :OIDP oonoP oPIm oouoP oonoP oonoP oonom monoP oonoP oonoP oonom oouom POIoP PoaoP NOIOP oouoP NOIoP OOIOM NOIoP oonom oonom nanP POIOP :OIOP wouoP cocoP NOIOP PouoP oouom PonoP POIoP NonoP NouoP PouoP monoP POIoP PouoP NOIoP aatoP oonoP POIoP oonoP oouoP oouoP POIoP sonoP :OIOP NOIoP oonoP oouoP oonoP oonoP nonoP monoP monoP NouoP oonoP oonoP oonoP mouoP mouoP PPue ecloP oouoP PouoP PouoP POIoP oonoP POIoP :OIoP hleom sziom Hzxuom NOIOP mouoP NonoP PouoP oonoP oonoP NOIoP monoP oouoP PouoP PouoP oonoP NOIoP oouoP oonoP NOIoP POIoP NOIoP POIoP oonoP PPIoP ooooP NOIOP ooaoP ooaoP oonoP oonoP DOIDP hoaoP PouoP oOIoP mouoP POIoP NOIoP COIoP oouoP oonoP NetoP NOPN "oqum monoP mo N monoP mo N ooooP mo P oonoP no m oonom mo N ooaom mo m NoPm "QJmHm aouoP P3 P oonoP mo N :OIoP mo N NonoP Po P oonoP Po m oouom mo P oouoP 00 N PoPm Poqum POIOP Po P oonoP oP P oonoP mo m oouoP Po P oouoP 00 N coucP mo P ocuOM mo P NO:N Poqmmm oanP OP P oonoP mo P monoP Po N octoP oP P oonom 50 N oonom 00 P OOIOM oP P PemN uoqum :OIoP no P oouoP Po P oonoP NO P mouoP mo P oonoP so P mouoP mo P NOIoP mo P PONN "aqumm mouoP so P NovoP oP P NOIoP mo m NOIoP mo P Hzxnom Hzxuom H: x PP 0P 0P Po mo mo mP mo mo 00 mo mo mo mN mP wP 0P :0 PmN PON sz PNP PoP Pm oxN me PON amP PNP PoP Pm omN mNN PON sz PNP PCP Po ooN PmN PON sz PNP POP Po 0mm mNN oPN nmP PNP PoP Pm ooN mNN PON amP on PoaoP POIoP oonoP ooooP PPIoP oonoP NOIoP oosoP oouoP oonoP POIoP POIoP oonoP oonoP couoP oonoP coucP oonoP oonoP oouoP oonoP oonoP oouoP oouoP oonoP POIoP NOIoP POIOP monoP oouoP oouoP PPIOP monoP oonoP monoP concP oouoP ccncP Hleom NouoP oonoP ooaoP PPIOP POIOP NouoP oonoP NOIOP ocuoP NonoP monoP oouoP oouoP oonoP cocoP oonoP couoP oonoP oonoP oonoP oouoP oonoP oonoP oonoP PoaoP oonoP monoP NOIoP NOOOP oonoP oonoP OOIOP oouoP oonoP oouoP oouoP oonoP CCICP Hzxtom NonoP PocoP PetoP =PIoP PouoP PocoP oonoP mouoP PoIoP NOIoP zouoP oouoP PouoP POIoP oonom oonom oou0m oouom oonom oouom oonom oonom POIoP oouom NOIoP PPlo monoP oonoP POIoP oonom oouom PouoP PouoP oonom NouoP oouom oouom oonom szuom mPuoP oonoP oouoP oonoP NOIOP PouoP oonoP POIoP oonoP oonoP oonoP oonoP POIoP PouoP oouoP oonoP OOIOP oouoP oonoP oonoP oonoP oonoP oonoP oonoP PouoP POIoP ooooP oouoP NouoP oonoP oonoP NOIOP PouoP oonoP :ouoP oonoP oonoP oonoP Hzxuom NPIoP PonoP PonoP mouoP mouoP PouoP oouoP POIoP oonoP oouoP monoP oonoP POIoP oonoP oonoP oouoP oonoP oonoP oouoP oouoP oonoP oouoP oosoP oonoP oouoP mouoP oonoP mouoP oouoP oonoP oouoP mouoP oonoP PonoP POIoP ooucP oocoP oonoP Hleom Hleom H: z POPN NoaP NomP PomP PONP mPIoP 09 N NonoP mo P zouoP Po P NoPN Poqum monoP mo P PouoP mo P PoucP co m PouoP 00 P PouoP so P NouoP mo P NouoP Po P Poqum oouoP NP P mouoP oP P oouom mo N oouom Po P coucm Nc P oouom so P oou0m mo P "oquu oouom NP P oouom mo P oo-Om co m oouom oo.m oouom mo P oouom No m oouom mo P Poqum PouoP NP P mouoP oP P mouoP mo m NoaoP 00 P souoP mo P oouom Po N oouOm mo P Paqmum oonom mo P oouoP mo m oouoP co m mouoP mo P ccucm me P oonom mo P oocom co P Poqmnm PP oP mo mP NP PP Po co mo Po 0P 0P 0P 0P mo no mo PNP PoP Pm omN mNN PON nmP PNP PoP Pa ooN mNN PON nmP PNP PoP Po omN mNN oPN smP PNP PoP Pm owm mNN oPN :mP PNP PoP Po omN Pmm oPN nmP PNP PoP Po no A7a-2 POIOP POIOP mOIoP oouoP oouOP OOIOP OOIOP MOIOP POIOP OOIOP OOIOP NOIOP OOIOP OOtOP mosoP oOIoP oOIOP oPIN PPus POIOP OOIOP monoP PouoP PNuoP =o-oP oPu: oPuo NouoP PowoP oo-oP NOIOP mouoP NPlo PPIO POIOP OOIOP POIOP qulom NOIOP POIOP OOIOP OOtOP conoP OOOOP OOIOP oouOP MOIOP MOIOP oouoP NOIOP POIOP POIOP socOP mPIN oOIOP mouoP POIOP OOIOP OOIOP OOIOP PPuo hPlo NPUN mPuo sOIOP OOIOP NOIOP OOuOP moucP souoP oPIo OOIOP OOIOP ooooP POIOP Hleom NOIOP OOIOP POIOP OOIOP OOIOP POIOP OOIOm POIOP OOIOP OOIOP OOIOm OOIOv MOIOP POIOP POIOP sPIN sPIo mOIOP NOIOP NOIOP oouom OOIOP sOIoP mPaN oPIo aOIoP sOIoP NOIOP OOtOP POIOP POIOP MOIOP sPIoP mOIoP OOIOP mouoP POtOP Hzxuom OOUOP OOIOP monoP OOIOP POIOP POIOP sOIOP mouoP mOIOP NPlo OOIOP sOIOP OOIOP OOtOP OPIN oOIOP mOIOP oPIo sOIOP POIOP IO oonoP PPON PPIo OOIOP mPus sPIOP NP|OP oonoP POIOP NOIOP OOIOP sOIoP MOIOP NOIOP OOIOP OOIOP Hzxuom POIOP NOIoP sOIoP PoooP OOIOP oonOP MOIOP MOIOP POIOP POtoP OOIOP mouoP mOIoP OOIOP mOIOP NPIm NPos mPno sOIOP OOIOP Io NPuoP PoaoP moooP oPuN OPIN oOIOP OOtOP PonoP PonoP OOIOP OOIOP PPIn mPns POIOP NouoP NOIOP PouoP mo PouoP mo ocscP mo PouoP no oouoP NO oonom mo PonoP oo Nomm Poowum monoP Po NOIOP so sOIOP so oouom mo OOIOP so OOIOP so PONm uoqmnm NNMPMv—v- NNMPNP OOIom mo N PoPm Poqwum ooooP co OOIOP so NPuo mo PPIOP mo sOIOP No NOIOP mo OPIOP PO ONIs 0O Noms "QAMHm OOIOP co NonoP co NOIOP so OPIs oo NPIOP so POIOP Po OOIOP so souOP so Poms "oommm MOIOP mo NPno mo sOIOP so oPIo oo POIOP mo sOIoP co moooP so NONs "oqmum FNMv-v-Mv-v- POMPv-Mo—v- v-v-Mr-MMI— Hzxiom Hzxiom H3 1 oP NP NP oo so so so mP PP oo no so so no ON mP NP oo PO so mo NN PP mP NP oo OP so so ON mP PP mo PO so so ooN PmN PCN smP PNP PoP Pm ooN PON smP PNP POP Po POP com PmN PON smP NmP PNP POP no ooN PmN PON smP NmP PNP POP Po ooN PMN PON smP PNP POP Po on sOIoP oOIoP PPuo sOIoP oOIOP OOIoP POIOP PPIOP oOIoP NOIOP NOIOP OOIOP mouoP OOIOP NOIOP POIOP oonoP POIOP NOIOP NOIOP OOIOP OOIOP POIOP OOIOP OOIoP OOIOP OOIOP NOIOP OOIOP mOIoP oOIOP NOIOP POIOP OOIOP OOIOP OOIOP POIOP hleom sOIoP mOIOP MOIOP sPlo NOIOP NOIOP OOIOP OOIOP POIoP mOIoP oOIoP OOIOP sOIOP POIOP POIOP OOIOP mouoP OOIoP NocoP OOIOP OOIOm OOIOP OOIOP mOIoP NOIOP OOIOP POIOP POIoP PouoP OOIOP sPIs OOIOP ooIOP OOIOP OOIOP oonoP POIOP Hleom OOIOP POIOP POIOP OOIOP OOIOP OOIOP NOIOP sOIOP POIOP POtoP ¢PIOP OOUOP OOIOP NOIOP POIOP OOIOP POIOP OOIOP POIOP OOIOP OOIOP oouom MOIOP NOIOP OOIOP sOIoP OOIOP POIOP OOIOm POIOP NPIo oOIoP NOIOP oouom oouom OOIom POIOP szlom NonOP OOIOP NOIOP OOIOP sOUOP OOIOP POIOP moooP POIOP MOIOP OOIOP POIOP oOIOP OOIOr OOUOP POIOP OOIOP sOIoP OOIOP OOIOP OOIOP POIOP sOIOP POIOP OOIOP OOIOP OOIOP OOIOP OOIOP mOIoP OOIOP PouOP OOIOP OOIOP POIOP OOIOP NOIOP HzXIom OOIOP POIOP NOIOP NOIOP OOIOP NOIOP NOIOP NOIOP sOIOP mOIOP POIOP MOIOP soaoP NOUOP POIOP NOIOP OOIOP NOIOP OOIOP NOIOP PocOP moaoP NOtOP POIOP POIOP MOIOP OOIOP OOIOP POIOP POIOP oOIOP POIOP mOIOP O0-0P OOIOP OOIOP sOIOP POPs NoPm NOIOP OOIOP OOIOP NOIOP POIOP OOIOP NOIOP mouoP POIOP mOIOP sOIOP OOtOP POIOP POIOP oouoP OOIOP POIOP POIOP OOIOP OOIOP POIOP NOIOP OOIOP mouoP sOIOP NOIOP OOIOm OOIOm POIOP oonoP POIOP sOIOP POIOP OOIOM OOIOm OOIom no no mo oo mo no so mo no so oo oo PO mo Po so so mP oo oo NO oo Nosm "noun“ no mo co co PO 00 mo Posm "oomHm so no no no mo mo NO Pomm ”oomHh FPMFNMF PONs uoqmum PPMPNNF ”oownm NMMv-Mv-v- Nva-v-Nv-v- Nv-mv-mv-v- MOIOP oo P Poounm Hleom Hleow H3 1 OP mP PP mo Po oo mo sP NP PP OO Po mo oo ON PP NP OP OP no so no 0P ON sP NP PO mo mo sP oo oo mo mo mo NO :0 OON PMN PON smP PNP PoP Po com PMN PON smP PNP PoP Po ooN Pmm PON smP NmP PNP POP Po ooN Pmm PON smP PNP POP Pm OmN PmN PON smP PNP PoP Po ooN no A7a-3 OOIOP oooOP oouom oouom mo PmN oouoP Po-oP m 2P NoioP mouoP oouoP oosoP oo-oP oo-om ao N No PNP oouoP oouoP oouom oouoP oouoP oouom mo m oo PoN PouoP NouoP oouoP oouoP PouoP PouoP oo N mo PoP oouoP oouoP oouom oo-oP oo-oP oouom No P mo omP PouoP ooaoP oouom ooloP oonoP oolom oo P mo Po oonoP oo-oP oouom ooloP oouoP oo-om No P mo PNP NoPP Pooum oonoP oonoP oouom oouoP oo-oP PouoP oo P Po PoP oouoP oo-oP PouoP oo-oP oonoP oouom oP m oP Po oouoP oouoP oouom oonoP oouoP oo-om oo P PN ooN NoaP Pooum oo-oP oonoP oo-om oouoP oo-oP oo-om oo P oP mNN oonoP oouoP oolom ooloP oo-oP oouom oo m PP oPN oo-oP PouoP oouoP oouoP oonoP Po-oP Po P NP ooN PouoP PouoP oonoP No-oP oo-oP PouoP oo P oP omP NouoP oouoP oonoP NouoP monoP oo-oP oo m NP PmN oo-oP oo-oP PouoP PouoP oo-oP oo-oP so P oo PNP oo-oP PouoP oo-oP No-oP PouoP PouoP oo m oP PoN oouoP PouoP oouom oouoP oouoP Po-oP oo N no PoP Po-oP mo-oP PoloP oouoP NouoP Po-oP No N oo omP oouoP PouoP PouoP