UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Design of a sampling system for the larch casebearer Coleophora laricella Hbn Moody, Benjamin H. 1977

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

Item Metadata

Download

Media
831-UBC_1977_A1 M66.pdf [ 15.89MB ]
Metadata
JSON: 831-1.0094303.json
JSON-LD: 831-1.0094303-ld.json
RDF/XML (Pretty): 831-1.0094303-rdf.xml
RDF/JSON: 831-1.0094303-rdf.json
Turtle: 831-1.0094303-turtle.txt
N-Triples: 831-1.0094303-rdf-ntriples.txt
Original Record: 831-1.0094303-source.json
Full Text
831-1.0094303-fulltext.txt
Citation
831-1.0094303.ris

Full Text

DESIGN OF A SAMPLING SYSTEM FOR THE LARCH CASEBEARER (Coleophora l a r i c e l l a Hbn.) by Benjamin H. Moody B. Sc. F., University of New Brunswick, 1968 . M. Sc. F., University of New Brunswick, 1971 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY THE FACULTY OF FORESTRY We accept this thesis as conforming to the required standard i n THE UNIVERSITY OF BRITISH COLUMBIA January, 1977 Benjamin H. Moody, 1977 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of Brit ish Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the Head of my Department or by his representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of f oYgyf< \^  The University of Brit ish Columbia 2075 Wesbrook Place Vancouver, Canada V6T 1W5 ABSTRACT The problems that arise in developing a sampling design for the various l i f e stages of the larch casebearer, Coleophora l a r i c e l l a (Hbn), are treated in relation to the changes in population density and distribution throughout the l i f e cycle of the insect. The basic principles of population sampling are followed in respect to stratification of the sampling universe into its logical components. Replications and successive samplings are satisfied. The peculiarities of branch structure, with short shoots, long shoots and fascicles of needles provide c r i t e r i a for the ultimate sampling units. Sampling was conducted with relation to: position of trees in the stand (interior, edge, or 'open grown1 trees); different crown levels; different branches at the same level based on exposure to sky-light; different 6-inch (15cm) segments of a branch throughout its length and different stages in the insect 1s 1i fe eyele. The main purpose of this investigation was to test the hypothesis that shifts in population concentrations influence the accuracy of sampling at fixed points in the crown of a tree. If this be so, refinement of techniques may become possible. Of the theoretical distributions tested the negative binomial gave the best f i t to the data for a l l l i f e stages except the egg stage which approached the normal distribution. Analyses of two sample units, the needle fascicle and the 15~cm branch section respectively, revealed similar statistical distributions. The variance of the number of insects per needle fascicle calculated for each tree sampled was highly related to the mean. Therefore, approximate normality of the data was achieved by application of Taylor's power transformation, but with the modification of adding a "C" constant to the variable before raising it to the power of P,in the equation: Y. = (X. + C)'">, where Y. = transformed observation, i I I X. = original observation, C = 1 and p = (1 - 1/2b) where b^  is a constant derived by the method of least squares. The analyses of variance showed that tree-to-tree, vertical and horizontal position of the samples in the tree crown were the most consistent variables influencing the distribution of eggs, larvae and pupae, while exposure of the tree in the stand had l i t t l e effect. The distributions of eggs with relation to the quantity of needles, type of shoots, and condition of needles (oviposit ion sites) on the branch were examined. The determinant factors in insect distributions were also recogn i zed. A practical three-stage sampling design was developed by considering variations between trees, and vertical and horizontal strata within the tree crown. The f i r s t stage is concerned with selection of the tree(s), the second stage would be the crown level within a tree, and the third stage the branches within each crown level. The variable to be estimated should be the number of insects per needle fascicle or short shoots (spurs) in winter. Such a sampling design would provide estimates of population trend and mortality within a generation. - i i -TABLE OF CONTENTS D Page ABSTRACT . . i TABLE OF CONTENTS i i i LIST OF TABLES v i i i LIST OF FIGURES x i LIST OF APPENDI CES x i i i ACKNOWLEDGMENTS xv INTRODUCTION 1 Earl ier Work 4 Deficiencies of Former Methods 7 THE INSECT 10 Taxonomy 10 World Distribution 12 Significance as a Defoliator 14 Biological Characteristics of the Insect 16 Morphological Forms and Stages 17 Life History 19 Reproduction, Growth S Morphogenesis 19 THE HOST TREE 22 Host species and Susceptibility 22 The Host Tree in British Columbia 24 The Host Tree as a Sampling Unit 27 Morphological Characteristics of Larch 28 Deciduous Characteristics of Larch 30 Vegetative Cycle 30 - i i i -THE STUDY AREA, MATERIALS AND METHODS 31 Need for Choosing Sampling Area 31 Sampling Procedures 34 The Sampling Universe 35 Selection of the Sampling Unit 35 Timing of Sampling 36 Field Procedures 40 Assessing the Tree 45 Sampling for Morphological Characteristics of Branches 45 Distribution of Fol iage and Shoots 45 Defoliation Measurements 46 Rating Defoliation 4 6 Insect Counts and Accessory Information 47 Overwintering Larval Samples . .' ^ Checking Insect Counts 49 Rearing Methods 50 Experimental Observations 50 ANALYSIS OF DATA 5 2 Fitting the Distributions 53 Transformations 54 Taylor's Power Law . . . . . . . 55 Analysis of Variance or '' F '—Test 57 Regression and Correlation 58 The Number of Samples 6 0 63 Allocation of Optimum Sampling Effort Data Preparation and Analysis Procedures . 63 The Basic Unit of Sampling 63 - iv -RESULTS AND DISCUSSION 65 Frequency Distributions 65 The Variance - Mean Ratio 69 'k' as an Index of Aggregation 69 Discussion on Distribution 74 Transformation of Data 76 FACTORS AFFECTING THE DISTRIBUTION OF CASEBEARER 79 Source of Variation in Population Estimate. 79 Inter-Tree Variation 79 Intra-Tree Variation 80 1. Crown Level Variation 80 2. Exposed and Shaded Branches . 81 3. Main or Side Branches 87 k. Horizontal Crown Position 87 DISTRIBUTION OF THE POPULATION BY LIFE STAGES 89 The Egg Stage . 89 Distribution of Eggs 89 Current Growth vs. Adventitious Foliage vs.Old Growth Foliage.89 Egg Placement on the Needle Surface 91 Vertical Distribution of Eggs 93 Horizontal Distribution of Eggs .93 Relationship Between Degrees of Defoliation and Number of Egg per Fascicle 95 Egg Distribution in Relation to Adult Behaviour 98 - v -The Larval Stage 100 Larvae Prior to Winter Dormancy 100 Spring or Postwinter Larvae 101 Larval Variation Between Trees 101 Vertical Variation in the Tree Crown 101 Exposure 104 Main and Side Branches 106 Horizontal Crown Variation 106 Comparison of Branch and Branch Tip Samples 107 D i scuss ion 108 Reasons Underlying the Distribution of Larvae 109 Larval Behaviour and Distribution Effects HO The Pupal Stage 1 1 4 Vertical Distribution Horizontal Crown Position H4 Exposure H6 Factors Affecting Pupal Distribution H6 Larch Fol iage  Distribution of Larch Fol iage in the Crown Fascicles 1. Tree-to-Tree Variation 2. Crown Level Variation 1 2 2 3. Exposure I 2 4 k. Main and Side Branches I 2 4 5. Horizontal Crown Position ]Zk - v i -Other Factors of Foliage Affecting Insect Distribution. • • . 127 Number of Needles per Fascicle 128 Needle Length and Fascicle Weight. 129 Intra-branch Variation in Fascicle Size 130 Inter- and Intra-tree Variation 130 Discussion on Needle Fascicles 131 I nsect Mortal i ty 132 Insect Mortality and Effects on Distribution 132 Spatial Difference in Mortality 135 Vertical Distribution of Mortality 135 Horizontal Distribution of Mortality in the Crown- . . . . . . .135 Seasonal Fluctuation in Casebearer Population 136 Recommendations for Sampling- 139 The Egg Stage 139 The Larval Stage 140 Pupa 141 A General Sampling Design 142 Sample S ize 144 SUMMARY AND CONCLUSIONS 148 LITERATURE CITED 153 APPENDICES 159 - v i i -LIST OF TABLES Table Page 1 Frequency distributions of numbers of insects (larch casebearer) per fascicle for the l i f e stages sampled . . . 66 2 Effect of development of larch casebearer during a single generation on estimate of the parameter k_and b^  for its immature stages, Thrums, B.C. 197^"75 71 3 Estimates of k_ of the negative binomial for each insect stage per fascicle and per 15-cm branch section 72 4 Regression and correlation of log variance on log mean for a l l stages of the larch casebearer 73 5 The correlation between mean and variance for the l i f e stage of larch casebearer 77 6 Egg distribution by needle condition and on needle surface - 1974 91 7 Distribution of eggs on the needle (197*0 92 8 Distribution of C. larice!la egg per fascicle by crown levels as collected at Thrums 197^  and 1975 9^  9 Distribution of C_. lar ice! la eggs per 15_cm branch section by crown levels and years 3k 10 Number of larch casebearer by tree branch type and horizontal crown position per fascicle per 2-15 cm branch sections 96 11 Defoliation rating by tree, crown level and exposure at time of egg stage, 197*t 96 12 Moth activity as observed on 14-15 June 197^  at Thrums, B.C. 99 13 Average number of larvae per fascicle and per 15_cm branch section by tree and stand position 102 14 Avg.No.of casebearers/15 cm branch section collected 197^-103 15 Number of larvae per fascicle by casebearer stage, exposure and crown levels 104 16 Avg.No.of larvae per fascicle for 12 trees at Thrums by horizontal crown position . . . 107 - v i i i -17 Comparison of spring larvae per fascicle by branch tips and by the outer third of branch . . . . . 108 18 The number and percentage of normal and needle tip cases by tree for the prewinter larvae 197** 110 19 Densities of pupae by crown levels in l~lk and '75 . . . . 115 20 Densities of pupae by horizontal crown position jn',7^  &'75.115 21 Avg.No.of pupae per fascicle by year, position of trees in the stand and exposure 117 22 Analyses of variance of numbers of fascicles per I5~cm branch section 119 23a Analyses of variance of numbers of fascicles per 15~cm branch section collected for egg counts in 197^ and 1975 120 23b Analyses of variance of fascicles per I5~cm branch section from samples collected for larva^ and larva^. . . 121 2k Density & percentage distribution of needle fascicles/15_cm branch in 3 crown levels of western larch at Thrums,B.C.. 122 25 Number of fascicles and eggs per 15~cm branch section by year and crown level 123 26 Average number of fascicles per 15-cm branch section by collection period, crown level, exposure and branch type 125 27 Average number of fascicles and eggs per 15_cm branch section by year, exposure and branch type 126 28 Distribution of insects and appropriate fascicles per 15~cm branch section by horizontal crown position. . .127 29 Average number of needles per fascicle by tree and crown levels 128 30 Avg.needle length (cm) for samples taken from h trees in 1975 129 31 Avg. weight of fascicle as calculated with the use of Ives 0955) formula, for an:interior(1) and an edge(6) tree. . .130 32 Survival of C_. laricel la through one complete generation near Thrums, B.C. 197^-1975 131* - ix -33 Number of postwinter larvae and desiccated larvae by crown level, 1975. . . . . . . 134 34 Average number of pupae, adults and parasites per fascicle by crown level, 1975 137 35 Average number of pupae, adults, parasites per fascicle by exposure and branch type, 1975 • 137 36 Average number of pupae, adults and parasites per fascicle by horizontal crown position, 1975 137 37 Analysis of variance for three-stage sampling • • ' 143 38 The calculated required number of trees to be sampled by insect stage and sampling precision 145 39 Mean (x) and standard error of the mean (S—) for the various l i f e stages sampled in numbers of insects per fascicle 147 - x -LIST OF FIGURES F i g u r e Page 1 D i s t r i b u t i o n o f l a r c h c a s e b e a r e r i n Canada (from Webb and Quednau, 1971) 13 2 L i f e c y c l e o f C o l e o p h o r a l a r i c e ! l a ( a f t e r Webb 1953) . . . 20 3 L a r c h f o l i a g e showing l o n g s h o o t , s h o r t s h o o t s and f a s c i c l e s o f n e e d l e s 29 4 The s t u d y a r e a , a s t a n d o f w e s t e r n l a r c h near C a s t l e g a r , B r i t i s h Columbia 32 5 Map showing s t u d y a r e a s ( a f t e r Shepherd and Ross, 1973) . 33 6abc L i f e s t a g e s o f t h e l a r c h c a s e b e a r e r sampled a. Needle f a s c i c l e s showing s p r i n g f e e d i n g damage t o n e e d l e t i p s , and pupa 37 b. Eggs on a d v e n t i t i o u s new n e e d l e s . C o u r t e s y o f The P a c i f i c F o r e s t Research Centre. . . .38 c. O v e r w i n t e r i n g c a s e s on dormant l a r c h t w i g 39 7 Diagram showing r e l a t i v e p o s i t i o n s o f sample t r e e s a t Thrums, B.C 42 8 a. D i v i s i o n o f l a r c h crown v e r t i c a l l y and h o r i z o n t a l l y 43 b. Sample branch showing branch s e c t i o n s (6) s e l e c t e d 43 9 C o l l e c t i o n and R e a r i n g bag w i t h 15~cm branch s e c t i o n . . . 51 10 Frequency d i s t r i b u t i o n o f i n s e c t s per f a s c i c l e and per 15 _cm branch s e c t i o n : a) pupa '7^; b) pupa '75 . . . . 67 11 Frequency d i s t r i b u t i o n o f eggs per f a s c i c l e and per 15~cm branch s e c t i o n 68 12 The r e l a t i o n s h i p between mean number o f i n s e c t s p er f a s c i c l e and v a r i a n c e : a) L a r v a ^ ; b) Egg 70 13 Number o f i n s e c t s per f a s c i c l e by crown p o s i t i o n 82 - x i -14 Number of pupae per fascicle: a) by tree and crown position; b) by exposure, branch position and tree, 197^ + - -83 15 Number of eggs per fascicle: a) by tree and crown position; b) by exposure, branch position and tree, 1 9 7 4 . 8 4 16 Number of prewinter larvae per fascicle-: a) by tree and crown position; b) by exposure, branch position and tree, 1974 85 17 Number of insects per fascicle by exposure and branch pos i t ion 86 18 Number of insects per fascicle by horizontal crown pos i t ion 88 19 Four l5 -cm branch sections showing defoliation ratings 2, 4 , 6, 10 97 20 Number of casebearer per 15~cm branch section by crown levels . .105 21 Casebearer population per fascicle from May 197** to June 1975 138 - x i i -Appendix LIST OF APPENDICES Page 1 Pupa] data sheet and sampling notes 159 2 Form used for egg counts 161 3 Form used for larval counts 162 4 Tables 1-25 163 Table 1 Observed and expected frequencies for pupae per .163 fascicle per 15~cm branch section 2 Observed and expected frequencies for pupae per 164 15-cm branch section 3 Observed and expected frequencies for eggs per 165 fascicle per 15~cm branch section 4 Observed and expected frequencies for larvae 166 per fascicle per 15~cm branch section 1974-75 . • • 5 Observed and expected frequencies for larvae per 167 15-cm branch section 1974-75 6 Analyses of variance of eggs per fascicle by tree, 168 crown level, exposure, branch type and horizontal crown pos i t ion 7 Analyses of variance of prewinter and postwinter 169 larvae by tree, crown level, exposure, branch type and horizontal crown position 8 Analyses of variance of pupae by tree, crown level, 170 exposure, branch type and horizontal crown position. 9 Average number of insects per fascicle by tree and 171 position in the stand 10 Avg.No.of insects per fascicle by tree and crown 172 pos i t ion, 1974 11 Avg.No.of insects per fascicle by tree position 173 in stand and crown position, 1975 12 Avg.No.of insects per fascicle by tree position 174 in stand and exposure, 1 974 . . 13 Avg.No.of insects per fascicle by tree and 175 exposure, 1975 - x i i i : -14 Avg.No. of insects per fascicle by tree and 176 branch type, 197** 15 Avg.No. of insects per fascicle by tree and 177 branch type, 1975 16 Avg.No. of pupae per fascicle by tree, branch 178 type and horizontal crown position, 197** 17 Avg.No. of eggs per fascicle by tree, branch 179 type and horizontal crown position, 197** 18 Avg.No. of prewinter larvae per fascicle by 180 tree, branch type and horizontal crown position, 197** 19 Avg.No. of postwinter larvae per fascicle by 181 tree, branch type and horizontal crown position, April 1975* • • • • • 20 Avg.No. of pupae per fascicle by tree, branch 182 type and horizontal crown position, June 1975« • • • 21 Avg.No. of eggs per fascicle by tree, branch 183 type and horizontal crown position, 1975 22 Avg.No.of insects per fascicle by crown level 184 and exposure • 23 Avg.No. of insects per fascicle by crown level 184 and branch type 2k Avg.No. of insects per fascicle by branch type 184 and exposure 25 Average number of fascicles per 15~cm branch 185 section by tree and collection period 5 I Frequency distribution of fascicles per 15-cm branch sect ion . . . , 186 || Relationship between variance (S ) and mean (x) of fascicle per 15-cm branch section by tree. . . . . 186 - xiv -ACKNOWLEDGEMENTS Sincere thanks are due to-Dr. K. Graham, Professor of Forest Entomology, University of British Columbia for guidance and advice; to Dr. A. Kozak for advice on statisti c s ; to the other members of my advisory committee, Drs. A.D. Chambers, J.H. Meyers and P.G. Haddock for helpful criticism of the preliminary draft of the thesis; and to Dr. D.A. Ross, Research Scientist, and other members of the staff at the Pacific Forest Research Centre, Victoria, B.C. for their interest and support of the project. I express sincere thanks also to the Faculty of Forestry, University of British Columbia; the National Research Council of Canada; and the Canadian Forestry Service, Pacific Forest Research Centre, for financial assistance and support. - xv -INTRODUCTION It has become virtually axiomatic that the design of popula-tion sampling systems can be prescribed within the framework of certain general principles. First, the sampling universe must be defined. Then that universe must be subdivided into component "strata" having known, apparent, or presumedly distinct features. Samples must be taken so as to represent either an average condition or a particular one. A sufficient number of samples must be taken to express with an acceptable level of confidence the particular parameters which are being sought. A principle to be recognized is that population sampling systems must be designed differently in detail to serve different purposes. The purposermay be to achieve a sensitive means of detecting low population densities, to express relative levels of populations in time and place or to detect relative density patterns. It may seek to estimate absolute populations per unit of terrain, or saturation density of a population relative to the available food supply. Another recognized principle is to define times in the l i f e cycle when the population is stable numerically or spatially, or presents a particular feature of interest or feasible opportunity. On the other hand a complete sequence of insect stages from egg to adult may be required for l i f e table purposes. - 2 -A particular guideline is the definition of appropriate sampling universes where the insect will be found at different times during its l i f e . This necessitates an understanding of the biology of the specific insect, and cannot be derived from general knowledge. Every species of insect presents a rather special problem which differs in various ways from every other insect. This follows from the fact that it has a distinct pattern of l i f e cycle and habits as it progresses ontogenetically from egg to adult. Each species reveals certain consistent tendencies in the general location of its various l i f e stages as they shift from one microhabitat to another under the dictates of their genetic code and internal influences. However, the loci of population concentrations within any given microhabitat vary according to circumstance. Such shifts of population concentrations within different parts of a microhabitat can reduce the accuracy of a sampling system which does not provide for this factor. Sampling design must recognize special features of the substratum on which insects occur. On a tree, special features of crown form, branches, twigs and foliage must be taken into account. It was in the light of these considerations that it has been necessary to devise a sampling design that meets the specific problems of measuring populations of the larch casebearer Coleophora l a r i c e l l a (Hbn) Coleophoridae: Lepidoptera. In accordance with the foregoing principles, the l i f e history and habits of the larch casebearer form an essential basis for the design of a sampling system explicitly applicable to this species. An account of the biology of £. lar icella is presented in the f i r s t section following this INTRODUCTION, being so placed in order that the information is close - 3 -at hand, but will not interrupt the continuity of the statements concerning the problems of sampling. The larch casebearer is an important forest defoliator of various species of larch (Larix spp) throughout the world, including western and eastern North America. The casebearer was f i r s t discovered on western larch (Larix occidental is Nutt.) in Idaho, U.S.A. in 1957 and in southern British Columbia in 1966- Suitable sampling techniques are now required to study the population of this insect and the effects of native and/or introduced control factors. Sampling procedures heretofore available for the larch casebearer left certain matters unresolved for their direct application to western larch: (i) they were based on relatively meagre samples, (ii) they were not suitably stratified into the different segments of diverse and widely varying sampling universes, ( i i i ) they related to species of larch other than L_. occ i denta 1 i s growing in other biogeographica1 regions, (iv) they were based on a fixed design of sampling which did not provide for the distortions of the spatial distribution of the insect populations under various influences. It is the purpose of this investigation to test the hypothesis that shifts in population concentrations of the larch casebearer influence the accuracy of sampling at fixed points in the crown of a tree. As a corollary, the further hypothesis would follow that sampling accuracy would improve by a shift of sampling intensity according to circumstances noted above. - 4 -The pursuit of these objectives was perceived as requiring procedures to: (1) ascertain the distribution, shifts in distribution, and vari-a b i l i t y of insect stages in the tree crown, (2) observe the behaviour of a l l insect stages as it affects population distribution, and (3) develop an efficient, s t a t i s t i c a l l y reliable sampling system for the larch casebearer on western larch in B.C. The purposes of such a system would be to develop an efficient sampling for biological evaluation of infestations at different times, places and circumstances. Such a design would serve the needs of life-table studies. The design would also assist in the formulation of important forest resource management decisions. As a possible improvement on existing sampling designs a more accurate yet economical design would assist in the evaluation of the effectiveness of any control methods. A prospective application now in sight would be in the assessment of attempts at biological control which is expected to follow the proposed release of parasites. The experimental sampling for this study was done in immature stands of western larch, infested with the casebearer, near Thrums and Castlegar in the Nelson Forest District, B.C. in 1974 and 1975-Earlier Work Various methods have been employed to sample larch casebearer population and infestation. Jung (1942), Burst and Ewald ( 1 9 5 5 ) , Eidmann (1965) and Schindler (1968) in Europe; and Webb (1953), Sloan (I965) and others in North America, are writers who have related larval - 5 -infestation to the needle fascicles of larch which, depending on a tree's intensity of growth, may contain up to kO needles each. Webb (1953) in eastern North America used population intensities' in his study to investigate short-term relationships between the casebearer (insect) and tamarack, J_. laricina (host). For intra-stand studies counts were made on the number of casebearers at three crown levels, as found on shoot growth of the previous year bearing 100 lateral spurs. Usually two or more shoots were included, these selected in order from the terminal toward the base of the branch, and spurs were counted beginning from the tips of the shoots. He found a highly significant variation in population density between crown levels, greater than that between trees, with larvae per spur shoots ratios usually highest at the base of the crown and de-creasing toward the tree top. Most of the other workers sampled casebearer populations using the method of Webb (1953). For example, Sloan (1965) sampled the casebearer in Wisconsin, using Webb's method and taking 3 branches of each of three crown levels. The data were used to illustrate the seasonal fluctuations in populations. However, unlike Webb (1953) who used the analysis of variance, Sloan claimed that the data were d i f f i c u l t to analyze with normal parametric st a t i s t i c s , as the sampling procedures included several populations about which very few statistical assumptions were made. Therefore, the non-parametric Friedman test was used for data analysis. Non-parametric statistics are useful when the nature of the theoretical distributions was not specified; thus, it is a comparison of ranking and signs. 