Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Analysis of some factors associated with distribution and intensity of attack by cone and seed insects… Kozak, Antal 1963

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

Item Metadata

Download

Media
831-UBC_1963_A1 K6 A5.pdf [ 10.63MB ]
Metadata
JSON: 831-1.0105603.json
JSON-LD: 831-1.0105603-ld.json
RDF/XML (Pretty): 831-1.0105603-rdf.xml
RDF/JSON: 831-1.0105603-rdf.json
Turtle: 831-1.0105603-turtle.txt
N-Triples: 831-1.0105603-rdf-ntriples.txt
Original Record: 831-1.0105603-source.json
Full Text
831-1.0105603-fulltext.txt
Citation
831-1.0105603.ris

Full Text

ANALYSIS OF SOME FACTORS ASSOCIATED WITH DISTRIBUTION AND INTENSITY OF ATTACK BY CONE AND SEED INSECTS IN DOUGLAS FIR by ANTAL KOZAK B.S.F., The Uni v e r s i t y of B r i t i s h Columbia., 1959 Faculty of Forestry (Sopron) M.F.j The Uni v e r s i t y of B r i t i s h Columbia, 1961 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF PH.D. i n the Faculty of Graduate Studies We accept t h i s thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA October, 1963 I n p r e s e n t i n g t h i s t h e s i s i n p a r t i a l f u l f i l m e n t of the requirements f o r an advanced degree at the U n i v e r s i t y of B r i t i s h Columbia, I agree that the L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r reference and study. I f u r t h e r agree that per-m i s s i o n f o r extensive copying of t h i s t h e s i s f o r s c h o l a r l y purposes may be granted by the Head of my Department or by h i s representatives,. I t i s understood that copying, or p u b l i -c a t i o n of t h i s t h e s i s f o r f i n a n c i a l gain s h a l l not be allowed without my w r i t t e n p e r m i s s i o n . • Department of Graduate Studies The U n i v e r s i t y of B r i t i s h Columbia,, Vancouver 8 , Canada. Date September 26, 1963 PUBLICATIONS Kozak, A., S z i k l a i , 0., G r i f f i t h , B.G., and Smith, J.H.G., 1963. Va r i a t i o n i n Cone and Seed Y i e l d from Young, Open-grown Douglas F i r s on the U.B.C. Research Forest. Univ. of B.C., Fac. of For. Res. Paper 57, 8p. Kozak, A. and Munro, D.D., 1963. An I,B.M. 1620 Computer Programme to F i t Frequency D i s t r i b u t i o n s . For. Chron. 39(3): 377-378. The U n i v e r s i t y FACULTY OF of B r i t i s h Columbia GRADUATE STUDIES PROGRAMME OF THE FINAL ORAL EXAMINATION FOR THE DEGREE OF DOCTOR OF PHILOSOPHY of ANTAL KOZAK B.S.F., The Un i v e r s i t y of B r i t i s h Columbia (Sopron D i v i s i o n ) 1959 M.F., The Uni v e r s i t y of B r i t i s h Columbia, 1961 MONDAY, SEPTEMBER 23, 1963, at 2:30 P.M. IN ROOM 235, FORESTRY AND.GEOLOGY BUILDING COMMITTEE IN CHARGE Chairman: F,H. Sdward K. Graham L. Schwartz B.G. G r i f f i t h J.H.G. Smith P.G. Haddock .0. S z i k l a i External Examiner: R.W. Stark U n i v e r s i t y of C a l i f o r n i a Berkeley ANALYSIS OF SOME FACTORS ASSOCIATED WITH DISTRIBUTION AND INTENSITY OF ATTACK BY CONE AND SEED INSECTS IN DOUGLAS FIR ABSTRACT An an a l y s i s was made of the d i s t r i b u t i o n of and i n t e r a c t i o n s between three major species of insects i n cones of Douglas f i r . This was based on over 3,500 cones from 93 trees i n 1961 and 4,000 cones from 97 trees i n 1962. High v a r i a t i o n i n damage was found among trees for each of the three insect species studied (Contarinia oregonensis Foote, Megastigmus spermotrophus Wacht1., and D i o r y c t r i a  a b i e t e l l a D0 & S.). In C_„ oregonensis t h i s v a r i a -t i o n was s i g n i f i c a n t l y related to the height of the trees and dates when cones became pendent. The per-centage of f i l l e d seeds and average cone siz e of the trees were Important in M. spermotrophus. The average cone s i z e of the trees and duration of vegetative bud f l u s h i n g were s i g n i f i c a n t l y asso-ci a t e d with D„ a b i e t e l l a . The w i t h i n t r e e - v a r i a t i o n of damage by C_. oregonensis was h i g h l y s i g n i f i c a n t and consistent from tree to tree and year to year. Damage i n -creased, s i g n i f i c a n t l y from the bottom to the top of l i v e crown. Suggestions are made to help increase the seed y i e l d by chemical co n t r o l of insects by s e l e c t i o n of uninfested or s l i g h t l y i nfested trees by sequen-t i a l sampling, and by s e l e c t i o n of cone and seed insect r e s i s t a n t trees for seed orchards. Although no tree was wholly r e s i s t a n t to a l l three ins e c t s , the best, with only 4.8 percent of i t s seeds damaged, was so much better than average that i t may be of considerable economic importance. The great range i n apparent res i s t a n c e of trees to attack by cone and seed insects should be recognized i n further studies and attempts be made to determine b i o l o g i ~ c a l l y sound reasons f o r the observed d i f f e r e n c e s . GRADUATE STUDIES F i e l d of Study? Forestry S t a t i s t i c a l Methods i n Forestry Research J.H.G. Smith Forest Tree Seed 0„ S z i k l a i Problems i n Forest Mensuration J.H.5. Smith Directed Studies i n Insect Ecology K. Graham Related Studies? Mathematical S t a t i s t i c s Computer Programming Miss L. Schwartz Miss C. Froese ABSTRACT ANALYSIS OF SOME FACTORS ASSOCIATED WITH DISTRIBUTION AND INTENSITY OF ATTACK BY CONE AND SEED INSECTS IN DOUGLAS FIR An analysis was made of the d i s t r i b u t i o n of and interactions between three major species of insects i n cones of Douglas f i r . This was based on over 3,500 cones from 93 trees i n 1961 and 4,000 cones from 97 trees i n 1962. High v a r i a t i o n i n damage was found among trees for each of the three insect species studied (Contarinia oregonensis Foote, Megastigmus spermotrophus Wachtl., and D i o r y c t r i a a b i e t e l l a D.&S.) In _C. oregonensis t h i s v a r i a t i o n was s i g n i f i c a n t l y r e l a t e d to the height of the trees and dates when cones became pendent. The percentage of f i l l e d seeds and average cone size of the trees were important i n M. spermotrophus. The average cone size of the trees and duration of vegetative bud flu s h i n g were s i g n i f i c a n t l y associated with _D. a b i e t e l l a . The within t r e e - v a r i a t i o n of damage by _C. oregonensis was highly s i g n i f i c a n t and consistant from tree to tree and year to year. Damage increased s i g n i f i c a n t l y from the bottom to the top of l i v e crown. Suggestions are made to help increase the seed y i e l d by chemical con t r o l of insects by s e l e c t i o n of uninfested or s l i g h t l y infested trees by sequential sampling, and by s e l e c t i o n of cone and seed insect r e s i s t -ant trees f o r seed orchards. Although no tree was wholly r e s i s t a n t to a l l three insects the best, with only 4.8 per cent of i t s seeds damaged, was so much better than average that i t may be of considerable economic importance. The great range i n apparent resistance of trees to attack i i i by cone and seed insects should be recognized i n further studies and attempts be made to determine b i o l o g i c a l l y sound reasons for the observed d i f f e r e n c e s . ACKNOWLEDGEMENTS Acknowledgement i s made to The University of B r i t i s h Columbia for prov i s i o n of laboratory and computing f a c i l i t i e s , and to the National Research Council for f i n a n c i a l help which aided m a t e r i a l l y i n t h i s research. The writer i s greatly indebted to Drs. Kenneth Graham and J. Harry G. Smith for t h e i r advice and encouragement throughout the work. Special thanks are due to Dr. B. G. G r i f f i t h for h i s permission to use the 154 trees of his project, established i n 1957, to study the phenological and growth c h a r a c t e r i s t i c s of Douglas f i r and also for making h i s data a v a i l a b l e to the author for analysis. The author also wishes to acknowledge the help and advice of members of The University of B r i t i s h Columbia Computing Center and of Dr. C. Froese and Mr. H. Dempster i n p a r t i c u l a r . Thanks are extended to the following persons who have contributed i n various ways to completion of t h i s t h e s i s : Dr. P. G. Haddock, Dr. J. W. Wilson and Mr. 0. S z i k l a i , Faculty of Forestry, Dr. L. Schwartz, Department of Mathematics, Mr. J. Walters, Research Forester, Mr.*' A.F. Hedlin, Forest Biology D i v i s i o n , Department of Forestry, Mr. L. Safranyik and Mr. G. Lesko, graduate students, Miss E. Taylor, stenographer, and Mrs. Mona Lambden, technician. i v CONTENTS Page ACKNOWLEDGEMENTS VII TABLES VIII FIGURES XII LIST OF APPENDICES XIV INTRODUCTION 1 MATERIALS AND METHODS 3 1. Cone and Seed Production C h a r a c t e r i s t i c s of the Sampled Trees 3 2. Sampling of Cone and Seed Insects of Douglas f i r . . . . 9 3. Description and Biology of Cone and Seed Insects Studied 10 D i o r y c t r i a a b i e t e l l a , D & S 11 Megastigmus spermotrophus, Wachtl 1'2 Contarinia oregonensis, Foote 14 Contarinia washingtonensis , Johnson 17 4. Description of Sample Trees .'17 5. B i o l o g i c a l Studies of Cone and Seed Insects of Douglas F i r - - 19 Time of emergence 20 Cone bagging experiment 21 Trapping experiment 22 6. S t a t i s t i c a l Techniques Used 22 Analysis of variance 22 Page Duncan's new multiple range test 23 Frequency d i s t r i b u t i o n s and goodness of f i t 23 Regression and c o r r e l a t i o n analysis 25 Sequential sampling 27 F i t t i n g a common k to a series of negative binomial d i s t r i b u t i o n s 29 N-stage sampling, sample size and optimum a l l o c a t i o n . 31 Use of IBM 1620 e l e c t r o n i c computer 33 EXPERIMENTAL RESULTS 34 1. Analysis of Amount of Cones Damaged 34 Trees with three crown levels sampled 35 Trees with two crown levels sampled 41 Analyses of year to year differences 46 Analysis of damage by d i r e c t i o n i n the crown 51 Analysis of between tree v a r i a t i o n 55 Analysis of D i o r y c t r i a a b i e t e l l a damage 60 Analysis of Megastigmus spermotrophus damage . . 65 Analysis of Contarinia oregonensis damage. . . . 66 Analysis of undamaged cones 68 Analysis of Contarinia washingtonensis damage. . 69 Relationships among the dependent v a r i a b l e s . . . 69 2. Analysis of Amount of Seeds Damaged 71 Trees with three crown levels sampled 72 Trees with two crown levels sampled 79 Analyses of year to year differences 82 Analysis of damage by d i r e c t i o n i n the crown 89 Analysis of between tree v a r i a t i o n 89 v i Page Analysis of D i o r y c t r i a a b i e t e l l a damage 95 Analysis of Megastigmus spermotrophus damage . . . 100 Analysis of percentage of f i l l e d seeds damaged by Megastigmus 102 Analysis of Contarinia oregonensis damage . . . . 103 Analysis of percentage of undamaged seeds 105 Relationships among the dependent v a r i a b l e s . . . . 106 Transformations 108 3. Interactions of Cone and Seed Insects of Douglas f i r . . . 108 In t e r r e l a t i o n s between D i o r y c t r i a and Contarinia. . . . 110 I n t e r r e l a t i o n between D i o r y c t r i a and Megastigmus. . . . 116 I n t e r r e l a t i o n between Megastigmus and Contarinia. . . . 117 4. Parasites of Cone and Seed Insects 119 5. Time of Emergence 120 6. Time of Attack 122 7. Trapping Experiment 122 8. Frequency D i s t r i b u t i o n s and Sequential Sampling 124 9. Optimum a l l o c a t i o n of Sample Sizes 136 DISCUSSION 147 1. Preventing of F e r t i l i z a t i o n 152 2. Reducing of Quality of Extracted Seeds 153 3. Improving Seed Production 155 4. Need for Further Studies 161 CONCLUSIONS 1 6 3 APPENDICES 1 6 5 LITERATURE CITED 1 7 4 LITERATURE CONSULTED 178 v i i i TABLES Number , Heading page 1 Flower and cone production per tree 4 2 Cone production by various classes of Douglas f i r 5 3 Influence of crown p o s i t i o n on the cone production of trees . . 6 4 P r e c i s i o n of tree c h a r a c t e r i s t i c s measurements 18 5 Analyses of variance of 1961 cones (trees with three crown le v e l s sampled) 36 6 Percentages of damaged and undamaged cones i n 1961 (means from analyses of variance of Table 5) 37 7 Analyses of variance of 1962 cones (trees with three crown le v e l s sampled) 38 8 Percentages of damaged and undamaged cones i n 1962 (means from analyses of variance of Table 7) 39 9 Analyses of variance of 1961 cones (trees with two crown leve l s sampled) 42 10 Percentages of damaged and undamaged cones i n 1961 (summary of analyses of variance of Table 9) 43 11 Analyses of variance of 1962 cones (trees with two crown le v e l s sampled) 44 12 Percentages of damaged and undamaged cones i n 1962 (means from analyses of variance of Table 11) 45 13 Analyses of variance of damaged and undamaged cones by tree, year, crown p o s i t i o n and l e v e l 47 14 Percentages of damaged and undamaged cones by year, crown l e v e l , crown l e v e l and p o s i t i o n and tree (summary of analyses of variance of Table 13) 48 15 Changes i n i n f e s t a t i o n class from year to year on the same i n d i v i d u a l tree 50 16 Analyses of variance of 1961 cones by tree, d i r e c t i o n , crown p o s i t i o n and l e v e l 52 17 Analyses of variance of 1962 cones by tree, d i r e c t i o n , crown p o s i t i o n and l e v e l 53 i x Number Heading Page 18 Percentages of damaged and undamaged cones by d i r e c t i o n . . 54 19 Analyses of variance of 1962 cones by eight d i r e c t i o n s , crown l e v e l , and p o s i t i o n 56 20 Percentages of damaged and undamaged cones by eight d i r e c t i o n s on the cone 57 21 Independent tree v a r i a b l e s used i n multiple c o r r e l a t i o n and regression analysis of 1961 data (93 trees) 58 22 Independent tree v a r i a b l e s used i n multiple c o r r e l a t i o n and regression analysis of 1962 data (97 trees) 59 23 Description of dependent variables 61 24 Co r r e l a t i o n and regression c o e f f i c i e n t s of the analysis of 1961 cones 62 25 Co r r e l a t i o n and regression c o e f f i c i e n t s of the analysis of 1962 cones 63 26 C o e f f i c i e n t s of determination for 1961 and 1962 64 27 Relationships among the dependent variables i n 1961. . . . 70 28 Relationships among the dependent variables i n 1962. . . . 70 29 Analyses of variance of damaged and undamaged seeds i n 1961 (trees with three crown le v e l s sampled) 73 30 Percentages of damaged and undamaged seeds i n 1961 (trees with three crown l e v e l s sampled) 74 31 Analyses of variance of damaged and undamaged seeds i n 1962 (trees with three crown l e v e l s sampled) 76 32 Percentages of damaged and undamaged seeds i n 1962 (summary of analyses of variance of Table 31) 77 33 Analyses of variance of damaged and undamaged seeds i n 1961 (trees with two crown le v e l s sampled) 80 34 Percentages of damaged and undamaged seeds i n 1961 (summary of analyses of variance of Table 33) 81 35 Analyses of variance of damaged and undamaged seeds i n 1962 (trees with two crown l e v e l s sampled) 83 X Number Heading Page 36 Percentages of damaged and undamaged seeds i n 1962 (summary of analyses of variance of Table 35) 84 37 Analyses of variance of damaged and undamaged seeds by tree, year, crown p o s i t i o n and l e v e l (trees with two crown leve l s sampled) 86 38 Percentages of damaged and undamaged seeds by year, crown l e v e l and p o s i t i o n and tree (summary of analyses of variance of Table 37) 87 39 Changes of i n f e s t a t i o n class from year to year on the same i n d i v i d u a l tree 88 40 Analyses of variance of 1962 seeds by tree, d i r e c t i o n , crown p o s i t i o n and l e v e l 90 41 Analyses of variance of 1961 seeds by tree, d i r e c t i o n , crown p o s i t i o n and l e v e l 91 42 Percentages of damaged and undamaged seeds by d i r e c t i o n . . 92 43 Analyses of variance of 1962 seeds by eight d i r e c t i o n s , crown l e v e l and p o s i t i o n 93 44 Percentages of damaged and undamaged seeds by eight d i r e c t i o n s on the crown 94 45 Description of dependent variables 96 46 Co r r e l a t i o n and regression c o e f f i c i e n t s of the analysis of damaged and undamaged seeds i n 1961 97 47 Cor r e l a t i o n and regression c o e f f i c i e n t s of the analysis of damaged and undamaged seeds i n 1962 98 48 C o e f f i c i e n t s of determination for 1961 and 1962 99 49 Relationship between the dependent variables i n 1961 . . . 107 50 Relationship between the dependent variables i n 1962 . . . 107 51 Summary of the best and worst ten trees i n percentage of cones damaged 109 52 Frequency d i s t r i b u t i o n s of number of seeds per cone damaged by Contarinia oregonensis, Megastigmus spermotro- phus and t o t a l numbers of damaged seeds and undamaged f i l l e d seeds 125 x i Number Heading page 53 Limits f o r sequential sampling of cones 1266 54 Limits f o r sequential sampling plans, sampling trees within stand 135 55 Components of variance of the damage of Contarinia oregon-ensis and Megastigmus spermotrophus and on the undamaged f i l l e d seeds, expressed as a r c s i n square root 145 56 Optimum a l l o c a t i o n of sample sizes i n a four stage sampling 146 57 Damages, f i l l e d seed and cone production on the best f i v e trees (averages of 1961 and 1962 observations) x i i FIGURES lumber Heading To follow pagi 1 D i s t r i b u t i o n of cone and seed insect damage i n Douglas f i r cones 116 2 D i s t r i b u t i o n of cone and seed insect damage i n Douglas f i r cones (middle crown) 116 3 D i s t r i b u t i o n of cone and seed insect damage i n Douglas f i r cones (upper crown) 116 4 D i s t r i b u t i o n of cone and seed insect damage i n Douglas f i r cones (outside crown) 116 5 D i s t r i b u t i o n of cone and seed insect damage i n Douglas f i r cones (inside crown) 116 6. Time of emergence of. Contarinia oregonensis and Megastig-mus spermotrophus i n 1962 122 7 Time of attack of Contarinia oregonensis and Megastigmus  spermotrophus 124 8 Sequential graph f or sampling cones damaged by Contarinia  oregonensis 135 9 O.C. and A.S.N, curves f or sampling cones damaged by Contarinia oregonensis 135 10 Sequential graph f or sampling cones damaged by Megastigmus  spermotrophus 135 11 O.C. and A.S.N, curves f or sampling cones damaged by . Megastigmus spermotrophus 135 12 Sequential graph f or sampling t o t a l number of damaged cones 135 133 O.C. and A.S.N, curves for sampling t o t a l number o£ damaged cones 135 14 Sequential graph for sampling undamaged f i l l e d seeds . . 135 15 O.C. and A.S.N, curves for sampling undamaged f i l l e d seeds 135 16 Sequential graph f o r sampling trees damaged by Megastig-mus spermotrophus 145 17 O.C. and A.S.N, curves f or sampling trees damaged by Megastigmus spermotrophus 145 x i i i 18 Sequential graph for sampling trees damaged by D i o r y c t r i a  a b i e t e l l a 145 19 O.C. and A.S.N, curves for sampling trees damaged by D i o r y c t r i a a b i e t e l l a 145 20 Sequential graph f o r sampling trees damaged by Contarinia  oregonensis 145 21 O.C. and A.S.N, curves f o r sampling trees damaged by Contarinia oregonensis 145 22 Sequential graph f o r sampling trees for undamaged cones. 145 23 O.C. and A.S.N, curves for sampling undamaged cones. . . . 145 x i v LIST OF APPENDICES L i s t Heading Page A Map showing lo c a t i o n of 154 Douglas f i r trees studied. . 165 B Form on which the information was c o l l e c t e d 166 C Five colour classes of female flowers on Douglas f i r . . 167 D Counting of Megastigmus sperotrophus from one of the rearin g boxes. 168 E Rearing box and s o i l ^ temperature recorder 169 F Cones covered with cellophane bags 170 G Trapping frames on tree No. 63 171 H The output forms of the program f i t t i n g frequency d i s t r i b u t i o n s 172 I The percentage of seeds damaged by Contarinia oregonensis on four sides of the crown of f i v e trees 173 INTRODUCTION The increasing attention of recent years that i s being given to insects a f f e c t i n g seeds and cones of forest trees has not only shown the importance of these insects, but also has prepared the way for the d e f i n i t i o n of, and assault on new problems. There now e x i s t s a considerable body of l i t e r a t u r e on the biolo g i e s of various species of seed and cone insects a f f e c t i n g a wide range of tree species (Keen, 1958). There s t i l l remains a lack of knowledge of the patterns of d i s t r i b u t i o n within and between cones, and within and between trees. A gap also remains i n our knowledge of the factors a f f e c t i n g the d i s t r i b u t i o n and abundance of the various insect species within and between cones and trees. A better understanding of cone insect d i s t r i b u t i o n s and of factors of abundance would be of considerable value i n seed production p r a c t i c e . It would a i d i n the a l l o c a t i o n of seed c o l l e c t i o n e f f o r t according to the estimated damage, i t would aid i n forming decisions on control, and i t may provide indications f o r the s e l e c t i o n of genetic s t r a i n s that are less susceptible to attack and damage. Douglas f i r i s one of the important forest species on which a considerable b i o l o g i c a l l i t e r a t u r e on seed and cone insects has been published. It i s also one for which more e c o l o g i c a l information on these insects i s needed, while i t also o f f e r s some p o s s i b i l i t i e s for in v e s t i g a t i n g the ro l e of i n d i v i d u a l tree c h a r a c t e r i s t i c s i n s u s c e p t i b i l i t y . Even within a given age class on the same s i t e , Douglas f i r i s very v a r i a b l e i n the dates of beginning of seasonal growth. This i s obviously important f o r any insects whose es t a b l i s h -2 merit depends on a close synchrony with a s p e c i f i c stage of seasonal growth. Douglas f i r i s also v a r i a b l e i n cone colour, another factor which conceivably could a f f e c t the behaviour and egg d i s t r i b u t i o n of the i n s e c t s . In addition to the problems of d i s t r i b u t i o n and abund-ance of insect attacks, Douglas f i r cones o f f e r some opportunities to study i n t e r - and i n t r a - s p e c i f i c competition among in s e c t s . Douglas f i r cones represent a microhabitat, complete with food supply, i n which several species of insects co-exist. Their food requirements and habits of feeding d i f f e r considerably, but because of p r i o r occupancy by one species, or indiscriminate feeding by another, an antagonistic r e l a t i o n s h i p becomes possible. It i s the purpose of t h i s study to investigate, f o r one experi-mental area, some of the features of and factors i n the d i s t r i b u t i o n and abundance of insects a f f e c t i n g seeds and cones of Douglas f i r . 3 MATERIALS AND METHODS The study was i n i t i a t e d i n the University Research Forest at Haney i n the spring of 1961. One hundred and f i f t y - f o u r open-grown Douglas f i r trees, about 23 years o l d i n 1961, were studied. The Douglas f i r became established on the area a f t e r a w i l d - f i r e i n o l d logging slash i n 1930. Appendix A shows the d i s t r i b u t i o n of sample trees on the area. Tree phenology, flowering, cone production and some growth c h a r a c t e r i s t i c s of the trees have been studied by G r i f f i t h (1963) since May, 1957. Meteorological reports are also a v a i l a b l e from the area for several years i n the Annual Reports of the U n i v e r s i t y of B r i t i s h Columbia Research Forest. Since the o r i g i n of 154 trees studied i s natural regeneration, the v a r i a t i o n of t h e i r c h a r a c t e r i s t i c s i s wide and well suited for the study. The number of trees (154) sampled for the study was not s t a t i s t i c a l l y defined. These trees and t h i s number were used because of the large amount of information already a v a i l a b l e for them. 1. Cone and Seed Production C h a r a c t e r i s t i c s of the Sampled Trees The abundance of food supply plays an important part i n the l i v e s of f orest i n s e c t s . Because of t h e i r host tree s p e c i f i c i t y , t h i s i s e s p e c i a l l y true for cone and seed insects. The importance of a v a i l a b l e food supply may be indicated by the f a c t that i n years when the cone crop i s heavy, a r e l a t i v e l y small percentage of the seeds and cones are destroyed and i n years when the crop i s small, insects take a large portion of i t . 