oonoP oouoP mo-oP No P mo Po oouoP PonoP oo-oP ooioP oouoP oo-om mo N Po PNP . PoPP PoomPN oouoP oo-oP oo-om oouoP oo-oP oouom Po P NP PoP oo-oP PouoP PouoP PouoP oo-oP PouoP Po P mo Po Po-oP oouoP NonoP mouoP Po-oP oo-oP oo P oN ooN PooP PooNHm oouoP PouoP PonoP Po-oP oo-oP oouoP oo P mP PmN Po-oP ooioP oouoP No-oP oouoP oouoP oo m mP PoN oouoP oouoP PoloP oonoP NoloP oonoP oo P oP ooN oPuoP oouoP oo-oP mouoP oouoP No-oP oo P PP amP mouoP oouoP =o-oP oouoP oPuoP PouoP Po m mP PmN PouoP oouoP oouoP ooloP ooaoP oo-oP oo P oo NmP oouoP ooaoP PoioP oo-oP oouoP oouom Po N NP PoN oo-oP oouoP oonoP Po-oP oouoP oouoP :o m oo PNP PouoP oouoP NoioP NouoP PPuoP mouoP mo P mo amP oouoP NonoP oouoP oonoP oo-oP oouom no N Po PoP oo-oP Po-oP mouoP No-oP oouoP oo-om mo N Po PNP oouoP oouoP No-oP oouoP oocoP PouoP mo N oo Pm oouoP NouoP oouoP oouoP oouoP PouoP mo P Po PoP Pomo PoomPP oonoP oouoP oo-om oouoP oo-oP oouom =o m so Po NomP "ooqu Po-oP PP-oP oonoP oonoP oouoP oonoP oP P PP ooN PouoP zonoP Po-oP mouoP oo-oP mouoP mo P oN PmN ooaoP oonoP PouoP PoaoP ooioP oouoP Po P 2N ooN PouoP moloP No-oP NPuo oouoP oo-oP oo m =P PoN NoioP oouoP oo-oP oonoP PoloP ooloP mo m PP PmN NouoP mouoP PouoP oouoP oouoP souoP oo P oP nmP oo-oP oonoP PouoP oouoP NoloP ooloP mo N mP PoN oouoP ooioP oo-oP oo-oP oo-oP Po-oP mo m Po PNP oonoP monoP oo-om PouoP ooloP oonoP oo P oP omP oo-oP PonoP NoloP PouoP ooloP oo-oP so N mo PoP oouoP oo-oP oo-om PouoP oouoP PonoP oo m oo PNP ooaoP oo-oP oolom oo-oP souoP oo-oP :o N so Po NouoP PouoP oonoP Po-oP oo-oP oouoP so P oo PoP PoNo PoouPP NouoP oouoP NouoP oouoP oo-oP oouom oo P oo Po . NoNP Pooum PouoP ooioP oouoP mouoP mouoP mouoP oo P oP ooN oouoP oouoP PouoP oo-oP mo-oP PouoP mo P oo PmN ooloP NoloP NouoP PonoP oo-oP mouoP mo P mP ooN ocuoP PoioP NouoP oouoP Po-oP Po-oP Po m Po PoN NouoP mo-oP oouoP mo-oP =o-oP PouoP mo P NP PmN oo-oP PouoP oouoP oouoP moloP ooloP mo P oo omP PouoP PouoP mo-oP oPuoP NP-oP oonoP mo N PP PoN oonoP oouoP oouoP ooloP Po-oP ooaoP no m oo PNP NPuoP ooloP No-oP No-oP oP-oP NouoP me P Po omP oouoP ooloP oouom Po-oP NouoP PouoP mo N mo Po ooioP PouoP NouoP oouoP oo-oP oo-om oo N Po PNP Poom PooNPm oouoP oouoP oo-om moloP oouoP oouom oo P so PoP . oouoP oo-oP oouom oouoP oouoP PouoP oP P so Po oouoP NouoP Po-oP oonoP Po-oP oo-oP mo P oP ooN PoNP "oommo mo-oP oouoP PonoP ooloP mouoP PonoP mo P Po PMN NouoP ooioP oo-oP oouoP mouoP oouoP mo m oo PoN soloP oouoP PoloP PouoP oo-oP No-oP oo P oP ooN oo-oP mouoP ooioP oonoP mo-oP NouoP No P oo omP NouoP oouoP oouoP mo-oP PouoP mo-oP oP P oP mNN oouoP ooloP ooioP oouoP NouoP oonoP No m mo PNP NouoP oouoP PouoP mouoP oouoP ooioP Po m PP PoN oo-oP PouoP oo-om oouoP oouoP oouom mo N mo PoP Po-oP Po-oP oo-oP mouoP oouoP oonoP No P PP omP oouoP oonoP oolom oouoP oouoP NonoP :o N No Po NoPP PooNPm momm Pooqu szuom sz-om sz-om szuom sz-om sznom P: z xo oo szuom Pz¥-om sz-om szaom szuom sziom P: 2 mo oo A7a-4 OOIOP POIOP MOIOP sOIOP OOtOP POIOP monoP OOIoP OOIOP OOIOP OOIOP OOIOP OOIOP POIOP OPtOP oNIOP OOtOP OOIOP POIOP OOiOP OOIOP POIOP OOIOP POIOP OOIOP OOIOP POIOP OOIOP OOIOP OOIOP OOIOP oouoP OOIOP OOIOP conOP POIOP OOIOP OOIOP OOtOP NosOP moaOP OOIOP POIOP POIOP POIOP OOtOP POIOP PouoP NonOP POIOP mPIOP PPIOP NOIOP POIOP POIOP POIOP OOIOP NOIOP OOIOP NOIOP oouoP OOIOP POIOP oonOP NOIOP OOIOP OOIOP OOIOP oonoP NOIOP OOnOP POIOP Hzxuom sziom NouOP OOUOP OOIOP OOIOP NOIOP OOiOP POIoP NOIOP POIOP OOIOP POtOP OOIOP NOIOP POIOP monoP mPIoP OOIOP OOIOP OOIOP OOIOP POIOP monoP OOIOM oouoP POIOP oonom POIOP OOtom OO|OP OOIOm oooom OOoom OOIom POIOP oouom Oonom OOIOP oouOP oouOP OPIOP OOIOP OOIOP POIOP NOIOP OOIOP POIOP POIOP OOIOP NOIOP OO|OP POIOP sPIOP OOtOP OOIOP OOIOP OOIOP OOIOP POIOP sOIOP NOIOP monOP OOIOP OOIOP OOIOP NOIOP OOIOP OO|OP OOIOP OO|OP OOIOP OOIOP OOIOP hzxnom hleom NOIoP OOIOP POIOP POIOP sOIoP PoaOP PoaoP POIOP POIOP mouoP OOIOP OOIOP POIOP OOIOP OOIOP NPIOP monoP POIoP POIOP OOIOP OOIOP POIOP OOIOP POIOP OOIOP OOIOP POIOP POIOP OOIOP OOtOP OOIOP OOUOP OOIOP OOIOP OOIOP PonOP No-oP oo P Po oouom Po N oo PosoPPosmPP oouoP oo P mP oouoP mo m oP oonoP mo P oP oouoP Po N Po oouoP oo P so oonoP No N mo NoNoPPooNPP oo-oP oP P mP No-oP oP m oP oouoP oo P oP No-oP Po N Po oo-oP mo P oo oouoP No P Po PoNoPPosmPP oouoP oo P sP PPuoP mo m NP PouoP Po P oP oouoP oo N oP PouoP mo P oo oouoP oo P mo PoPoPuosum oouom sP P PP No-oP mo m PP oo-oP Po N sP oo-oP mo P mo oouoP so P NP momo PoomPN oo-om sP P oP oouoP mo m oP oouom so m oP NouoP mo P mP oouom No P oo oPiom mo N so oouom so N mo Nomo PoomPN oouom sP P PP PouoP mo m NP ooiom oo m NP NouoP so P so Pomo PosmPN st-om stuom H: 2 so OPP Po mNN PON smP PNP POP Po mNN PON smP PNP POP Po mNN PON smP PNP POP Pm omN PmN PON smP POP OoN PmN PON smP PNP POP Po ooN PmN PON smP oo OOIOP OOnOP OOIOP OOIOP NOIOP NouoP OOIOP NOIOP OOIOP oonoP OOIOP mOIOP OOIOP OOIOP OOIOP OOIoP OOIOP NOIOP POIOP OOIOP POIOP OOIOP OOIOP OOIOP OOIOP POIOP ooooP OO|OP OOIOP OOIOP OOIOP OOIOP OOIOP OOIoP Hzxuom ooaOP OOIOP oonoP POIOP POIOP NOIOP NOIOP PouoP oonom OOIOP NOIOP oOIOP POIOP MOIOP OOIOP OOIOP OOIOP OOIOP OOIOP OOIOP OOIOP OOIOP OOIOP POIOP OOIOP oonOP OOIOP OOIOP OOIOP OOIOP oouOP OOIOP OOUOP OOIOP Hleom POIOP POIOP oouom OOIOP NOIOP OOIOP oouom sOIOP OO|OP oouom sOIoP PPiOP NOIOP monoP oouom ooaom POIOP OOIOP mOIOP POIOP POIOP OOIom POIOP OOIOP OOIOM NOIOP OOIOm oouom oouom OOIom POIOP oouom POIOP oonom Hleom OOIOP OOIOP OOIOP POIOP OOIOP sOIOP POIOP OOIOP OOIOP OOIOP monoP NOIOP POIOP NOIOP OOIOP OOIOP OOIOP OOIOP OOIOP POIOP OODOP POIOP OOIOP OOIOP OOIOP POIOP OOUOP OOIOP POIOP OOIOP POIOP OOIOP OOIOP OOIOP Hzmlow OOIOP OOIOP OOIOP OOIOP OOIOP NOIOP OOIOP OOIOP OOIOP OOIOP OOIOP OOIQP OOIOP OOIOP OOIOP OOIOP OOIOP POIOP POIOP POIOP POIOP OOIOP OOIOP .OOIOP OOIOP OOIOP OOIOP OOIOP OOIOP OOIOP OOIOP OOIOP OOtOP OOIOP szlom Hzxoom H: x Pon uoqum PouoP so P oouom mo P oouom mo P Pomo Posum oouoP NP P Po-oP mo m oouoP mo m oo-om oo P mouoP mo P oouom mo P oo-om mo P NoNo "oomHo Po-oP NP P oouoP oP P oo-oP so m oouoP oo P oolom mo P oo-om so P oouom Po P PoNo PosmPo oouoP NP P NouoP oP P PoloP Po m ooioP oo P PouoP mo P No-oP mo P Po-oP mo P soPo Posmsm oouom NP P Po-oP oP P oolom Po N oouom Po P PouoP mo P oo-om mo P oouoP mo P moPo "ooum oo.om no P NoPo PoomHN oo-om so P OOIom mo P NosP "oomHm Po co mo mP mP PP oo mo mo mP IO PNP POP Po ooN PmN PON smP PNP POP Po OON mNN PON smP PNP POP Po omN mNN PON smP PNP PoP Po oaN mNN PON smP PNP POP Po Po Pm OoN oo A7a-5 No-oP oo-oP PonoP oonoP oouoP PouoP oo P mP mNN oo-oP NouoP oouom Po-oP No-oP oo-oP Po N PP oPN oo-oP oo-oP ooloP oo-oP PouoP oo-om oP P PP mNN PouoP oo-oP mo-oP Po-oP oouoP PouoP mo P oo smP PouoP Po-oP oo-oP oo-oP oouoP Po-oP oo m sP PoN oouoP Po-oP oo-om oo-oP oonoP oo-om so m oo NmP PouoP PouoP oo-om oouoP PouoP mo-oP mo P NP smP oo-oP oo-oP oo-om oouoP oo-oP mo-oP oo P Po oPP PonoP oouoP oo-oP oouoP PouoP oouoP oo P oo PNP oo-oP oo-oP oo-om oonoP oo-oP oo-om mP P so Po oo-oP oouoP oouom PouoP oo-oP Po-oP so P so PoP PossPPosmPP oouoP oonoP Po-oP oouoP PouoP oo-oP No P oo Po PoPNPPooNPP PouoP oo-oP oouoP PouoP oo-oP oouom oo N MP mNN ooioP oo-oP oo-oP oo-oP Po-oP oouom oP P sP PoN oo-oP oo-oP PouoP Po-oP oo-oP oouoP oo P oN mNN oo-oP oo-oP oo-om oo-oP oo-oP oo-om oP P oP smP oo-oP oo-oP oouoP ooloP Po-oP oo-oP oP N sP PoN oo-oP oo-oP oo-om oo-oP Po-oP oouoP mo m mo NmP mo-oP mo-oP oouoP oouoP No-oP oouoP oP P PP smP oo-oP ooioP oo-oN oouoP oo-oP oo-om oo P so PoP oo-oP oouoP oouoP oouoP mo-oP Po-oP so m PP NmP Po-oP oo-oP oo-oP oouoP oouoP PouoP No P mo Po oo-oP No-oP oo-oP PoloP Po-oP oouoP oo P oo PoP PoPmPPosuPP . moNPPuosNPP No-oP oouoP oouoP oo-oP oouoP ooioP oo P oP mNN oouoP Po-oP oo-oP oo-oP oosoP Po-oP Po P oo Po PonoP oo-oP oo-oP oo-oP Po-oP oouoP oo m sP PoN NoNPP osNPP mo-oP oo-oP No-oP oo-oP Po-oP oouoP so N mP smP oouoP oouoP PouoP No-oP Po-oP oo-oP so P so PNP