1 Population intensities is an expression of insect number in terms of available food, e.g. numbers of insects per needle fascicles (Morris 1955). - 6 -Also according to Sloan (1965) eggs were not found on the older branch growth. Consequently he sampled only the new growth. He is the only one to mention this, but did not state how he arrived at such a conclusion. Ciesla and Bousfield (197*0 developed a means of predicting potential defoliation by larch casebearer, by using counts of overwintering larvae to forecast regional or forestwide population and damage trends. For the expression of population densities of overwintering larvae they adopted Webb's (1957) index (number of larvae per 100 spur shoots) as the unit of measure in their study in northeastern Washington, northern Idaho and western Montana. The system consisted of k branch samples of 100 spur shoots each collected from mid crown of 10 dominant or codominant western larch , 30 to 50 f t . (10 to 16 meters) in height, for a total of 40 branches/plot (63 plots) during November and December. They assumed that the relationship with respect to population densities in the crown (upper, mid and lower) of eastern larch held true for larch casebearer in western larch. Because accurate counts of the small overwintering larvae require laboratory examination rather than fi e l d counts, a plan with a fixed rather than a variable number of samples (sequential sampling) was developed by Ciesla and Bousfield (1974). Andrews, et_ a_L (1967) stated that no standard method of sampling larch casebearer has been developed. Three methods for sampling larvae were used i n i t i a l l y in British Columbia by the Forest Insect and Disease Survey Team to compare results and establish a sampling pattern, they are: - 7 -(1) A sample of 5 randomly chosen 12-inch (30 cm) branches from the lower crown from each of 5 trees (in early June for final instar 1arvae). (2) The sequential sampling method used in New Brunswick (Webb, 1957) was tried at 5 locali t i e s . In this method 100 fascicles on a shoot constitute one sample unit. (3) A sample unit of ten 16-inch (40 cm) branches taken from the lower crown from each of 5 trees at 5 localities was taken in October (for prewinter larvae) to estimate the population density of the succeeding generation. No conclusive results were obtained as to the suitability of these methods for sampling the casebearer. Ultimately, the 18-inch (45 cm) branch sample normally used in insect surveys was adopted for use in the year 1968. Eidmann (1965) in an ecological study on the larch casebearer in south Sweden, found most eggs and prewinter larvae on shots of the preceding year. In spring, larval infestation was heaviest on the youngest growth. Jagsch (1973) constructed life-tables for the larch casebearer in the natural areas of distribution of Lar ix dec idua in the Alpine area of Styria, Austria. His sample unit consisted of about 150-200 terminal shoots, each about 30 cm long and taken from different parts of the tree, from different trees and from different parts of a stand. Deficiencies of Former Methods These methods do not clearly account for the highly significant variation in population between crown levels and the non-uniform dis-tribution within branches. - 8 -New shoots offer flexible sample size and ease of collection, but shoot numbers lack s t a b i l i t y . That is, the number of shoots can vary considerably with intrinsic factors such as flower production and effects of repeated defoliation. This unit of population intensity must be re-jected in absolute population work. A restricted part of the branch, such as the apical 18 inches {kS cm) of the branch in Canadian Forest Insect Survey work, 15 inches (38 cm) in the U.S.A. and the apical second year's growth by Webb and others for the larch casebearer provide f l e x i b i l i t y of sampling design and are easily collected. However, the proportion of the insect population using the sampling unit as a habitat is not constant. It is affected by biotic and abiotic factors, such as temperature differences and feeding preferences, resulting in a shift in center of activity of the population. Depending on time of collection, therefore, the apical portion of the branch may support different proportions of the total population (eggs, larvae or pupae). Therefore the use of partial branches may introduce a bias because of non-uniform distribution of casebearer on the twig. A restricted part of a branch does not lend itself to quantitative assessments of the total number of units per tree and per acre. Although it may be calculated, it would be very d i f f i c u l t to account for the basal components of the branches and for their insect population relevant to the tips. Also, 2nd year twigs lack stability i.e. they vary with the quantity and quality of new shoots produced. These weaknesses point to the need for sampling involving the whole length of the branch. While several population sampling procedures have been described above for estimating populations of the larch casebearer, the possible needs to shift the aim at different sampling targets at different times of the season and different stages of an infestation have not been explicit. - 9 -It might be expected that with an inflexibly prescribed design of sampling, the accuracy of sample estimates may change as the population shifts within the crown of a tree throughout a season. Inasmuch as one of the major requirements in sampling is to achieve the best gain in accuracy or sensitivity within the shortest time and least cost of labour, an understanding of the conditions affecting the results is desirable. Shifting of the distributions might be expected to result from mass movements of larvae upward, downward, inward, or outward on the crown, according to the particular larval instar, the state of hunger, diet, and day-by-day responses to light, temperature, humidity, wind, atmospheric pressure, gravity, residual amount of foliage, and differences in mortality. The relative shifts of population concentrations might be expected to differ between trees in different positions relative to the stand margins or center. If such shifts occur in population con-centrations, they should be recognizable in successive sample estimates of the population. Also, if shifts of concentration are measurable, they may be correlatable with observable factors, which in turn may provide some guidelines for sampling. - 10 -THE INSECT Taxonomy The larch casebearer, Coleophora la r i c e l l a (Hbn) belongs to the family Coleophoridae in the large group of microlepidoptera of the superfamily Tineoidea. Since this species of insect infests larch and the final larval stages live in a case, i t was given the name "larch casebearer". Also the word Coleophora means "bearing a sheath" and the word la r i c e l l a refers to "larch", the ending -el la meaning "small". Common names: In the literature of other languages, the larch casebearer is found under such common names as: Larchenminiermotte (German), larktradsmalen (Swedish), lariksmot (Dutch), porte-case du meleze (French), la coleofora del larice (Italian), la minador del alerce (Spanish). Synonymy: As an aid in the search of the literature the synonymy for £. l a r i c e l l a is listed in chronological order (after Sloan 1965): Tinea l a r i c e l l a Hbn. (1914); T. la r i c i n e l l a Bechst., Blum. (1816); Eupista l a r i c e l l a Hubner, Haploptilia l a r i c e l l a , Hubner (1825); J_. laricel la Hbn. Samm. Europ. Schmett (1827); Ornix argyropennel 1 a Treit (1834); Graci1laria larcella Zel1. (1838) ; Colephora l a r i c e l l a Zell (1839); T. larcel la Ratz (1840); H_. laricel la (Hbn.) Banks (1925) Eupista l a r i c e l l a (Hbn.) Pierce, Metcalf (1935); £• nigricornis Hein. and Wck., Tol1 (1944). -11-Related species (evidence and significance i f any): The taxonomy of the larch casebearer appears to be in doubt. In the Soviet Union, Fal'kovich (196 * 0 has reported three species of Coleophora on larch; C_. 1 arieel 1 a (Hbn) , which infests Larix decidua in western and eastern Europe (including the Carpathian region), and two new species described from adults of both sexes as £. s i b i r ica and datu r i ca spp. n. both previously identified as C_. 1 a r i eel la. £. sib i r i c a mostly infests j^. s i bi rica and occurs in European Russia and in Siberia (as far as the Baikal area), and £. dahurica feeds on L^. dahur ica in eastern Siberia and the Soviet Far East. Divergence from an original casebearer species appears to have resulted from the isolation of areas of different larch species in the Tertiary period. In Japan, Moriuti (1972) stated that for many years the larch casebearer has been misidentif ied as the European C_. laricel la. The Japanese species was described from the adults of both sexes as C_. long i s ignel1 a sp.n. H. Pschorn - Walcher (personal letter to Dr. R.F. Shepherd, 1974) speculated as to whether C_. laricel la is really a truly European species. He noted that "taxonomically" i t is quite isolated among the European Coleophoridae and has only 2 synchronized parasites as compared to 7 for the birch casebearer C_. fuscedinel la. The source of western U.S. and western Canadian population is unknown and may have been the U.S.S.R. by way of Japan (Shepherd and Ross, 1973). This taxonomic problem should be investigated by comparing the casebearers from Europe, Siberia, Japan and eastern and western North America for specific difference. However, in spite of these taxonomic - 12 -differences, there is no apparent difference in the l i f e histories and behaviour except those due to climatic differences. World Distribution The larch casebearer occurs in Europe from the French Alps and Italy northward through Austria, Germany and Holland, southern and central Russia to Finland and the Carpathians. According to the literature, l i t t l e is known about the casebearer in eastern Europe (e.g. Poland). The casebearer is present in Siberia, in northern China and Japan and probably wherever larches occur. It followed introduction of larch into Sweden and Great Britain. According to Jagsch ( 1 973 ) , in the Alps it occurs up to tree line, but mass infestations are rare above 1,600 m (6 ,300 feet) (Eidmann, 1965 ) . The f i r s t North American report was by Hagen (1886) from Northampton, Massachusetts on introduced European larch. The larch casebearer now occurs throughout the range of tamarack (Larix laricina (Du Roi) K. Koch) from eastern U.S. to central Minnesota; and in Canada (Fig. 1) from the Atlantic Provinces and as far north as 4 9 ° - 5 0 ° N. latitude in Quebec and Ontario. Recently i t has been discovered progressively westward from Lake Superior to the Manitoba border in western Ontario and in 1965 was dis-covered in the extreme southeastern corner of Manitoba (Webb and Quednau, 1971) . Figure 1. D i s t r i b u t i o n o f l a r c h c a s e b e a r e r fn Canada ( f r om Webb and Quednau, 1971) . - ]h -In 1957 the casebearer was discovered on western larch (_L. occ idental is Nu tt.) in north-central Idaho in northwestern United States, approximately 1,700 miles (2,735 km) from the last reported infestation in Minnesota (Denton and Tunnock, 1972) and in British Columbia in 1966. Its current distribution encompasses most of the range of western larch in Washington, Idaho and Montana and a strip of approximately 200 miles (322 km) along the international boundary in British Columbia (Webb and Quednau, 1971). The northern limit and altitudinal limit of the larch casebearer does not extend as far northward or upwards as the host range. Eidmann (1963) explains the northern limit of Z. l a r i c e l l a in Europe as being a factor of the vegetative period of the host. The casebearer is killed when cool summers are followed by early winters. Cool temperatures appear to be the indirect cause of the insects death as such low temperatures reduce the vegetative period to the extent that there is insufficient time for the insect to prepare for dispause. A normal dispausing larva is able to survive temperatures approaching -30C (-22F). Significance as a Defoliator The larch casebearer has been a perennial defoliator of larch in Europe for centuries and has always been considered a serious pest. Currently i t is the most serious pest of western larch. Severe defoliation or rather needle-mining of a healthy tree is i n i t i a l l y followed by refoliation later in the same spring. Defolia-tion in successive years causes reduced production of foliage and shoots, reduction of terminal and radial growth, death of scattered twigs and branches, to death of the tree. - 15 -Diameter growth can be drastically reduced by repeated case-bearer defoliation. Schwerdtfeger and Schneider (1957) calculated increment losses of 35 to 45 percent after several years of continuous severe feeding. Denton (1964) showed i n i t i a l results of repeated de-foliation on radial increment on western larch in Idaho. Prior to casebearer damage in 1956, the larch trees added 4.0 mm radial increment. After five years of severe defoliation increment had decreased to 1.0 mm, representing a 75 percent reduction in growth. During the same period annual growth of non-defoliated larch trees decrease from 4.4 to 3-** mm or a reduction of 23 percent. According to Tunnock, et a l . (1969) there is evidence of radial growth loss of up to Sk percent in areas of Idaho where severe defoliation has occurred for four or more consecutive years. Effect of defoliation on height growth can be severe. Ewald and Burst (1959) found that height increment was retarded 17 percent in four years of larch casebearer feeding. Larch, being a serai species exhibits a high degree of shade intolerance and it can survive only if it maintains a dominant position in the canopy (Roe, et a l . 1971). Repeated defoliation reduces the growth of larch and places it at a competitive disadvantage with its associated species. Accordingly, larch can lose its dominance in mixed stands and eventually its potential to recover even though the casebearers population may decline. Growth reduction and loss of tree vigor can have serious long-term forest resource management implications with or without mortality of larch. In eastern North America, several entomologists including Herrick (1912), Craighead (1950) and Dowden (1957) l i s t larch casebearer among the tree-killing insects. In Europe there are reliable reports of trees dying after casebearer defoliation, Jung (19^+2) and others cite - 16 -instances of larch mortality following outbreaks. Others, such as Webb (1953) suggest that it is unlikely that the casebearer alone can cause tree mortality of tamarack in the east, but concluded that it might be a contributing factor. In Europe, casebearer feeding reduces the vigor of the trees and allows the larch canker fungus (Dasyscypha  willkommi i (Hart.) Rehm.) to enter and eventually k i l l the tree. In 1967, ten years after the casebearer was discovered on western larch, mortality was reported in the St. Joe National Forest, Idaho. In 1968 aerial surveys confirmed that serious deterioration and tree mortality were prevalent within thousands of acres of larch forests in Idaho. However, a study to determine the cause did not confirm that larch casebearer was the sole cause of tree mortality but that it was definitely responsible for weakening and predisposing western larch stands to mortality (Tunnock, et a l . 1969)-Biological Characteristics of the Insect The significance of the biological characteristics of the insect as a background to sampling is summarized below: It is necessary to know what specific object is to be sought i.e., the insect during any or a l l of its morphological forms as they metamorphose throughout l i f e from egg to adult. It is necessary to define the l i f e cycle because this provides the guidelines for knowing when any particular stage will be found or what stage will be found at any prescribed time during the l i f e cycle. - 17 -It is necessary to have a knowledge of the behaviour and habits in order to know where to sample; so that the sampling efforts can be directed to specific points in the nabitat any time of the season, time .' of day or under various circumstances which might affect the distribution patterns. Morphological Forms and Stages Descriptions of adults, eggs, larvae, and pupae serve as a basis for specific search and identification of the pest. The l i f e stages of the larch casebearer were described by Webb (1953) • Adu11: This is a small silvery to greyish brown moth, with no conspicuous markings; wing expanse 9 mm (3/8 inch), narrow, fringed with long slender hairlike scales. Maie - dull slate-gray, the sides of the abdomen appearing almost straight and parallel, and the claspers give the anal extremity a bifurcated appearance. Fema1e - Lighter gray than the male also silvery on the vental part of the abdomen. Sides of the abdomen giving a more robust appearance. Tip of abdomen abruptly truncated. Egg: The eggs are hemispherical with 12 to 14 lateral ridges radiating from apex to base and under magnification resemble inverted j e l l y molds-. They are orange-yellow to clear gray just before hatching, however, reference to their color varies from cinnamon-rufous to reddish brown. Average diameter is 0.29 mm and average height 0.17 mm. The flat lower side is cemented to the needles by a transparent substance. - 18 -Larva: The larvae develop through four instars. First instar averaging 1.0 mm in length; mean head width of 0.11 mm; color honey-yellow, with fine ferruginous markings. Head pale brown, darkening along the epicra-nial suture, thorax slightly wider than abdomen; thoracic legs and small prolegs well developed; abdominal proleg not distinct. Second instar larvae averaging 1.4 mm in length, mean head width 0.15 mm, abdomen with brown pigmentation, thorax darking dull brown, head dark brown. Third instar larvae, average 1.8 mm in length; mean head width 0.24 mm, abdomen darker brown, head capsule black, inside a tubular case made from a hollowed-out needle which is yellowish in color and rectangular in shape. Fourth instar or mature larvae average 4.3 mm in length; mean head width 0.33 mm. abdomen liini'fiorm dull brown in color, head and thoracic shield black, divided along the mid-dorsal line, the larva is inside an enlarged tubular case. Pupa: The pupae are obtect, dark brown without distinctive markings or features; size varies from 2 to 4 mm (av. 3 mm) in length and slightly less than 1 mm wide; it is inside a greyish cigar shaped case about 4 mm long. Life History - 19 -The l i f e history of the larch casebearer has been studied in other regions of the world by many investigators, including: Reissig and Ratzeburg (1869) , Escherich (1931) and Jung (19^+2) in West Germany, Eidmann (1965) in southern Sweden, Herrick (1912, 1929, 1935) and Webb (1953) in eastern U.S.A. The larch casebearer has one generation each year. Adult emergence occur from mid-June to early July. The eggs are laid singly and scattered over the foliage in late June to July. The newly hatched larvae bore directly into the needles and start mining. Larval development to second instar larvae range from late July to late August. Third instar larvae, the casebearing and overwintering stage, occur from late August to early May. The fourth and final instar larvae feed throughout May, pupation begin in late May and extends into the middle of June (Fig. 2). Reproduction, Growth S Morphogenesis Reproduction: The adults emerge in June; normal sex ratio is 50:50 (Webb, 1953)- This bisexual ism requires mating and renders their f e r t i l i t y vulnerable to meteorological influences. The moths-are crepuscular, mating takes place 1 or 2 days after emergence and is stimulated by decreasing light. The female is pro-ovigenic having its f u l l complement of eggs at time of emergency. Peak oviposition occurs during the f i r s t week. According to Quednau (1967) in Quebec, Canada, the optimum temperature for egg laying was 70 to 80 F (21 - 26.7C) and oviposition by average-sized females was 66.6+ 3.1, range 48-111, eggs (compared to Webb's (1953) average of 50 eggs (maximum 133) deposited by f i e l d mated females in I nsect Stage Larva^ Pupa Adult Egg Larva 1-2 Larva. Winter Apr i 1 foliation of larch, resumption of feeding May June J u l y Aug . Sept. Oct. O Fig. 2. Life-cycle of Coleophora l a r i c e l l a (after Webb 1953)-- 21 -laboratory cages at room temperature). The average l i f e span of the females is 15 days (Quednau 1967). The embryonic development lasts 10-20 days, depending on the temperature, 12 days at 80-85°F (26-29C) according to Quednau (I967). Larvae: After hatching the f i r s t instar larvae bore through the under-side of the egg directly into the needles, the i n i t i a l faeces f i l l i n g the empty egg cases. During the f i r s t two larval instars, the cater-pill a r s mine the needles for about 6 weeks after which they moult to the third instar. The larva now makes a. case from the hollowed-out needle and continues to feed on the foliage in the late summer until the needle moisture'content f a l l s and nourishment decreases while the photoperiod shortens. Just before needle-fall the larva firmly attaches its case with s i l k threads to various parts of the branch. The winter dormancy lasts until bud-burst or until the appearance of new larch needles, i.e. until mid or late April in low-lying areas or until mid-May at higher elevations. The larva molts to the final instar, resumes feeding as a casebearer and enlarges its case. Before feeding, the casebearer fastens its case firmly to a needle with a pad of silk threads and mines the interior as far as the larva can reach without leaving its case. After a feeding period of 3 to k weeks, the larva attaches its case to a needle or center of a needle fascicle and pupates. The pupal resting stage last about 2 to 3 weeks, after which the moths emerge and the l i f e cycle is initiated again. - 22 -THE HOST TREE The variations in the distribution of the insect and its damage may be linked with some character of the environment including its host species. The larch casebearer occurs within certain approximately definable t e r r i t o r i a l limits with boundaries affected by host availability and climate. Therefore, it is necessary to know the distribution of the preferred host and of insect damage, as well as the distribution of certain other environmental factors such as, habitat type, stand age and density for the development of sampling designs to serve the various needs of resource management. Host species and Susceptibility Tree species of the genus Larix are the preferred hosts of the larch casebearer. All ages and vigor of larches are attacked. Young and old trees are attacked mainly at the intermediate altitudes 750~1300m (2460-4265 feet) (Eidmann I965) and below 3000 feet (1000 m.) in British Columbia. Greatest effects are probably on the younger trees which are in competition with associated species. Attacks on conifers other than species of Larix are rare, and then probably represent biological accidents or circumstance, such as attacks on understory trees during severe infestations. Peirson (1927) reported larvae feeding on white pine, Pinus strobus L. and f i r , Abies sp. where opening buds were preferred. In Europe, Van Poeteran (1933) and Schwarz (1933) found the larvae feeding in needles of Douglas-fir, Pseudotsuga menziesii (Mirb.) Franco, (Webb, 1953); Eidmann (1965) - 23 -reported that it also infests to some extent Douglas-fir in the neighbourhood of larch in south Sweden. Luitjes (1971) investigated the development of the larch casebearer on young Douglas-fir planted in 1965 under a thinned stand of 35-year old Larix kaempferi (1eptolep i s) in Holland. Despite a reported larval mortality of over 30% in July-October, 19% of the needles were damaged in 1966 and 26% in 1967• To date the larch casebearer has not attacked sub-alpine larch L_. lyal 1 i Pari, in British Columbia or western U.S.A., but this is possibly due to the extreme climate in its range at altitudes 7,000-10,000 feet (2133~30^+8 m) in the environment at timberline isolated from western larch. Early reports l i s t Japanese and Russian larches as resistant to the casebearer, but infestations have been reported on both. Limited studies on site differences have been carried out in Europe; Jung (19^*2) reported without explaining why that in a certain area one tree was attacked heavily while neighbouring trees were not (from Sloan, 1965)-This is probably due to the fact that the casebearer will concentrate on one tree until forced to disperse. Jung f e l t that sickly trees would be subjected to heavier attacks. Schimitschek (1963) indicated that the following stand conditions were suggestive of larch casebearer outbreak conditions: (a) in natural alpine areas, warm slopes; (b) effects on the physiological state of the tree (e.g. disturbance of water balance by grazing); and (c) forest history of the area (e.g. changes in environmental factors which affects the quality and quantity of foliage produced). These conditions are applicable also to outbreaks of other defoliators. - 2k -Schindler (1965) in Germany stated that damage by the larch casebearer is particularly severe in poor quality stands. The Host Tree in British Columbia The host tree in British Columbia is western larch (Larix  occidentalis). The mature western larch is a large tree 100-180 feet (30-55 m) in height and 3~k feet (1m) in diameter. In British Columbia its range is (restricted to the southeastern portion of the province: eastward from Okanagan Lake to the flank of the Rocky Mountains northward to Shuswap Lake and Columbia Lake. Altitudinal range is approximately 1,800-4,000 feet (549-1219m). Western larch occurs in British Columbia in the following biogeoclimatic zones of Krajina (1965): 1. the interior western hemlock zone only on the drier subzone (IWH ); a 2. the interior Douglas-fir zone (IDF); 3. the ponderosa pine-bunchgrass zone, in the moister pockets. Such phytosociological delineations are potentially important for purposes of population sampling designs which recognize the need to stratify sampling universes according to different expectations of insect distribution. The study area and most of the western larch occurs in the drier subzone of the Interior Western Hemlock zone which is characterized as follows (after Bell, 1965; Krajina, 1965): - 25 -CIimate Regimen: microthermal continental humid with no distinctly dry season Accumulated day degrees over 43F(6C): 1,500-3,000 per year Mean annual temperature: 42-46F (6~8C) Mean monthly temperature January 20-29F (-7 to -12C) July 65-68F (18-20C) Number of months above 50F (10C) 5 below 32F (0C) 3-4 " " frost-free days: 100-150 Mean annual precipitation: 20~38 inches (51~96cm) " " snowfall: 57~l60 inches (145-^06^) Seasonal occurrence of precipitation: Wettest season - winter (40% of total precipitation) Wettest month usually Dec. or Jan. 2.2-5.1 in. (56-129mm) Driest season - summer (20% ppt) Driest month usually July 0.9-2.4 in. (23"6lmm) Clouds: very common Prevailing winds are from the northwest and precipitation is characteristically orographic. Soils: zonal group - Minimal and orthic podzol humus - Ligno-mycelial mor This subzone is found between longitude 117°51 and 119° and latitude 49°10' and 51° , and altitude 1,400-4,400 feet (427-13^1 m) in the Selkirk and Monashee mountains of southeastern B.C. and to a lesser extent in the Rockies. - 26 -General climatic requirements of western larch are listed according to Krajina (1965) and may affect the insect population: (a) Western larch is adapted to the microthermal continental montane humid climate with a moderately long vegetative season. (b) It does not tolerate very humid climate even if its transpiration rate during the vegetative season is very high, especially in the spring or during refoliation in late spring after insect damage. (c) It tolerates an average annual precipitation of 28 inches (71cm) up to 35 inches (89cm) in a few larch stands, and a minimum of 18 inches (46cm). (d) Shade tolerance of larch is almost nil and stands have been maintained over the years by wildfire or by the clearcut systems of today. According to Krajina (1965) the nutritional requirements of western larch are moderate, comparable with those of Douglas-fir. It grows better in a rich supply of calcium and magnesium, its relation to nitrogen has not been studied (Krajina, 1965 ) . This subzone (IWH ) is one of the richest areas of various coni-a ferous tree species in British Columbia. Western larch is a dominant species of wide ecological amplitude and grows in mixture with other subclimax and climax species, such as, the climatic climax western hemlock (Tsuga heterophy11 a (Raf.) Sarg) , the edaphic climax western redcedar (Thuja plicata Donn) on wet sites, Douglas-fir (Pseudotsuga menziesii var. glauca (Mirb) Franco), grand f i r (Abies grandis (Dougl.) Lindl) - 27 -western white pine (Pinus monticola Dougl.), lodgepole pine (P_. contorta Dougl.). Associated hardwoods are; black cottonwood (Populus balsamifera L. ssp. tr ichocarpa) , trembling aspen (P_. tremuloides Michx) and birch (Betula  papyrifera March). The i n i t i a l spread of the larch casebearer was most rapid along the valley bottoms. The insect population decreases at about 2,500 feet (762m) in elevation with few casebearers above this altitude in B.C. (Shepherd and Ross, 1973). The Host Tree as a Sampling Unit As an insect sampling problem larch has certain noteworthy peculiarities in the physical form of the tree, its deciduous characteristics, and its effects on insect behaviour. These are such that recognition must be given to unequal and changing distributions of the insects to the extent that stratification of sampling design is required to meet the expecta t i ons. Expectations are that the insect population at any specific period of the season, or time of day, or state of defoliation will differ between base and apex of the crown and between tips and bases of branches within a crown. Also, the ultimate sampling unit must be the fascicle of needles during the growing season. However, as part of the population occurs also on the branches, sampling data must relate to specifically prescribed portions of branches and the short shoots associated therewith. - 28 -Morphological Characteristics of Larch Need 1es: The needles are light green and soft becoming harder later in the season, and yellow before shedding in the autumn. The needles are 2.5 to 50 mm in length, triangular in transverse section and with sharp-pointed tips. The needles of larch are either in fascicles of 16-40 needles clustered on short spurs, or as single needles arranged spirally along the growth of the current year only. Branches: The branches contain both long shoots or the current years shoot bearing single needles; and on the older shoots, short shoots bearing terminal clusters of needles: 1. The shorts shoots are referred to in the literature as "spur shoots", "spurs", "short spurs", "fascicles of needles" or "needle fascicles". In this study the short shoots w? thout its complement of needles are designated as spur shoots, and with needles as fascicles. The short spur shoots increase very slowly in length but have each year's growth marked externally by a distinct ring of leaf-scars. 2. The long shoots sometimes called terminal shoots or twigs have the needles scattered singly and spirally along its length. The base of each elongating shoot is surrounded by peripheral needles of the top of the previous year's growth (Fig. 3). - 29 -F i g u r e 3« Larch f o l i a g e showing long shoot, s h o r t shcots and f a s c i c l e s of needles. Buds: Buds a r e o f three k i n d s : 1. Terminal on the long shoots producing long o r short shoots. 2. A x i l l a r y on the long shoots, s o l i t a r y i n the needle a x i l s producing long o r short shoots but u s u a l l y the l a t t e r . 