4 The female buds of Douglas f i r can be recognized as early as November of the year p r i o r to the cone crop (Finnis, 1953). A l l e n (1941), F i n n i s (1953) and Johnson _et. _al. (1962) described methods by which the female cone buds can be recognized, and methods were des-^ cribed to determine the possible number of mature cones on a tree. Kozak, S z i k l a i , G r i f f i t h and Smith (1963) stated that there i s a highly s i g n i f i c a n t c o r r e l a t i o n between number of female flowers per tree counted e a r l y i n May and mature cone crop (r = 0.97). On the average 109 cones per tree were produced by 173 female flowers. It was found by them that the v a r i a t i o n of cones surviving varies from year to year. The average flower and cone production of the 154 trees for s i x years (from 1957 to 1962) i s given i n Table 1. Table 1. Flower and cone production per tree. In Average flowers number of cones 1957 * 134.0 1958 1.1 0.6 1959 259.0 134.0 1960 2.3 1.7 1961 277.0 121.0 1962 318.0 273.0 * not estimated Table 2 summarizes the cone production of various classes of trees. High v a r i a t i o n was observed i n cone production between trees (Kozak e_t. _al. , 1963). It was shown by a multiple c o r r e l a t i o n analysis 5 that d.b.h. (r = 0.50), tree height (r = 0.47), crown width (r = 0.34), number of branches i n f i r s t whorl above breast height (r = -0.21) and s i t e index (r = 0.18) a l l have s i g n i f i c a n t simple c o r r e l a t i o n s with number of cones produced by trees. Table 2„ Cone production by various classes of Douglas f i r . Average Number D.b.h. Age i n Class of number of of years inches years tree cones per with cones (1961) (1962) year* A l l 154 109.0 2.9 9.54 24.0 Worst 10 0.3 0.3 6.57 20,7 Best 10 459.0 4.1 13.40 27.2 Smallest 10 37.0 1.1 5.91 20.3 Biggest 10 320.7 4.3 15.30 28.7 Oldest 10 218.1 3.4 12.20 32.1 Best 1 650.0 5.0 15.30 28.0 Worst 1 0.0 0.0 8.40 25.0 * In some years the cone crop over 200 was assigned a r b i t r a r y numbers, e.g. Medium - 500 cones. The frequency of cone crops (or number of years with cones) was s i g n i f i c a n t l y correlated with tree d.b.h. (r = 0.41), tree height (r = 0^34), and crown width (r = 0.27). From these figures i t can be stated that both the number of years with cone crop and the size of the cone crop depend upon the size of trees. Kozak e_t. _al. (1963) reported Dr. Beaton's analyses showing that the f o l i a g e of the 10 worst cone producing trees contained 1.13 per cent nigrogen, which was a s i g n i f i c a n t l y smaller amount than 1.24 per cent contained by the 10 6 best cone producers. A number of trees did not hehave t y p i c a l l y . For example, 11 out of the 154 trees had a good flower crop i n 1962, but yielded a very poor cone crop. Since t h i s c h a r a c t e r i s t i c of trees changes from year to year, i t i s very l i k e l y that the l o c a l s o i l moisture and other c l i m a t i c factors a f f e c t e d them. Out of 31 trees with heavy flower crops only three produced heavy cone crops i n 1961. Nine medium and nine heavy crops were found out of 20 heavy flower crops.in 1962. As Table 3 indicates, most of the trees produce cones on the middle and upper t h i r d of the crown, and only a few of them produce cones on the lower t h i r d . Table 3. Influence of crown p o s i t i o n on the cone production of trees. 1961 1962 Crown l e v e l Percentage segments with of 154 cone crops Thirds Upper 79.9 83.8 Middle 70.1 79.2 Lower 22.1 46.1 It i s i n t e r e s t i n g to compare the abundance of cone crops i n d i f f e r e n t crown lev e l s of the two years. A much higher percentage of trees had cones i n the lower and middle thirds of t h e i r crowns i n 1962 than i n 1961. Sine the cone crop i n 1962 (318 cones per tree) was much better than i n 1961 (277 cones per tre e ) , Table 3 may indicate that i n better cone crop years many of the trees are loaded with cones from the bottom to the top of the l i v e crown and i n a medium cone year the trees w i l l produce cones mostly on the upper and middle thirds of the crown. Kozak _et. _al. (1963), published a table by which the number of cones per acre can be estimated as a function of d.b.h., crown width and tree height. This table was based on a cone production r a t i o , instead of actual number of cones, which was defined as the average number of cones per tree from 1957 to 1962 divided by H x CW (H = tree height, CW = crown width). The reason behind t h i s was that, since a l l trees were open-grown t h e i r l i v e crowns extended to the ground and were roughly c o n i c a l i n shape. Because of t h i s , the t o t a l crown volume varies as H x CW2. The average cone production r a t i o of the 154 trees was 0.005 with a standard deviation of 0.006. The r a t i o ranged from zero to 0.1 cones per unit of crown. A multiple c o r r e l a t i o n analysis showed that the cone production r a t i o increased s i g n i f i c a n t l y with tree height, and s i g n i f i c a n t l y decreased with number of branches at breast height. The average number of cones per tree and the number of years with cone crop per tree were also highly correlated with the r a t i o (Kozak e t . al., 1963). Since no c o r r e l a t i o n was found between crown volume and cone production r a t i o , the authors concluded that t h i s r a t i o i s determined primarily by i n d i v i d u a l tree d i f f e r e n c e s . The average number of extracted seeds per cone was 17.78 i n 1962 for 97 trees which agreed well with average of 19.3 for 183 trees of several coastal B.C. provenances studied by Robinson (1963). Garman (1951) reported 44 seeds per cone on the average, which i s much higher 8 than the figures above, but i n his' study the seeds were obtained by d i s s e c t i o n of cones. Robinson (1963) c a r r i e d out a multiple c o r r e l a t i o n analysis using t h i r t e e n v a r i a b l e s and 183 trees. He found that the number of seeds per cone increased s i g n i f i c a n t l y with the age and height of trees and with length and width of seeds. Kozak _et.. al. gave an equation by which the number of sound f i l l e d seeds can be calculated from the number of sound f i l l e d seeds on the l o n g i t u d i n a l cut surface of a cone: Y = 3.04 x -0.33. This equation based on 110 cones had a standard err o r of estimate of 0.94 seeds and simple c o r r e l a t i o n c o e f f i c i e n t ; r = 0.941, which i s highly s i g n i f i c a n t . The per cent of f i l l e d seeds per cone varies from tree to tree. The factors a f f e c t i n g t h i s v a r i a t i o n were tree height (r = -0.252), d.b.h. (r = -0.294), age (r = -0.316), date cones became pendent (r = -0.402) and crown width (r = -0.295). Since tree size and age are the factors which had most influence, and t h i s was unexpectedly negative, an explanation of t h i s w i l l be offered i n another chapter, a f t e r the discussion of cone and seed insect damage. It i s very l i k e l y that t h i s e f f e c t was not primary, but rather the r e s u l t of the bigger trees being much more exposed to cone and seed insect attack than the smaller trees. Kozak _et _al. (1963) stated that the quality of extracted seeds i s above the average qu a l i t y . An average of 36.7 per cent f i l l e d seeds per cone per tree was found for the extracted seeds of 44 trees. For these same trees 8.2 per cent f i l l e d seeds per cone per tree was found on the h o r i z o n t a l cut surface of cones. An equation i s av a i l a b l e to c a l c u l a t e the per cent of extracted f i l l e d seeds (Y) from the per cent of f i l l e d seeds found on the cut surface (X): Y = 15,000 + 4.280 x -0.099 x 2. xhe equation has a standard error of the estimate equal to 17.3 per cent and an'R of 0.648, which i s highly s i g n i f i c a n t . With c o n t r o l l e d p o l l i n a t i o n A l l e n and S z i k l a i (1962) increased the f i l l e d seeds per cone from an average of 1,1 to 21.4 with dry, and to 24.6 with wet p o l l i n a t i o n . 2. Sampling of Gone and Seed Insects of Douglas F i r For the present study, extensive sampling has been c a r r i e d out. More than 3,500 cones were sampled from 93 trees out of the 154 i n 1961. This sampling was repeated i n the summer of 1962 when approx-imately 4,000 cones were taken from 97 trees. Selection of sample trees was based on cone production. A l l of the trees with s u f f i c i e n t cone crops were sampled. The samples were drawn from three l e v e l s of the crown, the lower, middle and upper t h i r d s . Each l e v e l was separated into outside and inside halves of the l i v e crown. Six cones, two large, two medium and two small, were c o l l e c t e d randomly from each of the s i x sample locations on the south side of the crown. Only three crowns, one large, one medium and one small, were taken from those trees which had poor cone crop. Cones were c o l l e c t e d just a f t e r the seeds became milky ( l a s t week of August i n both years). The sample cones were taken from the clos e s t branch to the middle of each crown l e v e l , and from the close s t branch from the south side. From the middle and upper thirds of the crowns, the cones were obtained by climbing. The cones i n the lower crown l e v e l were usually within reach of the pruners. Five trees were s p e c i a l l y sampled i n each year. From these trees the sample cones were* taken from four sides of the crown, north, 10 west, and east, i n addition to south, from three crown l e v e l s and from outside and inside halves of the l i v e crown. To further evaluate side to side differences, sample cones were c o l l e c t e d from eight sides on one tree i n 1962, at 45 degree i n t e r v a l s of azimuth from north. Besides the s i x sample cones, one extra cone was taken from each sample place for studying the parasites of cone and seed in s e c t s . These cones were stored i n polyethylene bags, one cone per bag, and the number and species of emerged parasites were recorded the following spring. The other s i x cones c o l l e c t e d from each sample place were s l i c e d i n h a l f l o n g i t u d i n a l l y using the Winjum and Johnson (1960) modified cone cutter, and the number of damaged seeds was counted on the cut surface. The number of damaged seeds was observed with respect to three d i f f e r e n t species of cone and seed i n s e c t s : Contarinia oregonensis, Foote (Cecidomyiidae), Megastigmus spermotrophus (Wachtl., Torymidae) and D i o r y c t r i a a b i e t e l l a (D. and S. Phycitidae). In 1962 an a d d i t i o n a l species, Contarinia washingtonensis, Johnson, (Cecidomyiidae) was also studied. Appendix B shows the form used for data c o l l e c t i o n . It was planned to include the well-known Douglas f i r cone- moth Barbara colfaxiana (Kearf.) i n t h i s study, but the damage by t h i s insect was so rare that t h i s insect had to be ignored. 3. Description and Biology of Cone and Seed Insects Studied Keen (1958) l i s t e d 63 species of insects which have been found i n Douglas f i r cones. Seventeen are believed to be i n some measure damaging to the cones. The injuriousness of 13 of them i s uncertain; 33 are parasites and predators. He has also presented a useful key to help i d e n t i f y these insects. Johnson and Winjum (1960) set up a key for the s i x most important species of cone and seed insects, noting that many other species i n -habit Douglas f i r cones. Four of the cone and seed insects which were the basic material of the study now w i l l be described. D i o r y c t r i a a b i e t e l l a (D & S). This species has not been i n t e n s i v e l y studied on Douglas f i r . The work which has been done indicates the wide range of d i s t r i b u t i o n and habits of the ins e c t s . Munroe (1959) stated that D i o r y c t r i a  a b i e t e l l a may a c t u a l l y represent two or three species of insects rather than one. The wide range of d i s t r i b u t i o n of the species i s shown by the fac t that specimens i n the United National Museum are from Abies spp., Douglas f i r , and a l l Pinus spp. i n north and c e n t r a l America, B r i t i s h Columbia and Labrador south to Guatemala. This insect feeds on slash and longleaf pine i n F l o r i d a (Ebel and Merkel, 1957), i n the cones of red pine i n Ontario (Lyons, 1957) i n the cones of pine, spruce, true f i r s and Douglas f i r i n B r i t i s h Columbia (Ross, 1958), and i n Douglas f i r cones i n C a l i f o r n i a (Stevens, 1959). The larvae are also known to be shoot borers of pines and Douglas f i r , and occa s i o n a l l y feed on buds and fol i a g e of t h i s species. Almost no information i s av a i l a b l e on the e c o l o g i c a l factors a f f e c t i n g t h i s species. Because of i t s v a r i a b l e feeding habits i t i s very l i k e l y that they are independent of cone supply and competitors. The tolerance for a wide range of c l i m a t i c conditions i s confirmed by i t s wide d i s t r i b u t i o n . The l i f e h i s t o r y of D i o r y c t r i a i s not c l e a r yet. Keen (1958) stated that they have two generations per year. Lyons (1957) found only one generation per year i n Ontario. As Keen (1958) described, some eggs are deposited by moths reaching the adult stage i n October. More eggs are deposited by another group of moths which emerge i n May or e a r l y June. A f t e r hatching, the larvae bore through the scales and feed i n d i s c r i m i n a t e l y on scales, bracts and seeds. As the cones ripen, the larvae leave them and form t h e i r cocoons on the ground. Some of them pupate immediately and emerge i n October; the rest of the population (probably most of them) spends the winter as prepupal larvae, and pupates and emerges the following spring. No information i s a v a i l a b l e on diapause i n t h i s species. The eggs are 1 mm long, white, oval, and fl a t t e n e d with f i n e l y roughened surface. Larvae have f i v e i n s t a r s with 0.45, 0.71, 1.10, 1.35 and 1.70 mm head width, r e s p e c t i v e l y (Lyons, 1957). They are red or purple i n colour, sometimes with a greenish tinge. The absence of anal comb helps to d i s t i n g u i s h i t from the Douglas f i r cone phaloniid (Henricus fuscod-orsana (Kearf.)). The pupa i s 10-12 mm long with s t r a i g h t , slender caudal hooks. The forewing of adults i s predominantly grey, with white trans-verse zig-zag l i n e s . The wing spread of adults was described as 28-32 mm by de Dandt (1930) and 23-28 mm by Lyons (1957). Megastigmus spermotrophus (Wachtl.). The geographical d i s t r i b u t i o n of t h i s species follows well the d i s t r i b u t i o n of i t s host. I n f e s t a t i o n i s reported on Douglas f i r from 13 C a l i f o r n i a , Idaho, Colorado, Washington, B r i t i s h Columbia, Oregon and New Mexico (Keen, 1958). They were introduced i n infested seeds to Great B r i t a i n , Western Europe and New Zealand. This insect almost c e r t a i n l y feeds on seeds of Douglas f i r only. The l i f e h i s t o r y of Megastigmus i s well studied, but l i t t l e i s known about the e c o l o g i c a l f actors a f f e c t i n g the d i s t r i b u t i o n and i n t e n s i t y of attack. Hussey (1954) published a d e t a i l e d study on the l i f e h i s t o r y arid habits of th i s species. Mating takes place on the Douglas f i r needles soon a f t e r the adults emerge from the pupal stage. The female can lay f e r t i l e eggs without f e r t i l i z a t i o n , but a l l the adults from such parthe-nogenetic eggs are males. The act of o v i p o s i t i o n was described by M i l l e r (1916). 'He reported that the female rests on a cone scale with her head pointed toward the base of cone, drives her ovipositor', through the cone scales and deposits an egg i n a young seed. Two to f i v e minutes are required f o r o v i p o s i t i o n . According to Hussey (1955), normally only one egg i s l a i d i n a seed, but where there i s consider-able competition between the egg laying females f o r seed, as many as seven eggs are found i n one seed. Only one larv a develops to the adult stage when several eggs are l a i d within the same seed. No s p e c i f i c information i s av a i l a b l e on whether or not the female would lay an egg into an u n f e r t i l e or empty seed, although i t i s known that a p o t e n t i a l l y sound seed i s necessary f o r the development of a Megastigmus larva. It i s conceivable that the female selects by some means the f e r t i l i z e d seeds for o v i p o s i t i o n , which i s indicated by the fact that she spends a considerable time "choosing" the scale into which her o v i p o s i t o r i s inserted. 14 The o v i p o s i t i o n occurs when the cones are pendent and are 2-3 weeks o l d . Older or younger cones are safe from attack. Eggs hatch i n about 3-5 days. Larvae pass through f i v e i n s t a r s , which take s i x to eight weeks during July and August, and by the end of t h i s period the whole seed content i s eaten. Larvae remain i n the seeds over winter, pupate and emerge the following spring. The eggs are white, smooth and spindle-shaped, with a long pedicel at the anterior end. The larvae are up to 6 mm long, yellow white, fo o t l e s s grubs, with 14 d i s t i n c t body segments. The adults are small wasp-like insects with clear wings, having a well defined, clubbed stigma below the middle anterior margin of forewing. The females have a long, upcurved o v i p o s i t o r , 2-.5 - 4.6 mm long. Keen (1958) stated that less than ten per cent of the l a r v a l population become retarded as larvae i n each year and therefore emerge*1 as adults i n the second dr t h i r d spring. As Hussey (1955) reported th i s diapause can be broken by cold treatment. Koerber (1957) suggested 42 days cold treatment at 4° F to break the winter diapause. The Megastigmus larvae i n the seed, growing deep insd.de the cones, are very wellT.protected from parasites and predators. However, Hussey i n Scotland (1955) reported three species of c h a l c i d wasps, which are parasites of Megastigmus. He indicated that seed-eating rodents also destroy the larvae i n infested seeds. No such attacks have been report-ed i n the United States or Canada. There are no predatory insects described i n the l i t e r a t u r e which are known to prey on Megastigmus. Contarinia oregonensis (goote) The damage of Contarinia oregonensis i s one of the most serious to the seeds of Douglas f i r . The insect was recognized by M i l l e r i n 1914, but was not reported as causing serious damage u n t i l recently. Foote (1956) described the species from specimens taken i n Oregon i n 1916. Although the species was not described, Graham and Prebble (1941) observed the g a l l s i n cone scales of Douglas f i r . Rudinsky (1955) reported some midge damage i n scale tissue of Douglas f i r cones. Recently Hedlin (1958, 1959, 1961, 1962), Johnson (1962 a, 1962 b, 1962 c, 1963 a, 1963 c, 1963 d), Johnson and Heikkenen (1958), Johnson and Winjum (1960)and iPettinger and Johnson (1962) c a r r i e d out extensive studies on Contarinia oregonensis. Although numerous studies were done on t h i s problem, a large number of e c o l o g i c a l problems remain to be solved. Johnson and Heikkenen (1958) described the morphology and biology of the species. They have also stated that i n heavy i n f e s t a t i o n s the g a l l s may completely destroy or displace the seed; i n l i g h t e r i n f e s t -a t i o n the seed may not be damaged. However, the seed coat i s fused to the g a l l so the seed w i l l not f a l l from the cone when r i p e . Because of t h i s damage, as many as 99 per cent of seeds may remain i n the cone. They discovered that a good estimate of percentage of ga l l e d seeds within a cone can be obtained by the count on the cut surface of l o n g i t u d i n a l l y s l i c e d cones. Hedlin (1959) stated that "the cone midge i s the most important insect i n Douglas f i r cones'.1:' In one study he found 35 per cent of the cones to be inf e s t e d . According to Hedlin (1962), Contarinia oregonensis i s well adapted to i t s environment. Adults are able to o v i p o s i t under r e l a t i v e -l y unfavourable weather conditions. The larvae can survive even i n 16 water a f t e r leaving the cones. Johnson (1962 a) studied the over-wintering habit of the cone midges. He observed that they over-winter i n the l i t t e r under the tree from which they came, concentrating i n the base of male flowers and between the f a l l e n leaves of ground vegetation. No over-wintering larvae were found i n the mineral s o i l . The greatest concentration of cocoons was under the edge of the crown. Temperature v a r i a t i o n s between o o 4 and 10 C have no e f f e c t on the m o r t a l i t y of over-wintering larvae. The l i f e h i s t o r y of Contarinia i s described by many research workers. This can be summarized as follows: the eggs are deposited between the developing cone scales, during the period when the cones are open to receive p o l l e n . When the eggs hatch, larvae bore into the cone tissue and form polythalamous g a l l s on the scales near the seeds. The l a r v a l development i s completed by the time the cone matures. In the f a l l they leave the cone and over-winter i n the l i t t e r as prepupae and pupae. They emerge during the following spring as adults. A portion of the population remains i n diapause i n each year. Up to 52 per cent of the population (Hedlin, 1962) can remain i n the l a r v a l stage within the cocoon to emerge one or more years l a t e r . Because of t h i s , a good cone crop could be severely infested even when i t follows a poor or medium crop. The f r e s h l y l a i d eggs are white, translucent, and elongate "ovoid i n shape. The egg surface i s smooth and shiny. The larvae are f l a t -tened dorsoventrally, composed of 13 body segments and a r e t r a c t a b l e head, 3-4 mm i n length. They pass through three i n s t a r s . The f i r s t i n s t a r i s almost colourless, but the t h i r d i n s t a r i s orange-pink. Small hooks are v i s i b l e i n the second i n s t a r , and the anal hook i s f u l l y developed i n the t h i r d i n s t a r . The pupa i s s l i g h t l y shorter than the larvae. The adult c h a r a c t e r i s t i c s are v i s i b l e through the pupal skin. Thoracic horns, antennal horns and cephalic setae are of taxon-omic importance. The adult i s a d e l i c a t e insect with long legs and antenae. Its wings have reduced venation. Antenae are twelve-segmented. The abdomen i s bright orange i n colour. The females have long o v i p o s i t o r s . .Contarinia washingtonensis (Johnson). Since Contarinia washingtonensis i s a recently described species, not much information i s a v a i l a b l e on i t . Johnson (1963 a) described the species i n 1963 and reported the morphological c h a r a c t e r i s t i c s of the d i f f e r e n t stages. The female lays eggs when the cones are closed and pendent. The eggs are l a i d beneath the bracts of cones. The newly hatched larvae mine i n the cone scales and do not cause g a l l s l i k e Contarinia oregon- ensis larvae, when the cones are s t i l l greenish. No information i s a v a i l a b l e on e i t h e r the ecology of the insect, or on the factors a f f e c t i n g the insect i n the d i f f e r e n t stages of development. 4. Description of Sample Trees The following data were c o l l e c t e d to describe the sample trees: diameter at breast height (d.b.h.), height, crown width, crown density, crown length, s i t e index at 100 years, age, number of branches i n the f i r s t whorl above b.h., diameter of branches, pollen and flower product-ion, colour of female flowers, number of cones, cone length, cone width, phenological c h a r a c t e r i s t i c s , elevation and ground vegetation. Table 4 summarizes the p r e c i s i o n of these measurements. Table 4. P r e c i s i o n of tree c h a r a c t e r i s t i c s measurements. Object P r e c i s i o n D.b.h. Nearest tenth of an inch Height Nearest foot Crown width * Nearest foot Crown length Nearest foot Site index Nearest foot Diameter of branches Nearest tenth of a centimeter Cone length Nearest tenth of a centimeter Cone width Nearest tenth of a centimeter E l e v a t i o n Nearest foot * Two measurements were taken, one on north-south and another on west-east d i r e c t i o n . The following phenological c h a r a c t e r i s t i c s of the trees were av a i l a b l e from Dr. G r i f f i t h f o r 1958 to 1962: dates of i n i t i a t i o n and end of vegetative bud flus h i n g , duration of vegetative bud f l u s h i n g and date cones became erect. The data on date of cone er e c t i o n were c o l l e c t e d i n the period of 1958 - 1961, but the date cones became pendent was recorded only i n 1962. The number of female flowers was counted i n 1961 i n the cl a s s e s : 0, 0-50, 50-100, 100-150, 150-200. The larger numbers than 200 were estimated as medium or heavy. In 1962, the flowers were counted to the nearest 10 up to 200, and f o r those trees which had more conelets 19 than 200, the equation of Winjum and Johnson (1962) was used to estimate the number. The number of cones was counted up to 200 i n both years and numbers over 200 were estimated as medium and heavy i n 1961, and calculated by the Winjum and Johnson equation i n 1962. Because of the conceivable influence of cone colour on the a t t r a c t -ion and o v i p o s i t i o n behaviour of the insects, t h i s factor was taken into consideration. A f i v e - c l a s s system as worked out by Mr. 0. S z i k l a i and the author was used to describe the colour of female s t r o b i l i . The pure green was defined as class one, and the pure red as class f i v e . Three intermediate classes were set between the two extremes (Appendix C ) . The abundance of male s t r o b i l i was determined both i n 1961 and 1962 as none, very low, low, medium and heavy. For crown density a c l a s s i f i -c ation was worked out having f i v e classes i n i t from l i g h t to dense crown. The ground vegetation was described under each tree by the follow-ing a s s o c i a t i o n s : polystichum, o r t h i c blechnum, moss, vaccinium moss, o r t h i c s a l a l and l i t h o s o l i c s a l a l . 