No-oP oo-oP No-oP PouoP NouoP so-oP oo P oP mNN Po-oP oo-oP oo-oP oo-oP No-oP Po-oP Po P Po PoP Po-oP oN-oP oN-oP PouoP oo-oP NP-oP oP m PP PoN ooioP PouoP PonoP oonoP oouoP ooiom mo P oo Po souoP oo-oP oouoP Po-oP No-oP mo-oP oo P oP smP soNNPPosNPP oo-oP Po-oP PoloP PouoP oo-oP oouom oo m so NmP Po-oP so-oP Po-om oouoP oo-oP oo-om so P mo PoP oouoP NouoP oouoP PouoP PonoP PouoP me P sP mNN NouoP mouoP oonoP PouoP oonoP oouom co P mo Pm mP-oP Po-oP so-oP Po-oP mo-oP PPnoP oo m PP PoN PoNPPPosNPP mouoP mP-oP PPuoP oo-oP oo-oP No-oP no N oo smP oo-oP oo-oP oo-om Po-oP Po-oP oouoP oo P oo PNP mo-oP sonoP No-oP soloP mo-oP PouoP oP P NP mNN ooloP oo-oP ooloP oo-oP PouoP oonoP oo P mo PoP mo-oP mouoP ooloP Po-oP Po-oP ooioP oP m oP PoN oo-oP PouoP oo-oP oo-oP oo-oP oouom mo P oo Po PouoP No-oP oouoP Po-oP PouoP oouoP mo P oo smP NoNNPPosNPm so-oP Po-oP ooloP Po-oP Po-oP so-oP so P oo PNP oouoP Po-oP oouoP oouoP oo-oP oo-om oo P mo PoP ooioP Po-oP oouoP Po-oP oouoP oo-om oP P PP mNN Po-oP oouoP oo-oP Po-oP oouoP oouom No N oo Po oouoP PouoP PonoP PouoP PouoP PouoP mo m PP PON NoPPPPoomHm oo-oP PouoP No-oP moloP oo-oP PouoP mo P oP smP mo-oP so-oP Po-oP oo-oP so-oP oo-oP oo P Po PNP so-oP No-oP oo-oP oo-oP oo-oP Po-oP oP P NP mNN oo-oP oo-oP ooaom oo-oP souoP oo-oP oo P oo PoP so-oP so-oP mP-oP NoloP NogoP oonoP mo m mP PoN oonoP oo-oP oo-oP oouoP oo-oP oouoP so N Po Po oo-oP Po-oP oPuoP sonoP oo-oP No-oP mo P oo smP PoNNPPosNPP NouoP No-oP Po-oP oo-oP PoPoP Po-oP so P Po PNP oouoP NouoP oo-oP oouoP oouoP oo-om oo P so PoP NonoP NouoP monoP mouoP monoP oonoP oP P PP ooN oouoP oouoP NouoP oonoP PonoP oouoP No m =0 Pm oPuoP oouoP mo-oP No-oP souoP oP-oP oP P NP mNN PoPPPPosNPP NouoP Po-oP oo-oP oo-oP oouoP PouoP Po m Po PoN ooloP oouoP oo-oP oouoP NouoP Po-oP mo P Po smP No-oP oouoP oouoP mouoP mouoP sonoP oo P PP PmN NouoP ooloP oo-oP oo-oP Po-oP PouoP mo P mo PNP mo-oP PoloP moloP mouoP No-oP NouoP oP P NP PoN ooloP PouoP oo-om oo-oP oo-oP oo-om oo P mo PoP No-oP NouoP Po-oP oonoP mouoP NouoP mo P oP oPP oouoP oo-oP oouom oo-oP Po-oP oo-oP No P No Po NouoP oouoP oo-oP oo-oP oouoP NouoP oo N oo NmP NOPNPPoomHm POsOPPoomHm Hzxuom Hleom Hzmnom Hzxuom Hleom Payton H) x :0 oo Hzxnom Hzxuom Hzxnom Hzxtom Hzxnom Hzxuom H: x :0 oo A7a-6 OOIOP POIOP OOIOP OOIOP OOIOP OOIOM PO P PP OON CCIOP OOIOP OOiOm POIOP OOIOP OOIOP PO P mP OPN NOIOP OOIOP mOIOP mOIOP NOOOP NOIOP mo P PP OPP OOIOP POIOP OOIOP MOIOP OOIOP POIDP no N OP NmP OOIOP OOIOP OOOOm OOIOP OOIOP POIOP no N no OPP OOIOP OOIOP POIOP OOIOP OOIOP OOIom NO P 0O PO POmoPuoqum OOOOP POIOP NOIOP OOIOP OOIOP MOIOP oo P OP OON POIOP POIOP OOIOP NOIOP NOtoP OOIOP PO P mo PmN NOIOP POIOP OOIOP POIOP NOIOP OOIOP sO P mP oPN NoaOP sOIOP mOIOP NOIOP POIOP POIOP PO P NP smP POIOP sOIOP oouoP NOIOP nouoP mOIOP mo N NP NmP OOIOP OOIOP POIOP OOnOP POIOP POIOP mo N no OPP OOIOP NOIOP sOUOP oouOP POIOP OOUOP so N mo Pm . sOsmPuoquHm OOIOP PNIOP sonoP sOIOP mOIOP NOIOP PO P PP ooN POIOP NOIOP mOIOP OOIOP POIOP NOIOP mo P sP PMN POIOP OPIOP OOIOP .POIOP sOIOP OOIOP OP P PP OPN POIOP mOIOP OOIOP NonOP OOUOP mOIOP OP P OP smP POIOP POIOP OOIOm oOIOP sOIOP OOIOP so m mo NmP OOIOP OOIOP oouom OOIOP POIOP POIOP Po N no OPP OOIOP OOIOP oouom OOIOP OOIOP oouom mo m PO Po NOsmPuoome POIOP OOIOP OOIOP OOIOP OOIOP OOIOP mo P PP OON OOIOP OOIOP OOIOP OOIOP POIOP OOIOP so P mP OPN NOIOP rOIOP PonOP NOIOP .OOIOP POIOP mo P PP OPP sOIOP POIOP POIOP POUOP sOIOP oOIOP no N OP NMP OOIOP OOIOP OOIOm OOIOP OOIOP POIOP no N no OPP OOUOP OonoP OOIOm OOIOP OOIOP OOIOM Po N PO Pm PosmPuoqum oOIOP MOIOP souoP POIOP NOIOP OOoOP mo P PP OPN OOIOP sOIOP NOIOP POIOP oOIOP sOIOP OP P mo smP OOIOP NOIOP POIOP OOIOP POIOP OOtOm so m PO NmP OOIOP MOIOP OOIOP OOIOP mOIOP OOIOP no N co OPP OOiOP OOIOP POIOP OOIOP OouOP OOIOm 0O N mo Po POnmPuoqum oonOP OOIOP OOIOm oonOP OOIOP OOIOm mo P oo mNN POIOP OOIOP oooom OOIOP POIOP OOtOP PO N OP OPN OOIOP OOIOP OOtOP OOIOP POIOP OOIOP OP P 00 smP OOIOP OOIOP OOIom OOIOP OOIOP OOIOm mo m so NmP OOIOP OOIOP OOIOm OOIOP OOIOP OOoom so N so OPP OOUOP OOIOP OOIOm OOIOP OOIOP OOIOm NO P so Po NossPuoqum szlom Hleom Hleom Hzxnom sznom Hzalom H: x :u on A7a-7 NPOImN PsOImN moonmN mNOImN osOImN mNOImN sNOtmN mPoumN sOOImN POOImN moonmN PooumN PPOImN mOOImN ONOImN PPOImN mooumN PNOImN PmoumN sPOImN NNoamN oPoumN PPOImN PMOImN mOOImN OPOImN Pmoomw NmoumN wPOImN sPOImN oOOImN NOOImN szczm mPolmN mPoumN PmoumN Nmo-mN mmOImN omonmN NmOImN sPOImN oooumN PPOImN oooumN ooonmN NPOImN NOOImN PPOImN mmonmN oOOImN PmoumN OsonmN PmOImN PNoumN PMOImN PNOomN OPOImN mNonmN Nmoumm mmonmN sNoomN POOImN oPOImN oOOImN oooumN szlzm msonmN NNOImN mNoumN PmoumN oPoumN mOOImN mooumN mPoumN PPOImN ooonmN moonmN sOOtmN PPOImN oooumN mPoumN OsOImN OOOImN PPOImN mmoumN omoamN OMOImN POOImN MOOImN NoonmN PPOImN mooamm PPOImN NNoumN mOOImN NPOImN OPOImN mOOImN szlzm sNOImN mPPumN moonmm mmoumN omoumN OPOImN ONOImN PooumN mOOImN mOOImN wPOImN mOOImN OOOImN mOOImN ONOImN oPOImN PPOImN omOImN OOOImN NPOImN oPOImN smOtmN PmOImN sNOImN osonmN ssOlmN PPOImN mNOImN mOOImN mOOImN NOOImN PPOImN szl3m PmoumN PmoumN mo P msocmN PNoomN mo m sPNImN OsOImN mo N OOOImN mOOImN co P PONs Poouum NOPImN mmOImN mo P omoImN NmOImN no N POOImN mmoamN so N PsOImN OOOImN co P MOOImN mPonmN OOOImN NooumN OOOImN mOOImN NPOImN POOImN PPOImN NPOImN OOOImN POPs Poqmuu Po so so mP co MFFNPF moonmN no Nosm POAmHm mNOImN ONOImN mo P OsOImN NOOImN mo P omNImN POOImN co m omOImN mPOImN mo P Posm "oouum PmoumN PmPImN so P PmoumN POOImN mo P PPOImN sPOImN mo m Pomm "oomHm NsOIMN PsOImN mo P ONOImN ssOImN mo N NOOImN mmOImN mo N mPOImN NsOnmN OO P NOPm POAmHu NsOImN msOImN PO P mmoumN OOOImN mo N PsOImN NNonmN mo N NNOImN mMOImN PO P POPm Poqum POOImN OOOImN Po P NOOImN POOImN OP P PPOImN OOOImN mo N NOsN Poommm szlxm Pzaizm H: z oP mP PP mo sP NP PP mo ON NP OP OP mo OP ON sP NP sP oo oo sP PP OP OP com PmN PON smP omN PmN PON smP OON PmN PON smP NmP PNP omN PmN PON smP ooN PmN PON ooN PmN PON smP ooN Pmm PON smP OON mNN OPN OO PooumN PmoumN OOOlmN ONOImN sNOImN mmonmN PooomN sOOImN POOImN mwoumN smOImN PPOIOP ONOImN POOImN OOOImN OOOImN sPOImN OOOImN OOOImN mooumN moonmN PPOImN NPOImN mNOImN OOOImN sOOOmN PNOImN PPOIOP PPOImN OOOImN mPOImN mPonmN szlrm ooonmN ONOImN NPOImN OmOImN PPOImN omonmN mOOImN sOOImN PPOImN NNOImN NNOImN PNOIOP PNOImN mOOImN OOOImN oPormN moo-mN ooo-mN OOOImN OOOImN mooumm oOOImN OOOImN omoumN ONOImN oPOImN sPOImN oPOIOP NOOImN mOOImN MNonmN sOOImN Hlezm PoonmN oPoumN PPOImN oooumN oPOImN PsOImN sOOImN OPOImN OOOImN sNOImN ONOImN mPOIOP OPOImN NPOimN OOOImN moonmm mPO|mN OOOImN POOImN OOOImN OOOImN mOOImN NPOImN PPOImN MNOImN oOOImN POOImN NNOImN POOImN sOOImN oPonmN NOOImN szlzm NooumN sPonmN NPOImN mmoumN sNOIOP ooOImN MOOImN MOOImN OOOImN oPOImN PNOImN ONOImN OOOImN moosmN OOOImN OPoumN NOOImN OOOImN NOOImN POOImN OOOImN MOOImN PooumN PooumN mOOImN PNOImN mPoumN mPOImN ONOImN mNoamN ooonmN OOOImN szlzm NOsN PomN PONN NOPN POPN NOsP NomP POmP PONP NOOImN NOOImN PO P Poouum msOtmN PNOImN OP P noqmam PPOImN PPonmN 00 P PNOImN sPOnmN PO P msOIOP OMOIOP PO P PmOImN sNOImN mo P “oomHm oooumN ooonmN Po P soonmN ooo-mN oP P ooo-mN ooo-mN Po P mNocmN PPoumN oo P PmoumN oPoumN so m Poqmnm mPOImN ONOImN mo P mOOImN moonmN OO P PPOImN PPOImN co N OOOamN OOOImN oo P Poowmm PNOImN OPOImN NP P mooumN POOUmN OP P OOOImN OOOImN PO P Poownh OOOImN oooamN NP P OOOImN PooumN OO P OOOImN OOOImN OO m NOOImN OOOImN co m Poquum sOOImN MOOImN NP P moonmN ooonmN OP P OOOImN NOOImN mo m NNOImN wOOImN OO P mPOImN POOImN PO m Poome MOOImN POOImN mo P OOOImN mooumN mo m mPOnmN POOImN OO m OPOImN OOOImN mo P sOOImN moonmN PO m "oomHh szlzm szlzm H: 2 OP mP mP mP mP oP mN mP OP 0P MP mP NP PP PO mP oP Po oP OP 0P OP ON PP sP NP OP NN 0P mP OP OP :0 smP OON ooN mNN OPN smP OON mNN oPN smP NmP OON mNN OPN smP OoN mNN smP omN mNN OPN smP ooN mNN oPN smP NmP ooN PMN OPN smP NmP OO A7a-8 moonmm