3« From the p o i n t s o f short shoots producing f a s c i c l e s o f needles, or fl o w e r s I.e. megasporangiate and micro-s p o r a n g i a t e s t r o b i l i accompanied by s l i g h t e l o n g a t i o n o f the shoot, o r l e s s f r e q u e n t l y a long shoot w i t h s p i r a l l y arranged leaves e s p e c i a l l y under abnormal growth c o n d i t i o n s such as d e f o l i a t i o n . - 30 -Deciduous Characteristic of Larch Significance: Larches are the only conifers in the stands that are deciduous, and consequently there are no differences, other than seasonal ones, in the age of needles upon which the larvae can feed. Both of these factors simplify the situation with respect to possible qualitative nutritional differences in larval food (Heron, 1966). Also being naturally deciduous, larch is much more resistant to defoliation than other coniferous trees. Other conifers cannot replace foliage once it is lost except by new shoot growth, whereas larch is capable of producing two crops of needles on the same spur during a single growing season. Vegetative Cycle Larch foliage turns yellow in the f a l l , death and shedding of the needles soon follows, and the tree is bare throughout winter. Needles start changing color early in October and by mid-November most have been shed. New needles are produced the following spring, usually between mid-April and mid-May. The period between shedding of the old needles and production of the new foliage varies somewhat from year to year and with altitude or latitude. - 31 -THE STUDY AREA,. MATERIAL AND METHOD Need for Choosing Sampling Area The requirements for selection of stands for studying the factors which affect population sampling included homogeneous conditions and accessibility. The study area was an immature western larch stand which has been infested by larch casebearer since 1966. The site was typical of the areas infested by the casebearer in British Columbia as reported by the Canadian Forest Insect and Disease Survey. Local? ty: The main study area was a larch casebearer infested stand at Thrums near Castlegar in the Nelson Forest District, British Columbia. The stand was located on private property 1.1 miles (1.8 km) east of the Thrums elementary school off Highway No. 3- The site was a flat valley bottom at an elevation of 1,700 f t . (5l8m) with well drained sandy to gravelly loam. The larch stand was dense and the vegetation consisted of 80 percent western larch, 15 percent understory of redcedar (Thuja  piicata) and 5 percent Douglas-fir (Pseudotsuga menziesii), western hemlock (Tsuga heterophy11 a), grand f i r (Abies grandis) and aspen (Populus tremuloides). Shrubs included species common to most areas: Saskatoon berry (Amelanchier  a 1n i f o l i a Nutt), mock orange, (Philadelphus 1ew i s i i Pursh), snowberry (Symphoricarpus albus (L.) Blake), false box (Pachistima myrsinites (Pursh) Raf), Oregon grape (Mahonia nervosa Pursh), tufted phlox (Phlox caespitosa Nutt). Ground vegetation included: Mountain l i l y (Li 1ium montanum A. Nela.), bearberry or kinnikinnick (Arctostaphylos uva-urs? (L.) Spreng), bracken (Pteridium aquilinum pubescens Underv.), mosses such as, Pleurozium shreber? Mitt and grasses. - 32 -The study area, a stand of western larch near Castlegar, British Columbia. - 33 -Scale 1 miles 0 10 20 30 Range of Larix occidentalis Distribution of Coleophora laricella 1966-1968 Col lection Sites m Ww+i 1969-1970 1. Thrums 2. Sheep Creek 197! - 1972 Map showing study areas (after Shepherd and Ross, 1973). - 34 -The larch trees ranged in height from 30-60 feet ( 8 -15 m), 4-8 inches (10 -20 cm) in diameter at breast height (d.b.h.) with green crown length of over 80 percent of total tree length. The stand was about 30 years old. The other study area was located at Sheep Creek, 5 miles south of Salmo, British Columbia, just off Highway No. 6 and 21 .5 miles (3-**5 km) in a straight line southeast of the site at Thrums. The site was at an altitude of 2 ,000 feet A.S.C. (670m) on a level or gently sloping fluvial plain with soils predominantly a well drained Orthic Regosols. This stand consisted of clusters of larch trees mainly, typical of most larch stands in the region. The larch trees were immature with an average height of 45 feet (14m) and an average diameter at breast height of 6 inches (15 cm). Sampling Procedures The Need for Sampling. Sampling is necessary for the following reasons: (1) to establish species distribution; (2) to measure population densities and change; (3) to construct l i f e tables; (4) to make a biological evaluation of natural and a r t i f i c i a l (introduced parasites, pesticides, etc.) impacts on insect populat ions. - 35 -The Sampling Universe The 'universe' represents the habitat in which the insect population occurs and must be defined in terms of trees, foliage or forest floor. The species range may be the sampling universe but it is common to refer to stands or habitat within the insect's distributional range as the universe. Any homogeneous stand of forest may therefore be considered a universe, so as to avoid stratified random sampling from a heterogeneous universe. However, in devising sampling methods i t is necessary to describe the universe more exactly, such as, trees -- con-sidering position of trees in the stand (edge, interior, open grown), characteristics of trees as units (height, crown, age, etc.) and position in the crown (lower, middle, upper). Selection of the Sampling Unit It is assumed that the fascicles of needles provide the most appropriate unit. It is advantageous to base examination of foliage dis-tr i b i t i o n on the number of fascicles, and the distribution of the casebearer on the foliage„on the number of insects per fascicle. This unit of foliage is f a i r l y stable, small, reduces labour of recording in the f i e l d or laboratory, speeds calculation and contains eggs, a l l instars of larvae, and pupae. This unit closely satisfies the six c r i t e r i a for selection of the sampling unit laid down by Morris ( 1 955 ) , namely: (a) All units must have an equal chance of selection (b) The sample unit must be stable - 36 -(c) The proportion of insect population using the sample unit as a habitat must remain constant (d) The sample unit should be reasonably small so that enough units can be examined on a given plot to provide an adequate estimate of sampling variance (e) The sample unit should lend itself to estimates of absolute population (f) An important practical consideration is the ease of col 1ect ion. Timing of Sampling S ign i f i cance: After the sampling unit is selected, i t is necessary to decide how sampling should be timed in relation to l i f e history of the casebearer, its pattern of mortality and location of stages. Generally, a time interval is selected in which the insect is on oviposition sites, feeding sites, or i t is in a resting stage. This is preferable to a stage that is hidden (e.g. in the needle mines), relatively mobile, or easily disturbed (e.g. the moths). Sampli ng Intervals: Egg - at the time of completion of oviposition and commencement of hatching Larva (L^) - at the time of diapause and needle f a l l when larvae are firmly attached to the branch Larva (L^) - at the commencement of spring activity Pupa - just before emergence or after emergence. Here timing is important in assessing rates of mortality by parasites. - 37 -Fig. 6. a, b, c. Life stages of the larch casebearer sampled. a. Needle fascicles showing spring feeding damage to needle tips and pupa. b. Eggs on adventitious new needles. Courtesy of The Pacific Forest Research Centre. c. Overwintering cases on dormant larch twig. - ho -The calendar dates of sampling may vary widely from year to year. ' Therefore, the actual dates of sampling are dictated by prevailing rates of insect development. Determinants of these rates, may be obtained by spot sampling or by indices such as developmental curves or by phenology, such as degree days above a certain threshold temperature. Field Procedures Thrums (Fig. 7) Fifteen western larch trees were selected for casebearer sampling, 12 trees at Thrums (Fig. 7 ' ) and 3 trees at Sheep Creek. Each tree was numbered and tagged for future reference. These,trees were located with respect to position in the stand as follows: Interior stand trees (Nos. 1-4) Edge trees (Nos. 5"8) Open grown trees (Nos. 9-12) In clusters of trees (Nos. 13 — 15) ) Sheep Creek The few, scattered, 'open grown', trees were located along pathways into the stand. Increasing egg deposition sometimes occurs on trees along edges of stand openings. This is a function of (a) host discovery which is probably most evident when insect populations spread into new areas, and/or (b) more favourable microhabitat with more light, differences in quantity and quality of foliage, etc. on the exposed side. It is therefore suspected that edge trees tend to receive more eggs than do trees in the interior of the stand. - h] -The same trees were sampled through one insect generation. New trees were selected for the 1975 egg collection. Crown levels: The crown of each sample tree was visually divided horizontally into 3 levels of equal vertical length (lower, mid, upper (Fig. 8) ) and vertically into 2 halves (exposed and shaded). Two classifications of branches, relative to the degree of exposure to sunlight, were recognized: (a) branches fully exposed to light, such as, branches from isolated or open grown trees or from the sides of trees facing an opening, and (b) branches shaded by nearby adjacent trees, such as, in the interior of a dense stand; It was assumed that differences in light intensity affect variation in number of fascicles and number of insects per fascicle among the branches. Branches: To obtain information on the distribution of the casebearer on the branch, two whole branches were removed from each crown level, one from the exposed and one from the shaded side of the tree. The cardinal points were approximately: south for the exposed, and north for the shaded branches on stand edge and interior stand trees. Each entire branch was measured and cut into 3 equal sections. From each section two 6-inch (15 cm) lengths were cut at random, one from the main branch, and one from a lateral branch (Fig. 8 ) . This gave six 6-inch lengths per branch or thirty-six 6-inch samples per tree. The sample branch was removed by extendable pole-pruners with clamp attachments for the lowering of the branch with a minimum risk of dislodging insects from the sample. A 2k foot (7 -3 m) extension ladder was also used for reaching the taller trees. - kl -Larch Stand c ( ( ( Opening r Right-of-way for wire-line -To Castlegar Highway No. 3 -> To Nelson Figure 7. Diagram showing relative positions of sample trees at Thrums, B.C. F i g u r e 8. a. b. Division of tree crown ver t ica l l y and horizontally. Sample branch showing branch sections ( 6 ) selected. - 44 -Each sample unit was placed in a separate plastic bag 5x9 inches (12.7 x 23 cm), labelled, tied with "twistums", and transported to a laboratory at the University of British Columbia for cold storage at 5 C and counting or rearing of the various stages of the larch casebearer. Collections by two persons from the 15 sample trees from two areas took 3-4 man-days, depending on insect stage and weather. A return trip from laboratory to the f i e l d took 3-4 days. Collection dates, insect stages and sample sizes: 1. Pupal stage: 1 5 - 1 6 May 1974, 540 subsamples from 15 trees 2. Egg stage: 11-12 July 1974, 324 subsamples 9 trees 3. Initial larval stage: 2-3 November 1974, 432 subsamples from 12 trees 4. Final larval stage: 24-25 April 1975, 540 subsamples, from 15 trees 5. Pupal stage: 10-11 June 1975, 540 subsamples from 15 trees 6. Egg stage: 22-23 July 1975, 288 subsamples from 8 trees. Collections for the egg stage were taken in the fi e l d from 15 trees, but the high density of eggs and the low variability of egg counts allowed the use of fewer trees to obtain the.required level of prec i s ion. - 45 -Assessing the Tree The same branch samples taken for insect counts were utilized to determine the distribution of fascicles within the branch. As already described the foliated whole branch was divided into 3 equal lengths and one 6-inch (15 cm) section taken from the main axis, and one from a lateral of each of the 3 portions. It was required to know how the fascicles were distributed among the crown levels because there were differences in insect population density among the 3 crown levels. Accordingly, the total number of fascicles was recorded for each 15-cm sect ion. Sampling for Morphological Characteristics of Branches The following characteristics were studied by subsampling: (a) Number of needles per fascicle. This was measured by counting the number of needles on 10 fascicles per branch selected at random from the branch samples that were i n i t i a l l y used for insect population counts on June 1975, from trees Nos. 1, 3, 5, 6 and 7. (b) The longest needle per fascicle was measured to the nearest mm. Distribution of Foliage and Shoots The proportional distribution of foliage, needle fascicles, and shoots within the crown, and variations within various crown types, or locations in the stand were investigated for four crown types in accordance with Ives' (1959) classification for tamarack in eastern Canada, namely: - ke -Crown Type Character i st ic Ful1 crown Irregular S1ender High taper uniformly from bottom to top of tree e.g. Open grown trees a preponderance of branches on one side of the tree at different crown levels, e.g. tree along margin of stand trees with dead primary branches replaced by short branches originating from adventitious growth on the main stem; rare in western stands mainly in dense stands, considerable natural pruning from branch suppression e.g. interior stand trees. Defoliation . Measurements The assessment of damage to the trees in relation to defoliation necessitates recognition of, and allowance for, several determinant variables. These comprise: foliage age, location of foliage in the crown, time of defoliation and stage of leaf development. Accordingly, these variables were measured on the 15 sample trees. The larch casebearer feeds by mining the peripheral 1/3 or 1/2 of any needle and only occasionally the whole needle. The most noticeable and effective damage occurs in early spring at the time of needle growth. Therefore, June when larval feeding ceases, is probably the best time to estimate defoliation. Rating Defoliation Each 6-inch (15 cm) sample was given a numerical defoliation rating as follows: - kl -(1) For pupal samples 0 - Negligible - no visible damage 2 - Light - up to 25% of foliage damaged k - Moderate - 26-50% of foliage damaged 6 - Heavy - 5)"75% of foliage damaged 10 - Severe - over 75% of foliage damaged (2) For egg samples The estimate of defoliation was improved to give: (a) proportion of the number of needles per fascicle mined per sample unit (= Quantity). (b) proportion of the volume of needles per fascicle rendered non-functional by the insect per sample (= Volume). 0 = No Defoliation 1 = Negligible 2 = 5-15% 3 = 16-25% h = 26-35% 5 = 36-50% 6 = 51-60% 7 = 61-75% 8 = 76-85% 9 = 86-95% 10 = over 95% 11 = dead spurs Insect Counts and Accessory Information All data were recorded on field data sheets in a format directly convertible to automatic data processing (Appendix 1 _ 3 ) . - 48 -General Information, a l l insect stages. All data sheets included the following pertinent information: Tree No.^date of collection, area, crown class, crown level, exposure, and branch section. For each 6-inch (15 cm) branch sample the number of fascicles or live spurs in winter, dead spurs, and bases of side shoots were recorded. Pupal samples: Number of pupae on each 6-inch (15 cm) sample were counted and defoliation rating recorded. Records were kept also of moth and parasite emergence (Appendix 1). Egg samples: Sampling was carried out at about the time of completion of egg hatching. Eggs were counted and records kept of unhatched eggs, empty chorions, and ecluded eggs. The condition of eggs were distinguished as follows: Unhatched eggs - yellow-orange content will hatch eventually (live) - translucent white and 'collapsed', or discolored (dead) Empty chorions - transparent chorions, contents extracted probably by mites or hemipterous predators Ecluded or - grayish egg chorions often f i l l e d with frass giving a hatched eggs white green color i n i t i a l l y and darkening to a reddish brown color. The defoliation rating, number of eggs per fascicle, portion of needle on which egg was found (tip, mid and base), conditions of needle (sound or damaged), needle surface (upper or lower) on which egg was found, were also recorded. Counts were made by examination of needles under a dissection microscope, by rotating one fascicle of needles at a time. Before the task was completed most of the needles f e l l from the fascicle during the rotation process, and had to be examined individually. The procedure was - ks -tedious and time-consuming with the exception when only total egg counts were made. However, time could be saved by having an assistant do the record i ng. Overwintering Larval Samples This was the easiest and quickest insect stage to count, as most of the abscising needles had fallen off during collection. Counts were made of larvae on spu-rs, base of spurs, at nodes on bark and among the lichens when present, for each 15_cm branch sample. Checking Insect Counts Counts were checked to reduce percentage of insects missed, and depended on the relative concealment and size of each stage of the insect. In the prewinter and postwinter larval stages the insects were readily seen and checking was easy. After i n i t i a l counts the pupae were placed in individual p;lastic bags or vials for rearing and a second count was made after emergence of adults, the differences between counts were negligible. For eggs, which are small and easily missed, thorough checks were necessary. This was carried out through several re-examinations of the needles under the microscope or other magnifier. If the needles remained on the short shoot during examination for eggs, there was a 10 percent difference in egg counts between consecutive checks. However, when the needles dried out and f e l l - o f f the branches up to 20 percent difference in egg counts between consecutive checks occurred. - 50 -Rearing Methods Most of the rearing experiments were intended to yield information on casebearer or parasite emergence and therefore mortality of larvae or pupae on the different 15-cm branch samples taken from various strata in the tree crown. As most parasites emerges as adults during the pupal stage of the casebearer, it was generally found more satisfactory to rear mature larvae or pupae collected in the f i e l d , rather than earlier stage larvae. After i n i t i a l counting and removal of other insect species, each 15-cm section was placed in 5 x 9 inch (13 x 23 cm) plastic bags and tied at the neck with twistums. Rearing was carried out at room temperature at 20-23C (68-75F) and at 60-75% relative humidity. About 10% of the pupae collected were removed from the branch samples and reared as follows: (a) In shell vials 50 x 15 mm and 160 x 15 mm stoppered with cotton wool or cork, and (b) In gelatine capsules 20 x 5 mm. The other pupae were reared on the branch sections in 5 x 9 inch (13 x 23 cm) plastic bags in which the 15"cm samples were originally placed when collected i n the f i e l d . Experimental Observations Behavioural activities were observed in the f i e l d at the time of larval feeding as a casebearer, and at the time of adult emergence and mating (11-12 June 1974 and 15 June 1975). - 51 o Figure 9. Collection and rearing bag with 15_cm branch section. - 52 -ANALYSIS OF DATA The statistical techniques used for analyses of the data on each l i f e stage of the insect are briefly defined. Their relevance and limitations in describing population distribution and variability, and in estimating overall populations parameters are outlined. Frequency Distribution The frequency distribution of counts formed an important aspect of the quantitative studies of insect populations and received attention in the statistical analysis. Theoretical distributions have been fitted to the data for one or both of the following reasons: 1. To find a transformation in order to use the normal theory for statisti c a l analysis, such as the analysis of variance (it does not matter if the form of distribution fitted is particularly accurate (McGuire, et_ a_j_. 1957) ) . 2. To relate observed data to some theory of population growth or spread (requiring forms of distributions that are biological significant, such as the negative binomial (Anscombe, 1950), the Poisson (Skellam, 1952, McGuire, et a 1. 1957)-and Neyman type-A (Neyman, 1939). Determination of the theoretical distribution which best f i t s the set of observed values, and the chi-square test for goodness of f i t , required a modified form of the computer program written by Kozak and Munro (1963) at the University of British Columbia. This computer program fi t t e d the observed data to the four probabi1ity distributions commonly - 53 -recognized in forest sampling; the normal, Poisson, binomial and negative binomial. The parameters calculated from the data were: the mean, standard deviation, number of frequency classes, the value of probability P for the binomial distribution, and the constants k and P^ , for the negative binomial distribution. In its original form the program accepted a maximum of 2 0 frequency classes and this had to be increased to take 30 or more frequency classes for the present study. Fitting the Distributions The counts were analysed in a frequency distribution showing the number of 6-inch ( 15cm) branch sections containing 0 - 0 . 5 , 0 . 5 - 0 . 1 , 0.1-0.15, ...insects per fascicle for a given l i f e stage. If the insect is randomly distributed over the sampling universe, the distribution of insect per unit will approximate a Poisson series where the variance (S ) of the population equals the mean (x). The appearance of a frequency distribution may merely reflect an artifact of experimental design. For example, as the size of sample unit and/or population density increase, the zero values tend to disappear, and the distribution appears to approximate a normal bell-shaped curve. Often there are more zeros and high values than expected, and as a result 2 — S >x. This departure from randomness is referred to as "overdispersion". The negative binomial is the most useful distribution that has been applied for overdispersed insect counts. This distribution is described by 2 parameters, the mean and the exponent k_ which is a measure of aggregation. Generally values of k_ are in the region of 1 or 2 . As they - 54 -become larger (i.e. as S approaches x) the distribution approaches the Poisson. Fractional values of j< lead into the logarithmic series. The value of J< may be computed by several methods (Anscombe, 1949, 1950; Bliss and Fisher, 1953; Debauche 1962; Legay 1963 in Southwood I966; Katti and Gurland 1962). Only one method is presented here: -2 where S = variance where E = the sum of f.= frequency of the i 1 " ^ class m = number of frequency classes t h x.= mid-point of the i class (e.g. no. of insects) N = no. of observations (Ef.) 1 The efficiency of this estimate of J< is reliable only at low density popu1 at ions. Transformations m E i = l m E i = l (f.x.)' VN N-l It is often necessary to transform observations before analys so as to more nearly satisfy the assumptions of the usual statistical techniques. The normal or Gaussian distribution is not of interest as - 55 -means of deciding dispersion of insect population. Its importance lies in the fact that for most statistical methods the distribution must be normal and possess the associated properties that the variance is independent of the mean, and its components additive (Southwood, 1965). In order to meet the assumptions of analysis of variance, the data are transformed. Thereby the observed data are replaced by a function whose distribution is such that it normalizes the data or stabilizes the variance. Different kinds of transformation have been devised for the purpose, the applicability of which depends on the peculiarities of the data. Among the more usual transformations tried are: (a) 7x, the square root transformation (Bartlett, 1936) (b) Jx + 1/2 Bartlett (1936) (c) Log^^ (X + 1 ) , the logarithmic transformation (Williams, 1964) (d) XP, Taylor power law (Taylor 1961, 1965) Taylor's Power Law The distribution of individuals in natural populations is such 2 that the variance (S ) is not independent of the mean (m). Taylor (1961) 2 from the examination of several sets of samples found the S appears related to the nn as they tend to increase together when plotted and to obey a simple power law. c 2 b S = am where a and b are characteristics of the population in question. - 56 -The s e r i e s o f means and v a r i a n c e s n e c e s s a r y t o c a l c u l a t e a_ and £ was o b t a i n e d f r o m s e t s o f samples from d i f f e r e n t t r e e s . The 2 v a l u e s o f m and S c a l c u l a t e d from the raw d a t a a r e p l o t t e d on a l o g / l o g s c a l e . The v a l u e o f a_ and b_ a r e computed by l i n e a r r e g r e s s i o n i n l o g a r i thms. 2 l o g S = l o g a + b l o g m where l o g a_ and b_ a r e i n t e r c e p t and r e g r e s s i o n c o e f f i c i e n t r e s p e c t i v e l y . 2 T a y l o r (1961) has shown t h a t t h e r e l a t i o n s h i p S = a m g i v e s r i s e t o a system o f t r a n s f o r m a t i o n s d e r i v e d from t h e a p p r o p r i a t e v a r i a n c e s t a b i l i z i n g f u n c t i o n f ( m ) : f(m) = Q_jm" b / 2 dm ( T a y l o r , 1965) From t h i s , t h e q u a n t i t y t o be a n a l y z e d (Y) i s t r a n s f o r m e d from the o r i g i n a l c ount (X) by t h e e x p o n e n t i a l e x p r e s s i o n : Y = X P where X = the o r i g i n a l number, Y = t h e t r a n s f o r m e d v a l u e and p = (1 - 1/2b) I f p = 0 a l o g a r i t h m i c t r a n s f o r m a t i o n s h o u l d be used; I f p = 0.5 square r o o t s a r e a p p r o p r i a t e ; I f p =-0.5 r e c i p r o c a l s q uare r o o t s a r e r e q u i r e d ; and I f p =-1.0 r e c i p r o c a l s a r e t o be used (Southwood, 1966). The e x p o n e n t i a l e x p r e s s i o n was used i n t h i s s t u d y but w i t h t h e m o d i f i c a t i o n o f a d d i n g a "C" c o n s t a n t t o the v a r i a b l e b e f o r e r a i s i n g i t t o t h e power o f p: Y. = (X. + C ) P I I - 57 -This constant is needed when zero values are frequent in the data to be transformed and it could be between 0.5 and 2.0. After trying several values the constant of 1.0 was found to be the best for the present study. The use of transformations may lead to problems in the comparison of means, which may be based on different transformations. Justification exists therefore for not transforming the data unless it seriously violates the conditions necessary for the analysis of variance (LeRoux and Reimer, 1959)- For comparison, in this study analysis of variance was also computed on the untransformed data. The adequacy of a transformation in stabilizing the variance can be tested graphically or by calculating the correlation coefficient of the two terms (means and variances) (Harcourt, 1961b, I 9 6 5 ) ; and also by Bartlett's test for homogeneity of variances. Analysis of Variance or ' F ' Test The analysis of variance is used advantageously in research where quantitative data are measured, and permits determination of the spatial and temporal factors which exert a significant effect upon insect density. It is the process used for partitioning the sum of squares £ (x-x) into components which are thought to be related to differing causal circumstances. The objective is to test the hypothesis that a number of population means are equal. Therefore, the procedure is one of determining how much of the variation in the observations is - 58 -due to population differences, and how much to random variability. Comparison of the contribution of these 2 kinds of variations allows the determination of the importance of population differences. If the assumptions underlying the statistical techniques are not f u l f i l l e d , the test of significance will be affected. Four assumptions are usually necessary for the analysis of variance (Piatt and Griffiths, 1964) : 1. The experimental errors must be independent, may be f u l f i l l e d by assigning treatments at random. 2. The samples are from normally distributed populations, i f non normal can usually be rectified by transformation. 3. Variances within each treatment are equal, as the error variance in the analysis is a pooled error and each treatment contributes to i t . 