5. B i o l o g i c a l Studies of Cone and Seed Insects of Douglas F i r As i t was reviewed i n the d e s c r i p t i o n of four species studied, the morphology, anatomy, l i f e h i s t o r y and behaviour of cone and seed insects are f a i r l y w e ll known. In the present study, some observations were taken on the time of emergence of adults, time of attack, searching behaviour, abundance of p a r a s i t i c insects and competition between cone and seed in s e c t s . 20 Time of emergence: To f i n d some r e l a t i o n s h i p between the time of emergence of cone and seed insects and the s o i l temperature, eight rearing boxes, describ-ed by Johnson and Winjum (1960) were l a i d around a tree, two facing each d i r e c t i o n of north, west, south and east. One of the two i n each d i r e c t i o n was placed f i v e feet from the stem, well within the crown proje c t i o n ; the other was placed f i v e feet outside of the edge of the crown pr o j e c t i o n . Four other boxes were put under two trees on d i f f e r e n t elevations and aspects. This rearing experiment was i n i t i a t e d i n September 1961 and observation of time of emergence was c a r r i e d out i n the spring of 1962. Emerged insects were counted every day during c r i t i c a l periods (Appendix D). S o i l cages were set up close to the boxes to check whether or not the time of emergence i n boxes d i f f e r s from the time of natural emergence. The s o i l temperature was recorded beside the boxes from the 25th of A p r i l to the end of July (Appendix E ) . The d a i l y maximum, minimum and the average a i r temperature, as well as r e l a t i v e humidity, p r e c i p i t a t i o n and t o t a l hours of sunshine are av a i l a b l e from a weather s t a t i o n , which i s located about a mile from the area studied. The boxes were l e f t i n the f i e l d u n t i l the end of July, 1963 to obtain the percentage of insects which remained i n diapause. Cone bagging experiment: It was desired to f i n d the time of attack of cone and seed insects by bagging the cones. There are three d i f f e r e n t ways to carry out the bagging experiment. A number of cones can be bagged before the insects are emerged, then a few of them are removed p e r i o d i c a l l y . One other method i s . t o bag a c e r t a i n number of cones from time to time while the insects are a c t i v e . The t h i r d method i s s i m i l a r to the f i r s t one, but i n t h i s the cones, are released f or a c e r t a i n period of time only for insect attack. In t h i s experiment 20 cones were covered with cellophane bags on 25th of A p r i l , 1962, on each of the two trees selected f or t h i s study (Appendix F ) . One of these two trees was e a r l y and the other was late i n flowering. Two of these bags were removed every three days u n t i l the 25th of May, and then at two-week i n t e r v a l s u n t i l the end of June. The cones were tagged by date of removal of bags. At the same time the two bags were removed, three of the open cones were bagged, and tagged by date. A l l of these cones were c o l l e c t e d at the end of August and the number of damaged seeds was counted f o r each cone on the l o n g i t u d i n a l cut surface. Trapping experiment: It was desired to asc e r t a i n the role of f l i g h t behaviour on the d i s t r i b u t i o n of i n f e s t a t i o n within the crown. For t h i s purpose screens treated, with a non-drying adhesive were used as traps to catch the f l y -ing insects when they were searching for cones i n which to o v i p o s i t . This adhesive material used was "Stickem Special", a commercial product (piade by Michel and Petten Co., Oakland, C a l i f . ) , which was found to be very u s e f u l by Chapman (1962). One square foot of 16- mesh "Lumite" screen was framed, well smeared with stickem sp e c i a l and hung i n the f o l i a g e . Six trees were selected for t h i s experiment, two of them with 22 heavy cone crops, two without any cones, and two with medium cone crops. Three frames were hung on each tree, -one i n each of the three crown l e v e l s sampled, on the south side of the crown (Appendix F ) . In one tree four frames were hung i n the upper t h i r d of the crown and one was put on each of., the south, west, north and east sides. The number of the trapped Contarinia oregonensis was counted on 25th of May, 1962 and the Megastigmus spermotrophus on 16th of June. No D i o r y c t r i a were caught by the traps. 6. S t a t i s t i c a l Techniques Used  Analysis of variance: The term analysis of variance re f e r s to a general method of s t a t i s t i c a l inference. In general i t consists of a body of tests of hypotheses and methods of estimation, using s t a t i s t i c s which are l i n e a r combinations of sums of squares of the observed values. By analysis of variance the t o t a l variance of observed values can be analysed into i t s component fact o r s , the r e l a t i v e importance of which can then be tested against the r e s i d u a l variance. The r e s i d u a l variance i s c a l l e d experimental error, which i s the not accounted for variance of the factors contributing the t o t a l variance of experiment. Tukey (1949) stated: "perhaps more than any other s t a t i s t i c a l technique, the analysis of variance i s a l l things to a l l people". However, because of i t s many possible forms which depend on the structure of the process being analysed, the p r i n c i p a l d i f f i c u l t i e s are i n deciding which i s the appropriate form. Before using the analysis of variance i t i s necessary to consider the v a l i d i t y of the assumptions which are underlying the method. These assumptions are as follows: 1. The various e f f e c t s are fixed and the errors are ad d i t i v e . 2. The errors are normally d i s t r i b u t e d . 3. The errors are non-correlated. 4. The errors are the same from one experimental unit to another, regardless of the treatment used. Duncan's new multiple range t e s t : Duncan (1951) developed a multiple comparison test to compare each treatment mean with every other treatment mean. This test consists of three stages and i t i s very time-consuming. In 1955, he modified his t e s t , combining the three steps into one, This modified test i s very widely used i n s t a t i s t i c a l analyses to test the s i g n i f i c a n t differences between the means of a factor which was s i g n i f i c a n t i n the analysis of variance. The test i s based on the following formula: LSR = SSR x S x where: LSR = least s i g n i f i c a n t range SSR = s i g n i f i c a n t studentized range (may be obtained from tables as the function x of number of means compared and the degrees of freedom of experimental e r r o r ) . S x = |^  (error mean square) J r r = number of observations from which a mean to be compared i s calcu l a t e d . Frequency d i s t r i b u t i o n s and goodness of f i t : ' S c i e n t i f i c l i t e r a t u r e shows an increasing number of references to the use of various frequency d i s t r i b u t i o n s to describe q u a n t i t a t i v e l y the dispersion of organisms i n nature. The frequency d i s t r i b u t i o n helps to describe c e r t a i n b i o l o g i c a l phenomena, such as clumping or contagion and changes i n density (Graham, 1963). The frequency d i s t r i b u t i o n i s also the basis f o r determining the type of transform-at i o n to va l i d a t e the use of analysis of variance i n summarization of counted data. Many entomological studies are based upon a count of the number of i n d i v i d u a l insect species i n each unit area, host or time, f o r example, the number of sawfly cocoons per square foot of forest s o i l , the number of needle miners per twig, the number of bark-beetle g a l l e r i e s per unit area of bark, or the number of infested seeds per cone, e t c . When the sample contains a large number of units, the units may be scored i n a frequency d i s t r i b u t i o n showing the number of units with one, two or any number of insects, g a l l e r i e s or number of damaged seeds. Forest entomological data w i l l generally f i t one of the four following d i s t r i b u t i o n s : binomial, negative binomial, Poisson and normal (Waters, 1955). To f i t these common d i s t r i b u t i o n s the author and D. D. Munro (1963) wrote an I.B.M. 1620 s e l f contained, Fortran II computer program. In t h i s program the parameters of data are c a l c u l -ated, then the expected frequencies f o r each class are calculated using the appropriate mathematical formulae. To test the goodness of f i t of the observed frequencies, the Chi-square values are calculated f o r each frequency class and t o t a l Chi-square values f o r each d i s t r i b -u t i o n . Degrees of freedom are also tabulated (Appendix H). The program accepts a maximum of 20 frequency classes but t h i s may be 25 increased up to the capacity of the computer memory. This program was used to f i t a l l of the frequency d i s t r i b u t i o n s discussed i n t h i s t h e s i s . Time required to complete compiling and l i s t i n g the program i s 11.7 minutes. To complete one analysis, that i s , to c a lculate the parameters, f i t the four d i s t r i b u t i o n s and calcu-late chi-square values requires approximately 2.4 minutes. Regression and c o r r e l a t i o n a n a l y s i s : In regression analysis i t i s considered the problem of estimating the value of some dependent va r i a t e Y, on the basis of information on one or more other f i x e d values x, .... x L . The dependent va r i a b l e i s understood to have a p r o b a b i l i t y d i s t r i b u t i o n , while the independent v a r i a t e or v a r i a t e s do not have a p r o b a b i l i t y d i s t r i b u t i o n since they are f i x e d . Regression i s the equivalent of the term function i n mathe-matics. A regression can be l i n e a r or c u r v i l i n e a r , and i s c l a s s i f i e d by the exponent of the v a r i a b l e s . Linear or c u r v i l i n e a r regressions may come from a b i v a r i a t e or multivariate population. The c e n t r a l problem of the theory of regression i s to f i n d a l i n e or curve which represents the j o i n t p r o b a b i l i t y d i s t r i b u t i o n best, in an appropriate sense, usually i n the sense of least squares. Randomness i s e s s e n t i a l for the p r o b a b i l i t y theory, applying the regression a n a l y s i s . Also i t i s desirable to have independence between the f i x e d v a r i a t e s . In addition, the dependent va r i a b l e population has to be normal and has to have a common variance. Generally, a multiple regression equation can be given as: 26 Y = a + b x Xj_ + b 2 X 2 + b i Xj_ where: a = i n t e r c e p t y = dependent v a r i a b l e = independent v a r i a b l e b^ = -regression c o e f f i c i e n t The standard e r r o r of estimate (SEE) of the r e g r e s s i o n l i n e i s the square root of r e s i d u a l variance (RV), which i s the standard d e v i a t i o n of the b i v a r i a t e or m u l t i v a r i a t e j o i n t d i s t r i b u t i o n . The r e s i d u a l variance can be given as: SSY - (SP X]_ Y) 2 / SSX X - (SPXiY) 2 / SS^ RV = n - m - 1 where: SSY = sum of squares of the dependent v a r i a b l e SPY = sum of products of the dependent and independent v a r i a b l e s SSX^ = sum of square of the independent v a r i a b l e s n = t o t a l number of samples taken from the population m = number of independent v a r i a b l e s The e f f e c t i v e n e s s of the r e g r e s s i o n l i n e can be te s t e d by a n a l y s i s of v a r i a n c e , i n which a s i g n i f i c a n t variance r a t i o i n d i c a t e s a s i g n i f i c -ant change of the dependent v a r i a b l e w i t h the change of independent v a r i a b l e s . The order of importance of the independent v a r i a b l e s can be determined by the c o e f f i c i e n t of determination (CD). CD i s the deter-mination of r e l a t i v e importance of the v a r i a b l e s i n the c o n t r i b u t i o n of removed variance. In formula: b i SPYXi CDi = x 100 SSY During the routine, elimination of variables i n each t r i a l , the one with the most negative or the least p o s i t i v e CD i s eliminated f i r s t . The c o r r e l a t i o n c o e f f i c i e n t i s the measure of association between the dependent and independent v a r i a b l e s . If the population i s b i -v a r i a t e the c o r r e l a t i o n c o e f f i c i e n t i s c a l l e d simple ( r ) ; for a m u l t i v a r i a t e population, the c o e f f i c i e n t i s c a l l e d the multiple corre-l a t i o n c o e f f i c i e n t (R). Generally the c o r r e l a t i o n c o e f f i c i e n t i s calculated as: The higher the absolute value of R, the better i s the a s s o c i a t i o n of v a r i a b l e s . There are a number of other formulae a v a i l a b l e for the c a l c u l a t i o n of c o r r e l a t i o n c o e f f i c i e n t s and a number of other tests to test the measure of the a s s o c i a t i o n of the v a r i a b l e s . Sequential sampling: Sequential analysis i s a method of s t a t i s t i c a l inference whose c h a r a c t e r i s t i c feature i s that the number of observations required by the procedure i s not determined i n advance of the experiment. In the sequential procedure, sampling units are examined u n t i l the cumulative r e s u l t s f a l l into one of the classes distinguished by previously R where: -1.0 ± R ± + 1.0 determined l i m i t s . Because of th i s s i m p l i c i t y of the method i t i s very economical to use for forest insect sampling. This sampling method enables forest entomologists to determine the current status of known pests, where and when outbreaks are going to occur, and i f , and where, d i r e c t control measures are needed. For determining whether or not control measures are necessary, considerable time and expense can be saved i f a r b i t r a r y l i m i t s are set for just two class e s : "needing c o n t r o l " and "not needing c o n t r o l " . In order to develop an appropriate sequential sampling plan, i t i s necessary to know the d i s t r i b u t i o n involved. Sequential plans are defined by one or more pairs of p a r a l l e l decision l i n e s which usually serve, i n a forest entomological case, as c r i t e r i a for d i s t i n g u i s h i n g between dens i t i e s of i n f e s t a t i o n . Since a l l the units of the sample area are not examined, and only a random sample i s taken two types of errors can be made, e.g.: Type 1. The p r o b a b i l i t y of l a b e l l i n g an i n f e s t a t i o n medium when i t i s a c t u a l l y a l i g h t i n f e s t a t i o n . Type 2. The p r o b a b i l i t y of l a b e l l i n g an i n f e s t a t i o n l i g h t when i t i s a c t u a l l y a medium i n f e s t a t i o n . For sampling i t i s necessary to set up class l i m i t s which w i l l separate the d i f f e r e n t l e v e l s of i n f e s t a t i o n . To set up these l i m i t s the investigator has to be f a m i l i a r with the ecology, l i f e cycle and habits of the insect as well as with the economical measuring of the damage. The formulae to cal c u l a t e the equation of decision l i n e s can.'be derived from the formula of the appropriate d i s t r i b u t i o n , using the p r o b a b i l i t y r a t i o test (Wald, 1947). In addition to the decision l i n e s the operating c h a r a c t e r i s t i c (OC) curve i s a c h a r a c t e r i s t i c of the sequential analysis from which curve the power of the test, or i n other words the p r o b a b i l i t y of c o r r e c t l y c a l l i n g the samples, can be estimated as a function of i n t e n s i t y of damage. The sequential analysis does not specify a f i x e d sample size, but an average sample number (ASN) for examining the i n f e s t a t i o n at a s p e c i a l l e v e l can be given by a curve. This average sample number i s d i r e c t l y r e l a t e d to the allowance between classes and the desired degree of r e l i a b i l i t y . Narrowing the allowance or lowering the r i s k of error w i l l increase the average sample number. The formulae to f i n d the values of OC and ASN curves can also be derived from the d i s t r i b u t i o n function of the appropriate frequency d i s t r i b u t i o n . F i t t i n g a common k to a series of negative binomial d i s t r i b u t i o n s ; Since the counts for cone and seed insects were c o l l e c t e d from d i f f e r e n t trees a combined d i s t r i b u t i o n was needed to estimate the f i e l d population. This problem was solved by f i t t i n g a common k to a series of negative binomial d i s t r i b u t i o n s , because with one except-ion a l l of the d i s t r i b u t i o n s discussed i n t h i s paper were found to be negative binomial. The p r o b a b i l i t y function of' the negative binomial d i s t r i b u t i o n i s determined as: Px (k + x - 1) j R x x ! (k - 1) I g k where: x number of counts i n the sampling unit R p/g P m/k 30 g = 1 + p k = a parameter of the d i s t r i b u t i o n , or the measure of dispersion The negative binomial i s an extension of the Poisson series i n which the population mean i s not constant, but varies continuously i n a d i s t r i b u t i o n proportional to that of Chi-square ( B l i s s and Owen, 1958). There are a number of ways to estimate the common k for a series of d i s t r i b u t i o n s . B l i s s and Owen (1958) described two approaches to the problem. The most common of these two i s the regression method which was used i n th i s thesis because of i t s s i m p l i c i t y and accuracy for the problem. They proved the theory and derived the formulae to f i t a regression l i n e and 1/b of t h i s regression l i n e i s the estimate of common k (b = slope of the l i n e ) . Two var i a b l e s , x and y, are calculated from the parameters of each d i s t r i b u t i o n involved i n the combining process. X and y are defined as: x = m.2 s^/n", y = s2 - m? where: s^ = variance of the d i s t r i b u t i o n m = mean value of d i s t r i b u t i o n n = t o t a l number of counts i n the d i s t r i b u t i o n These two variables are calculated for each d i s t r i b u t i o n , and the equation l i n e i s calculated which should have a- zero intercept. Weighting was suggested by the authors when an increasing variance i s observed by increasing x values. In t h i s case the values of x should be weighted by the inverse variance. To s t a b i l i z e the variance logarithmic transformation i s needed i n some cases. The other method to f i t a common k f o r a number of d i s t r i b u t i o n s i s described by B l i s s (1953), which i s based on the theory of maximum l i k e l i h o o d . This method i s much more complicated and usually i s used only to f i n d a common d i s t r i b u t i o n of the d i f f e r e n t treatments i n an experiment. N-stage sampling, sample size and optimum a l l o c a t i o n : In some sampling schemes, the sampling units are i n groups of equal or unequal sizes and the groups, rather than the units, are sampled. Data that can be c l a s s i f i e d by a system consisting of a . unique order of c l a s s i f i c a t i o n c r i t e r i a , each c r i t e r i o n being applic-able within a l l categories of'the preceding c r i t e r i o n are i n an n-stage or "within-within-within" c l a s s i f i c a t i o n . Sampling the equal or un-equal sized groups of sampling units of these c l a s s i f i c a t i o n s i s c a l l e d n-stage or multistage sampling. Before the discussion of sample size i t has to be emphasized that multistage sampling i s usually less precise than random sampling for a f i x e d number of sample observations. The advantage of the method i s that by reducing the cost of observation i t obtains the desired p r e c i s i o n at a lower cost. Sample size i n multistage sampling i s determined from the variance of the sample mean formula (S^-). s 2 S2- n + + y n l n 2 n n 0.. .n 1 u2 n where: n. n sample size f o r nth stage n variance of nth stage The nature of the variance of the mean formula i s such that the same standard of error can be attained by using various combinations of n-s. The best choice of these combinations depends, n a t u r a l l y , on the cost of sampling within each category of c l a s s i f i c a t i o n , and on the magnitude of the variance obtained within each of the stages. In other words, the optimum a l l o c a t i o n of resources a v a i l a b l e f o r sampling i s a function of the components of variance and the money available to carry out the sampling operations. The problem may be approached from two points of view. The magnitude of S 2- can be s p e c i f i e d and th i s obtained at the minimum cost possible and secondly the t o t a l expenditure can be fi x e d and the smallest possible S 2- value obtained for the s p e c i f i e d expenditure. The two approaches b a s i c a l l y are the same since they want to minimize the(S2-)_x (cost) product, that i s : S 2- x C = ( s i 2 C-L + s 2 2 c 2 +....+ n i n 2 ^r^) X (C;L nl+c2^i^2+'• • + c n I 1 i r i 2 which i s equivalent to: 2 s 2 ^ CL S 2- x C = (SL2C-L + s22°2 + + s n 2 c n ) + n 2 s i 2 ° 2 + ^2 • • • • S l 2 c n n 2 " * n n + s2 c n *•* n 3 n 4 *•* n n + + s 2 n - l c n V It can be shown that t h i s expression has i t s minimum value when; c l s 2 2 / c2 s 3 2 , / c n - l s n 2 n 2 = ll C 7 I 7 2 - 5 n3 !y— and n„ c 3 s 22 c n s 2 n - l To f i n d the most economical values of n^ from n^, n^ and n n and the t o t a l sum of money ava i l a b l e (C) the following equation i s used: C = c l n l + c2 n2 n l + + c n n l n 2 n n 1* C-L, c 2 and c n may be given i n absolute monetary units or i n r e l a t i v e 33 r a t i o estimates for i t , which does not a f f e c t the c a l c u l a t i o n of optimum a l l o c a t i o n . Use of IBM 1620 e l e c t r o n i c computer: An IBM 1620 e l e c t r o n i c computer was used for almost a l l of the ca l c u l a t i o n s and s t a t i s t i c a l analyses presented i n t h i s study. Approximately 100 hours computer work was used for these c a l c u l a t i o n s , which i s roughly equivalent to 12,000 hours work on one of the best desk c a l c u l a t o r s . This volume of c a l c u l a t i o n would be a f u l l time job for a man for more than s i x years, even i f he worked without making mistakes. Fourteen programs for the use of the IBM 1620 were written by the author i n Fortran for various s t a t i s t i c a l analyses. Most of these were used s u c c e s s f u l l y to analyse the data c o l l e c t e d i n t h i s thesis and copies of the programs are a v a i l a b l e i n the U n i v e r s i t y of B r i t i s h Columbia Computing Center. EXPERIMENTAL RESULTS The analyses of data c o l l e c t e d were c a r r i e d out i n two separate groups. The cones were recorded as damaged or not damaged regardless of the number of seeds damaged by the d i f f e r e n t insect species. These data were analysed i n the f i r s t group of analyses. Then the t o t a l number of seeds and the number of damaged seeds were counted on the l o n g i t u d i n a l cut surface of each cone. Data c o l l e c t e d from the cut surfaces were analysed i n the second group of analyses. In both of the groups the analyses were directed towards showing differences, i f any, within trees and between trees. The crown l e v e l , outside versus inside crown (crown position) and d i f f e r e n t d i r e c t i o n s i n the crown were tested for within tree v a r i a t i o n . The difference between the two years of 1961 and 1962 was also tested. It was postulated for tests of s i g n i f i c a n c e that there was no difference between the d i f f e r e n t parts of the l i v i n g crown, between tree, and between the two years sampled. 