mOOImN NoonmN OOOImN POOImN OOOImN PPOImN mOOImN sOOumN POOImN NPOImN NOOImN PPOImN PPOImN PNOImN NNOImN PNOImN ONOImN mOOImN mPOImN oNOImN NOOImN OOOImN POOImN MOOImN mPOImN sOOImN mooomN NmPumN mPOImN oPOImN oOOImN NOOImN szl3m sOOImN POOImN OOOimN POOImN NOOImN OOOImN mPoumN mNOImN POOImN mOOtmN POOImN OOOImN OPOImN NNOImN NPOImN PNOImN ONOImN OPOImN NPOImN NOOImN OPoomN OOOImN mooumN OOOImN mooumN wOOImN mOOImN mOOImN PPOImN sooumN sOOImN mOOImN NOOImN Hzxnzm mooumN POOImN OOOImN OOOImN POOimN POOImN NPOImN oOOImN POOImN oooumN mOOImN OOOImN oOOImN ONOImN oPOImN mOOImN ONOImN ONOImN mOOimN MOOImN mPoumN OPoumN NOOImN NOOImN OOOImN oOOImN mOOImN oOOtmN moonmN POOImN oOOImN POOImN PooomN szszm NOOImN mOOImN POOImN POOImN PooumN NOOImN ONOImN sNOImN sOOImN sOOImN POOImN sOOImN oooumN MOOImN POOImN msOImN NmOImN mNOImN OOOImN NOOtmN POOImN mOOtmN sOOImN OOOImN OOOImN sooumN sOOImN OPOImN PmOImN PPOImN OmOImN OOOImN mNOImN sznzm OOOImN MOOImN OO P POOImN POOImN mo m OOOImN POOImN co m NOsP Pooumm NOOImN MOOImN PO P NOOImN NOOImN co m OOOImN OOOImN co m POsP "Osman OPOImN sOOImN OO P omoumm mmoomN Po m OOOImN NOOImN PO N NOmP Poamum OOOImN OOOImN PO P mOOImN mOOImN MO m NNOImN ONOImN mo N sOOImN NOOImN co P mOOImN NPOImN so N NONP uoqmmh POOImN PNOImN mo P PsOImN PMOImN mo P mmOImN PNOimN OO N PONP ”OAMHu NPOImN PPOImN oo P oooumN OOOImN OP P wOOImN PPOImN co m oooumN oOOImN NO P OPOImN oOOImN co m NOPP noawum NOOImN OOOImN oo P POOImN OOOImN oo P OOOImN oooumN co m mOOImN mOOImN co P mOOImN NOOImN mo m POPP "oqumm mPOImN oO NOOImN oO mOOImN OO mOOImN oo POOImN co PNOImN so POmo "Osman PNoumN moonmm OOOImN omoumm mPoumN NPOImN enr-wwnw-P szlzm szlzm H: z mP sP OP NP OP 0P NP sN PP mP OP NP ON mP PP oo oo :0 OON PmN oPN ooN PmN OPN OON PmN PON ooN PmN PON smP NmP OON Pmm PON OoN mNN OPN smP NmP ooN mNN OPN smP NmP OON PmN PON smP NmP PNP OO mmOImN ONoumN moonmN mNOImN oOOImN PNOnmN PmoumN NOOImN NNOImN mOOImN mPOImN sOOImN oooumN PNOImN OPOImN PNOImN oOOImN smouoP mPPumN PsoumN oPoumN PPoumN PPoumN mNOImN PPNImN moPumN NOOImN sPOImN sOOImN PsOImN mNOImN mmoumN mmoumN szlzm sPOImN NmoumN mOOImN mPocmN OOOImN sOOImN NPOImN POOImN mNoamN OPOImN PNOImN POOImN PNOImN sNOImN NOOImN PsOImN POOimN osOIOP MPOImN PoOImN mOOImN PNOImN MNOImN PmoumN mmPomN mmOImN wPoumN mPOImN sOOImN oPOImN OMOImN PNOImN omoomm HszIm POOImN mOOImN NOOImN NPoumN POOImN POOImN moonmN oooumm PNOImN POOImN NPOImN mOOImN NNOImN sOOImN NOOImN OPOtmN PPOImN smPumN oOOImN POPImN PNPImN sPOImN POOImN soonmN NOPImN PoOImN PsOImN PPOImN NOOImN ONOImN sPOImN oOOImN PmOImN Hlezm mOPumN PNOImN mmonmN mNonmN mNOImN ONOImN PPOImN PPOImN PmocmN mPonmN NsOImN OPOImN PPOImN ssOImN sPOImN MNOImN msoumN ssOIOP NOOImN PPNImN OPNImN OPOImN mOOImN ssOImN ooonmN OsPImN msleN ONOImN PmoumN NMOImN mmoumm NPOImN omPImN szlzm PMOImN oPouMN OP P oPoumN NOOImN mo P mPoumN NOOImN so m OPOImN mOOImN co P PONO Poqmnm PNOImN moonmN OO P mOOImN mOOImN mo P mPOImN ooonmN PO m Posm "oqunm OPOImN ONOImN mo P NPOImN OPOImN mo P omOImN NNOImN mo m momm Poqmnm oPOImN OmonmN OO P smOImN PsOImN mo P mNOlmN osOImN mo m OMOImN OPOImN OO P Nomm "quuu sNOImN PmOImN PO P NmOImN PPOImN so N PmOImN OMOImN so P PsOIOP OPO0mN sNPImN omOImN ONOImN sPOImN PNOImN POOtmN PoleN mPPImN OPOImN ONOImN Noms PONm uoqmnh PsPImN OO P PNPImN PsOImN omPImN PMOImN NPOImN mmoumm mPOImN NNOImN ommnmm sOOImN POOImN so P mo m mo P NO P mo m Poounh OO P 0O P so m OO P so P Po m Poms ”oqmum woOImN mPPImN mo P PsOImN PMOImN mo m ooleN smOImN so m PsOImN mPOImN OO P szlzm szl3m H3 z NONs "oqmmm PP ON sP OP OP OO PO OP PO oo OP NP NP mo mP PP mo ON PP mP NP mo PO NN PP mP NP oo OP ON mP PP mo OmN PmN PON smP OON PmN PON ooN PmN PON OON PmN PON smP ooN PON smP omN PMN PON smP NmP PNP OON PmN PON smP NmP PNP ooN PmN PON smP OO A7a-9 NOOImN OOOImN NOOImN oPOImN I mPoumN oNOImN NPOImN sOOImN OOOImN OOOImN oooumN OPOImN mooamm OOOImN PoonmN PNOImN NNOImN msOImN POOImN omoumm omOImN omoumN POOImN MMOImN mOOImN nsOImN oooumN PPouOP OoOImN OOOImN OmoumN mPoumN qulzm NooumN OOOImN moonmN NOOImN sNOImN sPOImN OOOImN NOOImN NOOImN OOOImN POOImN PNOImN MOOtmN NOOImN NOOImN NNOImN NmonmN OmOImN POOImN PPOImN OPOImN OPOImN POOImN PPoumN OPOImN PmoumN mmoumN NPOIOP moonmN OPoomN NNOImN soonmN szuxm PoocmN NOOImN mOOImN NOOImN sPOImN mOOImN POOImN mOOImN OOOImN OOOImN mOOImN OOOImN MOOImN POOImN POOImN NmOImN sOOImN omoumN OOOImN PNOImN OPonmN OPOImN OOOImN mPoumN MNOImN oNOImN mPOImN OOOIOP mOPcmN moonmN PPoumN sOOImN Hlerm sOOImN sOOImN POOImN PNOImN OOOImN POOImN OOOImN NPOImN POOImN PooumN OOOImN OPOImN OOOImN OOOImN oooumN OmPImN msOImN msoumN OPOImN PmOImN ONOImN mPOImN mOOImN omOImN mooumm POOImN sPOImN omOIOP mPOImN mmonmm PsOImN OOOImN Hlezm NOOImN POOImN OP P PooumN mOOImN OO m NPOImN mOOImN mo P PPOImN POOImN mo P PONNPuoqum omonmN OPOImN OP P oooomN OOOImN OP P mOOImN OOOImN PO m NOOImN PooumN mo P NOPNPuoqum PooumN MOOImN OP P sOOImN OOOImN co m OPOImN sOOImN mo P POPNPuoqum PPOImN MOOImN PP P POOImN sOOImN OO P OOOImN sOOIMN OP N PPOImN OOOImN OP P MONPPnoqmuh NPPImN OOOImN sP P mPonmN MNOImN OO P OmOImN OOOImN OP m OOOImN sOOImN OO P PONPPPOJmHm PPOImN ONOImN NP P mNOImN POOImN OP P MOOImN OOOImN OP m NOOImN PPOImN mo P NOPPPPquHm sMOImN osOImN NP P POOtmN OOOImN OP P NPOImN mOOImN mo m OPOImN NOOImN mo P POPPPuoqum sMOIOP ONOIOP MP P POOImN omOImN OO P NPOImN OOOImN OP P PNOImN OMOImN mo P POsOPuoquHm POOImN POOImN OO P NONOPuoqum hzxuzm Hleum a) : oP PP OP PO PP NP PO PO PP sP NP oP ON sP PP PP OP PP OP MP NP OP oo oP NP mP oo oP PP NP OP mP :0 mNN PON smP PNP OON mNN PON smP mNN PON smP OON mNN PON smP OON mNN PON smP OON mNN PON smP OON mNN PON smP OON PmN PON OPP mNN OO PPOImN PPOImN OOOImN OOOImN POOImN OOOImN ONOImN mmoumw PPOImN mOOImN PooumN OOOImN OOOImN POOImN OOOImN oooumN POOImN MOOImN OOOImN OOOImN OOOImN OOOImN soonmN msoumN OOOImN OmonmN osOImN sNOtmN NOOImN PPOImN PNOtmN mPOImN szl3m mOOImN ooonmN NOOImN NooumN OOOImN PPOImN smoumN oPoumN POPImN sOOImN sOOImN POOImN sOOImN mOOImN OOOImN POOImN OOOImN OOOImN OOOImN NOOImN OOOImN ooonmN sOOImN sPoumN NPOImN sNOImN NsOImN sPOImN sOOImN sPOImN POOImN ooonmN sz03m OOOImN POOImN NooumN NOOImN POOImN POOImN ONOImN PNOImN OPOImN POOImN POOImN NOOImN OOOImN POOImN OOOImN mOOImN NOOImN POOImN OOOImN mooumN OOOImN NOOImN MOOImN mPOImN OPOImN NNOImN mPOImN mPOImN MOOImN POOImN PoonmN mOOImN sznzm Pmoomm mPOImN mOOImN OOOImN sPOImN mPOImN sOOImN POOImN mmOtmN mooumN OPOImN oooumN NPoumN POOImN OOOImN NOOImN POOImN sPOImN OOOImN NOOImN NOOImN OOOImN sooumN NOOtmN ONOImN wooomN NOOImN PPonmN NOOtmN sPOImN POOImN MOOImN szIIm sNOImN ONOImN mo m mOOImN sOOImN mo P NONOPuoqum PPOImN OPOImN MP P NOOImN mOOImN OP P OOOImN OPonmN OP m NNoumN oOOImN no P PONOPuoquHm OPoumN MOOImN mP P POOImN OOOImN oo P mPOImN mooumN mo m sPOImN PPOImN PO P mOOImN PooamN OO N POPOPqumHm OOOImN oooomN sP P OPoumN PPOImN mo m sooumN OOOImN PO N momo "oqmwu OOOImN OOOImN (P P OOOImN ooonmN mo m POOImN OOOImN so m OOOomN mOOImN mo P NOMO «oquum oooumN oooumN sP P PoocmN mooumN mo m ooonmN PooumN so m Pomo PosuPP NOOImN MOOImN NP P mPOImN OOOImN mo m MOOImN mooumN mo N NONO "quHh sOOImN PooumN NP P MOOImN NooumN OP P mOOImN mOOImN mo N PPOImN moonmN OO P PONO “oomum OOOImN OOOImN NP P oooumN oooumN OP P PooumN NoonmN co N mOOImN MOOImN 0O P sOPm ”quum szlmm szltm H: 2 OP OP OP mP OP OP PP sP NP OP OP PP PP sP mP oP mP PP NP PP sP OP OP mP oo oP sP PP PON smP omN mNN PON smP OON mNN PON smP PNP OON PmN PON OON Pmm OPN smP OON PmN OPN OoN PMN OPN ooN mNN OPN smP ooN mNN OPN smP OO A7a—10 OOOImN OOOImN oooumN mooumN mOOImN mmoumm OmonmN NOOImN OOOImN szsxm OOOImN POOImN OOOtmN OOOImN OOOImN mPOImN mmOImN ooonmm oOOImN Hzxczm OOOImN OOOImN OOOImN sOOImN POOImN OOOImN mNOImN mOOImN mOOImN szsrm OOOImN OOOImN OOOImN POOImN MOOImN mmonmm OoOImN ooonmN msOImN szlxm POsP OOOImN OOOImN NP P OooumN OOOImN OP P OOOImN OOOImN 0O P Poommm