4. Treatment and environmental effects must be additive, i.e. the treatment and replication effects in a randomized block design must not interact. It should be noted that it is not certain that a l l of the assumptions are met, even with transformation. Quenouille (1950) observed that, in t-test and the variance ratio test, it is usually more important to meet assumptions of (2) and (3) only. Regression and Correlation Calculations of regression and correlation were applied for describing the statistical relationship between means and variances, to determine i f data transformation was necessary. There are several - 59 -simple methods of studying relationships among variables (the scatter diagram, freehand trends, and the method of selected points). However, for statistical analysis, the method of least squares for f i t t i n g re-gression lines is most reliable. Regression offers a useful approach to the study of simultaneous variation of 2 (or more) variables. With regression the m variables are differentiated into (m-1) independent variables and one dependent variable. The problem is to find the values of a_ and J), in the equation: Y = a + b„X + b X + + b X 2 1 2 2 mm which minimizes the sum of the squared deviations between predicted and A observed values of Y where; Y is the predicted value of Y, a_ is the intercept which fixes to position of the line, and b. are the regression coefficients. The correlation coefficient (r) measures the degree of linear association between 2 variables. That is, it gives an evaluation of the mutual relationship between 2 variables even when no cause-and-effeet relationship is known. The defoliation rating was recorded for the pupal and egg stages and these were correlated with the number of insects (eggs) per fascicle. The correlation coefficient can vary from -1 to +1. A coefficient of 0 indicates no linear correlation and a coefficient of 1 a perfect linear correlation. When variables are jointly affected because of external influences, correlation offers a logical approach to the analysis of data. In correlation analysis, random pairs of observations are assumed and information is obtained about a joint relationship between 2 variables. While in regression analysis only the dependent (Y) variable is assumed to be random, and regression almost implies a cause-and-effeet relationship. - 60 -Coefficient of Determination (r^): This coefficient is a measure of the amount of variation in Y attributable to the independent var iable X. The Number of Samples The number of samples required for estimating the mean densities of insect stages depends on the degree of precision or accuracy required, the amount of interbranch variance that exists within trees, the number of branches sampled per tree and the amount of variance that exists between trees in the:.stand. Several complicated methods for calculating sample size have been used, among which are the following five examples: (a) Where sampling is necessary at two levels, e.g. between and within trees, the number of units (Nt) that need to be sampled at the higher level e.g. tree (LeRoux and Reimer, 1959; Harcourt 1961a from Southwood 1965) is given by: (S2s/N ) + S 2 Nt = § P-(XD)2 2 where Ns = the number of samples within habitat unit (trees), S^  = variance 2 within the habitat unit (within tree variance), Sp = variance between habitat unit (= intertree variance), x = mean per sample (calculated from transformed data and given in this form), and D = the required size of the standard error expressed as a decimal fraction (0.1 normally) of the mean. - 61 -(b) If the dispersion of the population has been found to be well described by the negative binomial the desired number of samples is given by: N = 1/x + 1/k where x = mean, k = the dispersion parameter of the negative binomial, D = the required size of the standard error expressed as a decimal fraction of the mean (Rojas, 1964). To find the combinations of Nt (No. of trees) and Ng (No. of sample units per tree) that will provide equal sampling precision, Morris (1955) used: where St = the variance component of trees, 2 Sc = the variance component for sample units (branch) within trees, Sy = the standard error of the mean, set at various prescribed percentage of the mean. 2 2 St . N + Sc Nt = s (Sy) 2 . N 2 (c) Sampling can be based on the measurement of the frequency of occurrence of an organism, e.g. the frequency of occurrence of casebearers on a fascicle (Oakland, 1953; Henson, 1954). - 62 -An e s t i m a t e o f sample s i z e can be made by f i r s t o b t a i n i n g an a p p r o x i m a t e v a l u e o f the p r o b a b i l i t y o f o c c u r r e n c e o f an a t t r i b u t e (p) e.g. i f i t i s found t h a t 35% o f t h e f a s c i c l e s have c a s e b e a r e r s t h i s p r o b a b i l i t y i s 0.35. The number o f samples (N) i s g i v e n by: J. N = 1 P q D 2 where t = a q u a n t i t y depending on t h e no. o f samples and degree o f c o n f i d e n c e , and i s o b t a i n e d from t - t a b l e s , p = t h e p r o b a b i l i t y o f o c c u r r e n c e , q = 1-p, D = r e q u i r e d s i z e o f t h e h a l f c o n f i d e n c e i n t e r v a l about the e s t i m a t e d mean. (d) I f i t i s found t h a t t h e f a s c i c l e s a r e d i s t r i b u t e d d i f f e r e n t l y i n the d i f f e r e n t p a r t s o f t h e . h a b i t a t , they s h o u l d be sampled w i t h p r o b a b i l i t y p r o p o r t i o n a l t o t h e v a r i a n c e s (Henson, 1954). (e) F i n a l l y , when t h e number o f t r e e s per p l o t t o be sampled i s v e r y l a r g e , the r e q u i r e d number o f t r e e s t o be sampled f o r each s t a g e can be c a l c u l a t e d (from ana 1 y s i s o f v a r i a n c e ) a s : MS t • N t - - M D 2 where Nt = r e q u i r e d sample s i z e (no. o f t r e e s ) , D = d e s i r e d s t a n d a r d e r r o r o f the mean i n the u n i t s o f o b s e r v a t i o n s (not i n p e r c e n t ) , 1 = l e v e l s w i t h i n each t r e e , r = samples (branches) w i t h i n each l e v e l , MS = v a r i a n c e components f o r t r e e s . - 63 -Allocation of Optimum Sampling Effort The findings in this study are integrated into a plan for future sampling of the larch casebearer. It deals with how to take the samples, how many are needed, and where these samples should be taken to get the most precise estimates with the available resources. The following questions were asked: 1. What precision was obtained? 2. What intensity of sampling will be needed in future work to obtain a precision of 10 or 20 per cent of the mean? 3- How should these samples be distributed on a tree crown and among the trees on a plot? h. How many trees are needed per plot? Data Preparation and Analysis Procedures Larch casebearer data were punched on IBM cards and analyzed on the University of British Columbia IBM 370 Model 168 electronic computer, using several programs including the following: UBC MFAV Analysis of Variance/Covariance. This program computes an analysis of variance for a wide variety of designs. MFAV can also perform Duncan's Multiple Range Test or test contrasts on the means for a particular source of variation. Degrees of freedom, sums of squares, means of squares, variance ratios (F-test) and means of variables are tabulated, involving a maximum of 9 factors with up to 50 levels in each factor. Provision is made for any model, equal or unequal replications and selection of any error term desired. - 64 -UBC (Forestry) MREG Multiple Regression. This MREG program is in Fortran IV for IBM 7040 Data Processing System (Kozak and Smith 1965) modified for the IBM 370/168 System. A series of separate tabular values is calculated and a maximum of 70 variables may be analyzed at one time. The Basic Unit of Sampling The basic unit of sampling was the 6-inch (15 cm) branch section and development of the sampling technique was based on analyses of inter- and intra-tree variability of the number of insects per fascicle/15~cm branch  sect ion (q). This is a compound variable as both the numbers of insects (Y) and the numbers of fascicles (x) per 15 cm. branch section are variables. The individual observations are ratios of two variables (q. = Y./x.). The ratio estimator used was the "means-of-ratios" estimator (q = Z n q./n). r 1 In general, the sampling mean and variance of ratio estimators are biased, the bias, however, is usually negligible for large samples. Ratio estimates of population means or totals may or may not be more "stable" (less variable) than the corresponding values obtained by simple expansion (i.e. by computing Y = EnY./n as opposed to Y = q x x). The variability of Y relative to s i r r that of Y s will depend on the sign and size of the correlation coefficient between Y. and x. and the coefficient of variation of these variables. 1 1 The use of ratio estimators in this study did not affect the statistical analyses, but would affect the calculation of total number of insects and fascicles per tree. - 65 -RESULTS AND DISCUSSION Frequency Distributions The distribution of the larch casebearer in the volume of the tree crown could be in the form of gradients from top to bottom and from periphery to center. These gradients are mainly the result of behaviour towards light, gravity, temperature, moisture or available sites for feeding or oviposition. However, within this overall gradient the insect may be distributed either randomly or contagiously. Several ways of determining the spatial distribution patterns of the larch casebearer were carried out, namely, frequency distributions and chi-square tests, the ratio of the variance to the mean, k_ parameter of the negative binomial and b_ of Taylor power law. The larch casebearer counts were summarized by insect stages showing the number of 6-inch (15cm) branch sections within the density class limits of 0-0.05, 0.05-6.10, 0.10-0.15, ... etc. casebearers per fascicle "(Table 1). The frequency distributions of the number of insects (except egg stage) per fascicle did not follow the normal distribution, but were strongly skewed toward the left (Figs. 10, 11). Observed data fitted the negative binomial distribution for larval and pupal stages and approached the normal distribution for the egg stage (Appendix k, Tables 1-6). In the process of f i t t i n g the negative binomial, any "extra" modes were assumed to represent random variation. - 66 -Table 1. Frequency distributions of numbers of insects (larch casebearer) per fascicle for the l i f e stages sampled. Class Limi ts Pupa 1974 Number of Insects per Fascicle Pupa 1975 Egg 1974 1975 Larva 1974 3 Larva, 1975 * o.oo- .05 255 19 17 103 204 316 • 05- .10 79 4 7 42 78 74 .10- .15 58 17 9 49 68 47 .15- .20 30 10 7 31 33 25 .20- • 25 33 10 9 31 29 17 • 25- • 30 23 16 17 34 22 14 • 30- .35 14 20 11 24 19 10 .35" .40 9 10 11 13 14 3 .40- • 45 11 13 12 14 8 9 .45- • 50 8 10 8 13 6 4 .50- .55 13 27 18 9 16 5 • 55- .60 0 20 7 8 6 4 .60- • 65 2 9 10 7 7 2 .65- • 70 0 18 8 5 2 2 • 70- .75 1 10 10 4 3 0 .75- .80 0 15 10 6 2 3 .80- .85 0 12 14 1 1 0 .85- • 90 0 8 6 1 2 0 .90- • 95 2 11 7 4 1 0 • 95- 1.00 0 1 3 0 0 0 1 .00- 1.05 1 16 13 10 10 1 1.05- 1.10 0 4 8 2 9 1 1.10- 1.15 0 4 10 1 0 0 1.15- 1.20 0 3 3 2 0 0 1 .20- 1 .25 0 3 2 2 0 1 1.25- 1.30 0 5 4 2 0 0 1.30- 1.35 1 2 8 0 0 0 1.35- 1 .40 0 2 2 2 0 0 1.40- 1.45 0 1 4 1 0 0 1.45- 1.50 0 24 33 1 1 0 2 Total: 540 324 288 432 540 540 0/ -(a) 2 0 0 o c <u Z3 cr <D 1 0 0 1 0 Pupae l~]h 2 5 2 2 0 6 1 2 0 51+ I . I . I 0 1 2 3 1 + 5 6 7 8 9 No. pupae per 15-cm branch 0 . 0 2 5 0 . 2 2 5 0 . U 2 5 No. pupae per fascicle 0 . 5 7 5 u c rj C T <U l_ U . 3 0 1 2 5 6 2 0 3 1 5 « * 1 0 5 5<» <b> Si. Pupae '75 2 0 3 1 0 3 0 . 0 2 5 0 . 2 2 5 0 . 1 ( 2 5 Figure 10. Frequency distribution of Insects per fascicle and per 15-cm branch section, a) Pupae x~Jk\ b) Pupae '75. - 68 -Eggs per fascicle 1974 1(7 r 40 >-O c cr 3 0 2 0 1 0 5 0 . 0 5 0 . 4 5 0 . 8 5 1.2 5 1.55 Eggs per 15-cm branch section 27 20 r 1 r 1 , '' 1 1 1 1 1 . 1 1 1 1 1 1 . 1 I ' ' 1 I . 0 2 it 6 8 1 0 1 2 14 1 6 1 8 Class Interval F i gure 11. Frequency distribution of eggs per fascicle and per 15-cm branch section. - 69 -The Variance - Mean Ratio 2 The variance (S ) of the number of insects per fascicle calculated for each tree sampled was related to the mean (x) , as in a Poisson form of distribution (Fig. 12a). However, in most trees larch casebearer densities exceeded variance, and more tree samples were found in both t a i l s of the frequency polygon than are expected in a Poisson 2 distribution (where the S = x). 2 The variance/mean (S /x) ratios (an Index of Dispersion) of the independent variables (egg, larvae, and pupae per fascicle) were low, indicating the uniformity of spatial pattern. This was due to the fact that the total number of insects per 15-cm. branch section was divided by the number of fascicles on that section. When the 15-cm. branch section was made the sample unit (numbers of insects per branch section), the variance/mean ratios of a l l stages of the casebearer were high, indicating the aggregative nature of the data. For the egg stage (1975) the variance and mean per tree was largely independent, and the distribution approached normality. This was confirmed when variance was plotted over the mean (Fig. 12b). The counts for the egg stage in 197** apparently fitted the negative binomial but approached the normal distributions as indicated by the chi-square tests. 'k' as an Index of Aggregation The frequency distributions of the various stages of the larch casebearer were aggregated or clumped among trees as indicated by j< of the negative binomial distribution (Table 2). The value of k as a measure of - 70 -a. Pupa ' 7 5 Variance on Mean 0 . 1 4 0 0 . 1 1 2 0 . 0 8 4 0 . 0 5 7 0 . 0 2 9 1 -_L _1_ 0 . 0 2 9 0 . 0 5 3 0 . 0 7 6 0 . 0 9 9 0 . 1 2 3 0 . 1 4 6 b. Egg ' 7 5 0 . 4 4 9 0 . 3 9 0 0 . 3 3 2 r 0 . 2 7 4 f 0 . 2 1 6 t *-= > s - j r ^ s r - i ^ r r-^r-= r—J= 0 . 5 2 3 0 •. 6 2 8 0 . 7 3 4 0 , 8 3 9 0 . 9 4 5 1 , 0 5 0 Figure 12. The relationship between mean number of insects per fascicle and variance, a) Pupa llk; b) Egg. - 71 -dispersion can range from zero where aggregation is extreme, to infinity which defines a purely random distribution of counts. In practice however, any large value of k_ indicates an approach to randomness (Waters, 1 959 ) . The higher j< values (3 .3 ) for the egg stages at low population density reflects an i n i t i a l random tendency in new infestations. The lower Rvalues for larvae and pupae reflect a later aggregative tendency even in light infestations, and the retention of this characteristic due to mutual attraction at the particular times of collection (Table 3 ) . Possible causes will be noted subsequently in this manuscript. Table 2. Effect of development of larch casebearer during a single generation on estimate of the parameter k_ and _b for its immature stages, Thrums, B.C. 1974-75-Mean density Value : Of Date Stage Recorded per fascicle k_ b^  May 1974 Pupa 0.1218 0.450 1.3577 July 1974 Egg 0.6365 3.347 1.1873 Nov. 1974 Larva (L ) Fall 0.3H5 0.630 1.8149 April 1975 Larva (L^) Spring 0.1855 0.475 1.8570 June 1975 Pupa 0.1067 0.257 1.7190 July 1975 Egg 0.7132 3-245 0.4944 Unfortunately, k is somewhat unstable, as it frequently increases with the mean. Therefore, i t is advisable to stratify f i e l d data wherever possible in order to improve estimation of j< (Harcourt, 1963) -- 72 -Table 3- Estimates of J< of the negative binomial for each insect stage per fascicle and per 15-cm. branch section. Unit of Egg Egg Larva^ Larva^ Pupa Pupa Observation 1974 1975 1974 1975 1974 1975 Fasc icle/15-cm. branch_ sect ion x 0.629 0.713 0.300 O.I85 0.122 0.101 k. 3.347 3-245 0.630 0.475 0.450 0.258 15-cm. brancjn sect ion x 7.856 8.535 3-104 1.835 1.378 1.093 k 2.5H 2.584 1.069 0.891 0.720 0.507 It is evident that the Rvalues are affected by size of the observational unit (Table 3 ) . This additional source of heterogeneity may be attributed to the different number of fascicles per unit branch length at the various locations on the branch, and the aggregation habits of the prewinter and postwinter larvae and the pupae. When the negative binomial series was fitted to counts of casebearer per fascicle per 15-cm branch section, and counts per 15""cm. branch section, there were no significant differences between observed and expected values. ;l t is concluded that the negative binomial gives an adequate ; description of the frequency distribution of counts of prewinter and postwinter larvae and pupae of the larch casebearer. Taylor (1961 and 1965) contends that the sta t i s t i c b_ is a true "index of aggregation" and that a_ depends largely on sampling or computing characteristics. The index of aggregation b_ is a true population st a t i s t i c describing an intrinsic property of the organisms with a continuous graduation from: - 73 -near regular (b<l) through random (b = 1) to highly aggregated (b>l) In the present study the aggregative tendency of postwinter larvae (L^)and pupae appeared to be similar as the Rvalues for larvae^ and pupae (1975) did not differ significantly (Table 4). The t-tests calculated as per example below,'indicated that the 'b' values for al l l i f e stages sampled with the exception of the egg stage were significantly greater than 1. Thi s i result indicated high aggregation in al l larch casebearer 1 ife stages sampled except i eggs. Table 4 . Regression and correlation mean for a l l stages of the of log variance on larch casebearer. log Regress ion log on log x Var iable a b r P = 1 -1/2b Pupa (1974) -0.433 1 .358 O.96O 0.321 Egg (1974) -0.502 1 .205 0.816 0.398 Larva L^ 0.306 2.351 0.906 -0.175 Larva L, 4 0.102 1.857 0.906 0.0715 Pupa (1975) 0.197 1.711 0.927 0.1445 Egg (1975) -0.490 0.494 0.414 0.753 Example: Calculation of t-test for prewinter larvae. t = where b = regression coefficient bb G = 1, the number to be compared with and SE, = standard error of b b _ 1.857 - 1 _ .8570 = . „ q * ~ 0.2408 .2408 - 74 -From t-table with 13 df (f - 2 ), t = 3-012 at p = 0.01 so that b = 1.85 differs significantly from 1. The egg counts for 197** showed a diminished aggregative tendency and approached a random distribution as b_was only slightly significantly greater than 1; whereas eggs collected in 1975 tended toward a regular distribution. According to Taylor (1965) the power law appears to hold good down to low densities (x>l) in material examined. This implies that populations aggregated at low density tend to become regular when density increases, or vice versa. According to Taylor the law may break down eventually, or perhaps the concept of aggregation (S > x) is inappropriate at these low levels. Discussion on Distribution The larval and pupal data fitted the negative binomial, and the value k gave a measure of dispersion. The smaller the value of k, the greater the extent of aggregation, whereas a large value (over about 8) would have indicated a Poisson (random) distribution. The distribution of the egg stage in 1974 more closely fitted the negative binomial distribution than the normal distribution, but in 1975 i t followed the normal dis-tribution more closely. This was probably due to the much heavier defoliation in 1974 which resulted in the departure of suitable needle fascicles on branches from the normal distribution (discussion under the section on foliage). - 75 -The aggregation recognized by the negative binomial was due partly to the active aggregation by the larval stages prior to and after winter dormancy, and just prior to pupation, and partly to some heterogeneity of the environment such as distribution of needle fascicles, microclimate and natural enemies. The suitability of the negative binomial in describing frequency distributions of larch casebearers does not mean that aggregation is in any sense explained. As the negative binomial has been deduced from a number of widely contrasting hypotheses regarding the mechanism of dispersal, some examples are given by Waters and Henson ( 1 9 5 9 ) • The negative binomial distribution may have arisen through the aggregation tendency of prewinter larvae, postwinter larvae and pupae, or through statistical artifacts. Biological aggregation was caused by preferential responses to external stimuli, inter or intra-specific interactions, or reproductive behaviour as discussed below. Statistical artifacts may arise through sample unit size, shape and density (Waters and Henson, 1 9 59 ) or by combining samples from a number of random or non-random distributions (Bliss, 1 9 5 8 ) . Accordingly, the explanations for aggregation in the various insect: stages are discussed below. Egg Aggregation. Aggregation could be due to a behavioural cause, as the females tended to deposit a number of single eggs close to one another on the same needle or needle fascicle. Some needles contained up to h eggs, and fascicles up to 8 eggs in 1 9 7 4 . Alternatively, aggre-gation might be owing to the heterogeneity of the environment, in which only certain areas are suitable for oviposition, as the females con-centrated their egg deposition on undamaged needles of old growth. - 76 -However, when defoliation is light or negligible eggs are deposited singly and scattered on fascicles over the entire branch: Larval Aggregation. The prewinter larvae are highly aggre-gated as they settle down for winter dormancy. Shoot spurs contained up to 9 larvae per spur in moderate population densities. The postwinter larvae were s t i l l aggregated after winter mortality and one week of insect activity. The larvae spend the f i r s t week molting and occasionally enlarging their cases. Later on they become more dispersed as feeding progresses. Pupal Aggregation. Although the larch trees were lightly to moderately defoliated, between 70 and 8S% of the fascicles contained no pupae. It is apparent that there'was either heavy mortality or aggregation in the late larval stage. Although pupal numbers were low on most trees, mature larvae were sufficiently attracted to each other prior to pupation, so that, there was occasionally more than one pupae per needle fascicle. Up to 5 pupae per fascicle were observed. Transformation of Data Before statistical tests, such as the analysis of variance, could be applied to the data of insect stages, the highly skewed frequency distributions had to be transformed to meet the assumptions of the tests. This was necessary because the frequency distributions of the number of insects per fascicle, the variable to be analyzed, did not follow the normal distribution (except eggs 1975) . Secondly, transformation was - 77 -required because the variance of the number of insects per fascicle calculated for each tree sampled was highly related to the mean. The most obvious departure from normality was the strong correlation between the variance and the mean (Table 5 ) . The larger the absolute value of r_, the closer the points will f i t the line. Table 5- The correlation between mean and variance for the l i f e stages of larch casebearer. Stage Correlation coefficients (r) Untransformed Pupae 1974 0 .969 * Egg 1974 0.816V: Larvae 1974 0 .946 * Larvae 1 975 0.688* Pupae 1 9 75 O . 6 8 5 * Egg 1 975 0 . 2 0 3 ns * Significant (P < 0 . 0 1 ) ns = not significant (P > 0 . 0 5 ) The exponential transformation x P suggested by Taylor (1961) , where p is the Taylor power for transforming aggregated biological data, is the most applicable in these situations, but with the modification of adding a "C" constant to the variable before raising it to the power of J D: Y. = (X. + C) P. This constant is needed when zero values are frequent in the data to be transformed and it could be between 0 . 5 and 2 . 0 . The constant of 1.0 was found to be the best f i t in the present study. In the above transformation p_ is defined by the relationship between the mean and the variance. The values of p for the present study were: - 78 -1974 1975 Pupa 0 . 3 5 3 8 ( 0 . 3 ) 0 .1445 ( 0 . 1 ) Egg 0 . 4 0 6 3 ( 0 . 4 ) 0 . 7 5 3 0 ( 0 . 7 ) Larva 0 . 0 9 2 5 ( 0 . 1 ) 0 . 0 7 1 5 ( 0 . 1 ) The above transformation eliminated the dependency of the variance on the mean and also tended to normalize the frequency distributions. - 79 -FACTORS AFFECTING THE DISTRIBUTION OF CASEBEARER Source of Variation in Population Estimate The significant variables (branch section, level and trees) generally had very high F-ratios and their consistent significance in most analyses (Appendix *t, Tables 6, 7 and 8) substantiated the fact that these three variables contributed most to the estimated density variation of eggs, larvae and pupae of the larch casebearer among or within trees. Population means are expressed in terms of the original variates. Inter-Tree Variation For a l l three l i f e stages the number of insects per fascicle was significantly different (0.01 probability level) from tree-to-tree, so that the total insect population sampled was different from tree-to-tree. This difference could be due to some trees having more fascicles than other trees and therefore fewer insects per fascicle, but will be discussed under the section on foliage. The average number of insects per fascicle was slightly greater on the edge trees than on the interior or open grown trees, but this difference originated from one tree only (tree No. 8) for prewinter larvae and postwinter larvae and (tree No. 5) for pupae collected in 197*+ - Open grown trees appeared to have more eggs than trees in the other stand positions in both 197*t and 1975 (Appendix k, Table 9 ) . The tree-to-tree variation within the groups was more significant for a l l stages than the variation between categories of trees related to position in the stand (i.e. interior stand trees, edge trees, open grown trees, and clusters). - 80 -Intra-Tree Variation 1. Crown Level Variation The three crown levels, namely the lower, mid and upper third, showed s t a t i s t i c a l l y significant differences in insect densities for al l three insect stages (Fig. 13 and Appendix 4, Tables 6, 7, 8 ) . For the pupal stage in 1974 the number of insects per fascicle was significantly (P = 0.01) higher in the rhid crown level than in the upper crown level, but no differences was found between the lower and the upper levels. There was no significant differences between levels for the pupae in 1975. For eggs in both 1974 and 1975 the density was significantly (P = 0.01) higher in the mid and upper levels than in the lower level on a l l trees; but no difference was found between the mid and upper levels. Pooled data for a l l trees showed the highest density on the upper crown level. For the prewinter larvae, the number of insects per fascicle was significantly (P = 0.01) higher in the mid crown level than in the lower level, but no difference was found between the mid and upper levels. This appeared to be similar to the egg stage, but when the densities were pooled for al l trees, density in the mid level was higher than in the upper level. However, when the 15-cm.branch was taken as the sampling unit, instead of the fascicle, the relationship between levels was similar to that of the egg stage. For the postwinter or spring larvae the number of insect per fascicle was significantly higher in the lower crown level than in the mid and upper levels when a l l the trees were pooled. This was attributable to the fact that a l l the edge trees in the stand had significantly (P = 0.01) higher numbers (especially tree No. 8) in the lower level than in the mid and upper levels; whereas on the average, the interior stand and open grown trees had slightly more insects on the average in the upper and mid levels. There were no differences between the mid and upper levels When the trees were pooled. - 81 -These conclus ions can be drawn from the comparison of the average densities per crown level (Fig. 13) but they are not true for every tree (Figs. 14a, 15a, 16a and Appendix 4 , Tables 10, 11). In fact, there was a significant interaction between tree and crown level for the larvae (prewinter and postwinter), indicating a different trend from tree to tree. The trend for the postwinter larvae seems to be affected by position of trees in the stand as explained above. While for a prewinter larvae, the lower level had the least density on 8 of 12 trees. For the egg and pupal stages the interaction between a l l trees and crown levels was not significant, but the trends were different from one tree to the other. However, for eggs, 8 of 9 trees in 1 9 7 4 , and 7 or 8 in 1975 had fewer eggs per fascicle in the lower than in the mid or upper 1evels. 2. Exposed and Shaded Branches This is more important at mid and low levels of interior stand trees or edge trees rather than open grown trees. The number of prewinter and postwinter larvae, pupae (1974) and eggs (1975) did not, on the average, differ significantly between the exposed and shaded branches when the data for a l l trees were pooled. However, densities were slightly higher on the exposed branches for larvae and pupae ( 1 9 7 4 ) , and on shaded branches for eggs (1975) on 7 of the 9 trees. The number of eggs ( 1 9 7 4 ) and pupae ( 1 9 7 5 ) per fascicle was significantly (P = 0.01) higher on exposed branches than on the shaded ones (Fig. 13, Appendix 4 Tables 6 , 8 ) . However, the tree-to-tree variation between the exposed and shaded branches (Figs. 14, 15, 16 and Appendix 4 , Tables 1 2 , 13) indicates that the trend toward higher or lower numbers :between the exposed or shaded branches, respectively, is not consistent from one tree to another. - 82 -\ , • ,_ Lower Mid Upper Fig. 13. Number of insects per fascicle by Crown Position. Figure A. Figure B. ig. .14. Number of pupae per fascicle; a) by tree and-CTOwn-posi-t-ion-;-" b) by exposure, branch position arid tree, 1974. F i g . 