1. Analysis of Amount of Cones Damaged It was not possible to take samples from a l l s i x sample locations on each tree. In most of the cases the lower crown l e v e l was not sampled because of the lack of cones. The same problem'was found i n a few cases with the middle crown l e v e l . Because of these, the trees were separated into three groups: trees with three, two and one crown l e v e l s sampled. Two of these three groups were used to show the within and between tree differences of damage i n analysis of variance. The analyses of variance were computed for three f a c t o r s : crown l e v e l , outside versus inside crown and tree. The number of damaged cones out of s i x was analysed separately for Contarinia oregonensis, Megastigmus  spermotrophus and D i o r y c t r i a a b i e t e l l a i n 1961. Besides these the number of cones damaged by Contarinia washingtoriensis was also analysed i n 1962. Further, the number of undamaged cones was analysed i n both years. Trees with three crown l e v e l s sampled: Table 5 summarizes the analyses of variance of trees with three crown lev e l s sampled i n 1961. The mean values of t h i s set of analyses were given i n Table 6. A l l the analyses of damaged and undamaged cones were c a r r i e d out with using the v a r i a b l e of number of damaged or undamaged cones, but the mean values of the analyses were transformed to percentages for the reader's convenience. Tables 7 and 8 summarize the analyses of variance and the mean values for 1962, respectively, f o r those trees with three crown lev e l s sampled. As indicated i n Table 5 and 7, the number of cones damaged by Contarinia oregonensis varied s i g n i f i c a n t l y from tree to tree i n 1961, and i n 1962 from l e v e l to l e v e l , from outside to inside halves of the crown, and from tree to tree. • Although the l e v e l and outside versus inside crown differences were not s i g n i f i c a n t i n 1961, there i s a d e f i n i t e increase i n damage from the lower (69.6 per cent) to the middle (74.5 per cent) t h i r d of the crown. The difference was not so obvious between the middle and upper t h i r d of the l i v i n g crown, because the damage was 75.0 per cent i n theuupper t h i r d . In the out-side h a l f of the l i v i n g crown the damage was 70.3 per cent against the 75.8 per cent damage i n the inside h a l f . The damage i n 1962 was TABLE 5. Analyses of variance of 1961 cones (trees with three crown l e v e l s sampled). Damage by Undamaged C. oregonensis Megastigmus D i o r y c t r i a cones Source of v a r i a t i o n DF 1 MS F MS F MS F MS F Tree (T) 16 5.34 6 . 4 7 * * 19.01 12.75** 5.13 3.78** 1.00 1.71 Po s i t i o n (P) 1 2.83 3.42 1.41 - 0.16 - 0.24 Crown Level (L) 2 1.08 1.31 3.44 2.30 1.41 1.04 0.09 T x P 16 1.98 2.39* 0.58 - 1.09 - 0.52 T x L 32 1.37 1.65 1:15 - 1.54 1.14 0.23 P x L 2 2.77 3.35* 2.32 1.55 2.50 1.85 0.19 T x P x L 32 0.83 1.49 1.35 0.58 Tota l 101 1 The same notations w i l l be used i n a l l analyses of variance i n t h i s t h e s i s . DF = Degrees of freedom. MS = Mean square. F = F value of the variance r a t i o t e s t . * = S i g n i f i c a n t at 0.05 p r o b a b i l i t y l e v e l . ** = S i g n i f i c a n t at 0.01 p r o b a b i l i t y l e v e l . TABLE 6. Percentages of damaged and undamaged cones in 1961 (means from analyses of variance of Table 5). Crown All Ranges of tree Level Position over variation Damaee bv Lower Middle Upper Outside Inside average ,r> Maximum Minimum Percentages! C. oregonensis 69.6 74.5 75.0 70.3 75.8 73.0 94.4 50.0 Megastigmus 40.7 30.4 38.4 34.3 38.2 36.3 97.2 0.0 Dioryctria 18.6 24.5 18.6 21.2 19.9 20.6 50.0 0.0 Undamaged cones 8.3 6.8 6.8 8.2 6.5 7.4 22.0 0.0 1 Because some cones were damaged by more than one species of insect the percentages do not add to 100. Table 7. Analyses of variance of 1962 cones (trees with three crown l e v e l s sampled). Damage by Undamaged C. oregonensis Megastigmus C. washingtonensis D i o r y c t r i a cones Source of v a r i a t i o n DF MS F MS F MS F MS F MS F Tree (T) 28 8.99 10.07** 10.17 6.79** 5.30 4.52** 0.41 2.07* 2.62 5.75** Po s i t i o n (P) 1 8.30 9.30** 1.47 - 0.14 - 0.05 - 0.37 -Crown Level (L) 2 16.57 18.57** 3.93 2.62 0.21 - 0.25 1.25 2.69 5.92** T x P 28 1.57 1.76* 1.13 - 1.83 1.56 0.23 1.16 0.39 -T x L 56 1.47 :.1.65 1.49 - 1.18 1.00 0.23 1.15 0.52 -P x L 2 0.85 - 0.57 - 0.70 - 0.47 2.35 0.59 1.30 T x P x L 56 0.89 1.50 1.17 0.20 0.45 Tota l 173 OO TABLE 8. Percentages of damaged and undamaged cones i n 1962 (means from analyses of variance of Table 7). Crown Ranges of Level P o s i t i o n A l l tree Damage by Lower Middle Upper Outside Inside over average v a r i a t i o n Ma x imum Min imum Percentages C. oregonensis 56.9 65.5 74.7 62.1 69.3 65.7 94.4 22.2 Megastigmus 32.2 37.4 40.8 35.2 38.3 36.8 80.6 2.8 C. washingtonensis 35.1 35.9 33.9 35.4 34.5 34.9 58.3 0.0 D i o r y c t r i a 3.4 3.2 5.2 4.2 3.6 3.9 19.4 0.0 Undamaged cones 12.9 13.7 7.2 12.1 10.5 11.3 44.4 0.0 The r e s u l t s of the analyses of undamaged cones were quite d i f f e r e n t i n the two years. No s i g n i f i c a n t differences were detected e i t h e r between or within trees i n 1961. The analysis of 1962 data showed s i g n i f i c a n t v a r i a t i o n between trees and for crown l e v e l s . The Duncan's multiple range test indicated that there was no difference between lower (12.9 per cent) and middle crown le v e l s (13.7 per cent), but the number of undamaged cones was s i g n i f i c a n t l y less i n the upper crown l e v e l (7.2 per cent) than i n any of the other two. Trees with two crown l e v e l s sampled: The kind of analyses just discussed were repeated for those trees, which had samples from at least two crown le v e l s (middle and upper third ) and from outside and inside crown. Sixty-four trees were analysed f o r 1961 (Tables 9 and 10), and 66 for 1962 (Tables 11 and 12). Because the r e s u l t s of these analyses were s i m i l a r to those of the trees with three crown l e v e l s sampled, only the differences w i l l be discussed. The analysis of number of damaged cones by Contarinia oregonensis i n 1961 d i f f e r e d only i n one point from the analysis of trees with three crown l e v e l s . The outside versus inside crown by l e v e l i n t e r -action was not s i g n i f i c a n t when only two crown l e v e l s were used i n the a n a l y s i s . This indicates that the s i g n i f i c a n t difference i n the analysis of trees with three crown l e v e l s was caused mostly by the big difference between outside (62.7 per cent) and inside crown (76.3 per cent) i n the lower t h i r d of the l i v e crown. In the analysis of 1962 data the number of cones damaged by Contarinia oregonensis varied highly s i g n i f i c a n t l y from outside to inside crown i n the use of trees with three crown l e v e l s . This r e s u l t 41 s i g n i f i c a n t l y higher i n the inside crown (69.3 per cent) than i n the outside crown (62.1 per cent), which i s exactly the opposite trend of damage as found i n 1961. In 1962 the damage was s i g n i f i c a n t l y l i g h t e r i n the lower crown l e v e l (59.9 per cent) than i n the middle crown l e v e l (65.5 per cent) then s i g n i f i c a n t l y l i g h t e r i n the middle crown l e v e l than i n the upper one (74.7 per cent). These differences for crown l e v e l were tested by Duncan's new multiple range t e s t . In both years the number of damaged cones varied s i g n i f i c a n t l y from tree to tree. These v a r i a t i o n s w i l l be discussed i n d e t a i l l a t e r . Since the tree by crown p o s i t i o n i n t e r a c t i o n was s i g n i f i c a n t i n both ye a r S j the inverse i n f e s t a t i o n trend from outside to inside crown i n the two years was caused by the fact that the trees are d i f f e r e n t i n the trend of damage from outside to inside crown. The number of damaged cones by Megastigmus spermotrophus was found s i g n i f i c a n t l y d i f f e r e n t from t r e e to t r e e i n both 1961 and 1962. No s i g n i f -i c a n t v a r i a t i o n was found w i t h i n the crown of the t r e e S j though the damage showed an increasing t r e n d from the lower to the upper crown l e v e l i n 1962 and from outside to inside crown i n both years (Tables 6 and 8). S i m i l a r l y to that by Megastigmus the damage by D i o r y c t r i a a b i e t e l l a was found to be s i g n i f i c a n t l y d i f f e r e n t from tree to tree^ andno s i g n i f i c a n t differences were observed within the crown of the tree. The number of cones infested by Contarinia washingtonensis was observed and analysed i n 1962 only (Tables 7 and 8). This analysis showed a s i g n i f i c a n t v a r i a t i o n between trees. The within tree v a r i a t i o n of the damage was not s i g n i f i c a n t . Table 9. Analyses of variance of 1961 cones (trees with two crown lev e l s sampled). Damage by Undamaged C. oregonensis Megastigmus D i o r y c t r i a cones Source of Va r i a t i o n DF MS F MS F MS F MS F Tree (T) 63 3.95 4.25** 9.75 7.05** 3.32 4.38** 1.23 2.91** Posi t i o n (P) 1 0.02 - 1.13 - 3.29 4.34** 0.19 -Crown Level (L) 1 2.64 2.84 2.85 2.05 0.10 - 1.13 2.66 T x P 63 1.64 1.76* 0.91 - 1.17 1.55* 0.64 1.52 T x L 63 1.33 1.43 1.61 1.16 1.32 1.74* 0.31 -P x L 1 0.02 - 1.13 - 0.66 - 0.10 -T x P x L 63 0.93 1.38 0.76 0.42 Total 255 Table 10. Percentages of damaged and undamaged cones in'1961 (summary of analyses of variance of Table 9). Crown A l l Ranges of tree Level Position over v a r i a t i o n Damage by Middle Upper Outside Inside average Maximum Minimum Percentages C. oregonensis 73.9 77.3 75.8 75.5 75.6 100.0 20.8 Megastigmus 32.9 36.4 33.6 35.8 34.7 95.8 0.0 D i o r y c t r i a 19.4 18.7 20.9 17.2 19.1 54.2 0.0 Undamaged cones 9.4 7.2 7.8 8.7 8.3 50.0 0.0 Table 11. Analyses of variance of 1962 cones (trees with two crown l e v e l s sampled). Damage by Undamaged cones C. oregonensis Megastigmus C. washingtonensis D i o r y c t r i a Source of v a r i a t i o n DF MS F MS F MS F MS F MS F Tree (T) 65 6.95 8.41** 7.67 6.17** 5.81 5.16** 0.69 2.19** 2.00 6.10** Posi t i o n (P) 1 5.47. 6.62* 0.85 - 0.74 - 0.05 - 0.00 -Crown Level (L) 1 14/; 5 6 • 17.63** 0.64 - 0.00 - 0.24 - 5.66 17.62** T x P 65 1.37 1.65* 1.38 1.10 1.30 1.15 0.35 1.09 0.28 T x L 65 1.18 1.43 1.66 1.33 1.44 1.27 0.36 1.13 0.41 1.24 P x L 1 1.83 2.22 2.37 1.90 0.38 - 0.97 3.07 0.00 T x P x L 65 0.83 1.24 1.12 0.31 0.33 Total 263 TABLE 12. Percentages of damaged and undamaged cones i n 1962. (Means from analyses of variance of Table 11). Crown A l l Ranges of tree Level P o s i t i o n over v a r i a t i o n Damage by Middle Upper Outside Inside average Maximum Minimum Percentages C. oregonensis 66.4 74.2 67.9 72.7 70.3 100.0 4.2 Megastigmus 36.4 38.0 36.2 38.1 37.2 83.3 0.0 C. washingtonensis 35.3 35.3 36.2 34.5 35.3 :83.3 0.0 D i o r y c t r i a 5.0 6.0 5.8 5.3 5.5 25.0 0.0 Undamaged cones 11.5 6.6 9.1 8.9 9.0 66.7 0.0 46 became just s i g n i f i c a n t i n the two crown levels analysis (Table 11), with averages of 67.9 and 72.7 per cent i n outside and inside crown^ r e s p e c t i v e l y . The analyses of cones infested by Megastigmus gave s i m i l a r r e s u l t s to those when three crown le v e l s were sampled. In 1961 the analysis of number of cones infested by D i o r y c t r i a r e s u l t e d a highly s i g n i f i c a n t variance r a t i o for crown p o s i t i o n , i n a d d i t i o n to the highly s i g n i f i c a n t tree to tree v a r i a t i o n since 20.9 per cent of the cones were infested i n the outside crown against the 17.2 per cent i n the inside crown. Two of the int e r a c t i o n s , the trees by outside versus inside crown and trees by crown l e v e l , were also s i g n i f i c a n t at the 0.05 p r o b a b i l i t y l e v e l . In 1962 no differ e n c e was found between the two analyses of two and three crown l e v e l s . As Tables 7 and 11 indicate, no difference was found between the r e s u l t s of analyses of trees with two and three crown levels sampled i n the case of Contarinia washingtonensis i n f e s t a t i o n . In both cases only the between tree v a r i a t i o n was highly s i g n i f i c a n t . Analysing the number of undamaged cones of 1962, the two sets of analyses showed s i m i l a r r e s u l t s (Tables 7 and 11). In the analysis of 1961 data (Table 9), the between tree v a r i a t i o n became highly s i g n i f i c a n t , which was caused by the larger number of trees i n the two crown l e v e l a n a l y s i s . The trees by crown p o s i t i o n i n t e r a c t i o n a l s o showed s i g n i f i c a n t v a r i a t i o n i n the two crown l e v e l a n a l y s i s . Analyses of year to year d i f f e r e n c e s : Only 32 trees could be sampled with two crown levels ( middle and upper t h i r d ) j outside and inside crown and with s i x cones i n each sample l o c a t i o n i n both 1961 and 1962. The numbers of damaged and not damaged cones of these trees was studied by analysis of variance for four f a c t o r s : year, tree, crown l e v e l , and crown p o s i t i o n for Contarinia Table 13. Analyses of variance of damaged and undamaged cones by tree, year, crown position and l e v e l . Damage by Undamaged C. oregonensis Megastigmus D i o r y c t r i a cones Source of v a r i a t i o n DF MS F MS F MS F MS F P o s i t i o n (P) 1 0.00 - 0.02 - 0.02 - 0.00 -Tree (T) 31 7.40 10.37** 13.07 8.74** 2.06 4.70** 2.44 4.45** Year (Y) 1 4.52 6.33* 4.52 3.01 36.00 82.00** 0.88 1.'60 Crown Level (L) 1 4.00 5.60* 2.64 1.76 0.56 1.28 1.72 3.13 P x T ' 31 1.10 1.53 1.07 - 0.70 1.59 0.50 -P x Y 1 0.56 - 0.00 - 0.39 - 0.19 -P x L 1 0.39 - 0.56 - 0.39 - 0.03 -T x Y 31 2.95 4.13** 3.36 2.24* 1.73 3.94** 1.28 2.33* T x L 31 1.18 1.65 1.44 - 0.88 1.99* 0.40 -Y x L 1 0.06 - 6.25 4.17* 0.00 - 0.66 1.20 P x T x Y 31 1.48 2.07* 1.23 - 0.83 1.90* 0.35 -P x T x L 31 0.99 1.38 1.62 1.08 0.41 - 0.24 -P x Y x L 1 0.39 - 0.39 - 0.14 - 0.10 -T x Y x L 31 1.22 1.71 1.03 - 0.75 1.70 0.22 -P x T x Y x L 31 0.71 1.49 0.44 0.55 Total 255 •p-Table 14. Percentages of damaged and undamaged cones by year, crown l e v e l and p o s i t i o n and tree (summary of analyses of variance of Table 13). Crown A l l Ranges of Level Position over tree Year Middle Upper Outside Inside average v a r i a t i o n Damage by 1961 1962 Maximum Minimum Percentages C. oregonensis 77.6 73.2 73.3 77.5 75.4 75.4 75.4 97.9 27.1 Megastigmus 36.8 32.4 32.9 36.2 34.5 34.7 34.6 79.2 4.2 D i o r y c t r i a 18.2 5.7 11.2 12.8 11.8 ' 12.10 11.9 33.3 2.1 Undamaged cones 6.8 8.7 9.1 6.4 7.7 7.8 7.8 39.6 0.0 oregonensis, Megastigmus spermotrophus and D i o r y c t r i a a b i e t e l l a (Tables 13 and 14). The damage caused by Contarinia oregonensis was s i g n i f i c a n t l y d i f f e r e n t from year to year, (77.6 per cent for 1961 and 73.2 per cent for 1962). The highly s i g n i f i c a n t F-value for tree to tree v a r i a t i o n , and the s i g n i f i c a n t JF value f o r crown le v e l s agrees with the preceeding analyses (Tables 5, 7, 9, 11). The s i g n i f i c a n t f i r s t and second order int e r a c t i o n s (tree by year, tree by year by outside versus inside crown) w i l l be discussed l a t e r f o r each species. No s i g n i f i c a n t difference was found between the number of cones damaged my Megastigmus i n 1961 (36.8 per cent) and i n 1962 (32.4 per cent). The tree v a r i a t i o n was highly s i g n i f i c a n t i n a l l of the analyses discussed above (Tables 5, 7, 9 and 11). Also, the tree by year and l e v e l by year in t e r a c t i o n s were s i g n i f i c a n t . The greatest difference between the 1961 and 1962 damage was found i n D i o r y c t r i a . An average of 18.2 per cent of the cones was damaged i n 1961, more than three time the 5.7 per cent damaged i n 1962. The number of cones damaged by D i o r y c t r i a per tree resulted i n a highly s i g n i f i c a n t variance r a t i o which agrees well with the r e s u l t s of Tables 5, 7, 9 and 11. The tree by year i n t e r a c t i o n was highly s i g n i f i c a n t . Tree by l e v e l and tree by year by crown p o s i t i o n f i r s t and second order i n t e r a c t i o n s were just s i g n i f i c a n t . No s i g n i f i c a n t difference was found between the number of undamaged cones of 1961 (6.8 per cent) and 1962 (8.7 per cent). Although, the difference i s noticable, the high between tree v a r i a t i o n r e s u l t e d i n a high error term variance, and the analysis of variance did not show s i g n i f i c a n c e f o r year to year v a r i a t i o n . Table 13 shows two highly s i g n i f i c a n t and two s i g n i f i c a n t F -values f o r tree by year i n t e r a c t i o n . This may indicate that a large number of trees which were heavily infested i n 1961 became l i g h t l y i n f ested i n 1962, or v i c e versa. Because of t h i s , 63 trees from which some kind of samples were taken i n both years were c l a s s i f i e d as being l i g h t l y , moderately and heavily infested by the three d i f f e r e n t insect species. This c l a s s i f i c a t i o n was worked out for the observation of 1961 and 1962, separately. The c l a s s i f i c a t i o n was based on the l i m i t s of the average + 0.33 times the standard deviation, which resulted i n a f a i r l y s i m i l a r number of trees i n each c l a s s . In other words the tree was l i g h t l y infested i f the number of infested cones on i t was less than or equal to the average -0.33 times the standard deviation, and heavily infested i f the damage was more than or equal to the average +0.33 times the standard deviation, and the medium tree was between the two l i m i t s . The changes from 1961 to 1962 were recorded as same (='no change), minor change (from l i g h t to medium and vice versa and from medium to heavy and vice versa) and major change (from l i g h t to heavy and v i c e versa). The r e s u l t s of t h i s analysis are summarized i n Table 15. Table 15. Changes of i n f e s t a t i o n class from year to year on the same i n d i v i d u a l tree. Minor Major Same or Minor Damage by Same change change change per cent of trees D i o r y c t r i a 41.2 39.7 19.1 80.9 Megastigmus 63.7 20.6 15.7 84.3 C. oregonensis 52.5 36.5 11.0 89.0 As Table 15 shows, the percentage of major changes was not high, but i t was high enough to give s i g n i f i c a n t year by tree i n t e r a c t i o n . On the other hand, a l l of the highly s i g n i f i c a n t and s i g n i f i c a n t i n t e r -actions may be determined by the highly s i g n i f i c a n t tree v a r i a t i o n or the highly s i g n i f i c a n t year v a r i a t i o n i n the case of D i o r y c t r i a (Table 13). The s i g n i f i c a n t i n t e r a c t i o n of year by l e v e l of Megastigmus damage s i g n i f i e s that the damage i n 1962 was decreasing from middle to upper crown l e v e l (from 37.7 to 35.8 per cent) and increasing i n 1961 (from 28.2 to 36.6 per cent). S i m i l a r l y the s i g n i f i c a n t tree by l e v e l i n t e r a c t i o n indicates that the trend of i n t e n s i t y of attack changes from l e v e l to l e v e l as the year changes. -Analysis of damage by d i r e c t i o n i n the crown: As described i n the chapter on Materials and Methods, 5 trees were sampled i n both years from North, West, South and East sides to f i n d the differences i n damage of the various sides. Six cones were analysed for._C. oregonensis, Megastigmus and D i o r y c t r i a damage and for undamaged cones i n 1961 (Table 16). In addition the damage of washingtonensis also was analysed i n 1962 (Table 17). The mean values for four d i r e c t i o n s are summarized i n Table 18. D i r e c t i o n was not s i g n i f i c a n t i n 1961. The tree by d i r e c t i o n i n t e r a c t i o n was s i g n i f i c a n t i n the analysis of _C. oregonensis damage. As i t i s shown by a graph i n Appendix I the i n t e n s i t y of attack changes from side to side on d i f f e r e n t trees, but these differences average out. This i n t e r a c t i o n was not s i g n i f i c a n t i n the 1962 analysis (Table 17), but d i r e c t i o n showed s i g n i f i c a n t v a r i a t i o n i n _C. oregonensis. Analysing Table 16. Analyses of variance of 1961 cones by tree, d i r e c t i o n , crown p o s i t i o n and l e v e l . Dama£ ?e by Undamaged C. oregonensis Megastigmus D i o r y c t r i a cones Source of v a r i a t i o n DF MS F MS F MS F MS F Tree (T) 4 4.84 3.72* 30.07 17.68** 8.08 5.93** 2.49 4.38** Di r e c t i o n (D) 3 3.74 2.88 1.09 - 2.61 1.91 1.61 2.83 Po s i t i o n (P) 1 1.87 1.44 0.13 - 7.01 5.14* 4.41 7.76* Crown Level (L) 2 5.47 4.21* 0.93 - 2.03 1.48 4.43 7.80** T x D 12 2.99 2.30* 1.08 - 2.76 2.02 1.33 2.34* T x P 4 4.38 3.36* 1.03 - 1.47 1.07 0.28 - . T x L 8 1.64 1.26 0.96 - 2.08 1.52 0.77 1.34 D x P •3 0.21 - 2.51 1.47 3.54 2.60 0.36 -D x L 6 2.58 1.98 2.56 1.50 1.06 - 0.87 1.52 P x L 2 0.53 - 2.43 1.43 0.51 - 1.43 2.52 T x D x P 12 1.40 1.08 1.57 - 1.56 1.14 0.82 1.44 T x D x L 24 1.66 1.27 0.56 - 1.10 - 0.76 1.33 T x P x L 8 1.09 - 1.67 - 1.50 1.09 0.75 1.31 D x P x L 6 2.29 1.76 1.2! - 1.61 1.18 0.76 1.33 T x D x P x L 24 1.30 1.70 1.36 0.57 Total 119 Table 17. Analyses of variance of 1962 cones by tree, d i r e c t i o n , crown p o s i t i o n and l e v e l . Damage by Undamaged cones C. oregonensis Megastigmus C.washingtonensis D i o r y c t r i a Source of v a r i a t i o n DF MS F MS F MS F MS F MS F Tree (T) 4 2.72 2.30 10.94 11.58** 9.55 8.24** 0.78 2.44 1.60 2.67 Di r e c t i o n (D) 3 4.67 3.95* 0.38 - 1.52 1.31 0.25 - 0.58 -P o s i t i o n (P) 1 6.07 5.13* 0.53 - 0.01 - 0.41 1.28 4.03 6.76* Crown Level (L) 2 24.56 20.77** 2.56 2.70 0.91 - 0.23 - 3.63 6.09*' T x D 12 1.88 1.58 1.38 1.46 0.84 - 0.27 - 0.78 1.30 T x P 4 0.80 - 1.22 1.29 1.97 - 0.22 - - 0.10 -T x L 8 1.31 Is 10 1.28 1.35 0.98 - 0.44 1.38 0.57 -D x P 3 0.16 - 0.24 - 0.63 - 0.27 - 0.12 -D x L 6 0.69 - 1.57 1.66 0.45 - 0.18 - 0.38 -P x L 2 0.77 - 0.16 - 0.16 - 0.03 - 0.83 1.39 T x D x P 12 1.12 - 1.13 1.19 0.92 - 0.23 - 0.38 -T x D x L 24 1.29 1.09 2.69 2.84** 1.26 1.08 0.14 - 0.91 1.53 T x P x L 8 1.38 1.16 1.06 1.12 1.71 1.47 0.10 - 1.02 1.71 D x P x L 6 0.73 - 1.64 1.73 1.15 - 0.20 - 0.42 -T x D x P x L 24 1.18 0.94 1.16 0.32 - 0.60 Total 119 Table 18. . Percentage of damaged and undamaged cones by d i r e c t i o n . D i r e c t i o n North West South East Average Damage by 1961 Percentages C. oregonensis 70.0 67.2 60.6 57.2 63.8 Megastigmus 21.1 25.6 28.3 27.2 25.6 D i o r y c t r i a 20.6 18.9 30.0 22.2 22.9 Undamaged cones 12.2 21.1 15.6 13.9 15.7 1962 Percentages C. oregonensis 71.1 59.4 74.4 71.1 69.0 Megastigmus 32.2 32.2 28.3 31.7 31.1 C. washingtonensis 30.0 37.2 35.6 30.0 33.2 D i o r y c t r i a 2.2 3.3 5.6 5.0 4.0 Undamaged cones 11.7 15.0 9.4 12.8 12.2 55 th i s s i g n i f i c a n t v a r i a t i o n by Duncan's new multiple range test no difference was found between North (71.1 per cent), South (74.4 per cent) and East (71.1 per cent) sides, but the damage on West side (59.4 per cent) was s i g n i f i c a n t l y lower than any other. No other insect damage resulted i n s i g n i f i c a n t v a r i a t i o n for d i r e c t i o n i n 1962 (Table 17) and no explanation i s av a i l a b l e for these di f f e r e n c e s . One tree was sampled from eight sides i n 1962 at 45 degree;:inter-vals s t a r t i n g with North. The damage by _C. oregonensis, Megastigmus, _C. washingtonensis, D i o r y c t r i a and the undamaged cones were analysed on t h i s tree (Table 19, 20). Only the damage of _C. washingtonensis va r i e d s i g n i f i c a n t l y by d i r e c t i o n . The i n f e s t a t i o n was s i g n i f i c a n t l y more on two opposite sides S.W. (61.1 per cent), and N.E. (69.4 per cent) than the other d i r e c t i o n s . •Analyses of between tree v a r i a t i o n : With a very few exceptions a l l the analyses of variance resulted a s i g n i f i c a n t or highly s i g n i f i c a n t variance r a t i o for tree v a r i a t i o n i n damage. Because of the high v a r i a t i o n of the number of damaged cones from tree to tree, the factors which may a f f e c t t h i s v a r i a t i o n were studied by c o r r e l a t i o n analyses with the number of damaged cones per tree f o r each insect species. Four multiple c o r r e l a t i o n and regression analyses were made for 1961 damage of each of D i o r y c t r i a , Megastigmus, _C. oregonensis and undamaged cones as dependent v a r i a b l e s . The analyses were repeated for the 1962 data, including the damage of C. washingtonensis. The d e s c r i p t i o n of independent v a r i a b l e s of 1961 and 1962 are summarized i n Tables 21 and 22, res p e c t i v e l y . The Table 19. Analyses of-variance of 1962 cones by eight d i r e c t i o n s , crown l e v e l and p o s i t i o n . Damage by Undamaged C. ore gonensis Megastigmus C.washingtonensis D i o r y c t r i a cones Source of v a r i a t i o n DF MS F MS F MS F MS F MS F D i r e c t i o n (D) 7 0.64 - 2.71 1.62 1.85 7.41** 0.13 - 0.40 -P o s i t i o n (P) 1 1.69 1.24 1.69 1.01 3.52 14.08** 0.08 - 1.02 1.95 Crown Level (L) 2 6.90 5.09* 1.19 - 1.08 4.33* 0.02 - 1.90 3.64 D x P 7 1.64 1.21 1.26 - 1.33 5.32** 0.51 1.89 0.16 -D x L 14 0.44 - 2.35 1.40 0.63 2.52* 0.21 - 0.56 1.08 P x L 2 0.19 - 1.31 - 5.08 20.33** 0.27 1.00 1.52 2.92 D x P x L 14 1.35 1.67 0.25 0.27 0.52 Total 47 O N Table 20. Percentages of damaged and undamaged cones by eight d i r e c t i o n s on the crown. Direction Damage by N NW W SW S SE E NE Average C. oregonensis 69 .4 58.3 58.3 58.3 63.9 66. 7 52.8 58.3 60.8 Megastigmus 36 .1 38.9 30.5 38.9 19.4 55. 