OOOImN OOOImN PO P OOOImN oOOImN PO P PmonmN msOImN mo P PomoPuoomHm mmonmN OPOImN oo P sOOImN POOtmN PO P oPOImN OPonmN so P sOsmPuoqum Hzxnrm szl3m H3 2 mP oo oo OON mNN smP OON OPN OPP ooN Pmm OPN OO oPOImN sOOImN PsOImN msoumN PPOImN mOOImN OOOtmN sOOImN oOOImN oooumN moonmm mooumN OOOImN ooosmN moonmN ooonmN sOOImN sPOImN mPOImN OOOImN POOImN OOOImN OOOImN OOOImN NOOImN mOOImN oPonmN PsOImN PPonmN oooumN sznzm PsOimN oooumN PmoumN PmoumN oooumN NOOImN sPOImN mNOImN ooonmN PmoumN OOOImN mooumN OOOImN NOOImN NOOtmN oooomN NOOtmN sOOImN POOImN OOOImN POOImN OOOImN OOOImN NoonmN POOImN NOOImN ONOImN PNOImN PNonmN sOOImN sznzm PPOImN NmoumN NPOImN moonmN OPOImN NOOUmN POOImN OPOImN mOOImN NPPImN OPOImN sOOImN POOtmN POOImN sNOImN POOImN OOOImN OOOImN sooumN OOOImN POOImN OOOImN POOImN mooumN OOOImN NPOImN sPOImN ONOImN oPOImN POOImN Hzx03m OPOImN NmOImN PPOImN mmoumN ONOImN POOImN POOImN PNOImN ONOOOP sPOImN ONOImN soo-mN moo-mN PooumN PPOImN mPOImN OPOImN OPOImN moonmN POOImN PooomN OOOImN OOOImN mooumN NOOImN ONOImN NNOImN msOimN mPoumN NOOImN szlzm OOOImN OOOImN PO P NP sosmPPquHm owOImN mOOImN PO P PP mPOImN mooumN mo P sP ONOtmN sOOImN OP P PP PPOImN PNOImN OP P OP NosmPPquHm mOOImN MOOImN mo P PP NOOImN POOImN so P mP OOOImN PMOImN mo P PP PosmPnoome oPOIOP ONOIOP MP P sP NNOImN oPOImN mo P PP NNOImN mMOImN OP P mo POmmPuoomHm POOImN mOOOmN OP P PP POOImN NOOImN mo P oo PoonmN oooumN PO N OP NOssPuoqum OPOImN sNOImN mo P ON OOOImN oNOImN OO P mP mOOImN PPoumN PO N PP oooomN wooumN mo P OO PossPuoqum oPoumN OPOImN NP P sP POOImN POOImN co N mP oooumN PooomN OP P sP MOOImN OOOImN OP P OP POPmPuoqum OOOImN NOOImN OO P OP OOOImN POOImN co m sP OOOImN OOOImN so N MP sONNPuoqmwu sNocmN oPOImN OP P PP OPOImN POOImN mo P sP NmoumN PooumN oo m PP wOOImN wooumN mo N oo NONNPuoqum NOOImN POOImN NP P ON PONNPuoomHm szuzm szlzm H: x =0 smP OON PMN OPN smP ooN OPN OPP OON OPN smP omN mNN OPN ooN mNN OPN smP ooN mNN PON smP mNN PON smP OON mNN PON smP OmN OO Appendix 7b. Egg and larval count data. Each number is the number of insects found in 2 linear feet of row (= 1 square foot). FIELD/CH ------------------ EGGs / SQ.FT. ------------------------ DEGREE DAYS: 621 DATE: JUNE 5 7281 25 0 0 9 2 2 5 1 3 1 4 4 3 3 1 2 1303 11 19 11 24 9 16 16 7 12 27 9 14 8 11 21 15 1403 4 5 3 9 6 5 2 3 8 2 3 2 4 7 5 3 7203 16 25 16 5 13 12 8 9 9 18 11 8 4 10 11 4 7302 28 0 0 0 0 3 1 0 0 0 2 0 1 0 0 0 7401 24 O 5 1 0 2 3 1 0 0 4 2 0 0 0 0 7402 32 0 0 1 2 0 0 0 0 0 1 0 1 2 1 2 DEGREE DAYS: 725 DATE: JUNE 12 1303 7 8 5 10 16 7 10 17 5 4 2 0 0 0 2 3 1403 6 9 1 3 3 0 1 1 0 2 1 1 3 2 5 6 2103 8 4 11 11 3 22 10 13 7 6 11 5 2 5 2 '2 2401 7 8 17 17 21 19 0 0 1 2 3 0 0 2 0 6 3403 7 0 5 13 30 12 9 28 20 24 5 16 11 2 6 2 3403 6 7 2 3 2 2 7 5 3 1 1 6 7 3 1 7 6101 10 0 0 0 2 1 1 1 1 2 2 0 2 4 0 2 7203 20 0 0 0 0 0 0 3 5 6 4 4 5 1 10 14 10203 12 0 0 2 0 1 0 0 0 0 0 3 0 0 0 2 91101 30 0 0 0 1 1 0 3 0 1 2 91102 27 1 0 0 1 1 0 1 4 2 3 91103 33 0 0 0 0 1 0 0 1 0 0 91104 26 1 0 0 1 0 1 1 0 0 3 92101 33 0 2 3 2 2 3 4 2 2 5 94101 15 2 3 5 2 1 2 0 3 4 2 5 _ 3 3 1 2 95101 22 1 0 0 1 2 1 3 0 1 1 3 2 4 0 0 95102 22 2 1 1 2 0 3 1 4 0 0 1 2 3 4 1 95103 25 2 1 0 2 2 1 3 2 0 0 1 4 1 3 1 DEGREE DAYS: 804 DATE: JUNE 16 1403 13 1 3 1 2 2 0 0 2 1 2 2 4 1 4 1 2103 14 0 3 1 9 4 16 4 10 9 21 21 18 29 5 25 2401 14 15 4 0 70 1 0 4 0 2 4 1 0 6 1 7 3402 10 2 1 3 6 5 4 5 5 6 2 4 2 7 4 2 6101 15 2 1 1 2 7 8 4 5 6 5 2 4 6 3 1 7203 23 1 11 15 5 8 5 8 7 1 8 6 3 2 0 2 10203 15 0 0 0 0 0 0 0 0 3 0 0 0 0 0 1 91101 27 0 8 5 1 4 4 2 2 2 3 91102 29 1 1 2 0 0 1 0 1 1 6 91103 30 1' 0 0 0 0 0 0 1 0 0 91104 28 0 1 2 2 0 2 0 1 2 0 92101 27 3 1 6 4 5 5 7 5 4 1 92102 27 0 1 1 0 0 0 3 7 2 1 A7b-1 A7b-2 JUNE 16 (CONT.) DATE: 804 DEGREE DAYS: 10204 1410.1 00003 110013 11an02 00000 011120 2021'] DATE: JUNE 18 858 DEGREE DAYS: 2080-1112000 0300 30410132000 0001 6116311340000 0101.. 00861040000 0000 124021511000 11001 121011m0000200201040 00200490000600025101 01300343000301022010 11520172000100130000 002002M0000201220010 20n20264000100020000 40..”12140000100210110 20500330000310110000 20301241000100030001 41410031000200220001 61123891448777779007 11111111222222222332 33313133261134121123 00000000000000000000 34144122223111111111 11223670222111224555 1111999999999 DATE: JUNE 20 924 DEGREE DAYS: 001212100003000120120 102203000004201110000 200021000006010040010 001014000101000030220 012005000003000010001 001n01011002201032010 103403500002002020010 012402000004311130100 040301300006000010AUOO 110974200005300120100 3628”]011004000020020 334782m10203001102000 409120600007100210016 215616700005310011000 1 609014800006000000000 1 735334947077070770555 111112112322323223333 333131332561234121123 000000000000000000000 341441222221111111111 112236702221111224555 11119999999999 A7b-3 DATE: JUNE 23 1015 DEGREE DAYS: 2004000004101020000 1001320002010120000 1000016003100110000 0100”]0004001010000 0150501005001330000 001050M004200510000 0010001001111200000 0130015204000220100 1202111003100010000 00]”036101200220000 1445001203111000000 4110310302201430000 9300417005310510000 ] 9441104002201110000 2184408053303198005 2111122223333323443 3331313311234121123 0000000000000000000 3414412231111111111 1122367021111224555 119999999999 A7b-4 ---------------------- LARVAE / SQ.FT. FIELD/CH DATE: JUNE 5 621 DEGREE DAYS: 3200212 400011] 1 8400310 2400312 .I. 3100110 9300610 8400520. 5200200 ..lu 5100450 7000301.- 9300400 6200102 220052] ..ll 410001] 2 8200210 7201 1303 1403 7203 7302 740] 7402 DATE: JUNE 12 725 DEGREE DAYS: 432700203 4310 1|- qll ‘ 2 52871020m 6230 407000414 7134 .ll 2]] 2 907100202 0423 1 1| 1| 9712 254500302130522710] 1'14 1|. 42v10001021|202v404251 2309]50230064046299665 2 1.] 1|?v5441l4025041955116645 1 00%550300201231612] I 10m881301|1|20310521|2 1 2 31m384002000151l7434 ..l 645080105000204430] .l ..I 30%7015080200133105 1 5008224042001115213 52 .lu 3331331331234121123 0000000000000000000 3414441221111111111 1122336701111224555 19999999999 DATE: JUNE 16 804 DEGREE DAYS: 230020] 40117 1 0353000 2 31206 0320200 12.-I39 ] 0751-202 31228 1 1140700 1.0017 ..l..l. 1| 62041000202350100m 460220001014410005 22001010000431210w 00w400211132301008 2]”310002008200015 32w400000110400008 558330021026501411 .l 670740200113200117 1' ..l 331213312341211233 000000000000000000 414412211111111113 12236701111224.5551 19999999999 A7b-5 DATE: JUNE 18 858 DEGREE DAYS: 9024H208010 0121 50785100020 2104 20M68403101 0211 1013 72500400000 1 11 1 1132 ”0970702020100311511 1 12390503010102512101 1 2 90792300031000002210 21305400030100140001 1 1 00951400011012230101 1 4 1 80827403021100200004 1 1 20911300010010201002 1 1 2 33313133261134121123 00000000000000000000 34144122223111111111 11223670222111224555 1111999999999 DATE: JUNE 20 924 DEGREE DAYS: 600521001001002110111 Cv0_ln/_00000012001020222 1 309213002020100102102 1 9 6 19 7 0 14 20 21 15 «(v200000011300209‘v1u 1 8011000001001013 65001010201010310 3 0 12 1 238000050001005011 5 O 15 40m045000020001100002 41n050001011000113020 2031fi8020011111210211 1 605094012002001013032 ] 400455030001100111001 1 11 916253011010402002123 1 333131332561234121123 000000000000000000000 341441222221111111111 112236702221111224555 11119999999999 A7b-6 DATE: JUNE 23 1015 DEGREE DAYS: 3332020001000000000 3342220000000100110 4022220000002000002 1021700001200110110 0023630000000100100 1041320001001000010 1092M00000000010100 0020w10001100000000 1130510000000000000 0002910000000000002 2010510101000000000 2001010000000000000 3010010000000000100 0012110001100011000 3331313311234121123 0000000000000000000 3414412231111111111 1122367021111224555 119999999999 2 Appendix 7-C. Data for between + within-field log 5 x log i regression anaIysis (Table 9). A. 8401:; Degree Days Sweepnet Square Foot > 48°Fa 1111:161chb 169 52 m #Fieldsc 129;? 