15. Number o f eggs pe r f a s c i c l e ; a) by t r e e and crown p o s i t i o n ; . _i_ •. _ - . - — —b-)—by—exposure, -br-aneh--pos-i-t-ion-and - t r e e , ' 1 974. 0. tl , , a) \by . tree .and crown position; b) by exposure, branch position and tree, 1974. - 86 -- 87 -3. Main or Side Branches Significantly more pupae in 1974 and 1975 per fascicle occurred on the side branches than on the main branches, but significantly more overwintering larvae per fascicle occurred on the main branches than on the side branches. In 1974 the egg population was, on the average, higher on the main branches, while the postwinter larvae were more abundant on side branches, but these differences were not significant s t a t i s t i c a l l y on the two types of branches (Fig. 17, and Appendix 4 , Tables 14, 15). But in 1 9 7 5 . the number of eggs per fascicle was significa higher on the main branches than on the side branches. However, notwithstanding the overall trends, it would be a mistake to draw a general conclusion from the above results, because the trend changes from tree-to-tree (Figs. 14b, 1 5 b , 16b and Appendix 4 , Table 12), and for the prewinter larval stage, it changes for the three different crown levels (Appendix 4 , Table 9 ) • 4. Horizontal Crown Position Fig. 18, Appendix 4 , Tables 16, 17, 18, 1 9 , 20 and 21 indicate the trend of number of insects per fascicle by horizontal crown position (stem to periphery) for the three l i f e stages sampled. In general, i t can be concluded that the population density close to the stem is low for a l l stages, and increases for the mid and outer parts of the crown. However, this trend is inconsistent from tree-to-tree (Appendix 4 , Tables 16, 1 7 , 18, 19, 20 and 21). The 1 9 75 egg population was more evenly distributed through the horizontal crown position, possibly due to light defoliation which resulted in the availability of sound needles throughout the branches. - 88 -0.7 0.6 0-5 I No. of i nsects/ fasc i cle Side branch Main branch * Eggs Eggs O.k Main branch ' Larvae, L. 0.3 • * Larvae, L. 0.2 Side branch • ' Pupae 0.1 Side branch Pupae Main branch 0.0 I nner Mid Outer Sections from Stem to Outer Crown Fig. 18. Number of insects per fascicle by horizontal crown position. - 89 -DISTRIBUTION OF THE POPULATION BY LIFE STAGES The aspects of natural within-tree distribution of the larch casebearer investigated were: (a) distribution of eggs among certain categories of foliage; (b) horizontal distribution of different insect stages on the branch i.e. from stem to periphery; (c) vertical distribution of different stages. The Egg Stage Distribution of Eggs The location of 5 2 7 4 eggs was observed, 2 6 5 0 eggs on 324 six-inch branch samples taken from 9 trees in 1 9 7 4 , and 2 6 2 4 eggs on 288 six-inch branch samples taken from 8 trees in 1 9 7 5 -Current Growth vs. Adventitious Foliage vs. Old Growth Foliage Development of current shoots was severely inhibited by insect attack on the tree or by other environmental factors. Its peripheral location probably increases the chances of egg laying females alighting on i t . However, needle fascicles tended to receive relatively more eggs in comparison to current shoots in the proportion of about 7 5 and 25 percent. - 9 0 -Percentage of Eggs on: Current growth needles ... Less than 2 5 % per 1 inch branch length Old growth needles (fascicles) ... More than 7 5 % per 1 inch branch length. The ratio of current to old needles per inch branch length was 1 : 3 . Oviposition sites might be linked to: (l) needle shape, there being a greater width at the apical third of needles in fascicles at oviposition time; (2 ) single needles on current shoots are not yet fu l l y developed, and needles in fascicles on older growth offer a better standing platform for ovipositing fema1es. New adventitious needles produced after severe defoliation in early spring appeared to be the preferred oviposition sites. However, only one tree was found during this study with new needles in late spring. Compared with two neighbouring trees without adventitious needles, there were 3 - 5 times more eggs on fascicles of adventitious needles than on old needle fascicles. This preference for new needles was supported by the Forest Insect and Disease Survey Rangers and colour slides (Fig. 6 b ) borrowed from the Pacific Forest Research Centre, Victoria, British Columb i a. Also, sound needles were preferred to damaged needles. In 1974 when defoliation was moderate 8 9 . 6 % of the eggs were laid on sound needles and 1 0 . 4 % on damaged needles (Table 6 ) . In 1975 defoliation was light and therefore more sound needles were available for oviposition and only 5 . 2 % of the eggs were laid on damaged needles. It is also possible that the ratio of 1 0 . 4 / 5 - 2 merely reflect the ratio of damaged needles in 1974 and 1 9 7 5 . - 91 -Table 6 . Egg distribution by needle condition and on needle surface - 197**. Tree Total Need 1e Needle Surface Sound Damaged Lower Upper No. Eggs No. No. 2 215 203 12 5 . 6 2 0 9 6 2 . 8 3 2 18 208 10 4 . 6 204 14 6 . 4 4 2 2 2 184 34 15 . 6 203 15 6 . 9 5 173 148 35 1 9 . 0 172 11 6 . 0 6 224 191 33 14 . 7 211 13 5 . 8 7 3 5 5 3 3 9 16 4 . 0 338 17 4 . 7 9 3 0 0 296 4 1 -3 2 96 4 1.3 10 4 38 3 8 7 51 1 2 . 0 420 18 4 . 0 11 5 0 5 424 81 16 . 0 4 5 4 41 8 . 0 Total 2 6 5 0 2374 276 1 0 . 4 2511 139 5 - 2 Egg Placement on the Needle Surface Often more than one egg (up to 7 or more in severe infestation) is deposited on a needle surface, so that larval competition in the mines occurs. In 1 9 7 4 , only 2 . 9 percent of the total egg counts were deposited as more than one per needle (Table 7 ) . In the 1974 collections 3 . 2 percent of the eggs were on the upper and 9 6 . 8 percent on the lower needle surface. On the lower needle surface the incubation period is passed in saturated humidity as 9 5 ~ 1 0 0 - 92 -percent of the stomata on the average leaf are found on the underside and as a l l or most of the transpiration occurs through the stomata, the insect's eggs are in highly humidified atmosphere (DeLong, 1 9 7 1 ) . Tabl.e 7 . Distribution of eggs on the needle ( 1 9 7 4 ) . Tree Avg. Total Porti ion of Need 1e No. of eggs/needle No. Defoliat" Eggs Apex Mid Base 1 2 3 4+ 2 5 2 1 5 198 14 3 203 6 0 0 3 4 2 18 2 0 0 12 6 2 0 2 4 0 0 4 6 2 2 2 2 0 5 11 6 204 9 0 0 5 6 173 149 17 7 167 3 0 0 6 7 224 214 3 7 211 3 1 1 7 4 3 5 5 324 13 18 343 6 0 0 9 4 3 0 0 298 1 1 2 6 9 10 2 1 10 5 4 3 8 347 35 56 4 0 6 16 0 0 11 6 505 446 32 27 4 7 6 13 1 0 Average 2 6 5 0 2381 138 131 7 0 7 2 Percentage 8 9 - 9 5 . 2 4 . 9 2 . 6 0 . 3 Preference for the apical third of the needle probably originates from the characteristic resting position of the adult near the needle tip, as 8 9 . 9 percent of the eggs were laid on the apical third in 1974 (Table 7 ) more in 1 9 7 5 -- 93 -Vertical Distribution of Eggs The variation in density along the vertical height of the crown is shown in Tables 8 and 9 for 1 974 and 1 9 7 5 - The numbers of eggs per sample were light to moderate. The greatest density of eggs in both years was found in the upper tree crown, with the least density per needle fascicle in the lower crown. The percentage distribution of eggs per fascicle in the tree crown {25% lower, 35% mid and h0% upper crown levels) was similar for the two years although different trees were sampled. When egg density per 15-cm. branch section were used (Table 9 ) the higher density of fascicle per sample in the upper and mid crown levels were reflected in the higher distribution of eggs, in the upper crown which was double that of the lower crown level. Horizontal Distribution of Eggs Differences in egg numbers along branches from periphery to stem is shown in Table 1 0 . There is a significant difference between the section near the stem and the mid and outer sections; but no significant difference between mid and outer section for eggs collected in 1 9 7 4 . This is due partly to the more concealed situation of the inner branch section which reduces the chance of ovipositing females coming in contact with the fascicles in this section. It is due also to the lower density of needle fascicles on the inner section. Furthermore, the inner section was more heavily defoliated in 1 9 7 4 , and as there was no adventitious needle growth the number of prefer red. oviposition sites was considerably reduced. - 94 -Table 8 . Distribution of _C. lar icel1 a eggs per fascicle by crown levels as collected at Thrums 1974 and 1 9 7 5 -Crown Level No. of eggs/ fasc icle 1974 -% of total eggs No. of eggs/ fascicle 1975 -% of total eggs Lower 0 . 5 0 5 26 0 . 5 6 4 25 Mid 0 . 6 6 2 35 0 . 7 9 2 35 Upper 0.742 39 0 . 8 9 1 40 Total 1 . 9 0 9 2.247 Average 0 . 6 3 6 0 . 7 4 9 Table 9 - D i str i but ion of C. branch section by lar i c e l l a eggs crown levels and per 15-cm. years. Crown Level No. of eggs/ sect ion 1974 -% of total eggs No. of eggs/ sect ion • 1975 -% total of eggs Lower 5-741 23 5 - 5 0 0 20 Mid 8 . 3 8 9 34 1 0 . 2 0 8 37 Upper 1 0 . 4 6 3 43 1 1 . 7 7 1 43 Total 24 . 5 9 3 2 7 . 4 7 9 Average 8 . 1 9 8 9 - 1 5 - 95 -In 1975, actual defoliation was light and the distribution of eggs through the branch tended to be more uniform. The peripheral branch section contains slightly fewer eggs than the mid and inner sections due to the inclusion of a few new growth terminals from tree No. 12 on which only 0-3 eggs were laid. Relationship Between Degrees of Defoliation and Number of Egg per Fascicle A significant but weak negative correlation was displayed (r = -0.166 with 322 degrees of freedom) between the number of eggs per fascicle and the quantity of defoliation (proportion of needles per fascicle mined per sample unit) in 1974. Also, a significant negative correlation was found (r = -0.242 with 322 degrees of freedom) between the number of eggs per fascicle and the volume of defoliation (proportion of needles per fascicle rendered non-functional by the insect per sample). These two negative correlations seem to indicate that the adults select less defoliated branches for depositing their eggs, which results in stronger correlation between eggs and sound needle fascicles. Although the above correlations are highly significant s t a t i s t i c a l l y (0.01 probability level), for practical purposes, the quantity of 2 defoliation explains only 2.75 percent (r = 0.0275) and the volume 2 of defoliation explains 5.86 percent (r = 0.0586) of the variation of the density of eggs. Defoliation rating. On a scale of 1 to 10 the quantity of defoliation (proportion of needles per fascicle mined per sample unit) was moderate to severe in the Spring of 1974 (Table 11). There was no con-sistent differences in defoliation between tree crown levels. - 96 -Table 10. Number of larch casebearer by tree branch type and horizontal crown position per fascicle per 2-15cm. branch sections. Branch Section Stage Inner 1/3 Mid 1/3 Outer 1/3 Pupae 1974 0 . 1 6 9 0.184 0.311 Egg 197 1 * 1.035 1.459 1 .326 La rvae^ 0.539 0.663 0.668 Larvae, 4 0.231 0.346 0 .500 Pupae 1975 0.114 0.173 0.270 Egg 1975 1.581 * 1.518* 1.396 -'' The inner and mid sections had hi the upper crown levels. gher counts than the outer section in Table 11. Defoliation exposure at rating by tree, crown level time of egg stage, 197**. and Pos i t ion Defoli at i on Rat i ng - Quanti ty in Crown Tree lb. 2 3 h 5 6 7 9 10 11 LL 2 3 3 6 7 k 4 4 4 LS 8 4 4 6 1 6 7 5 6 ML 4 3 9 6 7 5 4 5 6 MS 3 3 7 9 4 4 6 4 6 UL 3 4 8 4 7 h 2 5 6 US 7 h 1 7 8 2 3 5 5 L = 1 owe r, M = m i d, U = upper crown level. L = exposed, S = shaded. - 97 -Fig. 19. Four 15_cm branch sections showing defoliation ratings 2, k, 6 and 10. - 9 7 a -- 98 -Egg Distribution in Relation to Adult Behaviour Adult emergence, behaviour and mating were observed in the f i e l d and laboratory. Moth emergence started briskly, reaching a peak on the second or third day. In the laboratory (temperature 22C and 7 5 % RH), emergence occurred throughout the day with the greatest number emerging at mid-morning. After a short flight the adult takes up its characteristic resting position on the tips of needles. This could account for the oviposition of over 8 5 percent of the eggs on the outer 1/3 of the needles. In the f i e l d , flight activity, mating and oviposition occurred about sundown. The moths are crepusular and fading daylight is essential for courtship and eventual mating. This is evident as it is d i f f i c u l t to obtain mating in the laboratory i f the transition from light to dark is made abruptly. Moths were active in the field on 14 and 15 July, 1974, and 15 June, 1 9 7 5 , at 6 : 3 0 p.m. Pacific Standard Time (PST) at about the time when overhead light intensities dropped to 2 0 0 ft.-candles. Peak activities were reached very quickly. At 7-*00 p.m., on the second day after emergence, most of the moths had paired off and mated. Mating continued until dark (9=30 p.m.) (Table 1 2 ) . This indicated a close association between diminishing light intensity and air temperature (down from 30C to 20C) and the inclination of the moths to mate. When the sun set behindsurrounding mountains and the air temperatures f e l l rapidly between 6 5 to 70F (18-21C) the stimulus for mating was triggered. There was a dramatic change in habit from the usually observed daytime indifference of one sex toward the other, to one of attraction, courtship and mating. - 9 9 -Table 1 2 . Moth activity as observed on 14-15 June 1974 at Thrums, B.C. T i me (PDT) Light f t.c. Temp. C Hum. % RH Act i v i ty 14 June 6:30 p.m. 7:00 7:30 7:50 8:00 8:30 9:00 9:10 9:30 200 150 60 a 1 most dark dark 23C 22C 67 6 5 High activity through the tree crown commencing f i r s t at the base. More than half the adults mating on needles. L i t t l e or no flight activity. L i t t l e act iv i ty. Some activity. Very l i t t l e activity, activity concentrated mainly in the upper level of crown. Very few moths active in tree tops. All flight activity ceased, some moths s t i l l in copu1 at ion. 15 June 5 : 4 5 a.m. 14C 2 pairs of moths mating. 2 fema1es in f1ight. - 100 -It is assumed that mating on larch is a behavioural adaptation tending to restrict the larch casebearer to this plant. Also, when the moths are disturbed during the day they quickly flutter back to the same branch or to another part of the host tree. Emergence from the egg is directly through the portion of the chorion attached to the needle into the leaf mesophyl1. This restricts the f i r s t and second instar larva to one needle with limited movement as a leaf-miner. The Larval Stage Larvae Prior to Winter Dormancy Emergence from eggs is followed by limited movement of early stage larvae. Eggs and f i r s t instar larvae are found exclusively on the needle on which the egg was laid. The second instar larva usually remain in the same needle, but occasionally, the larva may transfer from the f i r s t needle-mine to establish itself in other needles, especially when the egg is deposited on a damaged needle, or the larva is crowded in the mine. The third instar larva or case-bearing stage is capable of limited movement, and feeds on a number of needles (8 needles according to Eidmann^1965) prior to hibernation. Therefore, it is the third instar larvae that were sampled for information on the distribution of prewinter larvae at the time of dormancy. - 101 -Spring or Postwinter Larvae Activity of the postwinter larvae depended on the course of warming up of the weather in spring. At Thrums, the larvae began wandering around at about mid April in 1974 and 1975, at approximately the same time as the larch needles started to flush. Feeding is usually delayed for about a week and commences when the needles are 6-8mm in length. Sampling for larvae in early spring was carried out on 24-25 April 1 975 when larval wanderings had already commenced. Larval Variation Between Trees The number of prewinter and postwinter larvae per fascicle was significantly different (0.05 level) between trees. The averages showed slightly more larvae per fascicle on the edge trees than on the interior stand or open grown trees, but this difference originated from one tree only (tree No. 8 ) . The tree-to-tree variation within the edge trees was more significant than within the other stand position (Table 1 3 ) . Most of this difference was due to tree No. 8 which had a very high population, but was 1ocated in a different part of the stand and in mixture with other tree species. Vertical Variation in the Tree Crown The number of larvae per fascicle was significantly different between crown levels, but the significance of interaction between trees and crown levels (0.05 probability level) for the postwinter larvae indicated that the differences were not consistent. For the prewinter - 102 -Table 1 3 . Average number of larvae per fascicle and per 15_cm. branch section by tree and stand position. Tree Prewinter Larvae Postwinter Larvae per per per per Number fascicle branch section fascicle branch section 1 0.414 3 - 4 4 0 . 3 1 2 2.14 2 0 . 3 3 8 3 - 5 5 0 . 1 0 6 1 .25 3 0 . 2 0 3 2 . 0 5 0 . 1 2 3 1 .06 4 0 . 2 3 3 2 . 4 4 0 . 1 8 0 1 .72 Interior 0 . 2 9 7 2 . 8 7 0 . 1 8 0 1 .54 5 0 . 2 7 5 1 .80 0 . 2 8 2 1 .86 6 0 . 3 5 7 2 . 4 2 0 . 1 9 9 1 .97 7 0 . 211 2 . 7 2 0 . 1 8 9 0 . 9 7 8 0 . 7 5 4 10.14 0 . 5 1 3 6 .11 Edge 0 . 3 9 9 4 . 2 7 0 . 2 9 6 2 . 7 3 9 0 . 3 9 9 2 . 8 3 0 . 1 5 8 1 .39 10 0 . 2 5 1 2 . 8 0 0.187 2 . 2 8 11 0.147 2.42 0 . 1 5 1 2 . 7 2 12 0.155 2 . 2 5 0 . 0 7 2 1.19 Open 0 . 2 3 8 2 . 6 0 0.142 I . 8 9 - 103! " larvae, the population density was significantly higher in the mid leved'. than in the lower level, but no difference was found between the mid and upper levels. The above conclusions can be drawn from the averages but they were not true for every tree. This variation of larval density in crown levels is due mainly to variation in mortality and to a lesser extent fascicle variation in the tree crown. These are discussed under different sections. Larval migration from branch to branch is almost nil in light and moderate i nfestat ions. Table 14. Average No.of casebearers/15-cm.branch section collected 1 9 7 4 . Lower C rown Mid Levels Upper Egg 5.741 8.389 10.463 L 3 2.444 3.542 3.736 (.269) (.368) (.298)* L 4 1.7278 1.722 2.139 ( . 1 9 4 ) (.171) (.174)* P 7 5 0.767 1.211 1.533 ( . 0 8 0 ) (.108) (.091)* E " 7 5 5.500 10.208 11.770 ( ) = larvae per fascicle or spur shoot - = Higher number of fascicles/6-inch branch in the upper crown level than in the mid and lower levels. - 104 -As the n e e d l e s had f a l l e n o f f a t the time o f c o l l e c t i o n o f t h e p r e w i n t e r l a r v a e , and as t h e l a r v a e a t t a c h t h e m s e l v e s t o v a r i o u s p a r t s o f t h e b r a n c h , the number o f l a r v a e per 15 -cm. branch s e c t i o n may be a b e t t e r sample u n i t f o r t h i s s t a g e . T h i s would i n d i c a t e t h e t r e n d s e t from o v i p o s i t i o n , namely, t h e i n c r e a s i n g d e n s i t y o f l a r v a e from the lower t o t h e upper crown l e v e l as shown i n T a b l e 14 and F i g . 2 0 . Exposure The number o f l a r v a e ( b o t h p r e w i n t e r and p o s t w i n t e r ) per f a s c i c l e was not s i g n i f i c a n t l y d i f f e r e n t between t h e exposed and shaded b r a n c h e s . However, f o r t h e p r e w i n t e r l a r v a e i n t e r a c t i o n between t r e e s and ex p o s u r e s i s s i g n i f i c a n t a t t h e 0 .01 p r o b a b i l i t y l e v e l . T h i s i n d i c a t e s t h a t d i f f e r e n c e s between e x p o s u r e s i s not c o n s i s t e n t f r o m t r e e - t o - t r e e . T a b l e 15 shows t h a t the d i s t r i b u t i o n o f t h e l a r v a e on t h e shaded s i d e i n c r e a s e d w i t h h e i g h t i n t h e t r e e crown, t h i s c o u l d be a s s o c i a t e d i n i t i a l l y w i t h i n c r e a s e d sky l i g h t w i t h h e i g h t . On t h e a v e r a g e , t h e l a r v a e on t h e exposed s i d e o f t h e t r e e crown were d e n s e s t a t mid-crown l e v e l f o r p r e w i n t e r , and a t t h e l o w e r l e v e l f o r t h e p o s t w i n t e r l a r v a e ( T a b l e 15). T h i s i n d i c a t e s h i g h e r o v e r w i n t e r i n g m o r t a l i t y o r p r e d a t i o n i n t he mid crown. T a b l e 15. Number o f l a r v a e per f a s c i c l e by c a s e b e a r e r s t a g e , e x p o s u r e and crown l e v e l s . I n s e c t Crown L e v e l S t a g e Exposure Lower Mid Upper P r e w i n t e r L a r v a e - Exposed 0.301 0.429 0 . 2 7 7 Shaded 0.236 0.307 0.319 Average 0.269 0.368 0.298 P o s t w i n t e r Larvae*-- Exposed 0.222 0.206 0.140 Shaded 0.165 0.136 0.207 Average 0.193 0.171 0.173 12 t r e e s * * 15 t r e e s - 105 -Lower Mid Upper Fig. 20. Number of casebearer per 15_cm branch section by crown levels. - 106 -Larval distribution with respect to exposure was affected mainly by variations in mortality which appeared, on the average, to be evenly distributed between exposed and shaded branches within the same crown level, but differed between crown levels. Main and Side Branches Significantly more overwintering larvae per fascicle occurred on the main branches than on the side branches. Although the egg population was not different on the two types of branches. This dis-tribution is attributable mainly to active larval redistribution and variations in mortality. Migration to the firmer, more sheltered main branches, or mortality due to desiccation on the side branches (to be discussed) is probably the cause. For the postwinter larvae there is no difference between the main and side branches. As shown in Table 16 this is due to migration during the f i r s t week of spring activity to the outer third of the main branch and to the side branches which contain more fascicles per unit branch length. Horizontal Crown Variation For the prewinter larvae, there was a significant difference between horizontal crown position, with greater numbers being present on the mid-section of the main branches (Table 16). For the postwinter larvae, the significant difference in horizontal crown position was due to the larval migration towards outer third of the main branch, which had the greatest numbers, and the side branches. - 107 -Table 16. Average No. of larvae per fascicle for 12 trees at Thrums by horizontal crown position. Pos i t ion in Stand 1 nner Main Branch Inner S ide Branch Mid Outer Mid Outer 1nter ior L3 0.417 0 .254 0.393 0. 194 0 .209 0.314 L 4 0.140 0.-161 0 . 237 0.161 0. 166 0.215 Edge L 3 0.379 0.584 O.38O 0 .179 0 .288 0.335 L4 0.103 0.323 0.515 0 .218 0.276 0.340 Open S 0.304 0 . 2 5 8 0.337 0.143 0.143 0 .268 0 . 0 5 0 0.070 0.224 0.084 0.194 0.231 Average 4 0.367 0.449 0.362 0 . 1 7 2 0.214 0.306 0.098 0.185 0.325 0.154 0.212 0 .262 = prewinter larva L. = postwinter larva Comparison of Branch and Branch Tip Samples As most of the other studies on the casebearer involved samples of the tips of branches or current growth only, branch tips also were sampled for the spring larvae in the 1975 (Table 17)-- 108 -On the average, the number of larvae per fascicle on branch tips was not different from numbers on the outer third of the branch. Table 1 7 - Comparison of spring larvae per fascicle by branch tips and by the outer third of branch. Stand Larvae per Fascicle Position Outer 1/3 of branch Branch tips Interior (Trees 1-4) 0 . 2 3 7 0 . 2 3 4 Open (Trees 9"12) 0.224 0 . 3 1 3 * * trees nos. 1 0 - 1 2 only Discussion From the results there is evidence of major changes in the vertical distribution between egg and prewinter larval stages. The greatest number of eggs per fascicle was found at the upper level of the tree crown while that of the prewinter larvae was at the middle level. The shift of insect population could be the result of either movement, or greater survival of larvae, or both. The f i r s t two larval instars are in the needle-mines and the third instar larvae are capable of only limited movements prior to dormancy. Therefore, the relative shift of population concentration must be presumed attributable to either greater mortality in the upper levels or to differences in the number of spur shoots per linear unit of twig sample. - 109 -As shown earlier, if the 15_cm. branch sample is used as the sampling unit instead of the spur, the distribution of the larvae follows that of the egg. Most of the differences in larval density between crown levels appear to correspond to the difference in the number of spur shoots. Reasons Underlying the Distribution of Larvae The type or condition of the overwintering cases. Normally in constructing its case, the larva cuts off the tip of the needle. If the needle-tip remains attached to the case, or if the case is the needle-tip, then i t appears that the casebearer was forced to do so because of un-favourable conditions. Such conditions would be, the drying out of needles or the lack of readily available needles. It can further be supposed that the higher proportion of cases, with needle-tips or of needle-tips, associated with the overwintering larvae, reflects termination of prewinter activities earlier under unfavourable conditions. Table 18 shows the total number and percentage of larvae with cases constructed from needle tips for each sample tree and indicates an overall average of 14 percent of the total cases. Larvae in these cases were affected more adversely during the winter than larvae in normal cases and mortality was high. Most needle-tip cases occurred in the upper crown. The highest percentage (18-51%) was on the interior stand trees (nos. 1-4) in the upper crowns and overwintering in or among lichens on branches. - 110 -Table 18. The number and percentage of normal and needle tip cases by tree for the prewinter larvae 1974. Tree Overwintering Insect Cases Norma 1 Needle-t i p No. Total No. % No. % 1 124 102 82.3 22 17 . 7 2 128 112 87 .5 16 12.5 3 74 36 48.7 38 51.3 4 88 55 62.5 33 37 .5 5 65 62 95.4 3 4 .6 6 87 86 98.9 1 1.1 7 98 93 94.9 5 5.1 8 365 324 88.8 41 11.2 9 102 94 92 .2 8 7.8 10 101 86 85 .2 15 14.8 11 '87 81 93.1 6 6.9 12 81 73 90 .2 8 9-8 Total: 1400 1204 86 .0 196 14.0 Larval Behaviour and Distribution Effects Feeding, Migration and Orientation. The.prewinter larvae migrate to their hibernating sites. The postwinter larvae or spring larvae wander around during'feeding and occasionally also go to pupation sites. - I l l -At Thrums, third instar larvae were found feeding on the needles on 25 August 1 9 7 4 . No larvae were found in the needle mines on edge trees. Denton (1958) found larvae s t i l l actively feeding at the end of September in Idaho, USA. The brownish discolouration of damaged foliage was very noticeable at that time. He stated that as the older fascicles began to fade prior to being shed, the larvae moved out to the needles of new terminal shoots that were s t i l l green. There-fore, as many as 3 dozen larvae were found on a single 3~inch ( 7 . 