6 33.3 22.2 34.4 C. washingtonensis 41 .7 44.4 47.2 61.1 50.0 52. 8 47.2 69.4 51.7 D i o r y c t r i a 5 .6 2.8 8.3 2.8 2.8 5. 6 2.8 8.3 4.9 Undamaged cones 8 .3 11.1 13.9 5.6 8.3 8. 3 19.4 11.1 10.8 Table 21. Independent tree variables used i n multiple c o r r e l a t i o n and regression analysis of 1961 data (93 tr e e s ) . Variable Average Standard deviation Maximum>r value Minimum value XI = Site index (at 100 years) 156.24 28.40 200.0 90.0 X2 = Tree height (feet) 42.87 7.15 63.0 28.0 X3 = D.b.h. (inches) 9.71 2.54 18.0 5.6 X4 = Number of branches i n the f i r s t whorl above b.h. 7.48 2.51 15.0 4.0 X5 = Forest type (7 classes by quality = 1-7) 3.09 1.27 6.0 1.0 X6 = Number of cones 165.91 167.63 600.0 11.0 X7 = Crown width (feet) 20.85 3.57 29.0 13.0 X8 = Cone length (cm) 5.55 0.84 7.7 3.8 X9 = Cone width (cm) 2.03 0.22 2.6 1.6 XI0= Date cones became erect (days after f i r s t of A p r i l ) 29.43 4.64 •- 43.0 18.0 XI1= Elevation (feet from Loon Lake) 219.95 65.94 310.0 130.0 XI2= Size of pollen crop (5 classes, 0-4) 0.71 1.06 4.0 0.0 X13= Size of female flower crop (7 classes, 0-6) 3.82 1.80 6.0 1.0 XI4= Average flushing time (days a f t e r f i r s t of May) 36.08 5.95 58.0 22.0 X15= Duration of flushing (days) 47.19 5.44 60.0 36.0 XI6= Age i n 1961 (years) 19.58 6.65 34.0 12.0 XI7= Colour of female flowers (5 classes, 1-5) 2.99 1.05 5.0 1.0 XI8= Crown density (5 classes, 1-5) 3.24 0.98 5.0 1.0 X19= Percentage of f i l l e d seeds per acre (arcsin sq.rt.) 10.15 5.11 22.4 0.0 Table 22. Independent tree variables used i n multiple c o r r e l a t i o n and regression analysis of 1962 data (97 trees). Standard Maximum Minimum Variable Average deviation value value XI = Site index (at 100 years) 158.45 27.74 210.0 90.0 X2 = Tree height (feet) 43.74 7.62 63.0 28.0 X3 = D.b.h. (inches) 10.08 2.60 18.0 5.9 X4 = Number of branches in the f i r s t whorl above b.h. 7.79 2.82 18.0 4.0 X5 = Forest type (7 classes by quality, 1-7) 3.15 1.42 6.0 1.0 X6 = Number of cones 438.26 553.29 223.0 28.0 X7 = Crown width (feet) 20.89 3.78 29.0 13.0 X8 = Cone length (cm) 5.53 0.65 7.6 3.9 X9 = Cone width (cm) 2.10 0.22 3.0 1.7 XI0= Date cones became pendent (days a f t e r f i r s t of May) 18.81 5.35 30.0 7.0 XI1= Elevation (feet from Loon Lake) 212.89 63.02 310.0 130.0 X12= Size of pollen crop (-'5 classes, 0-4) 2.20 1.46 4.0 0.0 X13= Number of female flowers 503.10 613.84 243.0 6.0 X14= Average flushing time (days after f i r s t of May) 34.27 5.79 52.0 18.0 X15= Duration of flushing (days) 57.80 5.63 72.0 44.0 XI6= Age (years) 19.66 6.69 34.0 12.0 XI7= Colour of female flowers (5 classes, 1-5) 2.79 1.10 5.0 1.0 XI8= Crown density (5 classes, 1-5) 3.26 1.01 5.0 1.0 X19= Percentage of f i l l e d seeds per cone (arcsin sq.rt.) 14.80 4.88 27.1 3.9 dependent variables are described i n Table 23. Ninety-three trees were analysed i n 1961 and 97 i n 1962. Tables 24 and 25 summarize the simple c o r r e l a t i o n c o e f f i c i e n t s and multiple regression c o e f f i c i e n t s f o r 1961 and 1962, re s p e c t i v e l y . The c o e f f i c i e n t s of determination of the independent v a r i a b l e s are given i n Table 26. Analysis of D i o r y c t r i a a b i e t e l l a damage: In 1961 a p o s i t i v e s i g n i f i c a n t c o r r e l a t i o n was found between the number of infested cones per tree and the average cone width (r = 0.236) Two other factors were correlated with the damage of D i o r y c t r i a , the duration of vegetative bud f l u s h i n g (r = -0.245) s i g n i f i c a n t l y , and the per cent of f i l l e d seeds highly s i g n i f i c a n t l y (r =-0.371). With the multiple regression using 19 independent v a r i a b l e s 31.08 per cent of the v a r i a t i o n was accounted f o r (Table 26), This i s contributed mostly by three fa c t o r s , per cent of f i l l e d seeds, cone width and duration of vegetative bud f l u s h i n g . By the process of elimination the percentage of f i l l e d seeds was found to be the most important. The r e l a t i o n s h i p t h i s way does not s i g n i f y that the damage i s the dependent v a r i a b l e nor that the percent-age of f i l l e d seed per tree i s the independent v a r i a b l e . Therefore t h i s v a r i a b l e was not used to set up an equation to determine the number of cones damaged by D i o r y c t r i a . The percentage of f i l l e d seeds was affec t e d by the insect rather than vice versa. However, including t h i s v a r i a b l e and the other two most important variables a highly s i g n i f icant equation can be set up to calculate the number of cones damaged by D i o r y c t r i a (Y). (1) Y = 2.422 + 0.971 x 9 - 0.051 x 15 - 0.068 x 19 Table 23. Description of dependent vari a b l e s . Variables Standard Ranges in 1961 Average deviation Maximum Minimum Yl = Number of cones damaged by D i o r y c t r i a 1.31 1.06 4.50 0.00 Y2 = Number of cones damaged by Megastigmus 2.05 1.59 5.75 0.00 Y3 = Number of cones damaged by C.oregonensis 4.64 0.99 6.00 1.25 Y4 = Number of cones not damaged 0.44 0.54 3.00 0.00 In 1962 Yl = Number of cones damaged by D i o r y c t r i a 0.34 0.46 2.50 0.00 Y2 = Number of cones damaged by Megastigmus 2.08 1.41 5.75 0.00 Y3 = Number of cones damaged by C.oregonensis 4.37 1.30 6.00 2.50 Y4 = Number of cones .damaged by C.washington-ensis 2.30 1.40 6.00 0.00 Y5 = Number of cones not damaged 0.49 0.75 4.00 0.00 Table 24. Correlation and regression c o e f f i c i e n t s of the analysis of 1961 cones. Simple c o r r e l a t i o n c o e f f i c i e n t s P a r t i a l regression c o e f f i c i e n t s Variable for Yl-4 on XI-19 for estimation of Yl-4 from XI-19 Yl Y2 Y3 Y4 Yl Y2 Y3 Y4 XI -0.040 0.018 0.092 -0.086 0.001 0.000 0.007 -0.000 X2 -0.064 -0.091 0.249* -0.232* -0.018 0.022 0.005 -0.016 X3 -0.075 -0.186 0.148 -0.088 -0.076 -0.075 -0.001 -0.008 X4 0.154 -0.087 -0.146 0.068 0.013 0.011 -0.056 0.002 X5 0.034 0.059 -0.098 -0.059 0.024 0.038 -0.032 -0.065 X6 -0.138 0.018 0.004 -0.013 0.000 -0.001 0.000 0.000 X7 0.029 -0.071 0.099 -0.037 0.047 0.016 0.033 -0.001 X8 0.044 0.041 0.118 -0.145 -0.193 0.142 -0.011 -0/.014 X9 0.236* -0.226* 0.211* -0.176 1.362 -0.047 0.611 -0.216 XI0 -0.084 0.165 -0.050 0.023 -0.012 0.009 -0.020 0.006 XI1 -0.064 0.096 -0.226* 0.301** -0.000 -0.001 -0.001 0.002 XI2 -0.018 -0.262* -0.016 0.072 -0.029 0.013 -0.108 0.004 XI3 -0.163 -0.094 -0.086 0.096 -0.078 -0.007 -0.069 0.021 X14 -0.063 -0.020 0.164 -0.127 -0.012 0.006 0.025 -0.014 X15 -0.245* 0.021 -0.001 0.140 -0.048 0.004 -0.004 0.017 XI6 -0.047 0.007 0.131 -0.110 0.015 0.008 0.036 -0.009 XI7 -0.089 0.012 0.122 -0.104 -0.067 0.042 0.167 -0.029 XI8 0.004 0.044 -0.157 0.153 0.024 0.062 -0.154 0.073 XI9 -0.371** • 0.898** -0.172 -0.162 -0.072 0.274 -0.029 -0.025 Table 25. Correlation and regression c o e f f i c i e n t s of the analysis of 1962 cones. Correlation c o e f f i c i e n t s Regression c o e f f i c i e n t s Variable for Yl-5 on XI -19 for estimation of Yl-5 from X l - 19 Yl Y2 Y3 Y4 - Y5 Yl Y2 Y3 Y4 Y5 XI -0.205* 0.077 0.002 -0.136 0.052 0.000 0.005 -0.015 -0.003 0.007 X2 -0.006 -0.068 0.311** -0.045 -0.163 -0.018 -0.004 0.086 -0.048 -0.031 X3 0.086 -0.013 0.303** 0.021 -0.131 -0.006 0.045 0.079 -0.004 -0.023 X4 -0.115 0.059 0.040 -0.099 -0.047 -0.027 0.033 -0.026 0.093 0.009 X5 -0.171 0.252* -0.078 0.072 0.050 -0.041 0.070 0.072 -0.037 -0.029 X6 -0.064 0.082 -0.054 0.010 0.063 0.000 0.002 0.001 0.001 -0.001 X7 0.173 -0.119 0.220* 0.147 -0.103 0.027 0.002 -0.041 -0.081 0.017 X8 0.009 -0.263** 0.012 0.217* -0.049 -0.039 -0.421 0.062 -0.026 -0.023 X9 0.129 -0.346** 0.206* 0.380** -0.192 0.061 -0.530 0.658 2.246 -0.369 XI0 0.070 -0.169 0.333** 0.112 -0.181 0.002 0.053 0.059 -0.006 -0.026 XI1 0.007 0.191 0.003 -0.047 0.088 -0.001 0.005 -0.000 -0.000 0.002 XI2 -0.061 -0.022 0.076 -0.146 -0.039 -0.036 0.097 0.007 -0.274 -0.035 XI3 -0.049 0.037 -0.081 0.012 0.089 -0.000 -0.001 -0.001 0.000 0.001 XI4 -0.069 -0.267** 0.105 0.093 0.012 -0.016 -0.039 -0.028 0.029 0.040 XI5 -0.064 -0.257* 0.051 0.044 -0.025 -0.006 0.008 -0.011 -0.024 -0.007 XI6 0.185 -0.118 0.204* 0.058 -0.149 0.017 0.027 -0.069 -0.002 0.015 XI7 0.028 -0.228* 0.160 0.241* -0.032 0.007 -0.162 0.162 0.256 -0.027 XI8 -0.038 -0.083 0.248* -0.043 -0.051 -0.055 -0.207 0.172 -0.019 0.017 XI9 -0.277** 0.639** -0.392** -0.266** 0.192 -0.027 0.204 -0.067 -0.070 0.018 Table 26. Coe f f i c i e n t s of determination for 1961 and 1962. Dependent variables  Independent 1961 1962 variables Yl Y2 Y3 Y4 Yl Y2 Y3 Y4 Y5 XI -0.14 0. 01 1.74 0.14 -0.22 0.78 -0.07 0.82 1. 26 X2 0.79 -0. 90 0.87 4.86 0.18 0.14 15.70 1.17 5. 04 X3 1.36 2. 22 2.07 0.33 0.28 -0.84 4.77 -0.01 1. 03 X4 0.46 -0. 14 0.40 0.04 1.92 0.39 -0.22 1.35 -0. 16 X5 0.09 0. 18 -0.04 0.90 2.17 1.78 -0.06 0.37 -0. 27 X6 -0.13 -0. 20 0.00 -0.06 -0.04 5.02 -2.67 0.21 -3. 44 X7 0.46 -0. 26 1.18 0.01 3.94 -0.06 -2.60 3.21 -0. 88 X8 -0.66 0. 30 -0.11 0.30 -0.05 5.12 0.03 -0.26 0. 09 X9 6.60 3. 24 2.81 1.53 0.38 2.86 2.31 13.45 2. 06 XI0 0.43 0. 42 0.47 0.10 0.14 -3.34 8.08 -0.27 3. 30 XI1 0.12 -0. 51 1.74 8.31 -0.07 3.99 0.00 0.03 1. 48 XI2 0.05 -0. 22 0.18 0.50 0.69 -0.22 0.05 4.14 0. 26 X13 2.16 0. 07 1.08 0.67 0.19 -2.31 4.85 0.15 5. 92 X14 0.41 -0. 04 2.46 1.93 1.41 4.27 -1.33 1.12 0. 35 XI5 5.99 0. 02 0.00 2.39 0.48 -0.85 -0.23 -0.43 0. 13 XI6 -0.42 0. 02 3.15 1.22 4.73 -11.53 -7.28 -0.05 -1. 98 XI7 0.58 0. 00 1.49 0.58 0.04 2.79 2.13 4.70 0. 12 XI8 0.00 0. 01 2.38 2.00 0.46 1.23 3.32 0.06 -0. 11 XI9 12.88 79. 22 2.56 3.87 7.97 41.59 9.87 6.55 2. 22 Totals 31.08 83. 46 24.48 29.73 24.63 64.16 36.10 36.34 16. 43 The equation has a standard error of estimate of 0.945 cones and removes 23.3 per cent of the t o t a l v a r i a t i o n of the dependent v a r i a b l e . In 1962 the damage by D i o r y c t r i a was much lower (0.34 cones per tree) than i n 1961 (1.31 cones per t r e e ) . Probably t h i s i s the reason why the 1962 analyses gave d i f f e r e n t r e s u l t s from the 1961. In t h i s analysis the s i t e index and the per cent of f i l l e d seeds were negatively correlated with the damage. By the multiple regression equation with 14 v a r i a b l e s only 23.63 per cent of the v a r i a t i o n was removed. Because of the poor c o r r e l a t i o n s and the very low damage, no equation was set up to determine the 1962 damage of D i o r y c t r i a . Analysis of Megastigmus spermotrophus damage; In analyses of the 1961 data i t was found that cone width (r = -0.226) and size of pollen crop (r = -0.262) were negatively correlated with the number of cones damaged out of s i x by Megastigmus (Table 24). A highly s i g n i f i c a n t p o s i t i v e c o r r e l a t i o n was found between the damage by Megastigmus and the per cent of f i l l e d seeds per cone per tree. The l i n e a r multiple regression equation removed 83.46 per cent of the v a r i a t i o n of the dependent v a r i a b l e . The most important variables were percentage of f i l l e d seeds and cone width, r e s p e c t i v e l y . In 1962, cone length (r = -0.263), cone with (r = -0.346), average f l u s h i n g time of ..vegetative .buds " (r =-0.267), duration of vegetative bud f l u s h i n g (r = -0.257) and colour of female conelets (r = -0.228) a l l were negatively correlated with the amount of damage by Megastigmus. The percentage of f i l l e d seeds per cone per tree (r = 0.639) was p o s i t i v e l y correlated with the damage. The more s i g n i f i c a n t c o r r e l a t i o n i n 1962 may indicate that i n a better seed 66 year the insects are not depending on the trees with good seed crops as heavily as i n a f a i r l y poor seed year l i k e 1961. Thus i n a good seed year more of the tree factors besides the percentage of f i l l e d seeds are a f f e c t i n g the i n t e n s i t y of attack than i n a poor year. In 1962 the l i n e a r multiple regression equation removed 64.16 per cent of t h e ; t o t a l v a r i a t i o n of Megastigmus damage (Table 26). Percentage of f i l l e d seeds and cone length were found to be the most important v a r i a b l e s . In both years the percentage of f i l l e d seeds per tree per cone was found to be the most important v a r i a b l e . Using t h i s v a r i a b l e , the damage can be determined as: (2) Y = -0.775 + 0.279 x 19 i n 1961. SEE = 0.700 cones out of s i x . (3) Y = -0.659 + 0.185 x 19 i n 1962. SEE = 1.092 cones out of s i x . where Y = number of damaged cones out of s i x . X19 = a r c s i n square root percentage of f i l l e d seeds per cone per seed. The slopes of the two equations are s i g n i f i c a n t l y d i f f e r e n t , i n d i c a t i n g that the increase of damage as the percentage of f i l l e d seeds increases isi-much slower in. a •good;.seed year (1962) than i n a poor one (1961). Analyses of Contarinia oregonensis damage: Analysis of the 1961 data found a s i g n i f i c a n t p o s i t i v e c o r r e l a t i o n between the number of cones damaged by Contarinia oregonensis and tree height (r = 0.249) and cone width (r = 0.211). Negative s i g n i f i c a n t 67 c o r r e l a t i o n was found for elevation (r = 0.226). The l i n e a r multiple regression equation with 19 variables removed 24.48 per cent of the t o t a l v a r i a t i o n of the dependent v a r i a b l e . Tree height was found to be the most important v a r i a b l e , though i t was not highly s i g n i f i c a n t l y correlated with the dependent v a r i a b l e . Two variables, tree height and cone width together were highly s i g n i f i c a n t l y correlated with the number of cones damaged by Contarinia oregonensis (r = 0.322), and the equation was: (4) Y = 1.268 + 0.034 x 9 + 0.942 x 2 SEE = 0.955 cones out of s i x With t h i s equation and measurement of the average cone width (X9) and height (X2) of a tree, the number of cones out of s i x damaged by Contarinia can be estimated. In 1962, tree height (r = 0.311), D.b.h. (r = -.303), cone width (r = 0.206), date cones became pendent (r = 0.333), age (r = 0.204), crown density (r = 0.248) and percentage of f i l l e d seeds (r = -0.392) were s i g n i f i c a n t l y or highly s i g n i f i c a n t l y correlated with the cones damaged by Contarinia oregonensis. The l i n e a r multiple regression equation with 19 variables removed 36.10 per cent of the t o t a l v a r i a t i o n of damage. The percentage of f i l l e d seeds, tree height and the date cones became pendent contributed the most to variance. Since the percentage of f i l l e d seeds i s not l i k e l y determining the damage of Contarinia but the damage a f f e c t i n g the amount of f i l l e d seeds, the tree height was used to set up the equation to determine 68 the damage: (5) Y = 2.052 + 0.053 x 2 SEE = 1.232 cones out of s i x where: Y = number of damaged cones out of s i x by Contarinia oregonensis. X2 = tree height i n feet. A s i m i l a r equation can be set up from the 1961 data: (6) Y = 3.158 + 0.035 x 2 SEE = 0.972 cones out of s i x . Both the intercepts and slopes of the two equations (5 and 6) are s i g n i f i c a n t l y d i f f e r e n t . Analysis of undamaged cones: Two of the 19 variables were s i g n i f i c a n t l y correlated with the number of undamaged cones, the tree height (r = -0.232) negatively and the elevation (r = 0.301) p o s i t i v e l y . Since the damage of Contarinia oregonensis was the most dominant, t h i s resulted these two s i g n i f i c a n t c o r r e l a t i o n s . With 19 va r i a b l e s 29.73 per cent of the v a r i a t i o n of damage was removed by the l i n e a r multiple regression. In the elimination elevation was found to be the most important v a r i a b l e . This dependent v a r i a b l e does not give a useful equation, because the difference between the lowest and highest elevation was 180 feet only. None of the 19 variables was s i g n i f i c a n t l y correlated with the number of undamaged cones i n 1962, which indicates that the changes of d i f f e r e n t insect species are balanced out on the trees. The l i n e a r multiple regression explained 16.43 perccent of the v a r i a t i o n of the damage. Analysis of Contarinia washingtonensis damage: Only one year of observations was av a i l a b l e f o r t h i s a n a l y s i s . The cone length (r = 0.217), cone width (r = 0.380), colour of female flowers (r = 0.241) and percentage of f i l l e d seeds (r =-0.266) were s i g n i f i c a n t l y correlated with the number of cones damaged out of s i x by Contarinia washingtonensis i n 1962. With 19 variables 36.34 per cent of the v a r i a t i o n of damage was removed by the l i n e a r multiple regression equation. In the process of elimination cone length, percentage of f i l l e d seeds and the colour of conelets were the three most important v a r i a b l e s . Using the most important v a r i a b l e the damage can be described as: (7) Y = -2.787 + 2.410 x 9 SEE = 1.300 damaged cones out of s i x . where: Y = number of damaged cones out of s i x by Contarinia  washingtonensis. X19 = average cone length of the tree i n cm. Both the regression and c o r r e l a t i o n are highly s i g n i f i c a n t f o r equation 7. The equation removed 14.44 per cent of the v a r i a t i o n . Relationships among the dependent v a r i a b l e s : Table 27 and 28 summarize the simple c o r r e l a t i o n c o e f f i c i e n t s of the dependent variables of the two years. In 1961 i t was found that the number of cones damaged by D i o r y c t r i a was highly s i g n i f i c a n t l y c orrelated with the damage of Megastigmus (r = -0.409) and with the Table 27. Relationships among the dependent variables i n 1961. Simple C o r r e l a t i o n c o e f f i c i e n t s Yl Y2 Y3 Y4 Yl 1.000 -0.409** 0.043 -0.265** Y2 -0.409** 1.000 -0.129 -0.179 Y3 0.043 -0.128 1.000 .-0.558** Y4 -0.265** -0.179 -0.558** . 1 .000 Table 28. Relationshi ps among the dependent var i a b l e s i n 1962. Simple Co r r e l a t i o n c o e f f i c i e n t s Y l Y2 Y3 Y4 Y5 Yl 1.000 -0.232* 0.306** 0.122 0.295** Y2 -0.232** 1.000 -0.057 -0.201* 0.092 Y3 0.306** -0.057 1.000 0.160 0.776** Y4 0.122 -0.201* 0.160 1.000 0.442** Y5 -0.295** -0.092 -0.776* -0.442** 1.000 71 number of undamaged cones (r = -0.265). No c o r r e l a t i o n was found between the damage of Contarinia oregonensis and Megastigmus. The damage of Contarinia oregonensis was highly s i g n i f i c a n t l y correlated with the number of undamaged cones (r = -0.558). The damage of Megastigmus was the only one which had no s i g n i f i c a n t c o r r e l a t i o n with the number of undamaged cones. In 1962 the number of cones damaged by D i o r y c t r i a was negatively co r r e l a t e d with the damage of Megastigmus (r = -0.232), and with the undamaged cones (r = -0.295). Surpr i s i n g l y , p o s i t i v e and highly s i g n i f i c a n t c o r r e l a t i o n was found between the D i o r y c t r i a damage and the number of damaged cones by Contarinia oregonensis (r = 0.306). The r e l a t i o n s h i p was s i g n i f i c a n t and negative between the damage of Megastigmus and Contarinia washingtonensis (r = -0.201). The number of cones damaged by Contarinia oregonensis was highly s i g n i f i c a n t l y and negatively correlated with the undamaged cones (r = -0.776). A s i m i l a r r e l a t i o n s h i p was found between the Contarinia washingtonensis damage and the number of undamaged cones (r = -0.442). 2, Analysis of Amount of Seeds Damaged S i m i l a r l y to the analyses of number of damaged cones out of six, the percentage of infested seeds per cut surface of the cone was analysed for the 1961 and 1962 observations. Five separate analyses were run i n each case: the per cent of seeds damaged by Contarinia  oregonensis, Megastigmus spermotrophus (Megastigmus 1) and D i o r y c t r i a  a b i e t e l l a , the percentage of f i l l e d seeds damaged by Megastigmus  (Megastigmus 2) and the per cent of undamaged seeds. For a l l of the analyses the transformation of binomial percentages to the a r c s i n 72 square root percentage was used to normalize the d i s t r i b u t i o n . In the tables, summarizing the r e s u l t s , the averages are given in per-centages but the variances i n the analyses of variance tables are given i n the transformed form. Trees with three crown lev e l s sampled: Twenty trees which were sampled from three crown l e v e l s and from outside and inside crown (two crown positions) were analysed for the 1961 data (Tables 29 and 30). The analysis of Contarinia oregonensis showed s i g n i f i c a n t v a r i a t i o n for crown l e v e l and f o r tree by crown l e v e l i n t e r a c t i o n . Highly s i g n i f i c a n t v a r i a t i o n was found between trees. Table 30 shows that there was a s l i g h t increase i n the percentage of damaged seeds from the lower crown l e v e l (12.1 per cent) to the middle one (12.7 per cent), which i s not s i g n i f i c a n t . The upper crown l e v e l (15.2 per cent) was s i g n i f i c a n t l y more heavily i n f e s t e d . The highly s i g n i f i c a n t tree v a r i a t i o n i n damage w i l l be discussed l a t e r f o r each insect species. The s i g n i f i c a n t tree by l e v e l i n t e r a c t i o n indicates that even i f the general averages for crown l e v e l showed an increasing trend of i n f e s t -ation from the bottom to the top of the crown, the trend i s not the same for every tree. Two highly s i g n i f i c a n t F values were found i n the analysis of the percentage of seeds infested by Megastigmus. one for crown l e v e l and the other for tree v a r i a t i o n . No s i g n i f i c a n t difference was found between the lower (2.2 per cent) and upper crown lev e l s (2.1 per cent), but the damage i n the middle l e v e l (1.6 per cent) was s i g n i f i c a n t l y Table 29. Analyses of variance of damaged and undamaged seeds i n 1961 (trees with three crown levels sampled). Damaged by Undamaged Source of C. oregonensis Megastigmus 1 Megastigmus 2 D i o r y c t r i a seeds v a r i a t i o n DF MS F MS F MS F MS F MS F Crown Level (L) 2 73.41 3.83* 208.70 15.46** 4955.20 5.28** 17.66 - 22.64 -Posi t i o n (P) 1 0.32 - 17.50 1.30 304.00 - 5.75 - 8.96 -Tree (T) 19 177.17 9.25** 57.5 4.26** 2557.80 2.73** 266.03 3.96**195.90 7.39** L x P 2 33.91 1.77 17.1 1.27 1000.90 1.69 18.39 - 4.32 -T x L 38 38.85 2.03* 6.8 - 328.80 - 70.75 1.05 26.39 -P x T 19 24.19 1.26 13.6 1.01 492.10 - 45.80 - 26.86 1.01 Residual 38 19.15 13.5 938.20 67.22 26.51 Total 119 Megastigmus 1 Megastigmus 2 = percentage of seeds damaged by Megastigmus. = percentage of f i l l e d seeds damaged by Megastigmus Table 30. Percentages of damaged and undamaged seeds i n 1961 (summary of analyses of variance of Table 29). Crown A l l Range s of Level Position over tree average v a r i a t i o n Damage by Lower Middle Upper Outside Inside Maximum Minimum Percentages C. oregonensis 12.1 12.7 15.2 13.4 13.2 13.3 25.4 4.2 Megastigmus 1 2.2 1.6 2.1 1.7 2.3 2.0 14.0 0.0 Megastigmus 2 67.3 66.5 59.1 56.7 71.7 64.4 100.0 0.0 D i o r y c t r i a 3.7 3.5 2.9 3.5 3.2 3.4 15.1 0.0 Undamaged seeds 77.5 78.2 76.1 77.6 76.9 77.3 89.0 56.8 less than any of the other two. The percentage of f i l l e d seeds damaged by Megastigmus gave exactly the same r e s u l t s as the percentage of t o t a l number of seeds damaged by the same species. However, the v a r i a t i o n i n l e v e l s was d i f f e r e n t i n t h i s case, as the i n f e s t a t i o n was decreasing from the bottom of the crown to the top. No s i g n i f i c a n t difference was found between the lower (67.3 per cent) and middle crown le v e l s (66.5 per cent). The damage i n the upper crown l e v e l (59.1 per cent) was s i g n i f i c a n t l y 'l'qwer. than i n the other two l e v e l s . Only the tree to tree v a r i a t i o n resulted i n a highly s i g n i f i c a n t variance r a t i o i n the analysis of D i o r y c t r i a damage. The analysis of percentage of undamaged seeds resulted a highly s i g n i f i c a n t tree to tree v a r i a t i o n . This r e s u l t w i l l be discussed l a t e r . It was possible to sample 45 trees from the three crown l e v e l s and from outside and inside crown i n 1962. The analyses of these trees are summarized i n Tables 31 and 32. The r e s u l t of the analysis of Contarinia oregonensis damage was f a i r l y s i m i l a r to the 1961 an a l y s i s . A l l of the fac t o r s , crown l e v e l , p o s i t i o n and tree resulted highly s i g n i f i c a n t F values. P o s i t i o n i n crown was not s i g n i f i c a n t i n 1961. Two of the i n t e r a c t i o n s , tree by crown l e v e l and tree by crown po s i t i o n , were also highly s i g n i f i c a n t . As Table 32 indicates the damage showed an increasing trend from the bottom of the crown to the top. The percentage of damaged seeds was s i g n i f i c a n t l y greater i n the middle crown l e v e l (11.3 per cent) than i n the lower one (8.5 per cent), and again s i g n i f i c a n t l y more i n the upper crown l e v e l (15.8 per cent) than i n the middle one. The Table 31. Analyses of variance of damaged and undamaged seeds i n 1962 (trees with three crown levels sampled). Damage by Undamaged C. oregonensis Megastigmus 1 Megastigmus 2 D i