122.8 87 39 - .9927 -1 1269 91 24 -2 6076 -1.7868 101 54 -1.9663 -1.3426 110 8 -2.1918 -1.4357 121 6* .8463 -.4266 49 -1.4553 -1.1000 132 10* .8582 -.3513 15 - .8103 - .7111 154 42 .3127 -.0912 58 - .2066 - .5139 170 3* .5006 .0370 3* -1.4699 - .6425 201 34 .3599 -.0110 49 - .3783 - .5068 210 23 .2477 -.3465 12 - .8767 - .8163 225 27 .7844 -.4688 28 .-1.0590 - .7897 231 28 .2580 .0468 29 .0610 - .4028 290 57 .0947 -.0688 45 - .6174 - .6963 A7c-1 B. Degree Days > 48°Fa 621 725 804 858 924 1015 7 19 18 20 21 21 d #Fie1ds A7c-2 Larvae & Eggs - Square Foot Counts Larva1 109 s2 109 i 1.3227 .4021 1.8370 .6900 1.5353 .4783 1.4442 .4926 1.3169 .3910 .4341 -.1599 a Accumu1ated since 1 Apri1. b c 6-10 Samp1es of up to 10 1inear feet/fie1d. d 6 x 10-25 sweep samp1es/fie1d. 10-15 Square feet/samp1e. See Sect. 2.2.1. Eggs 109 52 lgg_g_ 1.5777 6981 1.3733 .5099 1.5548 .4819 .7371 1133 .8290 .1616 .6787 .0445 See Sect. 3.3.1. See Sect. 3.3.1. * Exc1uded in Tab1e 9 as being two sma11 a samp1e. Appendix 8--Pubescent Wheat Study Weather Data. Temperature records for three 1ocations near the study area with accumu1ated degree days (base 48°F) for each site and mean accumu1ated degree days for 811 three sites. R24 = rainfa11 during the previous 24 hours (inches/hours). Apri1. 1975 -New Car1is1e-- --South Bend--- ----Site Center ----- Mean Day Max Min d-Day Max Min d-Day Max Min d-Day R24 d-Day 1 33 24 -- 46 30 -- -- 2 29 18 -- 38 25 -- -- 3 25 18 -- 33 25 -- -- 4 27 16 -- 35 23 -- -- 5 29 16 -- 38 23 -- -- 6 30 18 -- 40 24 -- -- 7 39 18 -- 49 23 -- -- 8 50 ’18 -- . 52 25 1 1 9 43 34 -- 47 34 1 1 10 49 30 -- 50 30 1 1 11 41 26 -- 43 28 1 1 12 48 24 -- 50 23 1 ' 1 13 52 22 1 54 23 2 2 14 46 33 1 48 33 2 2 15 52 33 2 53 32 3 3 16 55 34 4 57 34 5 5 17 68 34 11 69 35 13 Insta11ed 12 18 67 62 27 69 57 28 .5/5 28 19 56 42 30 57 43 31 -- 31 20 47 40 30 49 36 31 Insta11ed -- 31 21 47 29 30 48 30 31 .1/.5 31 22 64 42 36 65 42 38 .1/1.5 37 23 70 50 48 68 53 50 .1/2 49 24 58 40 51 58 42 54 . -- 53 25 54 38 53 59 43 58 Ca1ibrated -- 56 26 61 33 57 62 35 63 62 32(4+est60) —- 61 27 49 41 57 51 42 64 48 40 64 .9/10 62 28 49 42 57 50 44 65 48 41 64 .4/2 62 29 76 40 69 75 41 76 76 40 76 -- 74 30 68 52 81 71 55 91 68 52 88 .1/.5 87 A8-1 A8-2 May, 1975 -New Car1is1e-- --South Bend--- ----Site Center ------ Mean Day Max Min d-Day Max Min d-Day Max Min d-Day R24 d-Day 1 60 44 86 61 45 97 60 42 91 -- 91 2 70 42 95 71 42 107 73 40 101 -- 101 3 62 51 103 65 53 118 61 49 108 .1/.5 110 4 56 44 106 57 43 121 55 41 110 -- 112 5 70 40 115 70 41 130 70 38 118 -- 121 6 65 52 127 68 51 142 65 50 128 .15/.5 132 7 66 45 137 68 47 152 66 44 136 -- 142 8 66 50 150 68 54 165 68 51 148 -- 154 9 75 52 166 75 53 181 76 50 163 -- 170 10 est180 76 47 195 73 48 175 -- 183 11 est194 77 46 209 78 42 188 -- 197 12 58 40 197 64 43 216 58 37 191 .1/.5 201 13 70 36 205 70 39 225 69 35 199 -- 210 14 72 41 215 72 47 237 72 ‘ 42 209 -- 220 15 60 44 220 59 45 242 57 40 212 -- 225 16 64 40 226 69 41 251 63 36 217 -- 231 17 . 78 40 239 75 40 262 80 40 231 -- 244 18 88 44 258 85 50 282 87 45 249 -- -263 19 92 57 284 91 62 310 90 60 276 (out for 290 20 94 63 314 91 67 341 94 62 306 repairs) 320 21 82 62 338 87 63 368 85 60 330 -- 345 22 72 60 356 73 61 387 70 56 345 -- 363 23 87 56 380 85 57 410 88 54 368 386 24 90 66 410 88 67 440 90 64 397 .05/2.0 416 25 88 63 438 86 65 468 87 62 423 .85/1.5 443 26 77 64 460 77 65 491 79 62 445 -- 465 27 70 50 472 72 54 506 71 46 456 -- 478 28 81 45 487 77 47 520 79 42 470 -- 492 29 79 57 507 80 58 541 78 55 488 -- 512 30 80 60 529 78 62 563 80 56 508 .4/3.0 533 31 68 52 541 68 53 575 68 46 517 -- 544 A8-3 R24 .05/.5 .35/3.5 .4/2.0 .6/4.0 .5/2.0 .4/1.5 June, 1975 -New Car1is1e-— --South Bend--- ----Site Center ----- Day Max Min d-Day Max Min d-Day Max Min d-Day 1 71 47 552 72 48 587 72 45 528 2 7O 46 562 65 49 596 68 44 537 3 74 49 576 74 51 610 76 48 551 4 83 52 596 82 55 630 82 52 570 5 80 62 619 79 61 652 81 60 592 6 70 56 634 71 58 668 73 59 610 7 60 45 639 60 49 674 58 48 615 8 68 44 648 68 45 683 68 42 623 9 72 45 659 73 48 695 73 42 634 10 80 52 677 80 55 715 82 50 652 11 84 63 703 80 64 739 83 64 678 12 76 56 721 76 59 759 75 55 695 13 88 57 745 83 58 781 86 56 718 14 82 58 767 81 60 803 84 56 740 15 71 60 785 71 59 820 71 57 756 16 78 52 802 77 56 838 76 52 772 17 80 69 828 80 69 864 78 66 796 18 91 64 858 88 66 893 88 62 823 19 90 72 891 87 73 925 92 70 856 20 93 70 925 90 71 957 96 68 890 21 92 66 956 88 60 983 92 64 920 22 94 68 989 91 71 1016 92 68 952 23 86 68 1018 84 69 1044 88 67 982 24 84 67 1046 84 68 1072 86 66 1010 25 79 68 1072 77 68 1096 79 66 1034 26 85 65 1099 82 66 1122 89 62 1062 27 90 64 1128 86 64 1149 91 62 1090 28 92 66 1159 89 67 1179 93 62 1120 29 90 66 1189 87 65 1207 92 63 1150 30 91 62 1217 88 63 1235 91 60 1178 Mean d-Day 556 565 579 599 621 637 643 651 663 681 707 725 1748 770 787 804 829 858 891 924 953 986 1015 1043 1067 1094 1122 1153 1182 1210 A8-4 Ni1es precipitation, 1975, in inches. Source: NOAA. 82 80:1. 090: 490.9 1 t .01 2 .77 . t 3 .21 .14 .29 4 .05 5 .01 .12 6 .46 7 8 t 9 t 10 11 .01 .50 12 .18 ' t 13 .36 14 t t .62 15 .05 .88 16 t 17 .42 18 1.40 ' 19 t 20 .14 t 21 .21 .30 22 .06 .05 t 23 .38 .08 24 .05 .22 .01 25 t .42 .04 26 27 .87 28 .49 29 .05 .12 30 .48 .35 Appendix 9. Examp1es of Fi1ter Operation. Appendix 9a. App1ication of fi1tering technique to 1974 Gu11 Lake data (Sawyer 1976). CROP HEIGHT --LARVAL DENSITY-- DD 0(Y) E(Y) 0(Y) E(Y) F(Y) FIELD: 911. 313 3.43 0.00 378 4.49 3.45 0.00 0.00 0.00 440 6.54 3.99 0.00 0.00 0.00 488 8.07 5.28 .06 0.00 .06 567 10.51 11.62 .15 .71 .43 663 13.50 16.71 .91 1.49 1 20 760 16.81 20.53 .74 1.89 1.31 835 16.18 22.68 .60 1.20 .90 926 18.58 24.18 .09 .48 .28 985 21.65 24.40 .03 .09 .06 1043 23.50 24.27 0.00 0.00 0.00 1157 29.65 23.93 .06 0.00 .06 1249 30.12 27.46 0.00 0.00 0.00 FIELD: 916. 313 8.10 0.00 378 3 98 8.11 0.00 0.00 0.00 440 5 71 6.08 .03 0.00 .03 488 7 32 8.96 .06 .39 .22 567 9 21 12.97 1.40 1.28 1.34 663 12 56 16.74 3.29 3.98 3.63 760 14 25 20.09 4.56 5.64 5.10 835 10 83 21.28 3.47 4.87 4.17 926 15 47 21.06 1.61 2.71 2.16 985 17 40 21.47 1.03 1.21 1.12 1043 18 94 22.51 1.06 .47 76 1157 24 13 26.00 11 0.00 06 1249 26 06 25.73 06 0.00 03 FIELD: 917. 313 9.00 0.00 378 4.13 9.01 0.00 0.00 0.00 440 5.71 6.62 0.00 0.00 0.00 488 7.99 6.21 .15 0.00 .15 567 10.12 12.03 1.72 1.04 1.38 663 12.09 16.72 8.26 4.10 6.18 760 15.08 19.85 8.02 9.61 8.81 835 14.49 21.56 3.16 8.36 5.76 926 17.17 22.90 2.64 3.55 3.10 985 19.76 23.12 1.29 1.71 1.50 1043 20.63 24.41 .71 .63 .67 1157 27.32 27.27 .06 0.00 .03 1249 28.03 27.65 0.00 0.00 0.00 Appendix 9b. App1ication of fi1tering technique to 1976 PWS data (Sawyer, unpub1ished data). CROP HEIGHT --LARVAL DENSITY-- DD 0(Y) E(Y) 0(Y) E(Y) F(Y) FIELD: 136. 282 4.00 0.00 330 6.00 4.02 0.00 0.00 0.00 354 5.00 5.01 .40 0.00 .40 380 8.00 6.68 .30 1.00 .65 399 10.00 8.58 .50 1.18 .84 499 16.00. 15.28 9.10 7.13 8.11 570 22.00 19.62 17.20 21.12 19.16 640 22.00 24.41 43.10 33.13 38 12 756 25.00 28 83 2.10 47.71 24 90 815 24.00 29 64 1.10 20.73 10 92 879 28.00 29.78 2.30 7.36 4 83 931 34.00 31.20 .30 2.89 1 60 1025 33.00 36.34 0.00 .31 16 FIELD: 236. 264 3.00 0.00 282 3.00 3.03 0.00 0.00 0.00 330 4.00 3.05 0.00 0.00 0.00 338 4.00 3.55 0.00 0.00 0.00 354 4.00 3.80 0.00 0.00 0.00 374 6.00 3.92 .10 0.00 .10 399 7.00 6.58 0.00 .36 .18 499 12 00 12.99 4.70 2.62 3.66 570 22 00 16.64 6.50 10.58 8.54 640 20.00 23.02 10.90 15.45 13.17 756 22.00 27.28 .80 17.69 9.24 815 18.00 27.52 .60 8.03 4.31 879 29.00 25.97 2.20 3.13 2.67 931 32.00 29.89 .40 1.61 1.00 1025 30.00 34.83 0.00 .16 .08 FIELD: 246. 330 5.00 0.00 354 4.00 5.00 0.00 0.00 0.00 374 4 00 4.51 0.00 0.00 0.00 399 4 00 4.27 .10 0.00 .10 499 12.00 10.49 6.00 2.23 4.12 570 16.00 15.47 12.40 12.29 12.34 640 18 00 19.65 20.50 24.49 22.49 756 15 00 24.88 5.20 33.53 19 36 815 14.00 23.03 4.20 18.50 11 35 879 24.00 21.97 5.30 9.07 7 19 931 28.00 25.61 2.80 4.77 3 78 1025 28.00 31.08 0.00 1.07 53 CROP HEIGHT 0(Y) DD FIELD: 337. 