7 c m ) new growth tip (Denton, 1958 ) . This was not observed in this study as the larvae hibernated in a variety of sites, such as, under or on lichens on most main branches, at the base of spurs, and on or under bark of branches. It is likely that larvae cease feeding by mid-September in response to decreasing photoperiod either directly or indirectly as a result of the host tree responding to photoperiod prior to leaf f a l l . Eidmann (1965) in Sweden, marked prewinter larvae and found migration had taken place in both directions to the periphery of the branch as well as to the stem, with the majority of larvae migrating toward the one year growth section. This is in agreement with the results of this study. The prewinter larvae are positively phototactic and negatively geotactic shortly before dormancy. Eidmann (1965) showed experimentally using sticks that most of the active larvae migrated upwards. However, the few that wandered downwards covered a greater distance. Therefore, the slope of the substratum seems to influence the direction of migration, and may account for migration towards the stem on branches at an angle - 112 -g r e a t e r than 90° t o t h e stem ( i . e . downward s l o p i n g branches) o r towards the main branch i f l a r v a e a r e on downward s l o p i n g s i d e b r a n c h e s . S p r i n g l a r v a e : O b s e r v a t i o n s on p o s t w i n t e r l a r v a l b e h a v i o u r i n A p r i l 1975 were made i n l a b o r a t o r y a t room t e m p e r a t u r e s (22C and 1 0 % RH) on p o t t e d p l a n t s o r f r e s h l y c u t l a r c h t w i g s . On a s l o p i n g branch s e c t i o n t h e l a r v a moved upwards u n t i l a branch spur was met and then outward and upward a l o n g a n e e d l e u n t i l i t reached', t h e t i p , then t u r n e d , s t r e t c h i n g i t s body'as i f measuring t h e d i s t a n c e from the n e e d l e t i p . The l a r v a then f a s t e n e d i t s e l f t o t h e n e e d l e and r e s t e d , t h i s o p e r a t i o n on the n e e d l e u s u a l l y t o o k . about 10 m i n u t e s . The m a j o r i t y o f the l a r v a e wandered about f o r more than 3 hours b e f o r e a t t a c h i n g t h e m s e l v e s t o a n e e d l e . Even w i t h o u t f e e d i n g t he l a r v a e can c o v e r c o n s i d e r a b l e d i s t a n c e a v e r a g i n g h cm per minute i n the l a b o r a t o r y e x p e r i m e n t s . Loos (1892) found c a s e b e a r e r s i n t h e f i e l d t h a t wandered a t l e a s t 5 m e t e r s , and saw some t h a t c o v e r e d k cms i n a m i n u t e . Webb (1953) o b s e r v e d marked l a r v a e i n the f i e l d and e s t a b l i s h e d a t o t a l m i g r a t i n g d i s t a n c e o f between 28 and 32 cms. ( d a i l y a v e r a g e was 4.8 cms) f o r normal l a r v a e b e f o r e p u p a t i o n , Eidmann (1965) i n h i s o r i e n t a t i o n e x p e r i m e n t s found 1 l a r v a wandered 6 3 cm down a 15° s l o p e i n 20 mins. I t was a l s o found t h a t n u t r i e n t d e f i c i e n c y and hunger s t i m u l a t e d , m i g r a t i o n ( t h i s may a c c o u n t f o r some o f the d i f f e r e n c e s between Webb 1953 and Eidmann I 9 6 5 ) • A l t h o u g h t h e l a r v a e a r e c a p a b l e o f t r a v e l l i n g some d i s t a n c e i t i s u n l i k e l y t h a t , i n l i g h t and moderate i n f e s t a t i o n s , they would t r a v e l beyond a b r a n c h . - 113 -It has been reported that the casebearer also feeds on young female strobili in the spring (Loos, 1892 and Eidmann, 1965). This was not observed in British Columbia but is worth bearing in mind as it could affect insect counts, distribution or survival. Loos (1892) mentioned a spring migration of casebearers to the interior of the crown. In contrast, Webb (1953) described a tendency to migrate to the tip of the twig. This study showed a migration to the outer third of the main branch and to the side branches. Orientation Experiment: Ten larvae were placed one at a time in a 2 cm diameter 20 cm long glass tubing, blackened for half its length and held at an angle of 30°. Each larva was placed in the blackened section on a strip of paper with a rough surface and the tube stoppered. The larvae were stationary for a while and when responding did so slowly, taking about 2 hours on the average to travel H cm to the lighted section against gravity. This indicated that the larvae were more strongly phototactic than geonegative. Compet i t ion: In the spring insect density becomes important. There is competition between larvae. If more than one larvae per needle (per spur in the laboratory, Quednau (1975), personal communication) is present the larvae will k i l l each other. When insects come together they spin s i l k threads tying their cases together and eventually leave the cases. This larval competition gives a density-control factor preventing overpopulation and starvation. This also explains some of the reasons for the movement of the early spring larvae towards the side branches and the outer third of the main branch where the numbers of needle fascicles are dense. - 114 -The Pupal Stage Vertical Distribution For the pupal stage in 1974 the number of insects per fascicle was significantly higher in the mid crown level than in the upper level, but no significant difference was found between the lower and upper levels. There was no significant difference between crown levels for the pupae in 1975, but, on the average, the density of pupae was highest in the mid crown level. When the 15~cm branch section was taken as the sampling unit instead of the fascicle, the relative distribution of pupae ( 1 9 7 5 ) conformed with that of the i n i t i a l egg ( 1 9 7 4 ) distribution in the crown (lower < mid < upper) see Table 19-Horizontal Crown Position In general, i t was concluded that the pupal population density on the branch section close to the stem was the lowest and increased toward the outer part of the crown (Table 2 0 ) . However, this trend was inconsistent from tree-to-tree for the open grown trees. In both years, the average number of pupae per fascicle and per branch section was 50 percent or more, greater for the outer sections of both main and side branches than for the inner sections, as shown in Table 2 0 . - 115 -Table 19- Densities of pupae by crown levels in 1974 and 1975. Level - 1974 Level - 1 975 Lower Mid Upper Lower Mid Upper Pupae per fascicle 0 . 2 6 9 0 . 3 6 8 0 . 2 9 8 0 . 0 8 0 0 . 1 0 8 0 . 091 per 15-cm section 1 . 233 1.611 1 . 3 78 O.767 1 . 1 6 7 1 . 607 Fasc icle per 15-cm section 1 1 . 3 9 12 .21 1 5 - 1 2 1 1 . 9 0 1 3 - 7 3 1 6 . 2 6 Table 20. Densities of pupae by horizontal crown position in 1974 and 1975. Main Branch Side Branch Inner Mid Outer Inner Mid Outer Pupae per fascicle (1) 0.058 0.070 0.121 0.111 0.114 0 . 1 9 0 (2) 0.069 0.082 0.105 0 . 0 4 4 0.091 0.165 per 15-cm ( 0 0.467 0.689 1 .522 1.37-8 1.544 2.840 (2) 0.400 0.511 1 .289 0.633 1.489 2.700 Fasc ic1e per section (1) 9.45 10.93 13.52 13.5 14.07 15.95 (2) 9.93 11.91 14.00 14.63 16 . 2 3 16.73 (.1) = 1974 (2) = 1 975 - 116 -Exposure Although the pooled values of pupal counts per fascicle on the exposed and shaded sides of trees in 197** were not s t a t i s t i c a l l y different, there were more pupae on the shaded side than on the exposed side of the trees. This was especially so on the edge trees where a strong dis-tinction in exposure could be made. The 1975 pupal counts showed the reverse trend with a significantly higher number of insects on the exposed branches than on the shaded ones (Table 21). Here again, a l l of the edge and interior stand trees showed a marked difference between exposed and shaded branches (Table 21). The results indicated a switch in the density of pupae from the shaded to the exposed branches or vice versa from year to year. This probably occurs when one side is defoliated more than the other and results in few or poor oviposition sites later on in the season. Factors Affecting Pupal Distribution Just before pupation the larva attaches the front-end of its case firmly to a needle. Pupal cases are found mostly in the center of a needle fascicle attached to the base of needles, less frequently on the outer portion of needles, and on or among lichens on the branches. They also pupate in other locations on the branches in instances of high population densities or other unfavourable conditions. - 117 -It appears that the casebearer pupates mainly in the neighbourhood of its last feeding place. However, it is not uncommon to find pupae on sections of twigs which have no damaged needle (also mentioned by Eidmann 1 9 6 5 ) . Table 21. Stand Pos i t ion Average No. of pupae per fascicle by year, position of trees in the stand and exposure. Pupae 1974 Exposed Shaded 1975 Exposed Shaded 1nter ior 0. ,106 0. 118 0. 137 0. 101 Edge 0. ,161 0. 202 0. 156 0. 067 Open 0. ,079 0. 091 0. 110 0. 076 C1usters 0. ,043 0. 052 0. 034 0. 030 Average 0. ,101 0. 120 0. 114 0. 072 The defoliation rating was recorded for the pupal stage and was correlated with the number of pupae per fascicle and the defoliation index. The correlation is significant s t a t i s t i c a l l y , meaning that the defoliation parallels the density of pupae, but the relationship explains 2 2 only 7.14 percent (r = 0.0714) and 18.04 percent (r = 0.1814) of variation of the defoliation index respectively in 1974 and 1 9 7 5 -Although pupal populations were low on most trees, it appears that the larvae had some mutual attraction for each other just before pupation. Up to 5 pupae per fascicle were observed in this study. - 118 -Larch Foliage Distribution of Larch Foliage in the Crown Distribution of the larch casebearer or defoliation can be misinterpreted or wrongly estimated if not considered in relation to distribution of foliage. The distribution of foliage (fascicles per 15~cm branch section) approaches normality as illustrated from branch samples collected for postwinter larval counts in Appendix 5, Fig. T. This is confirmed when variance is plotted over the mean (Appendix 5, Fig. II) which shows the variance and mean largely independent. It was therefore possible to apply analysis of variance and other statis t i c a l tests on the untransformed data, of fascicle counts. Fasc i cles 1. Tree-to-Tree Variation For a l l 6 sampling dates in 1974 and 1975 the average number of fascicles per 15~cm branch section was significantly different (0.01 probability level) from tree-to-tree (Tables 22, 23a, 23b and 2 4 ) . Thus the total number was different from tree-to-tree, but this would also depend on branch length and branches/tree. However, the averages (Appendix 4, Table 25) showed that the variation was due to differences between trees within the edge and open grown positions in the stand. The open grown trees also had more fascicles per branch section than the edge and interior stand trees. - 119 -Table 2 2 . Analyses of variance of numbers of fascicles per 15-cm branch section. Source DF May ' 7 4 (pupa) Mean Sq. F June ' 7 5 Mean Sq. (pupa) F Tree T 14 2 1 3 . 5 2 3 . 2 * * 2 7 5 . 4 3 0 . 8 * * Level L 2 6 9 2 . 6 7 5 . 1 * * 831 . 4 9 3 . 0 * * Exposure E 1 2.1 0 . 2 1 .2 . 1 Main & Side Branch M 1 1 3 8 8 . 8 1 5 0 . 6 * * 2 0 7 2 . 9 2 3 2 . 0 * * Parts, within M P 4 2 6 4 . 8 2 8 . 7 * * 240 . 2 2 6 . 9 " " ' T x L 28 3 7 . 2 4 . 0 * * 15 . 8 1 . 8 * T x E 14 2 0 . 0 2 . 1 * 13 . 6 1.5 T x M 14 10.1 1.0 13.1 1.5 T x P 56 7 - 4 0 . 8 1 7 . 4 1 . 9 * -L x E 2 6 . 9 0 . 7 14 . 0 1.6 L x M 2 3.1 0 . 3 8 2 . 1 9 . 2 * * L x P 8 7 - 2 0 . 7 11 .6 1.3 E x M 1 4 . 6 0 . 5 2 7 . 6 3.1 E x P 4 2 5 . 6 2 . 8 * 7 . 6 0 . 8 T x L x E 28 12.1 1.3 1 2 . 6 1.4 T x L x M 28 9.1 1.0 2 1 . 6 2 . 4 * * T x L x P 112 1 2 . 6 1 . 4 * 1 2 . 5 1 . 4 * T x E x M 14 14 . 3 1.5 14 . 5 1.6 T x E x P 56 7 - 5 0 . 8 1 2 . 8 1.4 L x E x M 2 6 . 6 0 . 7 4 . 7 0 . 5 L x E x P 8 4 . 8 0 . 5 1 3 . 0 1.5 TLEM 28 16 . 2 1 . 7 * 1 5 . 8 1 . 8 * ERROR 112 9 - 2 8 . 9 TOTAL 539 Significant within the 0 . 0 5 level. ** Significant within the 0 .01 level. - 120 -Table 2 3 a . Analyses of variance of numbers of fascicles per 15-cm branch section collected for egg counts in 1974 and 1975. July 1974 July 1 975 Source DF Mean Sq. F DF Mean Sq. F Tree T 8 2 1 2 . 2 2 0 . 1 * * 7 8 8 . 9 8.7** Level L 2 431 .1 4 0 . 8 * * 2 3 0 4 . 7 2 9 . 8 * * Exposu re E 1 1 8 . 8 1.8 1 16.5 1.6 Main & Side Branch M 1 1 4 1 8 . 8 I 3 4 . 3 * * 1 1148 .0 1 1 2 . 4 * * Parts, within M P 4 147.7 1 4 . 0 * * 4 5 1 . 7 5 . 1 * * T x L 16 2 8 . 6 2 . 7 * * 14 2 6 . 0 2 . 5 * * T x E 8 2 6 . 6 2 . 5 * 7 18.1 1.7 T x M 8 4 0 . 6 3 . 8 * * 7 3 5 - 1 3 . 4 * * T x P 32 1 0 . 5 1.0 28 1 2 . 5 1.2 L x E 2 1 .7 0 . 2 2 28.4 2 . 8 L x M 2 5.1 0 . 5 2 5 1 . 9 5 . 1 * * L x P 8 14.4 1.4 8 7 . 0 0 . 7 E x M 1 0 . 2 0 . 0 1 0 . 0 0 . 0 E x P 4 3 . 6 0 . 3 4 6.1 0 . 6 T x L x E 16 1 7 . 7 1.7 14 5 - 9 0 . 6 T x L x M 16 1 9 . 8 1 . 9 * 14 3 3 . 6 3 . 3 * * T x L x P 64 5 . 8 0 . 5 56 9 . 6 0 . 9 T x E x M 8 6.1 0 . 6 7 5 - 9 0 . 6 T x E x P 32 7.1 0 . 7 28 1 0 . 0 1.0 L x E x M 2 1 1 . 9 1.1 2 14.8 1.4 L x E x P 8 3 - 5 0 . 3 8 1 0 . 5 1.0 TLEM 16 9.4 0 . 9 14 7.3 0 . 7 ERROR 64 1 0 . 6 56 1 0 . 2 TOTAL 323 287 Significant within the 0 .05 level Significant within the 0.01 level - 121 -Table 23b. Analyses of variance of fascicles per 15"*cm branch section from samples collected for larva and larva Sou rce DF Nov. '74 <L3> DF Apr. ' 75 Mean Sq. F Mean Sq. F Tree T 1 1 2 9 9 . 2 4 3 . 7 * * 14 2 5 6 . 0 2 5 . 5 * * Level L 2 6 0 2 . 3 8 8 . 2 * * 2 404 . 5 4 0 . 2 * * Exposure E 1 2 1 . 3 3.1 1 6 0 . 7 6 . 0 * Main S Side Branch M 1 1 5 4 8 . 9 2 2 6 . 7 * * 1 1 9 6 8 . 4 1 9 5 . 8 * * Parts, within M P 4 1 0 9 - 0 1 5 - 9 * * 4 2 3 6 . 2 2 3 . 5 * * T x L 22 2 7 . 0 4 . 0 * * 28 1 0 8 . 7 1 0 . 8 * * T x E 1 1 2 3 - 3 3 . 4 * * 14 14 . 4 1 .4 T x M 1 1 1 7 . 2 2 . 5 * * 14 2 7 . 6 2 . 7 * * T x P 44 1 1.0 1.6 56 1 3 . 9 1.4 L x E 2 0 . 6 0.1 2 1 1 . 9 1 .2 L x M 2 2 7 . 9 4 . 1 * 2 4 . 2 0 . 4 L x P 8 1 0 . 5 1 . 5 8 6.1 0 . 6 E x M 1 0 . 0 0 . 0 1 2 6 . 2 2 . 6 E x P 4 11 .4 1 .7 4 1 9 . 8 2 . 0 T x L x E 22 1 0 . 4 1.5 28 1 8 . 7 1 . 9 * T x L x M 22 8 . 5 1 .2 28 1 3 . 4 1 .3 T x L x P 8 8 1 1 . 6 1 . 7 * * 112 1 7 . 6 1 . 7 * * T x E x M 1 1 2 0 . 2 2 . 9 * * 14 14.1 1.4 T x E x P 44 5 - 5 0 . 8 56 8 . 3 0 . 8 L x E x M 2 1.9 0 . 3 2 2 0 . 8 2.1 L x E x P 8 6 . 9 1.0 28 1 0 . 0 1 .0 TLEM 22 14 . 0 2 . 0 * 28 1 2 . 6 1.3 ERROR 88 6 . 8 112 1 0 . 0 TOTAL 431 539 * Significantly within the 0.05 level. ** Significantly within the 0.01 level. - 122 -2 . Crown Level Variation There were significant differences (P < 0 . 0 1 ) between the three crown levels in the number of fascicles per branch section (Tables 2 2 , 2 3 a , 2 3 b ) . The number of fascicles per branch section was significantly higher in the upper ( 38% ) than in the mid ( 33% ) and lower ( 29% ) levels, and significantly higher in the mid level than in the lower level. The above conclusions can be drawn from the averages (Table 2k) but they are not true for every tree. In fact there were significant interactions between tree and crown level for a l l sampling dates, indicating a different trend from tree-to-tree. For example, tree No. 5 was attacked by bark beetles on a few branches, and tree No.15 was heavily attacked by the larch casebearer for a number of years, resulting in branch die-back and the number of fascicles produced varied greatly. Also, flowering in the upper and mid crowns and heavier shoot production in the upper levels affected the number of fascicles per branch section on some trees, especially the open grown ones. Table 2k. Density's percentage distribution of needle fascicles per 15~cm branch in 3 crown levels of western larch at Thrums,B.C. Source Lower Mid Upper Total 1nterior * 9.0 10.6 10.2 (29.8) 30% 36% 3k% Edge * 7.3 8.7 12.5 (28.5) 26% 30% kk% Open * 11 . 9 - 13-4 15-2 (40.5) 29% 33% 38% CIuster * 11.6 11.7 13 . 9 (37.2) 31% 3 2 % 37% Average * 9 . 9 11.1 12 . 9 (33.8) 29% 33% 38% * Average dens it ies of needle fascicles per branch section - 123 -Crown level variation of fascicles in relation to egg  depos i t ion: Most of the vertical distribution of the casebearer in the tree takes place during egg deposition on the needle. Therefore, egg counts were compared with fascicle distribution in the three crown levels (Table 25). There was a strong correlation between number of fascicles and eggs per crown level. However, the figures indicate that a much higher number of eggs are laid in the upper crown than could be attributed to the higher density of fascicles alone. It is suggested that the adult behavioural preference for the more lighted upper level of the tree also plays an important part in the distribution of eggs in the tree crown. Table 25- Number of fascicles and eggs per 15_cm branch section by year and crown level • No.of fascicles/branch section No.of eggs/branch section Year Lower Mid Upper Lower Mid Upper 1974 11.39 12.21 15.12 5.74 8.39 10.46 1975 11.90 13-73 16.26 5.50 10.21 11.77 Although there are more fascicles per 15_cm branch section in the upper levels of the crown, the proportion of total foliage by volume or weight in the upper crown level is less than the mid and lower levels, due to the shape of the crown. Therefore, total number of insects per unit area or volume of foliage would be greater in the upper levels than in the mid or lower levels. - 124 -3. Exposure The number of fascicles per 15~cm branch section did not, on the average, differ significantly between the exposed and shaded branches (Table 2 6 ) . However the tree-to-tree variation between exposed and shaded branches indicated that the trend toward higher or lower numbers of fascicles per 15-cm branch section between exposed or shaded branches, respectively, was not consistent from one tree to another. This was attributable mainly to the open grown trees. A significantly higher deposition of eggs on the exposed than on the shaded branches bears out observations that the moths prefer the more lighted sections of the crown Table 2 7 . However, the reverse may be true when the exposed side is more heavily defoliated and no new adventitious needles are formed. 4. Main and Side Branches Highly significantly (P = 0 . 0 1 ) more fascicles per branch section occurred on the side branches ( 5 8% ) than on the main branches (42%) (Table 2 7 ) . Table 27 shows that proportionately more eggs were deposited on the main branches than on the side branch in 197** and 1 9 7 5 . 5 . Horizontal Crown Position Table 28 gives the average number of fascicles per branch section by horizontal crown position from stem to periphery. For a l l sampling dates highly significant differences (P = 0 . 0 1 ) in densities of fascicles occurred for a l l horizontal crown positions,with the lowest density close to the stem, and numbers increased for the mid and outer part of the crown. This difference was not so pronounced when only side branches were considered. The distribution of fascicles did not appear to - 125 -affect insect density which was consistently higher on the outer sections, except eggs which were scattered over the entire branch (Table 28 ) • Table 2 6 . Average number of fascicles per 15-cm branch collection period, crown level, exposure and sect ion branch by type. Col 1ect i on Crown Level Exposure Branch Period Lower Mid Upper Exposed Shaded Ma i n S i de 1974 May (P)* 11 .4 1 2 . 2 15.1 1 3 . 0 1 2 . 8 1 1 . 3 14 . 5 July (E) 1 0 . 4 1 2 . 6 14 . 4 1 2 . 7 1 2 . 2 1 0 . 4 14 . 6 Nov. (L ) 9 . 5 1 1 . 5 1 3 - 6 1 1 . 8 1 1 . 3 9 . 6 13 .4 1975 April (Lk) 9 - 9 11.1 1 2 . 9 1 1 . 6 1 1.0 9 . 4 1 3 - 2 June (P) 1 1 . 9 1 3 . 7 16 .3 1 3 - 9 1 3 . 9 1 1 . 9 1 5 . 9 July (E) . 1 0 . 4 1 2 . 9 1 3 . 8 1 2 . 6 1 2 . 2 1 0 . 4 14 .4 Average 1 0 . 6 13 . 3 14 . 4 1 2 . 6 1 2 . 2 1 0 . 5 14 .3 Percent 2 8 . 4 3 3 . 1 3 8 . 5 5 0 . 7 4 9 - 3 4 2 . 3 5 7 . 7 Life stage of larch casebearer present at that period. - 126 -Table 27 . Average branch number of section by fascicles and eggs year, exposure and per 15-cm branch type. Fasc icles/branch sect ion Eggs/branch section Exposure Exposure Year Exposed Shaded Exposed Shaded 1974 12.71 12.23 9.14 7.25 51% 49% 56% 44% 1975 12.65 12.17 8.99 9-33 5 1 % 49% 49% 51% Branch Branch Year Main Side Ma i n S ide 1974 10.38 14.56 7.46 8.93 41 .6% 58.4% 46% 54% 1975 10.41 14.40 8.85 9.47 42% 58% 48% 52% - 127 -Table 2 8 . D i str i but ion 15-cm branch of insects and appropriate section by horizontal crown fasc ic1es pos i t ion. per Unit/ Ma in Branch Side Branch Sect ion 1 nner Mid Outer Inner Mid Outer Pupa '7k 0 . 5 0 . 7 1 .5 1 .4 1 . 5 2 . 8 Fasc icle 9 .4 1 0 . 9 1 3 - 5 1 3 . 5 14.1 1 5 - 9 Egg Hk 4 . 8 8 . 0 9 . 6 7.6 9 . 8 9 - 3 Fascicle 8 . 9 10.1 12 .1 1 3 . 2 14.1 16.4 Larva^ 2 . 5 4 . 0 3 . 5 2 . 2 2 . 7 4 . 5 Fasc icle 8 .4 9 . 4 11.1 12 .4 13.2 14.6 Larva, k 0 . 6 1.2 2 . 2 1.4 2 . 2 3 .4 Fasc i cl e 8.1 9 .4 1 0 . 6 11 .2 13.3 1 5 . 0 Pupa ' 7 5 0 . 4 0 . 5 1 .3 0 . 6 1.5 2 . 7 Fasc icle 9 - 3 1 1 . 9 14 .0 14.6 16.2 16 .7 Egg ' 7 5 8 . 3 9 . 9 8 . 4 1 0 . 0 9 . 4 9 . 0 Fascicle Average 8 . 9 10 .4 1 2 . 2 13.1 14.3 1 5 . 5 Percent 1 1 . 9 1 3 . 9 16.3 1 7 . 6 1 9 . 3 2 0 . 9 Other Factors of Foliage affecting 1nsect Di str i but ion Foliage variation was also studied under the following headings: number of needles per fascicle, and needle weight per fascicle derived by using the regression equation of Ives (1955)> as it was not possible to obtain completely sound (undamaged) needle fascicles. Several factors could - 128 -influence the number or weight of needles per fascicle. Those that were thought important for study include: location of the fascicle within the branch, location within the vertical crown levels, defoliation and position of trees in the stand. Number of Needles per Fascicle The sizes of fascicles varied between different sections of the same branch. Therefore, 10 fascicle samples were selected from different parts of the branch for each crown level and exposure, to obtain an average number of needles per fascicle. The sampling date was that of the 1975 pupal collection (June 1 5 , 1 9 7 5 ) . On the average, the numbers of needles per fascicle increase with the height in tree crown levels (Table 2 9 ) , with the edge trees having slightly higher numbers per fascicle than the interior stand trees. Table 2 9 . Average number of needles per fascicle by tree and crown levels. Crown Level Tree Lower Mid Upper Range 1 26 31 33 9 - 4 5 3 2k 37 39 5 - 5 2 6 3 5 36 38 19 -51 7 32 kO k3 1 1 - 5 4 - 129 -Needle Length and Fascicle Weight There is considerable variation in needle length (Table 3 0 ) . There is a slight increase with height in the crown, and the open grown and edge trees (average 3 .40 cm in length) had longer needles than the interior stand trees (average 2 . 3 5 cm in length). An examination of needle length provides information on,foliage variation, defoliation effects on needle length coupled with effects on loss in photosynthesis and oviposition sites. Table 30 . Average needle length (cm) for taken from 4 trees in 1975. samples - • Tree Crown Level No. Lower Mid Upper Average 1 2.10 .2 .39 2.60 2.36 3 2.23 2.41 2.39 2.34 Average 2.16 2.40 2.49 6 3-31 3.63 3 . 5 2 3.49 7 3.01 3.34 3.26 3.34 Average 3.16 3.48 3-39 From a few preliminary samples it appeared that the regression equation of Ives (1955) for calculation of fascicle weight could be used. The formula is: Y = 2.988 + 0.04658X where y = weight of foliage in mg x = total needle length per fascicle in mm - 130 -Table 3 1 . Average weight of fascicle as calculated with the use of Ives ( 1 9 5 5 ) formula for an interior ( 1 ) and an edge ( 6 ) tree. Average Weight of fascicle (based on hO fascicles/level)mg Tree Crown Level No. Lower Mid Upper Average 22 32 38 31 6 51 56 60 56 Intra-branch Variation in Fascicle Size Therewas an indication that the fascicle size i.e. the number or weight of needles per fascicle increased from the distal (periphery) to the basal section of the branch, but this trendwas not consistent. Inter-and Intra-tree Variation The variation in fascicle sizes between trees was significant (P > 0 . 0 1 ) . This indicates that there may be considerable variation in fascicle size or number between trees, even when the trees appear similar. However, this difference is small between trees in the same stand positions. Table 31 shows that the weight of needles per fascicle varies significantly with crown level (lower < mid < upper) and stand position, being lighter in weight on the interior stand tree (31mg) than on the edge tree ( 5 6 m g ) . - 131 -Discussion on Needle Fascicles As was expected the results of the needle fascicle analyses have shown that the top levels, which receive more light and are younger, have more fascicles per unit of branch surface than the lower crown levels. Also the upper levels produce more new shoots, and although not the choice oviposition sites, the current shoots would provide good feeding sites for the final instar larvae the following spring. There is a similar variation in the horizontal distribution of foliage, with fascicles per unit branch section increasing from the basal part to the distal part of the branch. The differences between the various sections of side branches was not so pronounced, with the side branches having significantly more fascicles than the main branch. Analyses of the data have therefore indicated that branches should be sampled from a l l crown levels, because the amount of foliage per linear unit of branch differs for each level. However, to reduce the amount of work that would be involved, the mid crown level, which is about average, should be sampled. Alternatively, the side branches contain about the same number of fascicles per unit branch length regardless of crown level and horizontal crown position, and could be used for sampling. The above data and analyses of fascicles provided quantitative information on the number of eggs in relation to the number of fascicles per unit of branch size. This indicated that the availability of suitable : dviposition sites (i.e. Fascicles) may actually have a limiting effect on the size of insect population. But both the searching a b i l i t y of the insect and the suitability of available needles will affect the number of needle fascicles utilized for oviposition, - 132 -when fascicles are sparse in relation to the number of adults. The female moths, as observed in the f i e l d , have sufficient searching a b i l i t y to find any fascicle. However, a l l of the sites may not be suitable for oviposition at the time the insects are laying eggs. For example, needles may be heavily damaged (as shown earlier the insect prefers sound needles), adventitious needles may not be produced after heavy defoliation, or new shoots may not be sufficiently developed early in the season. Insect Morta1?ty Insect Mortality and Effects on Distribution As often occurs in sampling, mortality factors can be evaluated only by measuring insect population reduction at successive intervals during various developmental stages. In this study mortality is assessed by subdividing the l i f e stages into 4 periods: Period 1: Eggs to third instar or prewinter larvae. Table 32 shows a total mortality of 51 percent for this period in 197**. Mortality of eggs is usually not very high. No egg parasites were discovered. Mortality of 29 percent of the eggs examined in 1 974 was due to the unusually high percentage (2k%) of empty eggs, contents of which had been sucked out by a hemipterous bug. (Webb 1953 reported that k-kO% of the eggs were sucked out by Deraeocovis nubilis Knight with a mean of 10-15% in eastern Canada). The f i r s t and second instar larvae live in needle mines, and the chief cause of mortality is intraspecific competition in the mines and desiccation caused by dry weather conditions. At moderate or high densities, several eggs are frequently deposited per needle, up to 5 eggs/needle in this - 133 -study. If there are several larvae in a needle, those in the tip portion which is desiccated earlier, usually die. The third instar larvae prepare cases, and as free casebearing insects they feed a l i t t l e before dormancy, and as such, are subjected to death by predators. Predators are so few as to be negligible, and no parasite was reared or has been reported as k i l l i n g this stage. Premature needle f a l l in autumn may have some effect if the casebearers are s t i l l attached to them, or if the larva is unable to complete its case. Period 2: From third instar to fourth instar larvae. Mortality during this period for 1 9 7 4 - 7 5 was 41 percent (Table 3 2 ) . Overwintering mortality is due to cold or predation by birds. Table 33 shows the number and per-centage of desiccated or empty cases found in early spring. This accounted for 1^8 percent mortality; the majority of this was larvae whose cases consisted of needle-tips. Period 3' From fourth instar larvae to pupae. The mortality of casebearers for this period was 42 percent (Table 3 2 ) . This is due mainly to predators (including birds and insects) a flock of siskins (Spinus pinus Wilson) was observed on the 19 April 1974 at Sheep Creek feeding on casebearers for about 10 minutes in the mid crown area. Poor synchronization of the larvae emerging from dormancy in relation to the budding of the larch has been reported as contributing to mortality. Parasites play only a very minor role in mortality at this stage. Period 4: Pupae to Adults. Table 32 shows a mortality of 20 percent from the pupal to the adult stage. This is based on the rearings of pupae collected on 14 June 1975 just at the time of adult emergence. Therefore, mortality was due to hymenopterous parasites, mainly Spilochalcis sp. and Dicladocerus sp. - m -T a b l e 32. Surv i v a 1 o f C. 1 a r i c e l l a t h r o u g h one c o m p l e t e g e n e r a t i o n near Thrums, B.C • 1974 - 1 9 7 5 . Stage ! O f '0 S u r v i v a l o f an avg Devel opinent X M o r t a 1 i t y egg complement o f 50 Egg .6365 50 Larva-L l S L2... • 3250 51.1 L a r v a ( F a l l ) • 3 1 1 5 24. 5 L a r v a ( S p r i ng)' • 1855 . 1 2 6 5 40.6 14. 6 Pupa .1067 .0788 42 .5 8. 4 A d u l t .0849 .0218 20 .4 6. 