o r y c t r i a cones Source of v a r i a t i o n . DF MS F MS F MS F MS F MS F Crown Level (L) 2 949.83 50.70** 70.51 4.01* 1501.49 3.25* 0.37 925.95 45.44** Pos i t i o n (P) 1 436.56 23.30** 27.57 1.57 323.63 - 52.83 3.40 341.70 16.77** Tree (T) 44 494.30 26.83** 125.00 7.10** 1085.46 3.91**34.44 2.22** 511.85 25.12** L x P 2 43.18 2.30 5.50 - 95.21 - 53.46 3.44 38.85 1.91 T x L 88 36.16 1.93** 19.06 1.08 378.11 - 21.03 1.35 37.69 1.85** P x T 44 ,50.38 2.69** 17.87 1.02 311.38 - 19.18 1.23 41.05 2.01** Residual 88 18.74 17.60 461.49 15.54 20.37 Total 269 Table 32. Percentages of damaged and undamaged seeds i n 1962 (summary of analyses of variance of Table 31). Crown A l l Ranges of Level Position over tree averages v a r i a t i o n Damage by Lower Middle Upper Outside Inside Maximum Minimum Percentages C. oregonensis 8.5 11.3 15.8 10.3 13.2 11.7 54.6 1.0 Megastigmus 1 1.7 2.5 2.6 2.4 2.1 2.2 9.0 0.1 Megastigmus 2 24.3 36.0 35.2 29.9 33.5 31.7 96.1 0.4 D i o r y c t r i a 0.1 0.1 0.1 0.1 0.1 0.1 3.2 0.0 Undamaged seeds 90.0 87.2 82.4 88.0 85.3 86.6 98.1 40.0 outside crown (10.3 per cent) was s i g n i f i c a n t l y more l i g h t l y infested than the inside crown (13.2 per cent). Both of the s i g n i f i c a n t i n t e r -actions showed that even i f there i s a general trend of the i n f e s t a t i o n from the bottom of the crown to the top and from outside crown to inside, there are a number of trees on which t h i s trend d i f f e r s from the average. The analysis of percentage of damaged seeds by Megastigmus showed s i g n i f i c a n t v a r i a t i o n from l e v e l to l e v e l i n the crown and highly s i g n i f i c a n t tree v a r i a t i o n . The damage i n the lower t h i r d of the crown (1.7 per cent) was s i g n i f i c a n t l y less than i n the middle one (2.5 per cent). No s i g n i f i c a n t difference was found between the upper (2.6 per cent) and the middle crown l e v e l . The percentage of f i l l e d seeds damaged by Megastigmus resulted i n a s i g n i f i c a n t -F value for crown l e v e l and a highly s i g n i f i c a n t F--value for tree v a r i a t i o n . Duncan's new multiple range test showed that there i s no difference between the middle t h i r d of the crown (36.0 per cent), but the i n f e s t a t i o n i n the lower t h i r d (24.3 per cent) was s i g n i f i c a n t l y less than i n any of the other two. The percentage of seeds infested by D i o r y c t r i a varied highly s i g n i f i c a n t l y from tree to tree, but no s i g n i f i c a n t v a r i a t i o n was found within the crown of the trees. The analysis of percentage of undamaged seeds gave highly s i g n i f i c a n t variance r a t i o s for crown l e v e l , tree, crown p o s i t i o n , and for the interactions of tree by l e v e l and tree by crown p o s i t i o n . The order of means i n t h i s analysis i s exactly the reverse of the analysis of Contarinia oregonensis damage (Table 32). This i s quite understand-able since the damage of t h i s insect was the most dominant (11.7 per cent versus 2.2 per cent of Megastigmus and 0.1 per cent of D i o r y c t r i a ) . Trees with two crown lev e l s sampled: In 1961, 75 trees were sampled from two crown l e v e l s (the middle and upper t h i r d of the crown) and from two crown po s i t i o n s . The analyses of these trees are summarized i n Tables 33 and 34. Since the r e s u l t s are very s i m i l a r to those of Tables 29 and 30, only the differences between the two sets of analyses (trees with three and trees with two l e v e l s sampled) w i l l be discussed. The only difference found i n the analysis of Contarinia oregonensis i s that the v a r i a t i o n of crown l e v e l became highly s i g n i f i c a n t instead of s i g n i f i c a n t . In the two Megastigmus analyses the crown l e v e l did not show s i g n i f i c a n t v a r i a t i o n , which was s i g n i f i c a n t with three crown l e v e l s , although the trend of i n f e s t a t i o n was s i m i l a r (Tables 30 and 34). The analysis of percentage of f i l l e d seeds damaged by Megastigmus resulted i n a s i g n i f i c a n t F value for crown po s i t i o n , which was not s i g n i f i c a n t i n the analysis of trees with three crown l e v e l s sampled. An average of 61.4 per cent of f i l l e d seeds was damaged i n the outside crown and 72.6 per cent i n inside crown. No difference was found between the two analyses of D i o r y c t r i a  a b i e t e l l a damage. In the analysis of undamaged seeds with two crown l e v e l s , the crown l e v e l became highly s i g n i f i c a n t and the l e v e l by tree i n t e r a c t i o n s i g n i f i c a n t . This r e s u l t again i s exactly the opposite of the r e s u l t of analysis of Contarinia oregonensis damage as i t was discussed before. Table 33. Analyses of variance of damaged and undamaged seeds i n 1961 (trees with two crown levels sampled). Damage by ; Undamaged C. oregonensis Megastigmus 1 Megastigmus 2 D i o r y c t r i a cones Source of v a r i a t i o n DF MS F MS F MS F MS F MS F Crown Level (L) 1 294.25 9. 59** 11.03 - 1133.43 1.77 84.18 1. 45 346.60 9. 75** P o s i t i o n (P) 1 7.48 25.41 1.82 .35.16.5.0 5.48* 41.11 1.50 -Tree (T) 74 183.68 5. 99** 125.27 8.95** 3069.83 4.79** 209.55 3. 61** 228.59 6. 43** L x P 1 42.23 1. 38 19.13 1.37 473.00 - 0.99 25.00 -T x L 74 48.67 1. 59* 18.46 1.32 681.88 1.06 51.80 60.29 1. 70* P x T 74 43.55 1. 42 13.46 - 537.12 - 65.04 1. 12 44.23 1. 24 Residual 74 30.39 13.99 641.40 58.03 35.53 Tota l 299 00 o Table 34. Percentages of damaged and undamaged seeds i n 1961 (summary of analyses of variance of Table 33). Crown A l l Ranges of Level Position over tree Damage by Middle Upper Outside Inside average v a r i a t i o n Maximum Minimum Percentages C. oregonensis 13.9 16.4 14.9 15.3 15.1 43.4 1.4 Megastigmus 1 2.0 2.1 1.9 2.2 2.1 14.3 0.0 Megastigmus 2 70.2 63.8 61.4 72.6 67.1 100.0 0.0 D i o r y c t r i a 2.8 3.5 3.4 3.0 3.2 22.7 0.0 Undamaged seeds 79.1 76.1 77.7 77.5 77.6 94.6 37.6 82 Eighty-two trees were sampled from two crown le v e l s and from two crown positions i n 1962. The analyses of.these data are summarized i n Tables 35 and 36. As for 1961, only the differences w i l l be discussed here. No major difference was found between the two sets of analyses of Contarinia oregonensis. The only difference which was found i s that the tree by l e v e l i n t e r a c t i o n was just s i g n i f i c a n t i n the analysis of trees with two crown l e v e l s instead of highly s i g n i f i c a n t . In the analyses of Megastigmus 1 and Megastigmus 2, the l e v e l v a r i a t i o n was not s i g n i f i c a n t for the trees with two crown l e v e l s . The rest of the r e s u l t s were i d e n t i c a l . No difference was found between the r e s u l t s of the two sets of analyses of D i o r y c t r i a a b i e t e l l a . Only one change can be observed i n the analysis of undamaged seeds, which i s that the tree by l e v e l i n t e r -action was highly s i g n i f i c a n t when two le v e l s were analysed. This i n t e r a c t i o n was s i g n i f i c a n t only i n the 0,05 p r o b a b i l i t y l e v e l i n the case of three crown l e v e l s a n a l y s i s . Analyses of year to year d i f f e r e n c e s : Forty-eight trees were found with samples from at least two crown le v e l s (middle and upper le v e l ) and from two crown positions i n both years. These trees were analysed to f i n d out the year-to-year v a r i a t i o n of the percentage of damaged seeds. Those factors l i k e crown l e v e l , crown p o s i t i o n and tree which were tested before, were included i n these analyses to remove t h e i r e f f e c t from the t o t a l v a r i a t i o n , but they w i l l not be discussed here unless they are d i f f e r e n t from before. Table 35. Analyses of variance of damaged and undamaged seeds i n 1962 (Trees with two crown l e v e l s sampled). Damage by Undamaged Source of v a r i a t i o n DF C. oregonensis Megastigmus 1 Megastigmus 2 D i o r y c t r i a Seeds MS F MS F MS F MS F MS F Crown l e v e l (L) 1 919.70 43.29** 16.81 - 120.91 0.01 - 804.10 35.11* Posi t i o n (P) 1 521.52 24.55** 12.50 - 689.25 1.16 2.80 - 520.40 22.72* Tree (T) 81 440.39 20.73** 99.37 5.79* *1411.38 2.35** 30.29 1.73** 448.20 19.57*' L x P 1 24.45 1.15 0.09 - 3.73 73.66 4.20 2.20 -T x L 81 31.88 1.50* 21.45 1.25 468.68 22.03 1.26 30.26 1.32 P x T 81 48.69 2.29** 20.04 1.17 414.01 18.39 1.05 49.20 2.15*' Residual 81 21.24 17.16 600.80 17.53 22.90 Total 327 oo Lo Table 36. Percentages of damaged and undamaged seeds i n 1962 (summary of analyses of variance of Table 35). Crown A l l Ranges of Level Position over tree average v a r i a t i o n Damage by Middle Upper Outside Inside Maximum Minimum Percentages C. oregonensis 13.6 17.8 14.0 17.3 15.6 61.1 0.2 Megastigmus 1 2.4 2.6 2.6 2.4 2.5 17.6 0.0 Megastigmus 2 36.9 38.9 35.5 40.5 38.0 85.5 0.0 D i o r y c t r i a 0.2 0.2 0.2 0.2 0.2 3.7 0.0 Undamaged seeds 84.6 80.5 84.2 81.0 82.7 98.6 36.3 85 The analyses are summarized i n Tables 37. and 38. Analysing the damage of Contarinia oregonensis i t was found that the damage of outside crown (16.3 per cent) was s i g n i f i c a n t l y less than the damage of inside crown (18.9 per cent). The tree to tree and crown l e v e l v a r i a t i o n s were highly s i g n i f i c a n t . No s i g n i f i c a n t d i f f e r -ence was observed between the damage i n 1961 and 1962 (17.1 and 17.4 per cent, r e s p e c t i v e l y ) . Highly s i g n i f i c a n t v a r i a t i o n was found f o r the J i n t e r a c t i o n s of crown p o s i t i o n by tree, tree by year and tree by l e v e l , and s i g n i f i c a n t i n t e r a c t i o n f o r crown p o s i t i o n by year. The P x T and T x L inte r a c t i o n s were discussed already i n the analyses of 1961 and 1962 data. The crown p o s i t i o n by year i n t e r a c t i o n was s i g n i f i c a n t because of the non s i g n i f i c a n t v a r i a t i o n of crown p o s i t i o n i n 1961 became highly s i g n i f i c a n t i n 1962 (Tables 29, 31, 33 and 35). The highly s i g n i f i c a n t tree by year i n t e r a c t i o n w i l l be discussed l a t e r for each i n s e c t . The analysis of percentage of seeds damaged by Megastigmus gave a highly s i g n i f i c a n t variance r a t i o for tree v a r i a t i o n and a s i g n i f i c a n t r a t i o for tree-by-year i n t e r a c t i o n . No s i g n i f i c a n t difference was found between the damage i n 1961 (2.2 per cent) and i n 1962 (2.0 per cent). The analysis of percentage of f i l l e d seeds damaged by Megastigmus indicated highly s i g n i f i c a n t tree to tree v a r i a t i o n , and also the damage i n 1961 (70.1 per cent) was s i g n i f i c a n t l y more than the damage i n 1962 (33.5 per cent). In t h i s analysis the tree by year i n t e r a c t i o n was not s i g n i f i c a n t . The damage by D i o r y c t r i a i n 1961 (3.5 per cent) was highly s i g n i f i c a n t l y - h e a v i e r than i n 1962 (0.2 per cent). Also, the tree Table 37. Analyses of variance of damaged and undamaged seeds by tree, year, crown p o s i t i o n and l e v e l (trees with two crown levels sampled). Damage by Undamag ;ed C. oregonensis Megastigmus 1 Megastigmus 2 D i o r y c t r i a cones Source of v a r i a t i o n DF MS F MS F MS F MS F MS F P o s i t i o n (P) 1 182.65 7.90* 0.07 - - 66.09 - 18.98 - 162.80 6.29* Tree (T) 47 482.18 20.85**169.99 8.00** 3184.06 4.68** 140.90 4.34** 500.13 19.13** Year (Y) 1 6.11 - 9.50 44433.10 65.31**6814.84 210.05** 1548.40 59.86** Crown Level (L) 1 601.47 26.01** 1.61 - 1301.95 1.91 62.83 1.93 708.20 27.38** P x T 47 53.98 2.33** 17.39 - 41.02 - 43.58 1.34 65.91 2.54** P x Y 1 122.03 5.27* 27.61 1.30 477.07 - 3.50 - 30.50 1.17 P x L 1 52.08 2.25 0.92 - 5.09 - 22.25 - 28.90 1.11 T x Y 47 184.74 7.99** 42.70 2.01* 1100.80 1.61 105.60 3.25** 221.75 8.57** T x L 47 54.57 2.36** 19.40 - 505.71 - 41.41 1.27 46.91 1.81* • Y x L 1 3.35 - 25.66 1.20 1666.57 2.44 29.62 - 11.30 -P x T x Y 47 34.91 1.51 16.07 - 506.92 - 40.31 1.24 34.84 1.34 P x T x L 47 29.31 1.27 18.81 - 907.14 1.33 37.51 1.15 30.84 1.19 P x Y x L 1 4.46 - 9.22 - 16.50 - 5.83 - 49.60 1.91 T x Y x L 47 27.75 1.20 18.61 - 558.31 - 36.53 1.12 36.72 1.41 Residual 47 23.12 21.23 680.27 32.44 25.86 Total 383 Table 38. Percentages of damaged and undamaged seeds by year, crown l e v e l and p o s i t i o n and tree (summary of analyses of variance of Table 37). Crown A l l Range s of Level Position over tree Year Middle Upper Outside Inside average v a r i a t i o n Damage by 1961 1962 Maximum Minimum Percentages C. oregonensis 17.1 17.4 15.6 18.9 16.3 18.9 17.3 49.5 1.7 Megastigmus 1 2.2 2.0 2.2 2.1 2.1 2.1 2.1 8.7 0.1 Megastigmus 2 70.1 33.5 55.3 48.9 49.8 54.3 52.1 97.0 0.9 D i o r y c t r i a 3.5 0.2 1.2 1.6 1.3 1.5 1.4 7.1 0.0 Undamaged seeds 75.3 81.1 80.2 76.3 79.2 77.4 78.3 95.2 46.2 v a r i a t i o n and tree by year v a r i a t i o n were highly s i g n i f i c a n t i n the analysis of D i o r y c t r i a damage. The percentage of undamaged seeds was s i g n i f i c a n t l y less i n 1961 (75.3 per cent) than i n 1962 (81.1 per cent). The rest of the r e s u l t s of t h i s analysis i s i d e n t i c a l with the e a r l i e r analyses. Because of the highly s i g n i f i c a n t and s i g n i f i c a n t tree-by-year int e r a c t i o n s , the trees were c l a s s i f i e d as highly, moderately and heavily infested and, as i n Table 15, the minor and major changes from 1961 to 1962 were counted. The percentages of the changes are summarized i n Table 39. Table 39. Changes of i n d i v i d u a l i n f e s t a t i o n tree. class from year to year on the same Damage by Same Minor change Major change Same or minor change Per cent of trees D i o r y c t r i a 42.9 41.3 15.8 84.2 Megastigmus 1 52.4 39.7 7.9 92.1 Megastigmus 2 58.7 30.2 11.1 88.9 C. oregonensis 55.5 42.9 1.6 98.4 As Table 39 indicates, the changes of i n f e s t a t i o n class in percentage are not r e a l l y high, but they are high enough to give s i g n i f i c a n t variance r a t i o s for the year by tree i n t e r a c t i o n mainly because of the highly s i g n i f i c a n t tree v a r i a t i o n i n both years and secondly because of the changes i n i n f e s t a t i o n classes. 89 ^ Analysis of damage by d i r e c t i o n i n the crown; For those f i v e trees which were sampled from four sides of the crown for the damage of _C. oregonensis, Megastigmus 1, Megastigmus 2 and D i o r y c t r i a , the percentage of undamaged seeds was also analysed. These analyses are tabulated i n Tables 40 and 41 (for 1961 and 1962, respectively) and the mean values of the four d i r e c t i o n s are given i n Table 42c Only one insect species, Contarinia oregonensis, damaged s i g n i f -i c a n t l y d i f f e r e n t amount of seeds on d i f f e r e n t sides. This difference was found s i g n i f i c a n t i n 1962 and highly s i g n i f i c a n t i n 1961. However, the side e f f e c t was not the same i n the two years. In 1961 i t was found that the percentage of infested seeds increased s i g n i f i c a n t l y from South (7.7 per cent) to East (9.2 per cent), from East to North (11.4 per cent) and from North to West (12.6 per cent). In 1962, the i n f e s t a t i o n on West (7.9 per cent) was s i g n i f i c a n t l y l i g h t e r than on any other sides (N = 10.3, S = 10.2, E = 10.8 per cent). No s i g n i f i c a n t difference was observed between South, East and North. The s i g n i f i c a n t t r e e - b y - d i r e c t i o n i n t e r a c t i o n i n 1961 indicates that the trend of i n f e s t a t i o n from side to side was d i f f e r e n t on d i f f e r e n t trees. In analyses of the data c o l l e c t e d from a single tree from every 45 degrees of crown d i r e c t i o n i n 1962, none of the insect species damaged s i g n i f i c a n t l y d i f f e r e n t amount of seeds on the d i f f e r e n t sides (Tables 43 and 44). Analysis of between tree v a r i a t i o n : Since with a very few exceptions a l l the analyses of variance (Tables 29, 31, 33, 35, 37, 40 and 41) resulted i n a s i g n i f i c a n t or h i g h l y s i g n i f i c a n t variance r a t i o for the percentage of damaged seeds Table 40. Analyses of variance of 1962 seeds by tree, d i r e c t i o n , crown p o s i t i o n and l e v e l . Damage by Undamaged C. oregonensis Megastigmus 1 Megastigmus 2 D i o r y c t r i a cones Source of v a r i a t i o n DF MS F MS F MS F MS F MS F Tree (T) 4 83.80 7.29** 153.70 11.82* 2259.37 8.86* 71.24 4.85** 90.66 7.17* D i r e c t i o n (D) 3 48.94 4.26* 16.15 1.25 96.44 - 28.83 1.96 37.60 2.97 Po s i t i o n (P) 1 28.36 2.46 10.02 1655.44 6.49* 1.40 - 102.99 8.14* Crown Level (L) 2 350.69 30.54** 11.99 182.24 - 16.18 1.10 234.53 18.55*-T x D 12 31.08 2.70* - 355.81 1.39 20.63 1.40 21.41 1.68 T x P 4 13.06 1.13 19.11 1.48 "76;28 - 45.93 3.12* 4.83 -D x P 3 31.94 2.78 10.08 117.23 - 12.18 - 19.15 1.51 T x L 8 26.10 2.27 6.02 221.24 - 16.54 1.12 23.53 1.86 D x L 6 9.69 - 8.14 117.78 - 17.49 1.19 9.80 -P x L 2 18.45 1.60 12.91 174.10 - 0.77 - 17.44 1.37 T x D x P 12 14.60 1.27 8.02 171.40 - 15.97 1.08 10.19 -T x D x L 24 20.01 1.74 11.88 447.20 1.75 6.80 - 16.63 1.31 T x P x L 8 7.59 - 11.72 348.32 1.36 9.25 - 3.37 -D x P x L 6 12.99 1.13 12.92 601.63 2.36 11.40 - 8.81 -Residual 24 11.48 254.85 14.68 12.64 Tot a l 119 V O O Table 41. Analyses of variance of 1961 seeds by tree, d i r e c t i o n , crown p o s i t i o n and l e v e l . Damage by Undamaged C. oregonensis Megastigmus C. washingtonensis D i o r y c t r i a cones Source of v a r i a t i o n DF MS MS F MS F MS F MS F Tree (T) 4 288.80 11.24** 383.83 25. 79** 10950.60 10.09** 281.14 2C88* 426.33 15.98** Di r e c t i o n (D) 3 163.44 6.36** 30.63 2. 05 1164.92 1.07 236.77 2.42 2.56 -P o s i t i o n (P) 1 291.84 11.35** 12.08 - 510.60 - 141.34 1.44 341.67 12.81** Crown Level (L) 2 219.89 8.55** 5.14 - 103.20 - 119.91 1.22 338.83 12.70** T x D 12 40.77 1.58 11.62 - 531.37 - 132.04 1.35 66.94 2.50* T x P 4 82.86 3.22* 7.50 - 694.91 - 34.53 - 89.42 3.35* D x P 3 22.97 - 34.25 2. 30 1520.17 1.40 50.12 - 29.21 1.09 T x L 8 51.09 1.98 18.59 1. 24 593.20 - 87.74 - 53.93 2.02 D x L 6 31.72 1.23 20.29 1. 36 759.21 - 26.03 - 15.90 -P x L 2 71.33 2.77 23.04 1. 54 1349.04 1.24 35.66 - 50.03 1.87 T x D x P 12 47.89 1.86 13.20 1298.61 1.19 60.47 - 32.55 1.22 T x D x L 24 37.79 1.47 7.67 - 458.72 - 41.35 - 30.99 1.16 T x P x L 8 65.97 2.56* 16.02 1. 07 382.40 - 97.53 - 108.95 4.08** D x P x L 6 25.15 - 7.70 - 1082.48 - 48.95 - 16.40 -Residual 24 _25.69 - 14.88 1084.62 97.49 26.67 Total 119 Table 42. Percentages of damaged and undamaged seeds by d i r e c t i o n . D i r e c t i o n North West South East Average Percentages 1961 C. oregonensis 11.4 12.6 7.2 9.2 10.1 Megastigmus 1 0.8 1.8 1.3 1.2 1.2 Megastigmus 2 34.00 59.3 51.2 51.7 49.9 D i o r y c t r i a 3.2 1.4 5.6 3.2 3.2 Undamaged seeds 81.4 81.7 82.3 81.7 81.7 1962 C. oregonensis 10.3 7.9 10.2 10.8 9.8 Megastigmus 1 1.0 0.9 1.2 0.9 1.0 Megastigmus 2 27.1 26.2 20.8 24.7 24.7 D i o r y c t r i a 0.04 0.03 0.24 0.23 0.11 Undamaged seeds 87.5 89.4 86.9 86.6 87.7 V O Table 43. Analyses of variance of 1962 seeds by eight d i r e c t i o n s , crown l e v e l and p o s i t i o n . Damage by Undamaged C. oregonensis Megastigmus 1 Megastigmus 2 D i o r y c t r i a cones Source of v a r i a t i o n DF MS F MS F MS F MS F MS F Di r e c t i o n (D) 7 8.91 - 18.59 1.26 238.34 1.49 8.27 - 8.17 Pos i t i o n (P) 1 1.99 - 12.48 590.11 3.68 27.86 3.09 2.86 Crown Level (L) 2 1.95 - 30.62 2.05 147.21 - 97.57 10.82*"' * 1.57 D x P 7 42.58 2.47 2.30 81.27 - 5.95 - 38.29 2.47 D x L 14 16.09 - 10.12 266.53 1.66 4.74 - 14.43 P x L 2 17.37 1.01 11.15 29.45 - 3.77 - 15.23 Residual 14 14.87 160.38 9.01 15.45 Total 47 NO CO Table 44. Percentages of damaged and undamaged seeds by eight d i r e c t i o n s on the crown. D i r e c t i o n Damage by N NW W SW S SE E NE Percentages C. oregonensis 8.3 6.2 7.1 6.8 7.9 7.8 5.4 6.4 Megastigmus 1 0.5 0.8 0.6 0.7 1.1 1.2 1.0 0.6 Megastigmus 2 21.1 21.4 14.2 19.8 11.7 36.2 31.6 11.7 D i o r y c t r i a 0.40 0.11 0.23 0.03 0.03 0.32 0.05 0.41 Undamaged seeds 88.7 91.2 91.0 90.9 89.9 87.6 91.9 91.0 90.3 95 from tree to tree, multiple c o r r e l a t i o n and regression analyses were computed to show the factors which may a f f e c t t h i s v a r i a t i o n . Five multiple regression and c o r r e l a t i o n analyses were c a r r i e d out for each year. The percentages of damaged seeds per tree as dependent variables were analysed for Contarinia oregonensis, Megastigmus TL, Megastigmus _2, D i o r y c t r i a a b i e t e l l a and undamaged seeds. Nineteen independent variables were used i n the analyses which are described i n Table 21 for 1961 and i n Table 22 for 1962. The d e s c r i p t i o n of dependent variables i s given i n Table 45. The d e t a i l s of the analyses, the simple c o r r e l a t i o n c o e f f i c i e n t s and the multiple regression c o e f f i c i e n t s are l i s t e d i n Tables 46 and 47 for 1961 and 1962 r e s p e c t i v e l y . The c o e f f i c i e n t s of determination are given i n Table 48 for both years. Analysis of D i o r y c t r i a a b i e t e l l a damage; None of the factors out of the 19 analysed w a s ' s i g n i f i c a n t l y correlated with the percentage of damaged seeds by D i o r y c t r i a a b i e t e l l a i n 1962. This may be affected by the very low damage (0.3 per cent) i n that year. Two of the factors were s i g n i f i c a n t l y correlated with the per cent of damaged seeds i n 1961, the cone width (r = 0.229) and the duration of vegetative bud f l u s h i n g (r = -0.219). One of the factors, the percentage of f i l l e d seeds per tree was highly s i g n i f i c a n t l y correlated with the damage of D i o r y c t r i a (r .= -0.331). An amount of 30.18 per cent of the v a r i a t i o n of dependent va r i a b l e was removed by the l i n e a r multiple regression with 19 v a r i a b l e s . By Table 45. Desc r i p t i o n of dependent v a r i a b l e s . Variables Average Standard Range s In 1961 deviation Maximum Minimum Yl = Per cent of seeds damaged by Megastigmus 2. 1 0. 9 14.4 0.0 Y2 = Per cent of f i l l e d seeds damaged by Megastigmus 66. 6 20. 8 100.0 0.0 Y3 = Per cent of seeds damaged by D i o r y c t r i a 4. 4 1. 7 22.8 0.0 Y4 = Per cent of seeds damaged by C.oregonensis 16. 7 1. 5 46.7 1.4 Y5 = Per cent of undamaged seeds 76. 0 1. 8 94.9 37.6 In 1962 Yl = Per cent of seeds damaged by Megastigmus 3.3 10.7 17.6 0.0 Y2 = Per cent of f i l l e d seeds damaged by Megastigmus 31.3 10.1 100.0 0.0 Y3 = Per cent of seeds damaged by D i o r y c t r i a 0.3 0.4 9.7 0.0 Y4 = Per cent of seeds damaged by C.oregonensis 16.7 3.2 67.1 0.2 Y5 = Per cent of undamaged seeds 81.6 3.4 98.6 30.0 V O O N Table 46. C o r r e l a t i o n and regression c o e f f i c i e n t s of the analysis of damaged and undamaged seeds i n 1961. Simple c o r r e l a t i o n c o e f f i c i e n t s M u l t i p l e c o r r e l a t i o n c o e f f i c i e n t s for Yl-5 on XI-19 for Yl-5 on XI-19 Variable Y l Y2 Y3 Y4 Y5 Y l Y2 Y3 Y4 Y5 XI 0.024 -0.022 -0.041 0.043 -0.120 0.008 0.052 0.004 0.092 -0.109 X2 -0.073 -0.062 -0.036 0.208* -0.141 -0.001 -0.164 -0.010 -0.076 1.659 X3 -0.126 -0.087 -0.069 0.105 -0.029 -0.011 -0.041 -0.836 -0.237 0.742 X4 -0.092 -0.132 0.184 -0.077 -0.025 0.025 -0.652 0.195 -0.389 0.292 X5 0.059 0.198 0.123 -0.085 0.026 0.104 3.357 0.840 -0.215 0.056 X6 0.035 -0.002 -0.101 -0.107 0.153 -0.003 -0.016 0.000 -0.005 0.006 X7 -0.020 0.011 0.078 0.077 -0.069 0.099 1.009 0.511 0.391 -0.632 X8 0.007 -0.015 0.002 -0.079 0.083 0.464 1.447 -2.251 -1.510 2.598 X9 -0.268*' ''-0.241* 0.229* 0.063 -0.145 -3.748 -14.687 11.447 4.086 -8.733 X10 0.177 0.031 0.005 0.052 0.009 0:019 -0.598 0.063 -0.035 0.086 XI1 0.164 0.211* -0.078 -0.041 0.096 -0.000 0.037 -0.005 0.006 0.001 X12 -0.307*-v-0.286** 0.057 -0.089 0.058 -0.249 -3.634 0.257 -1.141 0.919 X13 -0.084 -0.047 -0.146 -0.108 0.115 -0.065 -0.327 -0.540 -0.336 0.339 X14 -0.050 -0.096 0.002 -0.207* -0.157 0.020 0.025 -0.039 0.270 -0.248 X15 -0.034 -0.116 -0.219* -0.037 0.171 -0.027 -0.418 -0.305 -0.038 0.282 X16 0.025 0.065 -0.019 0.133 -0.094 0.060 0.631 0.055 0.379 -0.480 KIVJ 0.009 0.060 -0.017 0.079 -0.050 0.149 1.413 -0.246 0.550 -0.367 X18 0.078 0.020 -0.026 -0.097 0.099 0.165 -1.961 -0.236 -1.222 1.287 X19 0.917* * 0.650** -0.331** -0.232* 0.373** 0.961 3.325 -0.425 -0.344 0.573 N O Table 47. C o r r e l a t i o n and regression c o e f f i c i e n t s of the ana l y s i s of damaged and undamaged seeds i n 1962. Simple c o r r e l a t i o n c o e f f i c i e n t s M u l t i p l e regression c o e f f i c i e n t s  f o r Yl-5 on XI-19 f o r Yl-5 on XI-19 Variable Yl Y2 Y3 Y4 Y5 Yl Y2 Y3 Y4 Y5 XI 0.027 -0.026 -0.026 -0.043. 