282 330 354 374 387 487 570 640 756 815 879 931 1025 FIELD: 346. 338 354 380 399 487 570 640 815 879 931 1025 FIELD: 524. 262 282 338 354 374 387 487 570 640 756 815 879 931 1025 .00 .00 .00 .00 .00 .OO .00 .00 .00 .00 .00 .00 E(Y) 12. 18. 21. 22. 25. 28. 31. mmpmm --LARVAL DENSITY-- 0(Y) E(Y) .60 2.10 3 4.10 4. 4.60 7 11.60 8 8.50 68 24.501 13. 26.901 21. 22.70 90. 3.80 47. .50 16. 0.00‘ 5. .10 0.00 .30 O. .80 3.40 1 1.00 13 8.00 22. 8.20 26. .50 17. 0.00 5. 0.00 1 0.00 0.00 0.00 O 0.00 0 0.00 0. .10 0. 0.00 . .50 2 3.80 5 12.10 9. 3.30 18. 8.90 10 .30 7 .10 2 .10 A9b-2 00 .78 .37 73 99 23 59 .13 F(Y) OOO ..a-a dWOOO-D-J .10 .22 .97 .02 .43 .02 .10 .48 .57 .59 .38 .30 .79 .39 :36 .60 .79 .82 .06 .OO .00 .OO .10 .39 .51 .69 .80 .69 .20 .18 CROP HEIGHT 0(Y) FIELD: 612. D0 380 399 487 570 640 756 815 879 931 1025 FIELD: 262 282 330 354 374 387 487 570 640 756 815 879 931 1025 FIELD: 330 354 380 399 499 1024. 1135. 6 6 6. 8. 3 2 E(Y) 12. 21 26 d—J 4550on .01 93 . .20 25. 25. .08 .98 30. 56 75 36 ~-LARVAL DENSITY-- 0(Y) O I 0000 00000 anew—4 A9b-3 E(Y) _.a_.| dwtONNLO-DOOOOO mebNO F(Y) —J—.l mmO-JON COCO c—l 10. .OO .00 .00 .00 .40 .67 .35 .18 .76 .07 .03 REFERENCES CITED Anderson, R. C. and J. D. Paschke. 1968. The biology and ecology of Anaphes flavipes (Hymenoptera: Mymaridae), an exotic egg parasite of the cereal leaf beetle. Annals Entomol. Soc. Am. 61: 1-5. Anderson, R. C. and J. D. Paschke. 1969. Additional observations on the biology of Anaphes flavipes (Hymenoptera: Eulophidae), with specia1 reference to the effects of temperature and super para- sitism on development. Anals. Entomol. Soc. Am. 62: 1316-1321. Anderson, R. C. and J. D. Paschke. 1970. Factors affecting the post re1ease dispersal of Anaphes flavipes (Hymenoptera: Eulophidae), with notes on its post re1ease development, efficiency, and emergence. Anna1s. Entomol. Soc. Am. 63: 820-828. Anscombe, F. J. 1949. The statistical analysis of insect counts based on the negative binomial distribution. Biometrics 5: 165-73. Baskervi11e, G. L., and P. Emin. 1969. Rapid estimation of heat accumu- lation from maximum and minimum temperatures. Ecology 50: 514-17. Balachowsky, A. S. 1963. The family Chrysomelidae. Entomologie Applique a' 1'Agricu1ture, Tome 1, Vol. 2. Masson et cie. Paris. 1391 pps. (Ref. in Wilson & Shade, 1964). Beall, G. 1941. The transformation of data from entomo1ogica1 field experiments so that the analysis of variance becomes applicable. Biometrika 32: 243-62. Bliss, c. 1. 1967. Statistics jflBiology. V01. 1. McGraw-Hill, New York. 558 p. . Bliss, C. I. and R. A. Fisher. 1953. Fitting the negative binomial distribution to biological data and note on the efficient fitting of the negative binomia1. Biometrics 9: 176-200. Bliss, C. I. and A. R. G. Owen. 1958. Negative binomial distribution with a common k. Biometrika 45: 37-58. Boswell, M. T. and G. P. Patil. 1970. "Chance Mechanisms generating the negative binomial distribution." From Random Counts jg_Scien- tific Work, Vol. 1, G. P. Patil. ed. Pennsylvania State University Press. pp. 3-22. Box, G. E. P. 1953. Non-normality and tests on variances. Biometrika, 40: 318. Box, G. E. P. and P. N. Tidwell. 1962. Transformation of the indepen- dent variables. Technometrics 4: 531-550. 8-1 8-2 Casagrande, R. A. 1976. Behavior and survival of the adult cereal leaf beetle, Oulema melanopus (L.) Ph.D. Thesis, Michigan State Univer- sity, East Lansing. 174 pps. Castro, T. R., R. F. Ruppel, and M. S. Gomu1inski. 1965. Natural his— tory of the cereal leaf beetle in Michigan. Michigan State Uni- versity, Agric. Exp. Stn., Quart. Bull. 47: 623-53. Caswell, H., F. Reed, S. N. Stephenson, and P. A. Werner. 1973. Photo- synthetic pathways and selective herbivory: A hypothesis. Am. Nat. 107: 465-80. Cohen, 0., 8. Foster, W. Helm, J. Tuccy. 1976. SPSS Regression Refer- ence. CDC publication number 76075800. Also, Northwestern Univer- sity Computing Center Document No. 287. Delong, D. M. 1932. Some problems encountered in the estimation of insect populations by the sweeping method. Ann. Entomol. Soc. Am. 25: 13-17. Draper, N. R., and H. Smith. 1966. Applied Regression Analysis. John Wiley and Sons, Inc. New York. 407 pps. Dysart, R. J., H. L. Maltby, and M. H. Brunson. l973. Larval parasites of Oulema melanopus in Europe and their colonization in the United States. Entom0phaga 18: 133-67. Fulton, W. C. 1975. Monitoring cereal leaf beetle larval populations. M.S. Thesis, Michigan State University, East Lansing. 128 pps. Fulton, W. C. and D. L. Haynes. 1975. Computer mapping in pest manage- ment. Environ. Entomol. 4: 357-60. Gage, S. H. 1972. The Cereal Leaf Beetle, Oulema melanopus (L.), and its interaction with two primary hosts: winter wheat and spring oats. M.S. Thesis, Michigan State University, East Lansing. 105 pps. Gage, S. H. and D. L. Haynes. 197 . Ecological investigations on the Cereal Leaf Beetle, Oulema melanopus (L.), and the principal larval parasite, Tetrastichus julis (Walker). Ph.D. Thesis, Michigan State University, East Lansing. 145 pps. Gage, S. H. and D. L. Haynes. 1975. Emergence under natural and manipu- lated conditions of Tetrastichus julis, an introduced larval para- site of the cereal leaf beetle, with reference to regional popula- tion management. Environ. Entomol. 4: 425-34. Gutierrez, A. P., W. H. Denton, R. Shade, H. Maltby, T. Burger, and G. Moorehead. 1974. J. Anim. Ecol. 43 (3): 627-640. Harcourt, D. G. 1963. Population dynamics of Leptinotarsa decimlineata (Say) in eastern Ontario. Can. Ent. 95: 813-20. 8-3 Hayman, B. I. and A. D. Lowe. 1961. The transformation of counts of the cabbage aphid (Brevicorne brassicae (L.)). N.Z.J. Sci. 4: 271-8. Haynes, D. L. 1973. Population management of the cereal leaf beetle. In Insects: Studies in Population Management. Geier, P. W., L. R. Clark, D. J. Anderson and H. A. Mix, eds. Ecol. Soc. Aust., Memiors 1, Canberra. pp. 232-40. Haynes, D. L., R. K. Brandenburg, and P. D. Fisher. 1973. Environmental monitoring network for pest management systems. Environ. Entomol. 2: 889-99. Haynes, D. L., S. H. Gage and W. C. Fulton. 1974. Management of the cereal leaf beetle pest ecosystem. Quaest. Entomol. 10: 165-76. Haynes, D. L. and S. H. Gage. 1972. Joint Conference Entomol. Soc. Can. and U.S.A., Montreal. (Ref. in Lee et al. 1976). Headley, J. C. 1975. The economics of pest management. Chapter 3 in Introduction to Insect Pest Management, edited by R. L. Metcalf and W. H. Luckman. John Wiley and Sons, New York. 587 pps. Helgesen, R. G. and D. L. Haynes. 1972. Population dynamics of the cereal leaf beetle, Oulema melanopus (L.): a model for age speci- fic mortality. Can. Entomol. 104: 797-814. Hildebrand, H. A. 1976. Nonlinear filters for estimating an insect population density. Ph.D. Thesis. University of Illinois, Urbana. 118 pps. Hodson, W. E. H. 1929. The bionomics of Lema melanopa (L.) in Great Briton. Bull. Entomol. Res. 20. 5-14. Holt, 0. A., R. J. Bula, G. E. Miles, M. M. Schreiber, R. M. Peart. 1975. Environmental physiology, modeling and simulation of alfalfa growth: 1.) Conceptual development of SIMED. Research Bul. gg7, Agric. Exp. Sta., Purdue University, W. Lafayette, Indiana. PPS- Hoxie, R. P. and Wellso, S. G. 1974. Cereal leaf beetle instars and sex, defined by larval head capsule widths. Ann. Entomol. Soc. Am. 67(2): 183-186. Hughes, R. D. 1955. The influence of the prevailing weather on the numbers of Meromyga variegata Meigen (Diptera, Chloropidae) caught with a sweepnet. J. Anim. Ecol. 24: 324-42. (Ref. in Ruesink & Haynes 1973). Jackman, J. A. 1976. A quantitative description of oat growth with cereal leaf beetle populations. Ph.D. Thesis, Michigan State University, East Lansing. 172 pps. B-4 Kalman, Ru E. 1960. A new approach to linear filtering and prediction problems. Trans ASME. J. Basic Eng. 820, 34-45. Kendall, M. G. 1948. The Advanced Theory of Statistics. 