7 T a b l e 33. Number o f p o s t w i n t e r l a r v a e and d e s i c c a t e d 1arvae by crown 1 e v e l , 1 9 7 5 -T r a p Crown L e v e l 1 1 C C Lower Mid Upper T o t a l s No. No. D e s i c . No . D e s i c • No . D e s i c . L i v e Des i c • 1 0 3 3 77 6 2 1 5 4 45 10 3 4 - - 38 4 4 8 7 13 62 28 13 15 20 222 48 18% 5 16 4 3 67 23 6 6 12 10 71 28 7 1 20 4 35 25 8 7 3 3 220 13 30 39 20 393 89 18% 9 2 - 1 50 3 10 2 - 1 82 3 11 10 13 9 98 32 12 1 1 - 43 2 15 14 11 273 40 13% 13 2 5 6 34 13 14 1 1 - 36 2 15 8 9 13 49 30 11 15 19 119 ^5 2 7 % Grand T o t a l s 69 83 7 0 1007 222 18% - 135 -Spatial Difference in Mortality Spatial difference in mortality depends mainly on the occurrence and behaviour of biotic and abiotic agents. Birds prefer particular parts of the crown?.. Parasites can be unevenly distributed spatially or females may not lay their eggs evenly, but frequently aggregated. Weather influences may differ between the various stand positions (edge, interior and open trees). Vertical Distribution of Mortality Total mortality based on the number of insects per fascicle was greatest in the upper crown for the egg to prewinter larval stages (Per iod 11) , as shown i n Append ix! 4, Tabl es 10 ,11 and Fig. 1 2 . For Period 2 , or overwintering stage, mortality was greater in the mid and upper levels than in the lower crown. For the 4 t h instar larval to pupal stages (Period 3) mortality was greatest in the lower and upper crown levels. Pupal mortality by parasites was greatest in the lower tree crown (Table 3 4 ) . Horizontal Distribution of Mortality in the Crown Although variation in the distribution along the branch is due mainly to migration or the availability of needle fascicles, variation in mortality could also play an important part (Table^36) . For example, lower mortality is caused by birds on the twig tips. This is attributable, either to higher insect density or possibly because birds do not prefer to s i t on the thin ends of twigs. Jagsch ( 1 9 7 3 ) noticed that bird - 136 -predation of larvae was usually more pronounced on thick branches which hang down from old trees. Table 36 shows that mortality due to parasites was greatest on the outer section of both the side and main branches. This is probably due to the higher density of larvae on the outer crown section, or the preference of the parasites for this crown section. This theory would account for the non significant difference in parasitism due to exposure (Table 3 5 ) . Seasonal Fluctuation in Casebearer Population Fig. 21 illustrates the seasonal fluctuations in population of larch casebearer in the study areas. Significant declines occurred in population between egg and larva^, the overwintering period (L^-L^) and larva^ to pupa. Very l i t t l e decline in numbers from pupa to adult stage occurred due to low percentage parasitism. - 137 -Table 3 4 . Average number of pupae, adults and parasites per fascicle by crown level, 1 9 7 5 -L e v e l Lower Mid Upper Pupae 0 . 0 8 0 0 . 1 0 8 0 . 0 9 1 Adults 0 . 0 7 1 0 . 1 0 0 0 . 0 8 3 Parasites 0 . 0 0 9 0 . 0 0 8 0 . 0 0 8 Table 3 5 - Average number of pupae, adults, parasites per fascicle by exposure and branch type, 1 9 7 5 . Exposu re Branch Exposed Shaded Main Side Pupae 0.114 0 . 0 7 2 O.O85 0 . 1 0 1 Adults 0 . 1 0 5 0 . 0 6 5 0 . 0 7 9 0 . 091 Parasites 0 . 0 0 9 0 . 0 0 7 0 . 0 0 6 0 . 0 1 0 Table 3 6 . Average number of pupae, adults and parasites per fascicle by horizontal crown position, 1 9 7 5 -Main Branch Side Branch Inner Mid Outer Inner Mid Outer Pupae 0 . 0 6 9 0 . 0 8 2 0 . 1 0 5 0.045 0 . 0 9 1 0 . 1 6 5 Adults 0 . 0 6 6 0 . 0 7 9 0 . 0 9 0 0.041 0 . 0 8 5 0.148 Parasites 0 . 0 0 3 0 . 0 0 3 0 . 0 1 5 0.004 0 . 0 0 6 0 . 0 1 7 - 138 -May June July - 1 9 7 4 -Nov Apr May June 1 9 7 5 — Ju 1 y Fig. 2 1 . Casebearer population per fascicle from May 1974 to June 1 9 7 5 . - 139 -Recommendations for Sampling The development of a sampling technique is straight-forward. All stages of the larch casebearer inhabit the same universe, namely the tree crown, and population densities of the different stages can be estimated by a suitable sampling unit (the needle fascicle) taken at appropriate times. There is one generation per year and no overlapping of the different stages.occur. The Egg Stage: Eggs are laid singly usually on the underside near the tip of the needle. The evacuated egg skins persist on the foliage for about 6 weeks or until the third instar larva begins constructing its case. Therefore, sampling can be delayed until nearly a l l of the eggs have hatched and thereby provide an estimate of: 1. the number of sterile eggs 2. the number of egg chorions from which the contents were sucked out by predators 3. the number of f i r s t instar larvae Egg samples, however, do present some d i f f i c u l t i e s . Eggs are very small (0.3 mm), almost transparent when hatched, widely, scattered, and relative to the number of needles that must be searched, they;are few in numbers at light population densities. Searching for eggs on foliage by microscopic examination is tedious and slow. This can be speeded up by rotation of fascicles with the underside of needles exposed under the microscope or magnifier before needles f a l l off. Extreme care in checking is required. - 140 -The numbers of eggs are usually high enough to warrant the use of only of few sample trees. The Larval Stage Egg hatch may be determined by removing the empty egg and looking for a mine entrance below i t , survival of the larva in the mine may be determined by s l i t t i n g open the mine and teasing out the larvae. Mortality of this stage is usually low, so that therefore, the dis-tribution in the crown would be similar to that of the egg stage. Prewinter Larvae: The overwintering larval stage is probably the easiest to sample because: (a) the timing is not c r i t i c a l (b) it is easy to count the number of insects (after leaf-fall) (c) the population is the most stable spatially (d) the frequency distribution of the number of larvae per fascicle (Table 1) is not as skewed as for the pupal stage There are disadvantages to sampling of this stage: (a) the degree of defoliation cannot be estimated, but damage can be estimated at the time of pupal samples (b) if not kept in cold storage, larval activity can commence again on warming up. The overwintering stage is recommended as the easiest jnsect stage to sample. Sampling of the spring larval stage requires c r i t i c a l timing because of larval activity, the egg stage is very d i f f i c u l t and time consuming to count, and sampling of the pupal stage also requires c r i t i c a l t imi ng. - 141 -Early Spring larvae: After the adults, this is the most active stage. If collection is required just before activity commences, timing can be very c r i t i c a l . Also, if larvae are active, some might be lost during collection. Pupa Pupal cases are usually firmly anchored with s i l k strands attached to foliage, at the base of needle fascicles, or under lichens, and thus persist for 2 weeks before adult emergence and for 6 weeks, or longer after emergence. Their aggregating habits makes the straw-brown cases easy to spot. Late larval populations can be estimated from samples of early pupae. Emergence of adults from pupae may be estimated by samples taken at the commencement of emergence (on the 14 and 15th June at Thrums in 1974 and 1975) and provides an estimate of the female and male moth popu-lation as well as those destroyed by parasites, predators and other causes. Also, adult emergence can be estimated by counting empty pupal cases collected well after adult emergence is completed. Defoliation can be estimated most accurately at this stage. The major disadvantage of this stage is that timing is very c r i t i c a l . The pupal stage is the most important if it is required to study the parasites affecting the casebearer, as a l l parasites reared to date emerged at the mature larval or pupal stages of the casebearer. - 142 -A General Sampling Design A practical sampling design for estimating the population density of a l l l i f e stages of the larch casebearer in a given stand would be a three-stage-'- sampling. The f i r s t stage would be the tree, the second stage would be the crown level within a tree, and the third stage the branches within each crown level. The variable to be estimated should be the number of insects per fascicle. In order to make this design practical, the number of samples to be taken in the second and third stage has to be fixed. Therefore, it is recommended that two crown levels should be sampled, the lower and the middle third of the trees, partly because the upper third is very d i f f i c u l t to sample and partly because the insect population was not significantly higher in the upper third of the crown than in the middle third, for any of the stages sampled. For the third stage, two branches should be taken randomly from each of the lower and mid crown levels. The present study didtnot indicate the necessity of stratifying the branches into two groups to represent shaded and exposed conditions (Appendix 4, Tables 12 S 13) when sufficient samples are included. From each branch three 15~cm lengths should be cut for observations of the number of insects and number of fascicles. The pooled sum of the number of fascicles per three 15-cm samples should constitute one observation. This way, one observation per branch will be the basic observation in the sample. The three 15-cm branch lengths should be randomly selected from the outside * Three-stage sampling is a statistical term, and it should not be confused with the different insect stages. Stage is underlined, when it is referred to sampling in this thesis. - 143 -two-thirds of the branch, since the insect population close to the stem was very low. Main and side branches should be given equal chance to be selected for the 15-cm length. Three 1 5_cm sections are recommended instead of one 18-inch (45 cm) section for easier handling and for better representation of the whole branch. The sum of these three counts should be used in the analysis of the data to avoid the very large number of zero observations which would result in a very skewed frequency distribution. Selection of two branches per crown level and two crown levels per tree will result in a fixed number of 4 observations per tree. The only sample size which has to be calculated in this three-stage sampling design is the number of trees to be sampled. The sample size can be calculated from either a pilot study or a previous analysis carried out under similar conditions. The data have to be analyzed by the following analysis of variance model: Table 37. Analysis of variance for three-stage sampling. Source of Components of Variation DF SS MS Variance Tree n-1 Level within tree n(Z--l) Samples within level nZ(r-l) Total nZr-1 SS, 2 2 2 MST a $ +ra L +ZaT SS. 2 2 M S L aS + r a L SS, MSS a s - ]hk -where n = number of trees in the sample I = number of levels within each tree r = number of samples (branches) within each level 2 0 " ^ = component of variance for samples within levels and 2 = component of variance for levels within trees 2 o"T = component of variance for trees Sample Size From the information above the required number of trees can be calculated-'- as: MST n1 = rl D2 where n1 = required sample size (number of trees) D = desired standard error of the mean in the units of observations (not in per cent) The procedure above was used to calculate the required number of trees to be sampled for the three different stages: * More complicated formula based on the components of variance (Table 3 7 ) is available in texts on sampling techniques, but it is not warranted here because the number of trees to be sampled in a given stand can be considered very large. - 145 -Table 38. The calculated required number of trees to be sampled by insect stage and sampling precision. Life Stage Sampli ng Prec i s ion of Insect 10% 20% Pupa 61 16 Egg 14 4 Larva^ 40 10 - ...a-The sampling precisions were calculated as the per cent of mean obtained from the present study. The analysis in Table 37 has to be calculated in untransformed data, because the variances have to be estimated for the actual observations. Analysis based on transformed data may under-estimate the number of trees to be sampled. The three-stage sampling design suggested above will give a reliable estimating system for the level of population density of the larch casebearer for a l l the stages covered in this study. The data obtained from the design should be analysed by the analysis of variance model given in Table 37 after using the exponential transformation ind icated earlier. The mean per third stage unit for a stand can be calculated as: n 1 r Z E E X X = 1 = 1 j = l k=l ' J l < nZ-r - 146 -th th X... = an observation from the i tree, j level and k branch n = number of trees sampled t = number of levels (2) sampled r = number of branches (2) sampled The standard error of the mean* is defined as: MS.J., n,Z and r are from Table 11 for untransformed data. From these statistics the confidence limits can be computed: t = student's t with nt (r-1) degrees of freedom for the desired probability level. For the present study the mean and the standard error of the mean are given in Table 39 . X + t S-x A more complicated formula is available for data when the number of trees in a stand is not very large. - 147 -Table 3 9 - Mean (X) and Standard error of the mean (S-) for the x various l i f e stages sampled in numbers of insects per fascicle. Life Stage of Insect Parameter 1974 Pupa Egg Larva, L„ 0 . 1 1 0 6 0 . 6 3 6 5 0 . 3 1 1 5 0 . 0 1 7 0 0 . 0 6 1 3 0 . 0 2 2 0 1975 Larva, L^ Pupa Egg 0 . 1 7 9 5 0 . 0 9 3 0 0 . 7 4 9 2 0 . 0 1 2 2 0 . 0 0 8 4 0 . 0 3 2 9 - 148 -SUMMARY AND CONCLUSIONS The study was carried out in immature western larch stands infested with the larch casebearer at Thrums near Castlegar and near Salmo in the Nelson Forest District, of British Columbia, Canada. The sites are typical of the areas of larch casebearer infestations in British Columbia as reported by the Canadian Forest Insect and Disease Surveys since 1966. The stands offered variability in habitat types and permitted the investigation of several factors within one general site. The investigations were based on two 1-year generations of the larch casebearer. Sampling by replication and stratification was conducted in relation to: position of trees in the stand (interior, edge and open-grown trees); different crown levels (lower, middle and upper); different branches at the same level based on exposure to sky light (exposed or shaded); different 15~cm segments (6) of a branch throughout its length; different stages in the insect's l i f e cycle (egg, larva and pupa). Individual subsamples taken to study insect distribution and to develop a sampling design included: (1) the pupal stage, May 197**, 540 subsamples from a total of 15 trees (754*pupae); (2) the egg stage, July 1 9 7 4 , 324 subsamples from a total of 9 trees (2,650 eggs); (3) the i n i t i a l larval stage, Nov. 1 9 7 4 , 432 subsamples from a total of 12 trees (1,400 larvae); (4) the final larval stage, April 1975, 540 subsamples from a total of 15 trees (1,007 larvae,,) - 149 -( 5 ) the pupal stage, June 1 9 7 5 , 540 subsamples from a total of 15 trees (642 pupae); and ( 6 ) the egg stage, July 1 9 7 5 , 2 88 subsamples from a total of 8 trees ( 2 , 6 2 4 eggs). Of the theoretical distributions tested (normal, binomial, Possion and negative binomial) the negative binomial gave the best f i t to the observed data for a l l l i f e stages except the egg stage which approached the normal distribution. The variance of the number of insects per needle fascicle, calculated for each tree sampled, was related to the mean. Therefore, approximate normality of the data was achieved by the application of Taylor's power transformation. The results of this study indicated that for a l l insect stages the number of insects per fascicle was significantly different from tree-to-tree. The averages showed slightly more insects per fascicle on the edge trees than on the interior or open-grown trees. This inter-tree variation indicates that several trees are required in sampling edge or open-grown and suggests that sampling based on interior stand trees alone will reduce this variation. The 3 crown levels showed s t a t i s t i c a l l y significantdifferences in insect population densities for a l l stages, but the trends were~ different from tree-to-tree. The differences in crown levels however were, on the average, highly significant for eggs in 1974 and 1 9 7 5 -The number of larvae (L^ and L^) and pupae per fascicle did not, on the average, differ significantly between exposed or shaded branches. However, the number of eggs per fascicle was significantly higher on - 150 -exposed branches than on shaded branches in 1974, whereas the reverse was evident in 1975. This may be interpreted as owing partly to the lower intensity of needle defoliation on exposed branches in 1974, as well as to the greater amount of light along the stand margins or openings. In 1975, higher infestation on exposed branches reduced the number of oviposition sites, so that more eggs were deposited on the shaded branches. Significantly, more pupae per fascicle occurred on the side (lateral) branches than on the main branches. More overwintering larvae per fascicle occurred on the main branches than on the side branches because: (1) greater number of needles were damaged by larval feeding on the side branches, thereby, leaving less food for the new larvae to eat; and/or (2) some intrinsic behaviour pattern whereby the insect sorts out the better overwintering shelter (under lichens, bark, etc.) offered by the main branch. This could also account for the fact that the sampling of the outer 18-inch (45cm) tips of branches by the Canadian Forest Insect and Disease Survey indicated a sharp decline in postwinter population in 1969. The cause of the decline was attributed to the severe winter (-40C in certain areas). The reason for the subsequent increase in populations (1970) was previously given as due to the release of the great potential of the insect to increase when climatic factors became favourable. As a result of this study, it is proposed that the apparent decline in 1969 and subsequent increase of casebearer populations was due.to the shift in densities from the side branches and branch tips to the main branches. The side branches are more subjected to the effects of severe winters than the main branches. Therefore, heavy mortality of the overwintering stage in 1969 did occur but mainly on the smaller side branches and tips - 151 -than on the main branch. This therefore indicates a need for a sampling design that would incorporate representation from both main and side branches in any sampling of the overwintering larval stage. The egg population was not significantly different on the two branch positions. Therefore, any portion of the branch can be sampled for eggs. In general, the population density close to the stem is low for al l stages, and increases toward the outer part of the crown. This is particularly so for the larvae (and therefore the pupae), as the number of needle fascicles (food) per linear unit increases with distance from the stem to periphery and on side branches. This finding indicates the need for sampling from the entire branch or making adjustment for the variation in insect distribution. However, in heavy infestations the distribution of eggs will follow that of the undamaged needles (if no refoliation occurs). The female adults deposit eggs singly and scattered over the tree crown with the result that, assuming that sufficient numbers of eggs are available, most needle fascicles receive some eggs. This represents a degree of randomness, which is important if mortality through competition in the needle-mines is to be avoided. Other habits of the adults responsible for distribution of eggs are: their tendency to concentrate oviposition in the most illuminated zones of the habitat, with the result that egg density is directly related to both position of trees in the stand (open grown trees had more eggs) and height in the tree. Spatial distribution is dominated by irregularity or heterogeneity of habitat, as the female adult seeks out undamaged needles to deposit her eggs. - 152 -Eggs are laid over the entire branch, and not only on the current shoot growth as reported by Sloan (1965) in Michigan, U.S.A. Egg-deposition site preferences for egg laying as found in the present study were: (1) Adventitious needles, by refoliation after heavy infestation. However, after 3 years of continuous refoliation the stand of trees would stop producing adventitious needles; (2) Undamaged old growth foliage; (3) Current shoot, which probably do not develop in time for egg laying, and the short spike-like needle shapes are not conducive to oviposition. Sloan also mentioned that adult females preferred current shoots of about 5"10 cm. in length. The mid-section of the branch had the greatest density of pupae. This indicates that most of the feeding by larvae commenced on the outer section in early spring. The pupal stage is the stage from which most of the natural insect parasites emerges. The most common indigenous parasites are, Dicladocerus westwoodii (Eulophidae) and Spilochalcis albifrons ( Cha1cididae). Dicladocerus adults appeared to lay eggs on nearby casebearer larvae on the same branch and on a l l portions of the tree crown. Spilochalc i s parasitizes casebearer larvae mainly in the lower crown level and on the two outer portions of the crown. This indicates the need for sampling mature larvae or pupae for parasites by taking samples from many different branches in the different vertical and horizontal positions in the tree crown. Finally, a sampling method which considers shifts in populations in the tree crown has been developed for the larch casebearer on western larch. - 153 -LITERATURE CITED Andrews, R.J., E.V. Morris and N. Bauman. 1 9 6 7 - Larch casebearer, Coleophora lar icella (Hbn.) pp . 1 3 5 _ 1 3 7 - J_n, Annual District Reports Forest Insect and Disease Survey British Columbia 1966. For.Res.Lab. Victoria, B.C. Inf.Rep. BC-X-11. Anscombe, F.J. 1 9 4 9 - The statistical analysis of insect counts based on the negative binomial distribution. Biometrics 5 . :165 "173 -Anscombe, F.J. 1 9 5 0 . Sampling theory of the negative binomial and logarithmic series distribution. Biometrika 3 7 - 3 5 8 - 3 8 2 . Bartlett, M.S. 1936. Square root transformation in analysis of variance. J. Roy. Statist. Soc. Suppl . (Ser.B) 3_: 6 8 - 7 8 . Bell, M.A.M. 1 9 6 5 . The dry subzone of the interior western hemlock zone. Part 1. Phytocoenoses. Ecol. of Western N.A. _l_:42-56. Bliss, C.I. 1 9 5 8 . The analysis_of insect counts as negative binomial distributions. Proc. int. Congr. Entomol . Vol. 2 pp . 1015 " " 1031 . Bliss, C.I. and R.A. Fisher. 1 9 5 3 - Fitting the negative binomial distribution to biological data. Biometrics 9 : 1 7 6 - 2 0 0 . Burst, R. and G. Ewald. 1 9 5 5 . Neue Utersuchung Uber die Biologie und die Bekampfungmog1ichkeiten der Larchenminiermotto. Allgem. Forstzeit. j _ 0 : 3 2 6 - 3 2 9 . Ciesla, W.M. and W.E. Bousfield. 1 9 7 4 . Forecasting potential defoliation by larch casebearer in the northern Rocky Mountains. J.Econ. Entomol . 67_:47-51 • Craighead, F.C. 1 9 5 0 . Insect enemies of eastern forests. U.S.D.A. Misc. Publ. 6 5 7 . p.490. Debauche, H.R. 1 9 6 2 . The structural analysis of animal communities of the s o i l . j_n_ Murphy, P.W. (ed.), Progress in Soil Zoology, pp.10-25 . DeLong, D.M. 1 9 7 1 . The bionomics of leafhoppers. Ann.Rev.Entomol. 1 6 : p.187. Denton, R.E. 1 9 5 8 . The larch casebearer in ldaho--a new defoliator record for western forests. Intermountain For.Exp.Sta., US For.Ser., Res. Note 5 1 , 6 p p . - 154 -Denton, R.E. 1964. The l a r c h c a s e b e a r e r i n w e s t e r n l a r c h f o r e s t s -N o r t h e r n Rocky Mountain Region - A problem a n a l y s i s . USDA, F o r e s t S e r v i c e , I n t e r m o u n t a i n F o r e s t and Range Experiment S t a t i o n , Ogden, Utah. 24pp. Denton, R.E. and S. Tunnock. 1972. L a r c h c a s e b e a r e r i n w e s t e r n l a r c h f o r e s t s . USDA F o r e s t S e r v i c e , F o r e s t P e s t L e a f l e t 96, 8pp. Dowden, P.B. 1957- B i o l o g i c a l c o n t r o l o f f o r e s t i n s e c t s i n t h e U n i t e d S t a t e s and Canada. J . F o r e s t . 55_: 723~726. Eidmann, H.H. 1961. Zur E n t w i c k l u n g von Epilamps i s boops Thorns, und C i r r o s p i l u s p i c t u s Nees, zwei P a r a s i t e n der L a r c h e n m i n i e r m o t t e C o l e o p h o r a l a r i c e l l a Hbn. Ent. T i d s k r . 82:52~55» Eidmann, H.H. 1965- O k o l o g i s c h e und p h y s i o l o g i s c h e S t u d i e n uber d i e L a r c h e n m i n i e r m o t t e C o l e o p h o r a l a r i c e l l a Hbn. S t u d i a f o r . Suec. no. 32, 226 pp. E s c h e r i c h , K. 1931. D i e F o r s t i n s e k t e n M i t t e l e u r o p a s . B e r l i n : V e r l a g . Paul P a r e y , 825pp. Ewald, G. and R. B u r s t . 1959- Untersuchungen uber L a r c h e n b l a s e n f u s s und L a r c h e n m i n i e r m o t t e . A l l g e m . F o r s t - u . J a g d z t g . 139: 173~l8l. F a l ' k o v i c h , M.I. 1964. C a s e b e a r e r s ( L e p i d o p t e r a : C o l e o p h o r i d a e ) damaging th e l a r c h i n t h e U.S.S.R. T h e i r d i s t r i b u t i o n and h i s t o r i c a l r e l a t i o n s t o ho s t p l a n t s . Z o o l . J . 43_(6) :851 -858. ( T r a n s l . from R u s s i a n ) . Can.Dep.For.Trans 1. 8, 20pp. Hagen, H.A. 1886. C o l e o p h o r a l a r i c e l l a HB. Very i n j u r i o u s t o L a r i x  europea i n Mass. Can.Entomol. 18: 125. H a r c o u r t , D.G. 1961 a. Design o f a s a m p l i n g p l a n f o r s t u d i e s on t h e p o p u l a t i o n dynamics o f the diamondback moth, P 1 u t e l l a  macu1i penn i s ( C u r t . ) ( L e p i d o p t e r a : P I u t e l 1 i d a e ) . Can.Entomol. 93:820-831. H a r c o u r t , D.G. 1961b. S p a t i a l p a t t e r n o f the imported cabbageworm, P i e r i s rapae (L.) ( L e p i d o p t e r a : P i e r i d a e ) , on c u l t i v a t e d C r u c i f e r a e . Can.Entomol. 93: 945~952. H a r c o u r t , D.G. I963. Major m o r t a l i t y f a c t o r s i n the p o p u l a t i o n dynamics o f the diamondback moth, P l u t e l l a m a c u l i p e n n i s ( C u r t . ) ( L e p i d o p t e r a : PI u t e l 1 idaeT"! Mem. Entomol . Soc. Can. 32: 55-66. H a r c o u r t , D.G. 1965• S p a t i a l p a t t e r n o f t h e cabbage l o o p e r , Tr i c h o p l u s i a  n i , on C r u c i f e r s . Ann.Entomol. Soc. Amer. 5_8^  89~94. H e a l y , M.J.R. and H.R. T a y l o r . 1962. T a b l e s f o r power-law t r a n s f o r m a t i o n s . B i o m e t r i k a 49= 557"559-- 155 -Heron, R.J. 1966. The reproductive capacity of the larch sawfly and some factors of concern in its measurement. Can.Entomol. 98: 561-578. Henson, W.R. 1954. A sampling system for poplar insects. Can.J. Zool . 32: 421-433-Herrick, G.W. 1912. The larch casebearer. Cornell Univ. Agr. Exp. Sta. Bui. 322, pp. 37-56. Ithaca, New York. Herrick, G.W. 1929. The larch casebearer. Proc. Nat'l. Shade Tree Conf. 5_: 67-68. Herrick, G.W. 1935. The larch casebearer Coleophora l a r i c e l l a Hbn. P. 138-140, _[_n_: Insect enemies of shade trees. Ithaca, N.Y.: Comstock Publ. Co., 417pp. Ives, W.G.H. 1955. Estimation of egg populations of the larch sawfly, Pristiphora erichsoni i (Htg.); Can.J. Zool. 33_: 370-388. Ives, W.G.H. 1959- A technique for estimating tamarack foliage production. A basic detailed population study of the larch sawfly. Can. Entomol. 91=513-519. Jagsch, A. 1973- Population dynamics and parasite complex of the larch casebearer moth European larch. App. Entomol. 73_: 1-42. Jung, W. 1942. Contributions to the knowledge of the larch casebearer (Coleophora l a r i c e l l a Hbn.). Z. ang. Entomol. 29_:475_517. (Trans 1 at ionJT Katti, S.K. and J. Gurland. 1962. Efficiency of certain methods of estimation for the negative binomial, and the Neyman type-A distributions. Biometrika 35:6-15. Kozak, A. and D.D. Munro. I963. An I.B.M. 1620 computer program to f i t frequency distributions. Forestry Chron. 39= 337-338. Kozak, A. and J.H.G. Smith. 1965- A comprehensive and flexible multiple regression program for electronic computing. Forestry Chron. 4j_: 438-443. Krajina, V.J. 1965- Ecology of forest trees in British Columbia. Ecol. of Western N.A. 2: 1-146. Legay, J.M. 1963 - A propos de la repartition de la cecidomyie du Hetre, Mi kiola fag i . Un exemple de distribution binomiale negative. Ann. Epiphyt. Cl4:49~56. - 156 -LeRoux, E.J. and C.Reamer. 1959- Variation between samples of immature stages, and of mortalities from some factors, of the eye-spotted bud moth, Spilonota ocellana (D. & S.) (Lepidoptera: 01ethreutidae), and the pistol casebearer, Coleophora  serratella (L.) (Lepidoptera:Coleophoridae), on apple in Quebec. Can.Entomol. 9]_: 428-449. Loos, C. 1891-92. Einige Beobachtungen uber Coleophora la r i c e l l a auf dem Schluckenauer Domanengebiete. Centralbl. f.d. Forstwesen. 17:375~379; 18:425"531 (from Webb, 1953). McGuire, J.U., T.A. Brindley and T.A. Bancroft. 1957- The distribution of European corn borer larvae, Pyrausta nubi1 a 1i s (Hbn.) in field corn. Biometrics 1 3'• 65~78. Morris, R.F. 1955- The development of sampling techniques for forest insect defoliators, with particular reference to the spruce budworm. Can.J.Zool. 33:225-294. Moriuti, S. 1972. Two new economically important species of Microlepidoptera infesting larch in Japan (Lepidoptera: Coleophoridae and Tortricidae). Kontyu 40:254-262. Neyman, J. 1939- On a new class of 'contagious' distributions, applicable in entomology and bacteriology. Ann.Math. Stat. J_0: 35-57-Oakland, G.B. 1953- Determining sample size. Can.Entomol. 85:108-113. Peirson, H.B. 1927- Manual of forest insects. Bull.No. 5 Maine Forest Service. 13PP-Piatt, R.B. and J.F. Grif f i t h s . 