0.041 0.067 0.008 0.011 -0.027 0.025 X2 0.040 0.077 0.040 0.326** -0.278** -0.000 0.132 -0.059 0.298 -0.287 X3 -0.057 0.075 -0.019 0.322** -0.278** -0.858 1.592 -0.490 1.116 -1.034 X4 0.045 0.140 -0.012 0.133 -0.125 0.955 0.288 0.083 -0.033 0.097 X5 0.050 0.109 -0.119 -0.080 0.068 0.117 1.230 -0.382 0.437 -0.399 X6 0.235* -0.071 0.142 -0.173 0.219* 0.004 0.023 -0.000 0.009 -0.007 X7 -0.027 0.027 0.084 0.246* -0.226* -0.251 -0.215 0.294 -0.158 0.121 X8 -0.165 -0.398** -0.093 -0.058 0.046 -6.279 -11.259 -1.844 -0.233 0.413 X9 -0.124 -0.230* 0.048 0.154 -0.170 7.891 -2.277 3.798 -1.508 0.738 XI0 -0.051 0.158 0.035 0.415** -0.397** 0.498 1.023 0.092 0.601 -0.524 XI1 -0.054 0.265** 0.058 0.045 -0.047 0.009 0.089 -0.004 -0.003 0.005 XI2 0.040 0.045 0.009 0.034 0.001 -0.495 0.731 -0.415 0.540 0.575 X13 0.222* -0.117 0.157 -0.194 0.237* 0.006 -0.025 0.003 -0.011 0.011 X14 -0.107 -0.049 -0.096 0.119 -0.092 -0.342 -0.614 -0.191 -0.461 0.479 XI5 -0.096 -0.053 -0.083 0.105 -0.088 0.171 0.404 -0.014 0.124 -0.109 XI6 -0.072 0.122 0.060 0.342** -0.297** 0.310 0.150 0.165 -0.025 0.068 XI7 -0.203 0.015 -0.097 0.168 -0.147 -3.078 1.216 -0.518 1.404 -1.254 XI8 -0.133 0.080 -0.144 0.158 -0.139 -2.689 -3.049 -0.953 -1.143 1.211 X19 0.259* 0.095 -0.069 -0.529** 0.522 1.056 1.054 -0.099 -0.785 0.823 Table 48. Coe f f i c i e n t s of determination for 1961 and 1962. Independent Dependent variables v a r i a b l e s 1961 1962 Yl Y2 Y3 Y4 Y5 Yl Y2 Y3 Y4 Y5 XI 0.10 -0.11 -0.05 5.22 4.50 1.24 -0.02 -0.13 0.31 0.27 X2 0.00 0.25 0.03 -1.56 -2.02 0.00 0.38 -0.31 7.10 5.88 X3 0.06 0.03 1.87 -0.87 -0.66 0.66 1.52 0.41 8.94 7.23 X4 -0.10 0.76 1.16 1.04 '-0.22 0.63 0.43 -0.04 -0.11 -0.33 X5 0.14 2.99 1.69 0.32 0.02 0.04 1.20 1.12 -0.47 -0.37 X6 -0.34 0.01 0.00 1.13 1.86 2.71 -4.48 -0.47 -7.86 -8.65 X7 -0.12 0.13 1.82 1.49 1.89 -0.13 -0.10 1.62 -1.40 -1.00 X8 0.04 -0.06 r0.04 1.38 2.19 3.54 14.42 1.94 0.08 0.12 X9 3.94 2.71 7.35 0.77 3:34 -1.13 0.56 0.69 -0.49 -0.26 XI0 0.27 -0.30 0.01 -0.11 0.04 -0.70 4.26 0.30 12.76 10.75 XI1 -0.01 1.83 0.3,4 -0.24 0.05 -0.16 7.29 0.26 -0.09 -0.13 XI2 1.46 3.88 0.20 1.49 0.68 -0.15 0.23 -0.09 -0.25 0.00 XI3 0.17 0.09 1.79 0.89 0.85 4.42 8.73 4.55 12.68 15.34 XI4 -0.10 -0.04 0.00 4.63 2.81 1.10 0.85 1.83 -3.05 -2.47 XI5 0.09 0.93 4.68 0.10 3.17 -0.48 -0.59 0.11 0.70 0.52 XI6 0.17 0.96 -0.09 4.62 3.63 -0.77 0.60 1.15 -0.54 -1.30 XI7 0.02 0.31 0.05 0.63 0.23 3.49 0.09 0.93 2.40 1.19 XI8. 0.22 -0.13 0.07 1.61 1.51 1.89 -1.21 2.41 -1.74 -1.67 XI9 81.39 39.06 9.27 5.68 13.26 6.99 2.39 0.57 19.37 20.31 Totals 87.43 53.36 30.18 28.26 37.19 23.21 36.58 16.89 48.34 46.18 NO NO 100 eliminating the least important variables i t was found that the percentage of f i l l e d seeds per tree was the most important v a r i a b l e . Since the c o r r e l a t i o n of t h i s v a r i a b l e with the dependent va r i a b l e was negative i t i s very l i k e l y that the insect affected the amount of f i l l e d seeds rather than vice versa. The other two most important variables were the duration of vegetative bud flushing and the cone width. The following r e l a t i o n s h i p can be given for these two v a r i a b l e s : (8) Y = -11.640 + 8.000 x 9 - 0.346 x 15 SEE = 7.3 a r c s i n square root per cent of seeds. where: Y = a r c s i n square root per cent of damaged seeds by D i o r y c t r i a . X9 = Cone width. X15 = duration of vegetative bud f l u s h i n g . The multiple c o r r e l a t i o n c o e f f i c i e n t (R) of equation 8 i s 0.333 which i s highly s i g n i f i c a n t . Analysis of Megastigmus spermotrophus damage: In 1961 the percentage of seeds damaged by Megastigmus was highly s i g n i f i c a n t l y correlated with cone width (r = -0.268), size of pollen crop (r = -0.307) and percentage of f i l l e d seeds per tree (r = 0.917). The multiple regression equation with 19 variables removed 87.43 per cent of the t o t a l v a r i a t i o n of the dependent v a r i a b l e . It was found by elimination that the most important variables were the percentage of f i l l e d seeds per tree, cone width and size of pollen crop. A useful equation can be given to estimate the damage of Megastigmus as: 101 (9) Y = 3.217 - 2.406 x 9 + 0.974 x 19 SEE = 2.2 a r c s i n square root percentage of damage by Megastigmus. where: Y ar c s i n square root percentage of damage by Megastigmus. XI9 a r c s i n square root percentage of f i l l e d seeds per tree. X9 cone width. The multiple c o r r e l a t i o n c o e f f i c i e n t of equation 9 was 0.923 which i s highly s i g n i f i c a n t . The equation using the most important v a r i a b l e alone was: For t h i s equation the c o r r e l a t i o n c o e f f i c i e n t was 0.919 which i s s i g n i f i c a n t i n the 0.01 p r o b a b i l i t y l e v e l . In 1962 the number of cones (r = 0.235), number of female flowers (r = 0.222) and the percentage of f i l l e d seeds per tree (r = 0.259) were s i g n i f i c a n t l y correlated with the percentage of seeds damaged by Megastigmus. It was found i n the elimination of the var i a b l e s that the most important variables were per cent of f i l l e d seeds per tree, number of female flowers and colour of female flowers, r e s p e c t i v e l y . The analysis resulted .in. a s i g n i f i c a n t equation to determine the damage as: (10) Y = 1.878 + 0.993 x 19 SEE = 2.2 a r c s i n square root per cent of damaged seeds by Megastigmus. (ID Y = -4.446 + 1.0112 x 19 where: SEE = 18.5 a r c s i n square root percentage of damaged seeds by Megastigmus. Y and X 19 = as i n equation 10. The c o r r e l a t i o n c o e f f i c i e n t of t h i s equation was found to be 0.259 which i s s i g n i f i c a n t at the 0.05 p r o b a b i l i t y l e v e l . There i s no s i g n i f cant difference between the slopes of equations 10 and 11, but the intercepts are highly s i g n i f i c a n t l y d i f f e r e n t . Analysis of percentage of f i l l e d seeds damaged by MegastigmusI The percentage of f i l l e d seeds damaged by Megastigmus was found to be s i g n i f i c a n t l y correlated with cone width (r = -0.241), elevation (r = 0.211), size of pol l e n crop (r = -0.286) and percentage of f i l l e d seeds per tree (r = 0.650) i n 1961. The l i n e a r multiple regression removed 53.36 per cent of the v a r i a t i o n with 19 independent v a r i a b l e s . The three most important variables were found as per cent of f i l l e d seeds, size of pollen crop and ground vegetation, though the ground vegetation was not s i g n i f i c a n t l y correlated with the dependent va r i a b l e In 1962, the cone length (r =-0.398), cone width (r = -0.230) and eleva t i o n (r = 0.265) were s i g n i f i c a n t l y or highly s i g n i f i c a n t l y c orrelated with the per cent of f i l l e d seeds. An amount of 36.58 per cent of the t o t a l variance of the dependent va r i a b l e was removed by the l i n e a r multiple regression. Cone length, elevation and date cones became pendent were found to be the three most important v a r i a b l e s i n the process of elimination. Since i t was impossible to f i n d a common important v a r i a b l e i n the two years, there i s no general equation given which would describe the percentage of damaged f i l l e d seeds by Megastigmus. However, i t can be stated that i n a poor seed year l i k e 1961 was, those trees which produce a high percentage of f i l l e d seeds are most damaged. This r e l a t i o n s h i p can be described as: 103 (12) Y = 16.628 + 3.590 x 19 SEE = 20.81 a r c s i n square root percentage of damaged f i l l e d seeds by Megastigmus. where: Y = a r c s i n square root percentage of f i l l e d seeds damaged by Megastigmus. X19 = a r c s i n square root percentage of f i l l e d seeds. The c o r r e l a t i o n c o e f f i c i e n t for th i s equation was 0.650 which i s highly s i g n i f i c a n t . In a good seed year, such as 1962, Megastigmus does not depend upon the f i l l e d seed production of the trees, since most of the trees have a f a i r amount of f i l l e d seeds. In these years the insects are affected mostly by the size of cones on the tree. This r e l a t i o n s h i p can be written as: (13) Y = 104.907 - 12.362 x 8 SEE = 18.79 a r c s i n square root percentage of damaged f i l l e d seeds by Megastigmus. where: Y = as i n equation 12 X8 = cone length i n cm. The c o r r e l a t i o n c o e f f i c i e n t of the equation was -0.398 which i s highly s i g n i f i c a n t . Analysis of Contarinia oregonensis damage: The percentage of seeds damaged by Contarinia oregonensis was s i g n i f i c a n t l y c orrelated with tree height (r = 0.232), average vegetative bud f l u s h i n g date (r = -0.207) and percentage of f i l l e d seeds (r = -0.232) i n 1961. Using 19 independent variables 28.26 per cent of the v a r i a t i o n of the dependent v a r i a b l e was contributed by the v a r i a b l e s . Percentage of f i l l e d seeds per tree, average f l u s h i n g time of vegetative buds, s i t e index at 100 years and age were the most important v a r i a b l e s . In 1962, tree height (r = 0.326), D.b.h. (r = 0.322), percentage of f i l l e d seeds per tree (r = -0.415) and age (r = 0.342) were highly s i g n i f i c a n t l y and crown width (r = 0.226) was s i g n i f i c a n t l y c orrelated with the percentages of seeds damaged by Contarinia. With the l i n e a r multiple regression equation 48.38 per cent of the v a r i a t i o n was removed using a l l the 19 v a r i a b l e s . The most important variables were percentage of f i l l e d seeds per tree, date cones became pendent, D.b.h and tree height r e s p e c t i v e l y . Only one of the most important v a r i a b l e s was common i n the two years, the percentage of f i l l e d seeds. However, i t i s very l i k e l y that the percentage of f i l l e d seeds had no e f f e c t on the abundance of the damage, but rather the presence of the insects a f f e c t e d the amount of f i l l e d seeds produced. The second most important v a r i a b l e was the date cones became pendent i n 1962. It was shown by Hedlin (1961), Johnson and Winjum (1960) and Johnson (1963 d) that the stage of cone development i s an important factor which a f f e c t s the Contarinia oregonensis damage. In the present study i t was found that the date cones became erect had no e f f e c t (Table 46), but the date cones became pendent was highly s i g n i f i c a n t l y correlated with the damage (Table 47). The r e l a t i o n s h i p for t h i s was found to be: (14) Y = 8.106 + 0.825 x 10 where: Y = a r c s i n square root percentage of seeds damaged by _C. oregonensis. X10 = date cones became pendent (daysafter May f i r s t ) . The equation has a standard error of estimate 9.58 a r c s i n square root percentage of seeds and r = 0.415 which i s highly s i g n i f i c a n t . Analysis of percentage of undamaged seeds: Only one factor, the per.cent of f i l l e d seeds per tree per cone, (r = 0.373) was highly s i g n i f i c a n t l y correlated with the percentage of undamaged seeds i n 1961. The l i n e a r multiple regression equation contributed 37.19 per cent of the v a r i a t i o n of the dependent v a r i a b l e s In the process of elimination the percentage of f i l l e d seeds, duration of vegetative bud flus h i n g , and average flushing time of the vegetative buds were the most important v a r i a b l e s . These three v a r i a b l e s contributed 20.28 per cent of the t o t a l v a r i a t i o n of the dependent v a r i a b l e . Tree height (r = -0.278), D.b.h. (r = -0.278), date cones became pendent (r = -0.397), age (r = -0.297), crown width (r = -0.226), and the size of female flower crop (r = 0.257) were s i g n i f i c a n t l y or highly s i g n i f i c a n t l y correlated with the per cent of undamaged seeds i n 1962. Most of these s i g n i f i c a n t c o r r e l a t i o n s are the inverse of the damage of Contarinia oregonensis, since the damage of t h i s insect species was dominant. As Tables 46 and 47 indicate i n both years the higher the per-centage of undamaged seeds, the more was the percentage of f i l l e d seeds. However, the s e l e c t i o n of trees f o r seed production cannot be based on t h i s because of the abundance of Megastigmus i s higher on those trees which have higher percentage of f i l l e d seeds. 106 Relationships among the dependent v a r i a b l e s : The re l a t i o n s h i p s between the dependent variables are summarized i n Tables 49 and 50 for 1961 and 1962 r e s p e c t i v e l y . In 1961 the per cent of seeds damaged by Megastigmus i s highly s i g n i f i c a n t l y correlated with the percentage of f i l l e d seeds damaged by Megastigmus (r = 0.781), percentage of D i o r y c t r i a damage (r =-0.296) and percentage of undamaged seeds (r = 0.332). The per cent of f i l l e d seeds damaged by Megastigmus was highly s i g n i f i c a n t l y correlated with the damage of D i o r y c t r i a (r = -.297) and s i g n i f i c a n t l y with the undamaged seeds (r = 0.212). The percentage of seeds damaged by D i o r y c t r i a was highly s i g n i f i c a n t l y correlated with the amount of undamaged seeds (r = -0.540), None of the other insect species damage had s i g n i f i c a n t r e l a t i o n s h i p with the damage of Contarinia oregonensis. The percentage of undamaged seeds was very highly affected by the damage of Contarinia oregonensis (r = -0.812). In 1962, the r e l a t i o n s h i p between the percentage of seeds damaged by Megastigmus and percentage of f i l l e d seeds damaged by Megastigmus was highly s i g n i f i c a n t (r = 0.779). Also the damage of D i o r y c t r i a (r =-0.213) and the undamaged seeds (r =0.201) were s i g n i f i c a n t l y correlated with the Megastigmus damage. A p o s i t i v e highly s i g n i f i c a n t c o r r e l a t i o n was found between the damage of D i o r y c t r i a and Contarinia oregonensis (r = 0.373). Both the damages of Contarinia and D i o r y c t r i a had a highly s i g n i f i c a n t r e l a t i o n s h i p with the percentages of undamaged seeds (r = -0.985', and r = -0.454 r e s p e c t i v e l y ) . These r e s u l t s w i l l be discussed i n more d e t a i l i n the chapter on competition. Table 49. Relationship between the dependent variables i n 1961. Cor r e l a t i o n c o e f f i c i e n t Yl Y2 Y3 Y4 Y5 Yl 1.000 0.781** -0.296** -0.151 0.332** Y2 0.781** 1.000 -0.297** -0.012 0.212* Y3 -0.296** -0.297** 1.000 0.129 -0.540** Y4 -0.151 -0.012 0.129 1.000 -0.812** Y5 0.332** 0.212* -0.540** -0.812 1.000 Table 50. Relationship between the dependent variables i n 1962. Corr e l a t i o n c o e f f i c i e n t Y l Y2 Y3 Y4 Y5 Yl 1.000 0.779** -0.213** -0.201 0.201* Y2 0.779** 1.000 0.035 0.119 -0.116 Y3 -0.213* 0.035 1.000 0.373** -0.454** Y4 -0.201* 0.119 0.373** 1.000 -0.985** Y5 0.201* -0.116 -0.454** -0.985** 1.000 108 Transformations: Because of the small number of factors found to be s i g n i f i c a n t l y -correlated as straight l i n e s with the damage of d i f f e r e n t insect species on d i f f e r e n t trees, a number of c u r v i l i n e a r r e l a t i o n s h i p s was t r i e d , such as parabolic, logarithmic and exponential. These trans-formations were applied separately f o r the dependent and independent va r i a b l e s , then for both together. Since no s i g n i f i c a n t improvement was found with these, and since the l i n e a r r e l a t i o n s h i p i s the simplest to work with, that w i l l be used to draw the conclusions which follow. 3. I n t e r r e l a t i o n s of Cone and Seed Insects of Douglas F i r To be precise, i n studying the i n t e n s i t y of damage of cone and seed insects between and within trees, the i n t e r r e l a t i o n s of them must be studied, because the l e v e l of i n f e s t a t i o n of a c e r t a i n species may be low for two main reasons: 1) the tree i s r e s i s t a n t to the insect, because of c e r t a i n phenological or other c h a r a c t e r i s t i c s of• the tree; 2) the low l e v e l of i n f e s t a t i o n might be caused by the competition of other insect species. Table 51 summarizes the damage on the ten best and ten worst trees selected by the percentage of undamaged cones. From these data i t i s c l e a r that the t o t a l damage of Contarinia oregonensis, Megastigmus  spermotrophus and D i o r y c t r i a a b i e t e l l a was more than 100 per cent on a number of trees. In many of the cases even the t o t a l damage of two of these species was more than 100 per cent. These r e s u l t s indicate that more than one species attack the Douglas-fir cone at a time which may r e s u l t i n some competition for space and food. Table 51. Summary of the best and worst ten trees i n percentage of cones damaged. Tree Per cent of cones damaged by Per cent of number Contarinia Megastig mus D i o r y c t r i a undamaged 1 + 2 2 + 3 1 + 3 1 + 2 + 3 1 2 3 cones Best ten trees 19 80.8 13.3 2.1 12.5 94.1 15.4 82.9 96.2 28 27.1 37.5 8.3 39.6 64.6 45.8 35.4 72.9 . 91 66.7 35.4 12.5 12.5 102.1 47.9 79.2 114.6 95 70.8 47.9 14.6 12.5 118.7 62.5 85.4 133.3 98 59.0 13.9 2.1 36.1 72.9 16.0 61.1 75.0 126 62.5 18.7 16.7 22.9 81.2 35.4 79.2 97.9 128 50.7 36.1 5.8 16.7 86.8 41.9 56.5 92.6 134 45.8 10.4 2.1 45.8 56.2 12.5 47.9 58.3 140 52.1 41.7 2.1 27.1 93.8 43.8 54.2 95.9 144 75.0 33.3 4.2 18.1 108.3 37.5 79.2 112.5 Averages 59.1 28.8 7.1 24.4 87.9 35.9 66.1 94.9 Worst ten trees 22 81.3 35.4 8.3 0.0 116.7 43.7 89.6 125.0 23 95.8 10.4 8.3 0.0 106.2 18.7 104.1 114.5 24 100.0 14.6 14.6 0.0 114.6 29.2 114.6 129.2 33 95.8 37.5 14.6 0.0 133.3 52.1 110.4 147.9 50 95.8 6.3 33.3 0.0 102.1 39.6 129.1 135.4 63 91.6 62.5 18.8 0.0 154.1 81.3 110.4 172.9 92 87.5 72.9 4.2 0.0 160.4 77.1 91.7 164.6 132 85.4 79.2 8.3 0.0 164.6 87,5 93.7 172.9 136 97.9 10.4 27.1 0.0 108.3 37.5 125.0 135.4 158 85.4 62.5 22.9 0.0 147.9 85.4 108.3 170.8 Averages 91.6 39.2 16.0 0.0 130.8 55.2 107.7 146.9 110 In 1961 one cone from each of the sample locations was taken randomly and the number of damaged seeds was counted separately i n each centimeter from the base to the top of the cone on the lo n g i t u d i n a l cut surface. The number of damaged seeds was observed with respect to three d i f f e r e n t species of cone and seed i n s e c t s ; Contarinia oregonensis, Megastigmus spermotrophus and D i o r y c t r i a a b i e t e l l a . Then, the number of damaged seeds was summarized and averaged by insect species along the axis of the cone (Figure 1), by insect species and crown l e v e l (Figures 2 and 3) and by insect species and outside- inside crown (Figures 4 and 5). Because of the small number of trees sampled from the lower crown l e v e l , the r e s u l t s of t h i s l e v e l were omitted. Each of Figures 1 - 5 contains seven curves: 1) damage by D i o r y c t r i a i n a l l cones; 2) damage by Contarinia i n a l l cones; 3) damage by Contarinia i n D i o r y c t r i a - f r e e cones; 4) damage by Contarinia i n Megastigmus-free cones; 5) damage by Megastigmus i n a l l cones; 6) damage by Megastigmus i n D i o r y c t r i a - f r e e cones; 7) damage by Megastigmus i n Contarinia-free cones. In t e r r e l a t i o n s between D i o r y c t r i a and Contarinia: A l l of Figures 1 - 5 show that i n the basal 55% of the cone the average number of seeds infested by Contarinia i s much higher i n D i o r y c t r i a -free cones than i t i s i n an unsorted batch of cones. As indicated by the curves, D i o r y c t r i a feeds mostly i n the basal 457. part of the cones. Because of the feeding habit of D i o r y c t r i a i n those cones i n which t h i s species i s present most of the Contarinia larvae are eaten by them i n the basal 45-557o part of the cones. Since D i o r y c t r i a a b i e t e l l a i s a phyto-phagous insect and does not depend on the Contarinia population, the Figure I- Distribution of cone and seed insect damage in Douglas-fir cones-Legend — Dioryctria damage - Contarinia damage in all cones - Contarinia damage in Dioryctna-free cones •- Contarinia damage in Meqastiqmus-free cones Megastigmus damage in all cones — Megastigmus damage in Dioryctria - free cones Megastigmus damage in Contarinia -free cones \ Base Cone Length in Percentage Top Figure 2 Distribution of cone and seed insect damage in Douglas-fir cones (middle crown) Figure 3 Distribution of cone and seed insect damage in Douglas-fir cones (upper crown)-Figure 4- Distribution of cone and seed insect damage in Douglas-fir cones (outside crown)-Figure 5- Distribution of cone and seed insect damage in Douglas-fir cones (inside crown) 116 i n t e r r e l a t i o n of them might be c a l l e d a s p e c i a l kind of predation, but i t i s rather a competition for food and space i n the cones, i n which competition Contarinia i s very badly i n h i b i t e d and D i o r y c t r i a i s not affected, or i f i t i s affected, the e f f e c t i s s l i g h t and cannot be measured by damaged number of seeds. The i n t e r r e l a t i o n between these two species was found to be s i m i l a r i n d i f f e r e n t parts of a tree (Figures 2 - 5 ) , meaning that the density of Contarinia population, which changes from l e v e l to l e v e l and from outside to inside crown, does not a f f e c t t h i s i n t e r r e l a t i o n . I n t e r r e l a t i o n between D i o r y c t r i a and Megastigmus: Koerber (1960) stated that Megastigmus spermotrophus fares badly i n the competition for the l i m i t e d food supply. The other insects, e s p e c i a l l y Barbara colfaxiana (Kft., Dethreutidae) and D i o r y c t r i a abiet-i e l l a may destroy nearly a l l of the Douglas f i r seed along with the Megastigmus larvae they contain. This would mean that the e f f e c t of D i o r y c t r i a and Megastigmus i s d i r e c t , s i m i l a r l y as i t was on Contarinia. This statement cannot be proved by the present study, because the r e s u l t s do not indicate such an e f f e c t . The average number of seeds damaged by Megastigmus i n a l l cones i s much lower than i n Dioryctria-:free cones, and t h i s difference i s not concentrated i n the basal part of the cones, as i t was found i n Contarinia, but continues from the base to the t i p of cones (Figures 1 - 5 ) . The difference between the two curves i s largest around 65 - 707o of the cone length from the base, where the peak of Megastigmus damage appears, and where the D i o r y c t r i a damage i s quite low. This r e s u l t c l e a r l y indicates that the e f f e c t of D i o r y c t r i a on 117 Megastigmus i s i n d i r e c t : the i n t e n s i t y of egg laying of Megastigmus females i s heavier i n those cones which are free from D i o r y c t r i a . The means by which the egg laying femaleccan detect the presence or absence of D i o r y c t r i a larvae i s not known. It i s very l i k e l y that the i n t e r a c t i o n of these two species i s competition for space i n which Megastigmus i s forced to f i n d a D i o r y c t r i a -free cone for o v i p o s i t i o n and the D i o r y c t r i a i s not affected, and p a r t l y i n predation i f they are feeding i n the same cone. The i n t e n s i t y of t h i s predation i s much smaller than was found i n Contarinia, because Megastig- mus has a chance to lay eggs i n D i o r y c t r i a - f r e e cones. Of course, because of the overlapping i n l i f e cycle of these two species a large number of Megastigmus might be preyed. No differences were found i n the i n t e r a c t i o n of these two insect species i n the d i f f e r e n t parts of the l i v i n g crown; therefore, the density of Megastigmus population, which changes from outside to inside crown, does not a f f e c t the D i o r y c t r i a larvae. I n t e r r e l a t i o n between Megastigmus and Contarinia: It i s very d i f f i c u l t to study the i n t e r r e l a t i o n between Megastigmus and Contarinia by the number of damaged seeds, because i t can very often be seen that a single seed i s damaged by both species. The growth and vigour of the larvae should be studied i n t h i s case to show the competi-tiv e r e l a t i o n s h i p between the two species, because as was shown by U l l y e t t (1950) for f i v e species of b l o w f l i e s , average size of i n d i v i d u a l s , fecundity and m o r t a l i t y a l l were affected because of competition for food. The i n t e r r e l a t i o n between Megastigmus and Contarinia i s always concentrated around a single seed, where only one Megastigmus l a r v a and a number.of Contarinia larvae are7present and i t i s very l i k e l y that there i s an upper 118 l i m i t to the number of Contarinia larvae around the seed which i s destroying the Megastigmus larva i n the competition. The damage curves (Figures 1 - 5) do not indicate d e f i n i t e differences i n damage of Contarinia between Megastigmus-free and a l l cones. This r e l a t i o n s h i p was found to be s i m i l a r i n a l l parts of the tree. However, the average number of seeds damaged by Megastigmus i n Contarinia-free cones i s higher i n a l l parts of the cone, than i n a l l cones. This difference i s much s l i g h t e r than was found i n the D i o r y c t r i a - Megastigmus i n t e r a c t i o n . Apart from t h i s , Contarinia has some i n d i r e c t e f f e c t on Megastigmus s i m i l a r to, but s l i g h t e r than D i o r y c t r i a . The competition of cone and seed insects can be seen not only i n cones but also i n trees. Tables 27 and 28 show that on those trees where the D i o r y c t r i a was abundant, a smaller amount of Megastigmus damage was found i n both years. This r e l a t i o n s h i p was s i m i l a r f o r number of cones damaged and for percentage of seeds damaged. The Contarinia oregonensis was not found to be affected by D i o r y c t r i a i n c o r r e l a t i o n analyses of the number of cones damaged on a tree. The same r e s u l t was found when the percentage of seeds damaged was analysed i n 1961, but i n 1962 the percentage of seeds damaged by Contarinia had a highly s i g n i f i c a n t and p o s i t i v e c o r r e l a t i o n with D i o r y c t r i a damage (Tables 49 and 50). Correlating the number of cones damaged by Megastigmus with the number of cones damaged by Contarinia, no s i g n i f i c a n t relationshipr, was found, but when the percentage of seeds was correlated the damage of Contarinia s i g n i f i c a n t l y reduced the damage of Megastigmus i n 1962 but no s i g n i f i c a n t r e l a t i o n s h i p was found i n 1961. 