3 Vols. Griffin, London. Kendall, P. G. 1948. On some modes of population growth leading to R. A. Fisher's logarithmic series distribution. Biometr1ka 35,6. (Ref. in Anscombe 1949). Kershaw, K. A. 1964. anntitative and Dynamic Ecology. American Elsevier, New York, 183 pps. Kleczkowski, A. 1955. The statistical analysis of plant virus assays: a transformation to include lesion numbers with small means. J. Gen. Microbiol. 13: 91-98. Kogan, M., W. G. Ruesink, and K. McDowell. 1974. Spatial and temporal distribution patterns of the bean leaf beetle, Cerotoma trifurcata (Forster), on soybeans in Illinois. Environ. Entomol. 3(4): 607- 617. Kuehl, R. O. and R. E. Fye. 1972. An analysis of the sampling distri- butions of cotton insects in Arizona. J. Econ. Entomol. 65(3): 955-60. Larsen, H. S. 1969. Introduction to Probability Theory and Statisti- ' cal Inference. John Wiley & Sons. New York. 387 pps. Lee, K. Y., R. 0. Barr, S. H. Gage, and A. N. Kharkar. 1976. Formulation of a mathematical model for insect pest Ecosystems--the cerea1 leaf beetle problem. J. Theor. Biol. 59, 33-76. Masayama, M. 1957. The use of sample range in estimating the standard deviation or the variance of any population. Sankhya 18: 159-62. Miller, C. D. G., M. K. Mukerji and J. C. Guppy. 1972. Notes on the spatial pattern of Hypera postica (Coleoptera: Curculionidae) on alfalfa. Can. Ent. 104: 1995-99. Morris, R. F. 1955. The development of sampling techniques for forest insect defoliators, with particular reference to the Spruce Bud- worm. Can. J. Zoology. 33(4): 225-293. Morris, R. F. 1960. Sampling insect populations. Ann. Rev. Entomol. 5: 243-64. Nie, N. H., C. H. Hull, J. G. Jenkins, K. Steinbrenner, and D. Bent. 1975. SPSS: Statistical Package for the Social Sciences. McGraw-Hill, New York. 675 p. Patil, G. P. and W. M. Stiteler. 1974. Concepts of aggregation and their quantification: a critical review with some new results and applications. Res. Popul. Ecol. 15: 238-54. B-S Pielou, E. C. 1969. Introduction to mathematical ecology. John Wiley and Sons, Inc. New York. 286 pps. Poole, R. W. 1974. An Introduction togguantitative Ecology. McGraw- Hill, New York. 532 pps. Pruess, K. P. and C. R. Weaver. 1959. Sampling studies of the Clover Root Borer. Ohio Agric. Exp. Sta. Bull. 827. Rabb, R. L. 1970. Introduction to the conference. 13 Concepts of Pest Management. Rabb, R. L. and F. E. Guthrie, eds. North Carolina State University, Raleigh, N.C. pp. 1-5. Ruesink, W. G. 1970. Sweepnet Determination of Population Densities of the Cereal Leaf Beetle, Oulema melanopgs (L.) N.S. Thesis. Michigan State University, East Lansing. 49 pps. ‘ Ruesink, W. G. 1972. The integration of adult survival and dispersal into a mathematical model for the abundance of the Cereal Leaf Beetle, Oulema melanopus (L.). Ph.D. Thesis, Michigan State University, East Lansing, 80 pps. ' Ruesink, W. G., and D. L. Haynes. 1973. Sweepnet sampling of the cereal leaf beetle, Oulema melanopus. Environ. Entomol. 2: 161-72. Sage, A. P. and J. L. Melsa. 1971. Estimation Theory with applications to communications and control. McGraw-Hill, New York. 529 pps. Sakawa, Y. 1971. Optimal filtering for linear distributed-parameter systems. IFAC Symp. on Control of Distributed Parameter Systems, Paper 13-7 pps. Sawyer, A. J. 1976. The effect on the Cereal Leaf Beetle of planting oats with a companion crop. M.S. Thesis, Michigan State University, East Lansing, 145 pps. Schillinger, J. A., and R. L. Gallun. 1968. Leaf pubescence of wheat as a deterrent to the cereal leaf beetle, Oulema melanopus. Ann. Entomol. Soc. Am. 61, 900-903. Searle, S. E. 1966. Matrix Algebra for the Biological Sciences (in- cluding Applications in Statistics). John Wiley and Sons, Inc., New York, 296 pps. Searle, S. R. 1971. Linear Models. John Wiley and Sons, Inc., New York. 532 pps. Shade, R. E. and M. C. Wilson. 1967. Leaf vein spacing as a factor affecting larval feeding behavior of the cereal leaf beetle, Oulema melanopus. Ann. Entomol. Soc. Am. 60, 493-496. Shade, R. E., H. L. Hansen, and M. 0. Wilson. 1970. A partial life table of the cereal leaf beetle, Oulema melanopus, in Northern Indiana. Am. Entomol. Soc. Am. 63: 52-9. B-6 Shepard, M., G. R. Carver and S. G. Turnipseed. 1974. A comparison of three sampling methods for arthropods in soybeans. Environ. Entomol. 3(2): 227-232. Simpson, G. G., A. Roe, and R. C. Lewontin. 1960. Quantitative Zoology. Harcourt, Brace; New York. 440 pps. Sokal, R. R., and F. J. Rohlf. 1969. Biometry. W. H. Freeman and Co. San Francisco. 776 pps. Southwood, T. R. E. 1966. Ecological methods. Chapman and Hall, Ltd. London. 391 pps. Steel, R. G. D. and J. H. Torrie. 1960. Principles and Procedures of Statistics. McGraw-Hill Book Company, Inc., New York. 481 pps. Stehr, F. W. 1970. Establishment in the United States of Tetrastichus julis, a larval parasite of the cereal leaf beetle. J. Econ. Entomol. 63: 1968-9. Taylor, L. R. 1961a. Aggregation, variance and the mean. Nature, Lond. 189: 732-35. Taylor, L. R. 1961b. Abstract of a presented paper. Biometrics 17: 498-99. Taylor, L. R. 1965. A natural law for the spatial disposition of insects. Proc. XIIth Int. Congr. Ent. (London, 1964) pp. 396-7. Taylor, L. R. 1970. Aggregation and the transformation of counts of Aphis fabae Scop. on beans. J. Appl. Biol. 65: 181-89. Thomas, G. B. Calculus and Analytic Geometry. 3rd edition. Addison- Wesley, Mass. 1010 pps. Tummala, R. L., W. G. Ruesink, and D. L. Haynes. 1975. A discrete component approach to the management of the cereal leaf beetle ecosystem. Environ. Entomol. 4: 175-186. Venturi, F. 1942. La "Lema melanopa L." Redia 22: 11-86. (Ref. in Shade and Wilson 1967). Vickeman, G. P. and K. 0. Sunderland. 1975. Anthropods in cerea1 crops - Nocturnal activity, vertical distribution, and aphid predation. J. Appl. Ecol. 12(3): 755-766. Wayman, P. A. 1959. A least squares solution for a linear relation between two observed quantities. Nature, Land. 184: 77-8. Webster, J. A., S. H. Gage, and D. H. Smith, Jr. 1973. Suppression of the cereal leaf beetle with resistant wheat. Environ. Entomol. 2: 1089-91. B-7 Wellso, S. G. l973a. Cereal leaf beetle: Larval feeding, orientation, development and survival on four small-grain cultivars in the laboratory. Ann. Entomol. Soc. Am. 66: 1201-1208. Wellso, S. G., R. V. Connin, and R. P. Hoxie. 1973b. Oviposition and orientation of the cereal leaf beetle. Annals Entomol. Soc. Am, 66, 78-83. Wellso, S. G. 1975. Cereal Leaf Beetle: relationships between feeding, oviposition, mating and age. Annals Entomol. Soc. Am. 68: 663-668. Wellso, S. G. 1976. Cereal Leaf Beetle: feeding and oviposition on winter wheat and spring oats. Environ. Entomol. 5: 487-491. Wiener, N. 1949. The extrapolation, interpolation and smoothing of stationary time series with engineering application. John Wiley and Sons. New York. Wiggins, S. C. 1956. The effect of seasonal temperatures on maturity of oats planted at different dates. Agronomy Journal 48: 21-25. Wilson, M. C. and R. E. Shade. 1964a. Adult feeding, egg deposition, and survival of larvae of the cereal leaf beetle on seedling grains. Purdue Univ. Agric. Exp. Stn. Res. Prog. Rep. 97. Wilson, M. C. and R. E. Shade. 1964b. The influence of various Gramineae on weight grains of post-diapause adults of the cereal leaf beetle, Oulema melanopa. Ann. Ent. Soc. Am. 57: 659-61. Wilson, M. C. and R. E. Shade. 1966. Survival and development of larvae of the cereal leaf beetle, Oulema melanopps (Coleoptera: Chrysomelidae), on various species of Graminae. Ann. Entomol. Soc. Am. 59: 170-3. Yun, Y. M. 1967. Effects of some physical and biological factors on the reproduction, development, survival and behavior of the cereal leaf beetle, Oulema melanopus (L.), under laboratory conditions. Unpublished Ph.D. Thesis, Michigan State University, East Lansing, Michigan. Diss. Abst., 28: 5068-8. -.— ‘h__-—