1964. Environmental Measurement and interpretation. Reinhold Pub. Corp. N.Y. 22. Quednau, F.W. 1967. Notes on mating, oviposition, adult longevity, and incubation period of eggs of the larch casebearer, Coleophora l a r i c e l l a (Lepidoptera:Coleophoridae), in the laboratory. Can.Entomol. 99j397~401. Quenouille, M.H. 1950. Introductory Statistics. Pergamon Press, London. 248pp. Ratzeburg, J.T.C. I869. Die Wa1dverderber und ihre Feinde 6^ 72, 160-162. Berlin, (from Webb, 1953). Reissig. 1869- Die Larchenmotte, Coleophora l a r i c e l l a Hb., (Tin. 1ar ic i nel1 a Bchst.) Zeit.f. Forest-und Jagdwesen. 1:129-137. (from Webb, 1953). - 157 -Roe, A.L., R.C. Shearer and W.C. Schmidt. 1970. Management of western larch. (Manuscript in preparation). From, Management of western 1 arch.... Northern Region. U.S.D.A. For.Service Handbook FSH2471.15RI. Rojas, B.A. 1964. La binomial negativa y la estimacion de intensidad de plagas en el suelo. Fitotecnia Latinamer. J_:27~36. Schimitschek, E. 1963- The influence of warm slopes and poor, degraded soil on the population density of primary needle-eating pests of the host plant. Zeit. Angew. Entomol. 53 '• 69~8l . Schindler, U. 1965- Zur Bekampfung der Larchenminiermotte. Forst Holzw. 20:348-353. Schindler, U. 1968. Population change in a typical perennial forest pest, the larch casebearer. (Transl. from German). Can.Dep. Fish.For.Transl. 00FF-111, 13p. Schwarz, I.H. 1933- Neue Scha'dlinge der Douglasie. Z.f. PIanzenschutz 43:417-418. Schwerdferger, F. and G. Schneider. 1957. Uber den Einfluss von Larchenminiermottenfrass auf Benadelung und Zuwachs der Larche. Forstarchiv 28_: 1 1 3-1 1 7. Shepherd, R.F. and D.A. Ross. 1973- Problem analysis:larch casebearer in B.C. Pac. For. Res. Cent. Int. Rept. BC-37-Skellam, J.G. 1952. Studies in statistical ecology. I. Spatial Pattern. Biometrika 39:346-362. Sloan, N.F. 1965. Biotic factors affecting populations of the larch casebearer, Coleophora l a r i c e l l a Hbn., in Wisconsin. Ph.D. Thesis, Univ. of Wis., Madison, Wisconsin. 193PP-Southwood, T.R.E. 1966. Ecological Methods. Methuen Co., London. 391pp. Taylor, L.R. 1961. Aggregation, variance and the mean. Nature Lond. 189:732-735. Taylor, L.R. 1965- A natural law for the spatial disposition of insects. Proc. XII. Int. Congr. Entomol. pp.396-397. Tunnock, S., R.E. Denton, C.E. Carlson and W.W. Jansen. 1969- Larch casebearer and other factors involved with deterioration of western larch stands in northern Idaho. USDA Forest Serv. Res. Pap. lnt-68. 10pp. - 158 -Waters, W.E. and W.R. Henson. 1959- Some sampling attributes of the negative binomial distribution with special reference to forest insects. Forest Sci. 5_:397"4l2. Webb, F.E. 1953. An ecological study of the larch casebearer Coleophora  laricel la Hbn. (Lep idoptera : Col eophor idae) . (Unpublished") Ph.D. Thesis, Univ. of Michigan. 212pp. Webb, F.E. 1957- Sampling techniques for the overwintering stage of the larch casebearer. Can. Dep. Agr. Bi-Monthly Prog. Rep. 1 3 0 0:1-2. Webb, F.E. and F.W. Quednau. 1971. Coleophora l a r i c e l l a (Hubner), larch casebearer (Lepidoptera:Coleophoridae). In Biological control programmes against insects and weeds in Canada 1959-1968. Tech. Comm. No.4, Comm. Inst. Biol. Contr. 266pp. Williams, C.B. 1964. Patterns in the balance of nature. In, Theoretical and experimental biology Vol. 3, Academic Press, New York. 325pp. Van Poeteren, N. 1933- Verslag over de werksaamheden van Plantenziekten kundigen Kienst in het jaar 1932. PI. Ziektenk. Dienst. 72. Wageningen. den Versl . - 159 -Appendix 1. Pupal data sheet and sampling notes. - 159a -LARCH CASEBEARER PUPAL DATA SHEET Tree No. Date Co l lec ted : Area: Crown Class : Crown I Level Exposure Branch Length Branch Sect ion 1 2 3 It 5 6 1 2 3 *» 5 6 1 2 3 k 5 6 1 2 3 1> 5 6 1 2 3 5 6 1 2 3 5 6 No. of Fasc ic les % Damage Q Vol No.of Pupae, No.Emerged Casebearer Tot JEmergence Date Remark: - 160 -LARCH CASEBEARER — SAMPLING NOTES TREE Nos. 1 - k Interior stand trees 5 - 8 Edge trees 9 - 1 2 Open grown trees 13 - 15 Open clusters near Thrums, B.C. near Sal mo, B.C. CROWN LEVEL U = Upper M = Middle L = Lower EXPOSURE S = sampled branch shaded e.g. by other tree(s) L = sampled branch not shaded e.g. facing out from stand BRANCH SECTION (15-cm or 6-inch length) Nos. 1, 2, 3 for main branch.starting from the stem k, 5, 6 for side branch starting from the stem DEFOLIATION RATING (see Text) 0 = Negligible -- no visible defoliation 2 = Light -- up to 25% defoliation k = Moderate — 26 to 50% defoliation 6 = Heavy -- 51 to 75% defoliation 10 = Severe -- over 75% defoliation - 161 -Appendix 2. Form used for egg counts. - 161a -EGG COUNTS — LARCH CASEBEARER JbATL COLLECT£bi AfiEA •• «J Ul 1 S P M UJ VI lis 11 s 1 5 S e EGG >u vl NO. Of L&C/FAS. PQRWN DF . NECbLB NtRbLE NEEDLE SURFACE CL. VOL. a: v: i-% s: I l l + 5" AP. Mm. BASE 1 * vi. a, ^j BL 1 I i 4 r (, BS 1 I I $ £ ML I 1 3 f S (, us/ z 3 f c out I 3 * C I 3 S (, • - 162 -Appendix 3« Form used for larval counts. - 162a -JiATB COLLECTED •• LARCH CASEBEARER — LARVAE DATA AREAi CAOIV/V CLASS: TAtE NO. T/\tB NO. 5: " «a VJ 5 Vj V O 5 Q VJ ; HI • Hi Appendix k. Tabl Table 1. Observed and expected frequencies for pupae per fascicle per 15-cm branch section. CI ass - 1974 -Frequency Di str i but ions - 1975 " Middle Obs. Neg. Binom. Chi-sq. Obs. Neg.Binom. Chi-sq. 0.025 255 260.8 0.13 316 323.4 0.17 0.075 79 94.8 2.64 74 72.0 0.05 0.125 58 56.0 0.07 47 39.1 1.59 0.175 30 36.9 1.30 25 25.4 0.01 0.225 33 26.1 1.80 17 17.9 0.04 0.275 23 18.8 0.91 14 13.1 0.05 0.325 14 13.8 0.00 10 9.9 0.00 0.375 9 10.3 0.17 3 7.7 2.85 0.425 11 7.9 1.22 9 6.0 1.47 0.475 8 6.0^  ) 4 4.7 0.12 0.525 13 4.8 y 12.94 5 3.8 0.37 0.575 7 3.6 4 3.1 0.27 0.625 2 2.5 0.10 0.675 2 2.0 0.00 0.725 0 1.6^ 0.775 3 1.4 0.825 0 1.1 0.875 0 0.9 0.925 0 0.8 r* 0.15 0.975 c 0 c vl 1.475 J ) 2 Total: 540 540 21.21 * 540 7.25 df 7 df 12 Significant (P = 0.01) T a b l e 2 . Observed and e x p e c t e d f r e q u e n c i e s f o r pupae per 15 -cm branch s e c t i o n . C l a s s Frequency D i s t r i b u t i o n s - 1974 - - 1 9 75 ~ M i d d l e Obs. Neg. Binom. C h i - s q . Obs. Neg. Binom. C h i - s q . 0 250 2 5 2 0.01 301 3 0 2 0 . 0 0 1 116 119 0 . 0 8 111 105 0 . 3 8 2 62 67 0.41 49 54 0 . 4 4 3 42 40 0 . 0 9 23 31 1.94 4 29 25 0.82 27 18 3 . 9 5 5 13 15 0 .31 8 11 0 . 9 8 6 14 10 2 . 0 8 5 7 0 . 6 2 7 4 6 0 . 6 5 2 5 1 .39 8 5 4 0 . 3 7 3 3> 9 5 2 2 . 7 0 5 2 2 . 8 9 10 5 1 T o t a l : 540 7 . 5 4 7 df 1 2 . 5 9 6 d f Table 3 - Observed and expected frequencies for eggs per fascicle per 15 -cm branch section. Class Frequency 1 Di str ibut ions - 1974 - - 1 975 -Middle Obs. Normal Neg.Binom. Chi-sq Obs. Normal Chi-sq. Neg.B inom. Chi-sq. 0 . 0 5 23 1 1 . 2 1 2 . 0 24 8 . 6 7 . 8 0 . 1 5 27 1 5 . 7 24 . 7 0 . 2 0 12 1 1 . 6 1.6 16 . 9 0 . 0 5 0 . 2 5 26 2 0 . 6 3 3 . 7 1 .77 26 15 . 0 8 . 0 24 .1 0.14 0 . 3 5 30 2 5 . 4 38 .1 1 .72 22 1 8 . 4 0 . 6 2 8 . 3 1.42 0 . 4 5 23 2 9 . 4 38 .1 5 . 9 7 20 2 1 . 5 0.1 2 9 . 7 3.17 0 . 5 5 47 3 1 . 9 3 6 . 3 3.14 25 2 3 - 9 0 . 0 2 8 . 9 0 . 5 3 0 . 6 5 27 3 2 . 5 3 1 . 5 0 . 6 5 18 2 5 . 3 2.1 2 6 . 6 2 . 8 2 0 . 7 5 25 3 1 . 0 2 6 . 7 0 .11 20 2 5 . 4 1.1 2 3 . 6 0 . 5 6 0 . 8 5 20 2 7 . 8 2 1 . 9 0 . 1 7 20 24 . 4 0 . 7 2 0 . 3 0 . 0 0 0 . 9 5 12 2 3 . 5 17 .3 1.66 10 2 2 . 2 6 . 7 17 . 0 2 . 9 2 1 .05 20 1 8 . 5 13 .9 2 . 6 2 21 1 9 . 3 0.1 14 . 0 3.46 1.15 7 13 . 8 1 0 . 4 1 .13 13 15 . 9 0 . 5 1 1 . 3 0.24 1 .25 8 9 . 6 8 . 0 0 . 0 0 6 1 2 . 5 3 . 4 9 . 0 1.02 1 .35 4 6 . 3 5 . 9 0 . 6 2 10 9 . 4 0 . 0 0 7.1 1.16 1.45 25 3 . 8 5.1 37 6 . 7 5 . 5 Total: 324 1 9 . 7 7 2 8 8 2 5 . 6 1 7 . 5 3 High values excluded from total. Table 4. Observed and expected frequencies for larvae per fascicle per 15-cm branch section 1974-75. r l Frequency Distributions L I 3 S S Prewinter Larvae Postwinter Larvae Middle Observed Ne§. Binom. Chi-sq. Observed Neg. Binom. Chi-sq. 0.025 103 104.7 0.02 204 206.7 0.03 0.075 42 58.9 4.85 78 85.5 O.65 0.125 49 43.4 0.73 68 55.1 3.00 0.175 31 34.0 0.26 33 39.4 1.05 0.225 31 27.8 0.37 29 29.9 0.03 0.275 34 23.2 5.00 22 23.3 0.07 0.325 24 19.5 1.05 19 19.5 0.01 0.375 13 16.4 0.71 14 14.9 0.05 0.425 14 14.1 0.00 8 12.1 1.39 0.475 13 12.1 0.05 6 9.9 1.55 0.525 9 10.5 0.20 16 8.3 7.05 0.575 8 9.1 0.13 6 6.8 0.10 0.625 7 7.9 0.10 7 5-7 0.29 0.675 5 6.9 0.51 2> v 4.7^  0.725 4 6.0 0.66 3 3-9 0.775 6 5.3 0.10 2 i 3.3 0.825 'A 4.6\ 1 30 2- 8 0.875 ' 39 4.0/ V 32.3 1.39 2 f 2 - k 1.475 1.0J • % 1.3J Total: 432 16.13 540 16.91 T a b l e 5. Observed and ex p e c t e d f r e q u e n c y f o r l a r v a e per 15-cm branch s e c t i o n 1 9 7 4 - 7 5 . C l a s s P r e w i n t e r Larvae Frequency D i s t r i b u t i o n s P o s t w i n t e r L a r v a e M i d d l e Obs. Neg. Binom. C h i - s q . Obs. Neg. Binom. Ch i - s q . 0 101 101 .4 0.00 199 199.1 0.00 1 90 80.6 1 .08 141 119.5 3.88 2 54 62.0 1 .03 74 76.0 0.05 3 54 47.2 0.97 39 49.3 2.15 4 37 35.7 0.04 23 32.3 2.67 5 22 26.9 0.90 21 21.3 0.00 6 24 20.3 0.68 10 14.0 1.16 7 10 15.2 1.78 8 9.3 0.18 8 8 11.4 1 .02 5 6.2 0.22 9 1 8.5 6.65 5 4.1 0 . 1 9 10 5 6.4 0.31 1 2.7 1.10 11 5 4.8 0.01 3 1.8 0.76 12 5 3.6 0.55 1 1.2 0.03 13 2 2.7 0.17 3 0.8 5.88 14 4 2.0 1.95 2 0.5 3.91 15 1 1.5 0.16 1 0.4 1.12 16 1 1.1 0.01 3 0.2 -k /V 17 1 0.8 0 . 0 3 0 0.2 0.16 18 8 0.6 T o t a l : 17.35 23.50 ** High v a l u e e x c l u d e d from t o t a l . - 168 -Table 6. Analyses of variance of eggs per fascicle by tree, crown level, exposure, branch type and horizontal crown position. Source DF Egg MS - 1974 F DF Egg -MS 1975 F Tree T 8 0.332 5.0** 7 1.229 6.8** Level L 2 0.899 13.5** 2 2.560 14.2** Exposure E 1 0.167 2.5 1 0.513 2.8 Main & side branch M 1 0.004 0.1 1 2.664 14.8** Parts, within M P 4 0.437 6.6** 4 0.298 1.6 T x L 16 0.081 1.2 14 0.428 2.4* T x E 8 0.073 1.1 7 0.125 0.7 T x M 8 0.070 1.1 7 0.149 0.8 T x P 32 0.063 0.9 28 0.335 1.9* L x E 2 0.013 0.2 2 0.010 0.1 L x M 2 0.149 2.2 2 0.480 2.7 L x P 8 0.052 0.8 8 0.350 1.9 E x M 1 0.052 0.8 1 0.434 2.4 E x P 4 0.057 0.9 4 0.030 0.2 T x L x E 16 0.073 1.1 14 0.675 3.7** T x L x M 16 0.091 1.4 14 0.263 1.5 T x L x P 64 0.062 0.9 56 0.180 1.0 T x E x M 8 0.119 1.8 7 0.086 0.5 T x E x P 32 0.040 0.6 28 0.211 1 .2 L x E x M 2 0.071 1.1 2 0.087 0.5 L x E x P 8 0.016 0.2 8 0.171 0.9 TLEM 16 0.109 1.6 14 0.134 0.7 ERROR 64 0.066 56 0.180 TOTAL 323 287 * Significant within the 0.05 level. ** Significant within the 0.01 level. - 169 -Table 7. Analyses of variance of prewinter and postwinter larvae by tree, crown 1evel, exposure, branch type and horizontal crown pos i t ion. Larvae^ Larvae^ Source DF MS F DF MS F TREE T 11 0.158 4. 5** 14 0.561 4.2** LEVEL L 2 0.502 14.4** 2 0 . 825 6.2** EXPOSURE E 1 0.047 1.4 1 0.472 3-5 MAIN & SIDE BRANCH M 1 0.460 13.2** 1 1.193 8.9** PARTS, within M P 4 0.053 1.5 4 1.230 9. 2** T x L 22 0.032 0.9 28 0.305 2 . 3 * * T x E 11 0.061 1.7 14 0.200 1.5 T x M 11 0.062 1.8 14 0 .152 1.1 T x P 44 0 .028 0.8 56 0.180 1.3 L x E 2 0.045 1.3 2 0.670 5.0** L x M 2 0.023 0.7 2 0.087 0.6 L x P 8 0.050 1.4 8 0.136 1.0 E x M 1 0.002 0.1 1 0.062 0.5 E x P 4 0.021 0.6 4 0 .235 1.7 T x L x E 22 0.043 1.2 28 0.140 1.0 T x L x M 22 0.024 0.7 28 0.169 1.3 T x L x P 88 0.040 1.2 112 0.186 1. 4* T x E x M 11 0.031 0.9 14 0.081 0.6 T x E x P 44 0.035 1.0 56 0.109 0.8 L-x E x M 2 0.030 0.8 2 0.012 0.1 L x E x P 8 0.022 0.6 8 0.211 1.6 TLEM 22 0.039 1.1 28 0.145 1.1 ERROR 88 0.035 112 0.134 TOTAL 431 539 Significant within the 0.05 level Significant within the 0.01 level - 170 -Table 8 . Analyses of variance of pupae per fascicle by tree, crown level, exposure, branch type and horizontal crown position. Pupae '75 Pupae '74 Source DF MS F MS F Tree T 14 0.126 4.5** 0.518 7.9** Level L 2 0.057 2.0 0.183 2.8-Exposure E 1 0.330 11.7** 0.163 2.5 Main S side branch M 1 0.155 5.5* 2.282 34.9** Parts, wi thin M P 4 0.351 1 2 . 5 * * 0.445 6.8** T x L 28 0.053 1.9* O.O83 1.3 T x E 14 0.038 1.3 0.019 0.3 T x M 14 0.079 2.8** 0.137 2.1* T x P 56 0.054 1.9** 0.074 1.1 L x E 2 0.024 0.8 0.010 0.1 L x M 2 0.234 8.3** 0.000 0.0 L x P 8 0.033 1.2 0.026 0.4 E x M 1 0.001 0.0 0.242 3.7 E x P 4 0.020 0.7 0.026 0.4 T x L x E 28 0.062 2.2** 0.086 1.3 T x L x M 28 0.034 1.2 0.061 0.9 T x L x P 112 0.031 1.1 0.057 0.9 T x E x M 14 0.022 0.8 0.032 0.5 T x E x P. 56 0.039 1.4 0.057 0.9 L x E x M 2 0.000 0.0 0.067 1.0 L x E x P 8 0.043 1.5 0.052 0.8 TLEM 28 0.044 1.6 0.052 0.8 ERROR 111 0.028 0.065 TOTAL 538 Significant within the 0.01 level* Significant within the 0.05 level. - 171 -Table 9. Average by tree number of insects per fascicle and position in the stand. Tree 1 N S E C T S T A G E 1 9 7 4 1 9 7 5 No. Pupa Egg Larva^ Larva, 4 Pupa Egg** 1 0.041 0 .414 0 . 3 1 2 0 . 1 0 2 -2 0 . 1 0 7 0 .424 0 . 3 3 8 0 . 1 0 6 0 . 0 7 5 0 . 8 1 4 3 0 .143 0 . 4 7 0 0 . 2 0 3 0 .124 0 . 1 5 3 * * 0 . 4 7 0 4 0 . 1 5 5 0 . 5 7 6 0 . 2 3 3 0 . 1 8 0 0 .142 0 . 6 1 5 1nter ior+ 0 . 1 1 2 0 . 4 8 5 0 . 2 9 7 0 . 1 8 0 0 . 1 1 8 0 . 6 3 3 5 0 . 3 4 5 0 . 5 6 6 0 . 2 7 5 0 . 2 8 2 0 . 1 5 8 0 . 7 8 3 6 0 . 1 7 3 0 . 6 0 3 0 . 3 5 8 0 . 1 9 9 0 . 0 7 5 0 . 6 9 2 7 0 .148 0 .748 0 . 211 0 .189 0 . 1 0 9 0 . 8 6 7 8 0 . 0 6 0 - 0 . 7 5 4 0 . 5 1 3 0 . 1 0 8 -Edge+ 0 .181 0 . 6 3 9 0 . 3 9 9 0 . 2 9 6 0 . 1 1 2 0 . 7 7 9 9 0 . 0 8 5 0 . 6 0 7 0 . 3 9 9 0 . 1 5 6 0 . 1 2 8 1 .103 10 0 . 0 9 9 0 . 9 2 2 0 . 2 5 1 0 . 1 8 7 0 . 1 3 4 -1 1 0 . 0 7 8 0 . 8 1 2 0 .147 0 . 1 5 1 0 . 0 9 5 -12 0 . 0 7 8 - 0 .155 0 . 0 7 2 0 . 0 1 8 0 .649 0pen+ 0 . 0 8 5 0 .783 0 . 2 3 8 0 .142 0 . 0 9 3 O.876 13 0 . 0 3 4 - - 0 . 0 6 3 0 . 0 3 8 -14 0 .037 - - 0 . 0 8 2 0 . 0 2 9 -15 0 . 0 7 1 - - 0 . 0 7 3 0 . 0 2 9 -C1 uster+ 0 .047 - - 0 . 0 7 3 0 . 0 3 2 -Stand Avg. 0 . 1 2 2 0 . 6 3 6 0 . 3 1 1 0 . 1 8 5 0 . 1 0 7 0 .713 + No samples taken. New trees selected in 1 9 7 5 -Average for position of trees in the stand. - 172 -Table 10. Average No.of insects per fascicle by tree and crown position, 1974. T r e e pupa Egg Larva, Crown Position Crown Position Crown Position No Lower Mid Upper Lower Mid Upper Lower Mid Upper 1 0 .023 0 .036 0 . 0 6 6 - * - - 0 .267 0 . 6 6 8 0 .308 2 0 . 1 2 9 0.081 0 . 111 0 . 170 0 . 5 5 2 0 . 5 5 1 0 . 2 9 3 0 . 3 2 3 0 . 3 9 8 3 0 . 1 2 9 0 . 1 5 5 0 .145 0 .534 0 .459 0 .417 0 . 2 1 2 0 . 1 6 3 0 . 2 3 5 4 0 . 1 1 8 0 . 1 3 5 0 .214 0 . 4 0 9 0 . 5 2 3 0 . 7 9 5 0 . 2 8 8 0 . 1 9 6 0 . 2 1 7 lnterior+ 0 . 1 0 0 0 . 1 0 2 0 . 1 3 4 0.371 0 . 5 1 1 O.588 0 . 2 6 5 0 . 3 3 7 0 . 2 8 9 5 0 . 2 8 2 0 .486 0 . 2 6 8 0.541 0 . 5 2 3 0 . 6 3 5 0 . 0 4 3 * * 0 . 2 0 9 * * 0 . 5 7 2 6 0 . 2 2 0 0 .189 O . l i o ' 0 . 570 0 . 5 9 2 0 .645 0 .305 0.561 0 .207 7 0 . 1 2 5 0 . 1 8 4 0 . 1 3 7 0 . 5 7 0 0 . 8 0 0 0 . 8 7 3 O . I85 0 . 2 6 5 0 . 1 8 3 8 0 . 1 2 1 0 .044 0 . 0 1 8 - 0 . 9 0 2 1 .023 0 . 3 3 7 Edge+ 0 .187 0 . 2 2 6 0 . 1 3 2 O.56O O.638 0 .718 0 . 3 5 9 0 .514 0 . 3 2 5 9 0 . 0 7 6 0 . 1 1 7 0 . 0 6 3 0 . 4 3 6 0 . 7 5 3 0 . 6 3 3 0 . 3 6 5 0 . 5 0 1 0 . 3 3 2 10 0 . 0 5 6 0 . 1 8 1 0 . 0 6 1 0 . 6 2 0 0 . 8 5 5 1 . 2 9 0 0 . 1 3 9 0 . 2 1 8 0 . 3 9 5 11 0 . 0 6 4 0 . 1 1 5 0 . 0 5 7 0 . 6 9 6 0 . 9 0 1 0 .842 0 . 1 1 2 0 . 1 7 0 0 . 1 6 0 12 0 . 0 9 4 0 . 0 7 2 0 . 0 7 0 - 0 . 1 1 8 0 . 1 1 9 0 . 2 2 9 0pen+ 0 . 0 7 2 0 . 1 2 1 0 . 0 6 3 0 . 5 8 4 O.836 0 . 9 2 2 O. I83 0 . 2 5 2 0 . 2 7 9 13 0 . 0 2 7 0 . 0 6 0 0 . 0 1 6 - - - -14 0 . 0 5 8 0 . 0 2 5 0 . 0 2 8 15 0 . 0 6 5 0 . 1 0 1 0 . 0 7 4 - - - -Cluster+ 0 . 0 5 0 0 . 0 6 2 0 . 0 3 9 - - - - - -Stand Avg. 0 .106 0.132 0.094 0 .505 0.662 0.742 0 . 2 6 9 O.368 O.298 * No samples taken. ** Shaded branches attacked by bark beetles. + Average for position of trees in the stand. - 173 -Table 1 1 . Average No. of insects per fascicle by tree position in stand and crown position, 1 9 7 5 -Tree Crown Position Crown Position Crown Position No Lower Mid Upper Lower Mid Upper Lower Mid Upper 1 0 . 2 7 9 0 . 2 7 8 0 . 3 7 8 0 . 061 0 . 1 8 4 0 . 0 6 0 4 -2 0 . 1 0 3 0 . 1 2 9 0 . 0 8 8 0 . 0 6 8 0 . 0 6 3 0 . 0 9 2 7 0 . 5 2 0 0 . 7 5 8 1 .163 3 0 . 0 6 4 0 . 1 2 3 0 . 1 8 4 0 . 1 3 9 0 . 2 1 2 0 . 1 0 8 * 0 . 4 4 4 0 . 4 3 2 0 . 5 3 5 4 0 . 2 9 7 0 . 101 0 .143 0 . 1 9 8 0 . 1 9 5 0 . 031 0 . 4 7 6 0 .519 0 .849 lnterior+ 0 . 1 8 6 0 . 1 5 8 O . I98 0 .116 O. I63 0 . 0 7 3 0 . 4 8 0 0 . 5 7 0 0 .849 5 0 .279 0 .384 0 .184 0 .279 0 .104 0.091 O.58I 0 . 6 6 3 1.104 6 0 . 2 6 6 0 . 1 7 7 0 . 1 5 3 0 . 0 6 1 0 . 1 0 7 0 . 0 5 7 0 . 3 6 8 1 . 012 O.696 7 0 . 3 0 8 0 . 1 2 1 0 .139 0 . 0 9 4 0 . 0 7 5 0 . 1 5 9 0 . 9 5 5 0 . 8 0 0 0 .846 8 0 . 6 5 3 0 . 5 7 1 0 . 3 1 6 0 . 0 5 3 0 . 1 0 7 0 .164 -Edge+ 0 . 3 7 6 0 .313 0 . 1 9 8 0 . 1 2 2 0 . 0 9 8 0 . 1 1 8 0 . 6 3 5 0 . 8 2 5 0 . 8 8 2 9 0 . 2 0 6 0 . 1 0 8 0 . 161 0 . 0 7 9 0 . 251 0 . 0 5 3 0 . 9 2 7 1.251 1 . 132 10 0 . 1 6 2 0 .144 0 . 2 5 5 0 .048 0 . 1 3 0 0 . 2 2 2 0 . 2 3 9 0 . 9 0 1 0 . 8 0 8 11 0 .018 0 .113 0 . 3 2 2 0 . 0 5 2 0 . 0 9 0 0 .144 12 0 . 0 2 7 0 . 0 6 3 0 . 1 2 7 0 . 0 3 1 0 . 0 0 7 0 . 0 1 5 " 0pen+ " 0 . 1 0 3 0 . 1 0 7 0 . 2 1 6 0 . 0 5 2 0 . 1 1 9 0 . 1 0 8 O.583 1 . 0 76 0 . 9 7 0 13 0 . 0 5 0 0 .098 0 .042 0 . 0 0 0 0.061 0 .052 -14 0 . 0 5 9 0 . 0 8 5 0 . 101 0 . 0 1 7 0 . 0 1 3 0 . 0 5 8 -15 0 . 1 3 5 0 . 0 7 3 0 . 0 1 2 0 . 0 1 3 0 . 0 1 7 0 . 0 5 6 - -Cluster+ 0 . 0 8 1 0 . 0 8 6 0 . 0 5 2 0 . 0 1 0 0 . 0 3 0 0 . 0 5 5 - - -Stand Avg.0.194 0 . 1 7 1 0 . 1 7 4 0 . 0 8 0 0 . 1 0 8 0 . 091 0 . 5 6 4 0 . 7 9 2 O.89I * New larch trees selected in 1 975 + Average for positions in the stand - 174 -Table 12. Average No. of insects per fascicle by tree position in stand and exposure, 1974. Tree Pupa Egg Larva No. Exposed Shaded Exposed Shaded Exposed Shaded 1. 0.026 0.056 — /V - 0.242 0.586 2 0.097 0.117 0.387 0.462 0.207 0.469 3 0.139 0. 148 0.574 0.366 0.312 0.094 4 0.161 0.150 0.701 0.451 0.299 0.168 1 nter ior+ 0.119 0.157 0.554 0.426 0.265 0.329 5 0.337 0.353 0.546 0.587 0.284 0.265 6 0.141 0.205 0.587 0.618 O.38O 0.336 7 0.128 0.169 0.795 0.700 0.133 0.289 8 0.040 0.081 - - 1 .074 0.434 Edge+ 0.161 0.202 0.643 0.635 0.468 0.331 9 0.087 0.084 0.653 0.563 0.484 0.315 10 0.080 0.118 1.000 0.844 0.271 0.231 11 0.077 0.080 0.903 0.727 0.158 0.136 12 0.073 0.084 - - 0.184 0.126 0pen+ 0.079 0.091 0.852 0.711 0.274 0.202 13 0.030 0.038 - - - -14 0.018 0.057 - - - -15 0.080 0.062 - - - -CI uster+ 0.043 0.052 - - - -Stand Avg, .0.101 0.120 0.683 0.590 0.336 0.287 * No samples taken + Average for position of trees in stand - 175 -Table 13- Average No.of insects per fascicle by tree and exposure, 1975-Tree Larva Pupa Egg* No. Exposed Shaded Exposed Shaded Exposed Shaded 1 0.260 0.364 0.117 0.087 0.744 0.884 2 0.081 0.132 0.102 0.047 0.410 0.531 3 0.099 0.149 0.165* 0.142 — /V ~i\ -4 0.155 0.205 0.163 0.127 0.598 0.631 I nter ior+ 0.149 0.212 0.137 0.101 0.584 0.682 5 0.453 0.112 0.235 0.081 0.882 0.684 6 0.227 0.171 0.090 0.059 0.603 0.780 7 0.236 0.142 0.154 0.065 - -8 0.472 0.555 0.144 0.072 0.789 0.945 Edge+ 0.347 0.245 0.156 0.069 0.755 0.803 9 0.193 0.123 0.182 0.073 1.048 1.160 10 0.265 0.109 0.196 0.071 - -11 0.105 0.198 0.054 0.137 - -12 0.068 0.077 0.009 0.026 0.575 0.723 0pen+ 0.158 0.127 0.110 0.076 0.811 0.941 13 0.077 0.049 0.038 0.038 - -14 0.086 0.077 0.037 0.022 - -15 0.069 0.077 0.026 0.031 - -Clusters- 0.077 0.068 0.034 0.030 - -Stand Avg. 0.190 0.169 0.114 0.072 0.706 0.792 A New trees selected in 1975-j - j - No samples taken. + Average for position of trees i n the stand. - 176 -Table 14. Average No.of insects per fascicle by tree and branch type, 1974. Pupa Egg- La rva Tree Ma i n S i de Main S ide Ma i n Side No. Branch Branch Branch Branch Branch Branch 1 0.045 0.038 - - 0.537 0.291 2 0.049 0.166 0.367 0.481 0.418 0.258 3 0.146 0.140 0.330 0.609 0.232 0.174 4 0.193 0.117 0.638 0.514 0.233 0.234 1nter ior+ 0.108 0.115 0.445 0.535 0.355 0.239 5 0.260 0.431 0.634 0.499 0.406 0.143 6 0.145 0.201 0.607 0.598 0.461 0.254 7 0.092 0.205 0.725 0.770 0.253 0.168 8 \ 0.063 0.058 - - 1.005 0.503 Edge+ 0.140 0.224 0.655 0.622 0.531 0.267 9 0.063 0.108 0.662 0.554 0.623 0.176 10 0.058 0.140 0.949 0.894 0.240 0.261 11 0.037 0.120 1.021 0.604 0.143 0.151 12 0.033 0.124 - - 0.160 0.150 0pen+ 0.047 0.123 0.877 0.684 0.291 0.184 13 0.026 0.043 - - - -14 0.007 0.068 - - - -15 0.026 0.116 - - - -Cluster+ 0.020 0.076 - - - -Stand Avg 0.083 0.138 0.659 0.614 0.393 0.230 + Average for position of trees in the stand. - 177 -Table 15- Average No.of insects per fasc i cle by tree and branch type - 1975. Tree Larva - 1975 Pupa - 1975 Egg - 1975* No. Ma i n Branch S i de Branch Ma i n Branch S ide Branch Ma i n Branch S i de Branch 1 0.367 0.257 0.092 0.112 0.970 0.658 2 0.093 0.120 0.053 0.097 0.538 0.403 ; 3 0.097 0.149 0.230 0.077* 0.707 0.522 4 0.159 0.202 0.171 0.118 - -1nter ior+ 0.179 0.182 0.136 0.101 0.738 0.527 5 0.318 0.246 0.208 0.108 0.784 0.781 "6 0.244 0.154 0.023 0.126 0.768 0.615 7 0.248 0.130 0.065 0.154 0.997 0.737 8 0.445 0.582 0.052 0.164 - -Edge+ 0.314 0.278 0.087 0.138 0.850 0.711 9 0.191 0.125 0.171 0.085 1.165 1 .042 10 0.134 0.240 0.062 0.205 0.870 0.429 11 0.097 0.212 0.073 0.118 - -12 0.043 0.102 0.003 0.032 - -0pen+ 0.116 0.170 0.077 0.110 1.017 0.735 13 0.053 0.072 0.022 0.054 - -14 0.105 0.058 0.024 0.035 - -15 0.058 0.088 0.029 0.028 - -C1 uster+ 0.072 0.073 0.025 0.039 - -Stand Avg 0.177 0.182 0.085 0.101 0.850 0.648 * New tree + Average for position of trees in stand - 178 -Table 16. Average No. of and horizontal pupae per fascicle by crown position, 1974. tree, branch type Tree Main Branch Side Branch No. 1 nner Mid Outer 1 nner Mid Outer 1 0.000 0.039 0.096 0.008 0.031 0.075 2 0.000 0.000 0.146 0.082 0.097 0.319 3 0.042 0.116 0.281 0.167 0.031 0.222 4 0.164 0.199 0.217 0.093 0.133 0.127 1nter ior+ 0.051 0.088 0.185 0.087 0.073 0.186 5 0.133 0.264 0.382 0.271 0.387 0.635 6 0.111 0.166 0.157 0.258 0.136 0.210 7 0.054 0.078 0.144 0.039 0.192 0.385 8 0.049 0.056 0.086 0.026 0.070 0.079 Edge+ 0.087 0.141 0.192 0.148 0.196 0.327 9 0.102 0.036 0.050 0.098 0.142 0.085 10 0.042 0.025 0.106 0.133 0.065 0.223 1 1 0.033 0.031 0.046 0.100 0.135 0.127 12 0.045 0.024 0.031 0.139 0.101 0.131 0pen+ 0.055 0.029 0.059 0.117 0.111 0.141 13 0.048 0.000 0.030 0.063 0.035 0.029 14 0.000 0.000 0.021 0.059 0.089 0.055 15 0.043 0.010 0.025 0.125 0.069 0.154 C1uster+ 0.030 0.003 0.025 0.082 0.064 0.079 Stand Avg. 0.058 0.070 0.121 0.111 0.114 0.190 + Average for position of trees in stand - 179 -Table 17- Average No. of eggs per fascicle by tree, branch type and horizontal crown position, 197*+ -Tree Main Branch Side Branch No. Inner Mid Outer Inner Mid Outer 1 — * v - - - - -2 0.243 0.375 0.484 0.622 0.440 0.381 3 0.236 0.260 0.49** 0.591 0.5**6 0.693 4 0.502 0.890 0.521 0.446 0.817 0.280 1nter ior+ 0.327 0.508 0.500 0.553 0.601 0.451 5 0.499 0.798 0.605 0.521 0.627 0.3*»9 6 0.292 0.627 0.902 0.401 0.706 0.689 7 0.451 0.813 0.91 1 0.644 1.030 0.637 8 Edge+ 0. 414 0.746 0. .806 0. • 552 0. 788 0. 558 9 0. 630 0.798 0. • 556 0. 566 0. 435 0. 663 10 0. 686 1.065 1. .096 0. .558 1. 106 1. 020 11 0. 822 1.186 1. .054 0. • 598 0. 613 0. 599 0pen+ 0. 713 1.016 0. .902 0. .574 0. 718 0. 761 Stand Avg. 0. 485 0.757 0. .736 0. .550 0. 702 0. 590 + No samples taken. Average for position of trees in the stand. - 180 -Table 18. Average No. of prewinter larvae per fascicle by tree, branch type and horizontal crown position, 1974. Tree Main Branch Side Branch No. 1 nner Mid Outer 1 nner Mid Outer 1 0.749 0.254 0.608 0.274 0.235 0.365 2 0.407 0.371 0.475 0.184 0.280 0.310 3 0.345 0.120 0.232 0.151 0.129 0.243 4 0.169 0.271 0.258 0.169 0.193 0.339 1 nter ior+ 0.417 0.254 0.393 0.194 0.209 0.314 5 0.618 0.252 0.347 0.126 0.038 0.266 6 0.153 0.979 0.252 0.126 0.451 0.186 7 0.168 0.479 0.113 0.131 0.228 0.147 8 0.579 0.627 0.808 0.333 0.436 0.741 Edge+ 0.379 0.584 O.38O 0.179 0.288 0.335 9 0.912 0.451 0.506 0.189 0.088 0.250 10 0.197 0.210 0.314 0.131 0.260 0.393 11 0.038 0.241 0.150 0.122 0.169 0.162 12 0.071 0.131 0.378 0.129 0.057 0.266 0pen+ 0.304 0.258 0.337 0.143 0.143 0.205 Stand Avg. 0.367 0.449 0.362 0.172 0.214 0.306 + Average for position of trees in the stand. - 181 -T a b l e 19- Average branch No. o f p o s t w i n t e r t y p e and h o r i z o n t a 1 l a r v a e per f a s c i c l e by crown p o s i t i o n , A p r i l t r e e , 1975. T r e e Main Branch S i d e Branch No. 1 nner Mid Outer 1 nner Mid Outer 1 0.465 0.219 0.417 0.289 0.265 0.216 2 0.049 0.065 0.164 0.057 0.159 0.143 3 0.024 0.157 0.114 0.068 0.118 0.250 4 0.021 0.204 0.252 0.231 0.121 0.253 1 n t e r i o r + 0.140 0.161 0.237 0.161 0.166 0.215 5 0.055 0.343 0.556 0.282 0.076 0.381 6 0.149 0.293 0.292 0.151 0.169 0.141 7 0.000 0.133 0.610 0.070 0.169 0.152 8 0.209 0.523 0.604 0.368 0.692 0.685 Edge+ 0.103 0.323 0.515 0.218 0.276 0.340 9 0.079 0.102 0.393 0.076 0.245 0.052 10 0.103 0.078 0.221 0.065 0.243 0.411 11 0.000 0.045 0.227 0.153 0.211 0.271 12 0.020 0.054 0.055 0.041 0.076 0.189 0pen+ 0.050 0.070 0.224 0.084 0.194 0.231 13 0.071 0.056 0.034 0.076 0.096 0.044 14 0.137 0.104 0.074 0.026 0.067 0.081 15 0.056 0.040 0.078 0.070 0.071 0.124 C1 u s t e r + 0.088 0.067 0.062 0.057 0.078 0.083 Stand Avg. 0.096 0.161 0.273 0.135 0.185 0.227 + Average f o r p o s i t i o n o f t r e e s i n the s t a n d . - 182 -Table 20. Average No. of pupae per fascicle by tree, branch type and horizontal crown position, June 1975• Tree Main Branch Side Branch No. 1 nner Mid Outer 1 nner Mid Outer 1 0.085 0.065 0.124 0.126 0.050 0.160 2 0.071 0.074 0.013 0.051 0.230 0.216 3 0.232 0.139 0.318 0.030 0.085 0.115 4 0.165 0.250 0.099 0.040 0.115 0.187 1 nter ior+ 0.138 0.132 0.138 0.062 0.120 0.169 5 0.000 0.458 0.167 0.073 0.086 0.164 6 0.000 0.012 0.058 0.098 0.059 0.221 7 0.052 0.100 0.042 0.000 0.273 0.189 8 0.032 0.040 O.O83 0.084 0.102 0.305 Edge+ 0.021 0.152 0.087 0.063 0.130 0.220 9 0.365 0.035 0.111 0.031 0.030 0.192 10 0.000 0.037 0.149 0.057 0.129 0.430 11 .0.000 0.01 1 0.208 0.026 0.286 0.040 12 0.000 0.000 0.009 0.000 0.026 0.071 0pen+ 0.091 0.021 0.119 0.028 0.117 0-183 13 0.000 0.000 0.066 0.007 0.093 0.061 14 0.000 0.000 0.072 0.023 0.008 0.072 15 0.033 0.000 0.055 0.022 0.005 0.055 C1 uster+ 0.01 1 0.000 0.064 0.017 0.035 0.063 Stand Avg. 0.069 0.082 0.105 0.045 0.091 0.165 + Average for position of trees in the stand. - 183 -Table 21 . Average branch No. of eggs per fascicle type and horizontal crown by tree, pos it ion, 1975. Tree Main Branch Side Branch No. 1 nner Mid Outer 1 nner Mid Outer 1 0.803 1.108 0.999 0.729 0.575 0.669 2 0.460 0.758 0.397 0.484 0.264 0.460 4 0.863 0.551 0.706 0.544 0.599 0.424 1nter ior+ 0.709 0.806 0.701 0.586 0.478 0.518 5 0.553 0.635 1 .185 0.833 0.529 0.982 6 0.553 0.946 0.807 0.550 0.477 0.819 8 1.272 0.848 0.876 0.906 0.721 0.584 Edge+ 0.793 0.808 0.956 0.763 0.576 0.795 9 1.312 1.266 O.9I7 1.181 1.013 0.930 12 1.095 1.380 0.134* 0.536 0.472 0.277 Open+ 1.203 1.323 0.525 O.858 0.742 0.603 Stand Avg. 0.861 0.936 0.753 0.720 0.582 0.643 * Low counts as this contained current year shoots. + Average for position of trees in the stand. - 184 -Table 22. Average No. of insects per fascicle by crown level and exposure. Exposed Shaded Crown Lower Mid Upper Lower Mid Upper Stage: Pupa (1974) 0.095 0.112 0.096 0.117 0.152 0.092 Egg 0.537 0.705 0.806 0.473 0.619 0.678 Larva^ 0.301 0.429 0.277 0.236 0.307 0.319 Larva^O'975) 0.222 0.206 0.140 0.165 0.136 0.207 Pupa 0.108 0.115 0.119 0.051 0.091 0.071 Egg 0.518 0.738 0.862 0.610 0.846 0.921 Table 23. Average No.; of insects per fascicle by crown level and branch type. Main Branch Side Branch Crown Lower Mid Upper Lower Mid Upper Stage: Pupa (1974) 0. 078 0. 097 0. 074 0. 134 0. ,167 0. 114 Egg 0. 480 0. 672 0. 826 0. 531 0. ,652 0. 659 Larva^ La 1^(1975) 0. 323 0. 538 0. 316 0. 215 0. ,198 0. 279 0. 218 0. 177 0. 134 0. 169 0. ,165 0. 213 Pupa 0. 091 0. 113 0. 047 0. 068 0. .093 0. 143 Egg 0. 630 0. 842 1. 077 0. 498 0. .742 0. 705 Table 24. Average^No. of insects per fascicle by branch type and exposure. Main Branch Side Branch Stage Exposed Shaded Exposed Shaded Pupa (1974) 0.083 O.O83 0.119 0.158 Egg 0.709 0.609 0.657 0.571 Larva 3 0.418 O.367 0.253 0.208 Larv a i t(1975) 0.193 0.186 0.160 0.179 Pupa 0.108 0.062 0.120 0.080 Egg 0.767 0.933 0.645 0.652 - 185 -Table 25. Average number of fascicles per 15-cm branch section by tree and collection period. Tree Average No. of Fascicles/15~cm branch section at No. P (1974) E ' (197^) L 3 P (1975) Avg. 1 10.4 9.9 9.8 12.3 10.6 2 11.3 12.7 10.5 10.1 12.1 11.3 3 10.7 12.5 10.2 9.4 11.2** 10.8 4 11.6 11.1 10.2 10.5 11.7 11.0 5 11.1 9.4 8.1 7.6 10.0 9-2 6 8.7 9-9 7.9 10.6 12.9 10.0 7 13-9 12.8 13.2 7-5 11.7 11.8 8 14.2 - 15.2 12.6 14.4 14.1 9 12.6 13.7 9.9 9.5 12.0 11.4 10 12.1 12.5 11.2 12.3 13.8 12.4 11 15-9 17.7 16.1 16.0 16.3 16.4 12 15.4 - 15-9 16.2 17.8 16.3 15 17.1 - - 13.2 16.7 15.7 16 12.5 - - 10.4 16.1 13.0 17 16.0 - - 13.8 19.5 16.4 Avg. 12.9 12.5 11-5 11.3 12.4 12.1 * No samples taken ** Different tree - 186 -Appendix 5. Fig. I. Frequency distribution of fascicles per 15-cm branch section. Fig. II. Relationship between variance (S ) and mean (x) of fascicle per 15-cm branch section by tree. - 186a -2 i* 6 8 1 0 1 2 1 4 1 6 1 8 2 0 2 2 2 k 2 6 2 8 Frequency d i s t r i b u t i o n o f f a s c i c l e s per 15"cm b r a n c h s e c t i o n . _.. ,1 : • 1 ' — 9 1 1 1 3 1 5 1 7 X R e l a t i o n s h i p between v a r i a n c e (S ) and mean (x) o f f a s c i c l e p er 15-cm branch s e c t i o n by t r e e . 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0094303/manifest

Comment

Related Items