4. Parasites of Cone and Seed Insects One cone from each sample lo c a t i o n of each tree was stored i n a polyethylene bag i n order to study the p a r a s i t i c insects emerging from the cones. Observations f o r t h i s study are available from 1961 only, because about 90 per cent of the stored cones of 1962 moulded and no parasites emerged from them. The poor r e s u l t s i n 1962 might be caused because at the time of c o l l e c t i o n , the cones were not as mature as i n 1961, although the c o l l e c t i o n was c a r r i e d out on the same calender date Two species, a Torymus sp. and a Platygaster sp. were the most numerous of the parasites that emerged from the cones. Both of them are parasites of Contarinia oregonensis. An average of 0.40 Torymus sp was found per cone, with a standard deviation of 0.64. The Polygaster  sp. was much more abundant with an average of 1.24 insects per cone, and a standard deviation of 1.36. If the two species were the same as described by Hedlin (1961), these r e s u l t s agree reasonably well with h i s . Hedlin found that 4.7 per cent of the Contarinia larvae were p a r a s i t i z e d by Torymus, and 25 per cent by Platygaster sp.. The number of parasites found per cone per tree was correlated with the percentage of seeds damaged by _C. oregonensis per tree. No s i g n i f i c a n t r e l a t i o n s h i p was found between the number of Torymus per cone per tree and the Contarinia damage. For Platygaster, i t was found that the number of parasites per cone per tree (y) was increased as: ( 1 5 ) y = e 2 ' m " 0 , 3 4 7 x + ° - 2 1 1 x 2 - 1 where: x = the a r c s i n square root percentage of seed damaged 120 by Contarinia oregonensis. The equation has a standard error of estimate of 1.30 parasites, and a multiple c o r r e l a t i o n c o e f f i c i e n t 0.260 which i s s i g n i f i c a n t at the 0.05 p r o b a b i l i t y l e v e l . Only seven Copidosoma sp., which i s a parasite of D i o r y c t r i a  a b i e t e l l a , emerged from the cones. Expressed as an average t h i s i s 0.016 parasites per cone. No other p a r a s i t i c insects emerged from the cones. 5. Time of Emergence In the spring of 1962 the times of emergence of Contarinia  oregonensis, Megastigmus spermotrophus and D i o r y c t r i a a b i e t e l l a were observed from 12 rearing boxes and from the ground at three locations. No difference was found i n time of emergence between the rearing boxes and ground. Because of t h i s , the counts from the rearing boxes only w i l l be discussed. These were much more accurate than observations of the ground below the study trees. The frequency of emergence of Contarinia oregonensis and Megastigmus  spermotrophus i s summarized by days i n Figure 6. Only one specimen of D i o r y c t r i a a b i e t e l l a emerged from the boxes i n the spring of 1962. On Figure 6, the d a i l y t o t a l hours of sunshine i s also plotted because i t seemed to have some e f f e c t on the time of emergence for both species. Of course, the average d a i l y s o i l temperature at two inches depth was highly s i g n i f i c a n t l y correlated with the t o t a l hours of sunshine and i t . i s - v e r y . l i k e l y that both the sunshine and s o i l temperature have the same<-effect on the emergence. No differences were found i n time of emergence between those rearing Figure 6 Time of emergence of Contorinia oregonensis and Megastigmus spermotrophus in 1962-a> e C Q> CP O X \— o XI > o o> c T3 cz o o a> CV mm. c mm. Ic CU in a. c -J o CO c H— o CO O i- c rj o ho o a o o \— I5r 10-Contarinia oregonensis Legend Average number of Contannio Total hours of sunshine 151— 03 jQ E C 0) 2 S 5;^ IO i f cu »-c ^ -1 o o !=; o 2 --o o Legend Average number of Megastigmus Total hours of sunshine Megastigmus spermotrophus 122 boxes which were on d i f f e r e n t elevation and aspects, perhaps because the difference i n e l e v a t i o n was only 200 feet. Also, the time of emergence was the same i n those boxes which were covered by the crown and i n those which were i n the open. 6. Time of Attack A cone-bagging experiment was run to f i n d some information on the time of attack of the cone and seed insects of Douglas f i r i n the spring of 1962. Figure 7 gives the time of attack on Contarinia oregonensis separately on an early flowering tree and on a late flowering tree. The two charts indicate that there was two days difference i n the beginning and s i x days difference i n the ending of attack between the e a r l y and late flowering trees. Also, the percentage of infested seeds on the e a r l y tree (8.2 per cent) was less than on the late one (14 per cent). Figure 8 shows the time of attack of Megastigmus spermotrophus. No difference was found between the late and early tree e i t h e r i n time or damage. No information i s a v a i l a b l e on the time of attack of D i o r y c t r i a , because of i t s very low l e v e l of damage i n 1962. 7. Trapping Experiment To study the i n t e n s i t y of searching of the cone and seed insects on d i f f e r e n t parts of the crown and on d i f f e r e n t trees, a trapping experi-ment was set up i n the spring of 1962. Analysis of variance was computed on the number of trapped insects for both Contarinia oregonensis and Megastigmus spermotrophus. The factors were analysed i n these Figure 7- Time of attack of Contarinia oregonensis and Meqostiqmus spermotrophus-Legend I cones bagged periodically l i i i I l I I i i i i i i i i l I I i i l I i I I I 25 27 29 I 3 5 7 9 II 13 15 17 19 21 Apri l May S Megastigmus spermotrophus to I I I I I I I I I I I I I I I I I I I I I I I I I I I 31 I 3 5 7 9 II 13 15 17 19 21 23 25 May June Date of cones bagged or opened 124 analyses of variance: group, trees within group and crown l e v e l . Group was based on cone production, with three l e v e l s : heavy, medium and l i g h t . None of the sources had s i g n i f i c a n t v a r i a t i o n i n the case of Megastigmus counts. For Contarinia, the counts were s i g n i f i c a n t l y less i n the lower crown l e v e l (8.8 insects per frame) than i n the middle crown l e v e l (12.0 insects per frame), and s i g n i f i c a n t l y more i n the upper crown l e v e l (14.5 insects per frame) than i n the middle crown level.. No difference was found between groups and between trees within group. 8. Frequency D i s t r i b u t i o n s and Sequential Sampling The frequency d i s t r i b u t i o n s of the number of seeds damaged on the lon g i t u d i n a l cut surface of the cones were tested by." chi-square analysis of goodness of f i t for Megastigmus and Contarinia. Total damage per cut surface and undamaged f i l l e d seeds per cut surface were also tested. The samples from f i v e trees (Numbers 5, 55, 56, 80, 126, i n 1961 and 30, 60, 63, 68 and 126 i n 1962) with 144 cones from each were used for t h i s a n a l y s i s . Four d i s t r i b u t i o n s , normal, Poisson, binomial and negative binomial were applied to the data. The best f i t i n a l l cases was found with the negative binomial. The parameter (k) and the p r o b a b i l i t y of f i t (p) of these d i s t r i b utions are tabulated i n Table 52. From each of the ten d i s t r i b u t i o n s of the same group a combined d i s t r i b u t i o n was calculated and the sequential sampling plans were set up (Figures 8, 10, 12, 14). The l i m i t s for the sequential sampling plans are" summarized i n Table 53. Table 52. Frequency d i s t r i b u t i o n s of number of seeds per cone damaged by Contarinia oregonensis Megastigmus spermotrophus and t o t a l numbers of damaged seeds and undamaged f i l l e d seeds. Tree number Contarinia damage k P Megastigmus damage k P T o t a l k damage P Undamaged k f i l l e d seeds P 5 1.623 0.19 1.138 0.90 2.806 0.85 3.968 0.99 55 0.642 0.58 5.042 0.89 3.311 0.14 1.244 0.19 56 1.492 0.78 0.857 0.99 3.057 0.72 0.866 0.84 80 0.671 0.52 4.322 0.99 5.807 0.15 0.610 0.99 126 1.326 0.09 1.146 0.99 4.053 0.18 0.560 0.93 30 1.697 0.79 2.987 0.72 5.264 0.68 7.937 0.34 60 1.734 0.28 1.863 0.99 4.772 0.82 2.141 0.19 63 2.054 0.82 1.117 0.94 5.425 0.80 1.539 0.99 68 1.531 0.09 1.033 0.98 7.228 •0.15 2.382 0.61 126 2.392 0.93 3.215 0.99 4.151 0.75 1.885 0.96 Combined 1.947 1.684 4.785 4.435 Expressed i n terms of the negative binomial d i s t r i b u t i o n with parameters p and k. 126 Table 53. Limits f o r sequential sampling of cones. Infestation l e v e l Light i, 'Medium Heavy Number of seeds per lo n g i t u d i n a l cut surface of a cone Contarinia damage 1 or less 2 - 4 5 or more Megastigmus damage 1 or less 2 - 4 5 or more Tot a l damage 2 or less 3 - 5 6 or more Undamaged f i l l e d seeds 2 or less 3 - 5 6 or more The calculated decision l i n e s f o r the sequential sampling plans are as follows: Light vs. Medium Medium vs. heavy Contarinia damage Y = 1.420 n + 5,491 Y = 4.468 + 32.479 Megastigmus damage Y = 1.421 n + 5.838 Y = 4.493 + 35.960 Total damage Y = 2.450 n + 8.206 Y = 5.450 n + 25.769 Undamaged f i l l e d seeds Y = 2.450 n + 8.424 Y = 5.478 n + 26.933 The operating c h a r a c t e r i s t i c (O.C.) and average sample number (A.S.N.) curves are given i n Figures 9, 11, 13 and 15, for the four sequential sampling plans. These plans, presented above, can be used to judge the l e v e l of i n f e s t a t i o n or the l e v e l of f i l l e d seeds production on any i n d i v i d u a l Douglas f i r tree, which i s selected f o r seed c o l l e c t i o n , or selected for studying the population l e v e l of Contarinia or Megastigmus. For sampling stands, 64 trees i n each year were randomly selected from those trees from which at least 24 cones were sampled. The Figure 8 Sequential graph for sampling cones damaged by Contarinia oregonensis-Number of cones examined — n Figure 9 0- C and A- S N curves for sampling cones damaged by Contarinio oregonensis Legend : P= probability of correct call N= average sample number P i-OO-L 0-50 0 0 0 OC- Curves \ Light vs \ Medium Medium vs Heavy '6 5 4* 6 Average number of seeds damaged A S- N- Curves N 100-50 Medium vs \ ^ Heavy Light vs- Medium 0-1 1 i 1 0 2 4 6 Average number of seeds damaged -i 8 Figure 10 Sequential graph for sampling cones damaged by Megastigmus spermotrophus-Number of cones examined — y Figure I Legend: •00-, 0-50-0 C and A S N curves for sampling cones damaged by Megastigmus spermotrophus P= probability of correct call N= average sample number 0 C Curves Light vs Medium Medium vs Heavy 000-0 T -2 - T -6 Average number of seeds damaged -i 8 N A S N Curves 100-50-i \ i \ i \ Medium vs y Heavy sv Light vs Medium 0 - r 2 — r 4 8 Average number of seeds damaged Figure 12 Sequential graph for sampling total number of damaged cones Figure 13 0 C- and A S N curves for sampling total number of damaged cones Legend P= probability of correct call N= average sample number 0 C Curves 1 0 0 0-50 0 0 0 0 N 150-100 5 0 -Light vs Medium \ i Medium vs i i Heavy \ \ T " 2 T " 4 8 Average number of seeds damaged A S N Curves i Medium vs Heavy Light vs Medium / \ / \ / \ / \ 3 4 5 Average number of seeds damaged Figure 14- Sequential graph for sampling undamaged filled seeds Figure 15 0 C-and A S N-curves for sampling undamaged filled seeds Legend • P= probability of correct call N= average sample number •00 0-50-000-0 C Curves i J Poor vs i Medium Medium vs Good "I 0 2 4 6 8 Average number of undamaged filled seeds N 150 100-50 A S N Curves Medium vs Good Poor vs „ Medium/ -0 T 0 I 1 i 2 4 6 8 Average number of undamaged fil led seeds 135 frequencies of damage on these trees stated as 0, 1, 2 24 out of 24 cones damaged were counted for Megastigmus, D i o r y c t r i a and Contarinia. The frequency d i s t r i b u t i o n of undamaged cones was also found. The damage of Megastigmus (p = 0.88, k = 3.573), D i o r y c t r i a (p = 0.62, k = 2.349) and undamaged cones (p = 0.72, k ='0.848) were found to be negative binomial d i s t r i b u t i o n s and the damage of Contarinia the normal d i s t r i b u -t i o n (p. = 0.68). with .mean ,17.61 and standard deviation 4.57 cones out of 24. The sequential sampling graphs for these four d i s t r i b u t i o n s are given i n Figures 16, 18, 20 and 22. The l i m i t s to set up these sampling trees within stands, are given i n Table 54. v Table 54. Limits f o r sequential sampling plans, sampling trees within stand. Light Medium Heavy Damaged cones out of 24 per tree Megastigmus damage 5 or less 6 - 12 13 or more D i o r y c t r i a damage 5 or less 6 - 12 13 or more Contarinia damage 5 or less 6 - 12 13 or more Undamaged cones 6 or less 7 - 13 14 or more The calculated decis as follows: Megastigmus damage D i o r y c t r i a damage Contarinia damage Undamaged cones l i n e s for the sequential Light vs. Medium Y = 5.466 n + 30.380 Y = 5.405 n + 39.644 Y = 5.500 n + 18.682 Y = 6.475 n +123.128 sampling plans are Medium vs. Light Y = 12.420 n + 123.128 Y = 12.302 n + 170.096 Y = 12.500 n + 18.682 Y = 14.012 n + 443.834 136 For these sampling plans, the operating c h a r a c t e r i s t i c and average sample number curves are plotted i n Figures 17, 19, 21 and 23. 9. Optimum A l l o c a t i o n of Sample Sizes Five trees which had.six cones sampled from each of the s i x sample locations i n both years of 1961 and 1962 were studied by analysis of variance for the percentage of seeds damaged by Contarinia oregonensis and Megastigmus spermotrophus and for undamaged f i l l e d seeds. Then the componentsof variance of a l l factors were calculated (Table 55) to help define the optimum a l l o c a t i o n of sample sizes i n the d i f f e r e n t stages of sampling. The stages i n the sampling were year, trees within year, crown l e v e l s within tree within year and crown positions within l e v e l within tree within year. The cost of unit sample i n d i f f e r e n t stages was defined as 100 monetary units f or year, 10 for tree, one for crown l e v e l and 0.1 for crown p o s i t i o n . For the c a l c u l a t i o n i t was supposed that 300 monetary units are a v a i l a b l e to carry out the sampling. The optimum a l l o c a t i o n of sample sizes i n the d i f f e r e n t stages i s summarized i n Table 56, for sampling Contarinia and Megastigmus damage and undamaged f i l l e d seeds. Study of the optimum a l l o c a t i o n of sample sizes f or D i o r y c t r i a has been omitted because of the very low damage i n 1962. The very high year to year v a r i a t i o n of undamaged f i l l e d seeds caused that only one tree i s required. The other two analyses determined a comparatively large number of trees to be sampled even with a r e l a t i v e l y high sampling price of a tree, which i s an i n d i c a t i o n of the high tree to tree v a r i a t i o n i n damage. Figure 16- Sequential graph for sampling trees damaged by Megastigmus spermotrophus-Number of trees examined — n Figure 17- 0 0 and ASN- curves for sampling trees damaged by Megastigmus spermotrophus Legend- P= probobility of correct call N= overage sample number 0 C Curves •00-r-0-50-o-oo-H Light vs Medium i i \ Medium vs \ Heavy \ i i i \ \ \ - r 9 II N 600-400-200-Average number of cones damaged A S N Curves Medium vs Heavy Light vs A Medium / \ — o-H-5 7 9 II Average number of cones damaged - i 13 Figure 18 Sequential graph for sampling trees damaged by Dioryctria abietella-Number of trees examined — n Figure 19 0 C and AS N curves for sampling trees damaged by Dioryctria abietella-Legend1 P= probability of correct call N= average sample number 0 C Curves N 600-400-200-0-+V Light vs-Medium Medium vs-Heavy 1 1 = ^ 5 8 II 14 Average number of cones damaged A S N Curves i Medium vs » Heavy Light vs- Medium ~r 5 T II 14 Average number of cones damaged Figure 20 Sequential graph for sampling trees damaged by Contarinia oregonensis Figure 21 0 C and A S N-curves for sampling trees damaged by Contarinia oregonensis Legend P = probability of correct call N= average sample number •00-0-50-o-oo-fV OC- Curves Light vs Medium \ Medium vs Heavy N 60 4 0 -20-5 8 11 Average number of cones damaged A S N Curves Light vs Medium / \ 14 Medium vs-Heavy 5 8 11 Average number of cones damaged 14 Figure 22 Sequential graph for sampling trees for undamaged cones-Number of trees examined — n Figure 23 OC and ASN- curves for sampling undamaged cones-Legend : P= probability of correct call N= average sample number OC Curves 100-0-50-o-oo-fV Poor vs Medium Medium vs Good 6 9 12 Average number of undamaged cones 15 N 1000-A-SN- Curves A / i Medium vs / i Good / » 500- A I I • \ I \ Poor vs- Medium 0-Hr 3 "i i ) 6 9 12 15 Average number of undamaged cones Table 55„ Components of variance of the damage of Contarinia oregonensis and Megastigmus spermotrophus and of the undamaged f i l l e d seed, expressed as a r c s i n square root. Source of v a r i a t i o n DF Year 1 Tree within year 8 Level within tree within year 20 P o s i t i o n within l e v e l within tree within year 30 Damage by Undamaged f i l l e d seeds Contarinia oregonensis Megastigmus spermotrophus Variance components Variance Components Variance Components of variance of variance of variance 151.80 288.70 212.3.7 -0.76 2.12 -3.10 Residual 300 249.54 19.79 130.81 130.81 2.10 133.62 75.86 62.84 41.45 -0.73 1.60 1.09 3.57 41.45 6145.4 179.4 64.9 115.2 35.3 34.81 3.18 4.19 13.31 35.32 Tota l 359 146 Table 56. Optimum a l l o c a t i o n of sample sizes i n a four-stage sampling. Source of v a r i a t i o n Year Tree Crown leve1 Crown p o s i t i o n Contarinia damage 2 6 4 8 Megastigmus damage 2 5 3 6 Undamaged f i l l e d seeds 3 1 4 6 Assumed unit cost of sampling 100 10 1 0.1 On changing the cost assumptions, the a l l o c a t i o n of sample sizes w i l l be changed. For example, decreasing the cost of year samples, the number of years taken w i l l increase and the sub-sample within year (tree) w i l l have a smaller sample size than before; the sample size of the sub-samples within trees w i l l not change. Generally, i t can be stated that lowering the price of a sample unit w i l l increase the sample size of the unit and w i l l reduce the sample size of the sub-sample within the uni t . DISCUSSION The r e s u l t s of the i n v e s t i g a t i o n center about four main species of i n s e c t s : D i o r y c t r i a a b i e t e l l a , Contarinia oregonensis, _C. washing- tonensis and- Megastigmus spermotrophus. They reveal information on the d i s t r i b u t i o n and abundance of the separate species, and on the e f f e c t s of competition within, and between species. In the two-year period of study the percentage of cones damaged by- D i o r y c t r i a dropped from 18.2 per cent i n 1961 to 5.7 per cent i n 1962. This change was much more pronounced when expressed by the mean percentage of seeds damaged; thus 3.5 per cent were damaged i n 1961 and only 0.2 per cent i n 1962. The reason for t h i s highly s i g n i f i c a n t change i s unknown, but i t i s very l i k e l y that the overwintering period was the part of the l i f e cycle when the pupal or prepupal population was very badly reduced. This can be stated because none of the traps caught a D i o r y c t r i a , and none of the s o i l cages yielded any emerged moths, and only one specimen was obtained from the 12 rearing boxes i n the spring of 1962. It can, accordingly, be concluded that a large portion of the D i o r y c t r i a population did not even emerge from the pupal stage. The cause of t h i s f a i l u r e to emerge might be assumed to be a t t r i b u t a b l e e i t h e r to various mortality factors such as parasites, predators, diseases or c l i m a t i c factors during the overwintering stage, or to retarded emergence caused by extended diapause for one or more years. However, only 10 per cent of the population was p a r a s i t i z e d i n the f a l l of 1961, which should not a f f e c t the next year's damage so badly. The p o s s i b i l i t y that emergence was merely delayed by diapause 148 was ruled u n l i k e l y by the t o t a l absence of emergence i n the rearing boxes i n the spring of 1963. It i s , of course, conceivable that the insects could remain i n extended diapause for more than one winter but i t seems u n l i k e l y that a measurable percentage would not emerge a f t e r one winter, unless a m o r t a l i t y factor then sets i n . The damage of D i o r y c t r i a showed highly s i g n i f i c a n t tree to tree v a r i a t i o n i n a l l of the analyses computed. The i n f e s t a t i o n ranged from 54.2 to 0 per cent of the cones damaged and from 22.7 to 0 per cent of the seeds damaged. In 1961 when the damage caused by t h i s species was considerable the i n t e n s i t y of attack was associated with „ cone length and duration of vegetative bud f l u s h i n g . Since no found s i g n i f i c a n t c o r r e l a t i o n was^between the cone length and duration of vegetative bud flushing, i t can be assumed that they affected the i n t e n s i t y of attack independently. Trees with longer cones had more damage than those with shorter cones. Since the D i o r y c t r i a l a r v a eats a large proportion of the cone by the end of i t s development, the small cone size may l i m i t the development of the larva because of i t s inadequate supply of food, and because i t dries out too r a p i d l y for the needs of the insect. The trees with a longer duration of vegetative bud f l u s h i n g had a smaller amount of damage than those with shorter duration of f l u s h i n g . This r e s u l t may indicate that the trees which s t a r t t h e i r l i f e function slowly i n the spring have i n s u f f i c i e n t food supply at the beginning of l a r v a l development and the larvae on them ei t h e r die or have reduced vigour. D i o r y c t r i a a b i e t e l l a i s not a s p e c i f i c Douglas f i r cone and seed insect. Because of th i s they do not depend on the cone production of Douglas f i r , nor on the s p e c i f i c c h a r a c t e r i s t i c s of the tree. This may explain why cone length and duration of vegetative bud flushing, factors which are rel a t e d to the food supply of the insect, were correlated with the amount of damage only. From the r e s u l t s of analyses, e i t h e r f o r percentage of cones or percentage of seeds damaged by D i o r y c t r i a i t can be seen that the damage was spread uniformly within the trees. The damage of D i o r y c t r i a reduces the volume of an attacked cone as well as the number of seeds i n a cone. As much as 50 per cent of the cone volume can be eaten by the end of l a r v a l development. The damage i s usually concentrated on the basal 55 per cent of the cones, where the scales are thicker (Figures 1-5). Generally only one larva matures i n a cone. If more than one larv a i s present i n a cone, i t i s thin and i t s development i s slow. The number of cones damaged by Contarinia washingtonensis was studied i n the summer of 1962. Approximately 35 per cent of the cones were infested i n that year. They a c t u a l l y do not reduce the cone volume but may cause cone scale necrosis (Johnson, 1963 b). A highly s i g n i f i c a n t tree to tree v a r i a t i o n was found i n the damage of t h i s species. The average cone length and width, and the colour of female flowers had a s i g n i f i c a n t e f f e c t on the tree to tree v a r i a t i o n . The trees with longer and wider cones had more i n f e s t a t i o n . This r e l a t i o n s h i p between the host and the insect may be associated with the space and food requirements of the insect. A higher percentage of cones was infested on those trees which had red or reddish conelets 150 than on the trees with green conelets. This r e l a t i o n s h i p may be explained by two d i f f e r e n t theories. The actual colour may a f f e c t the o v i p o s i t i n g female, or the d i f f e r e n t colour may be correlated with d i f f e r e n t chemi-cals having a d i f f e r e n t attractiveness to the insect. The number of cones damaged by Contarinia washingtonensis did not vary s i g n i f i c a n t l y within trees. This observation indicates good d i s -persion and/or searching power of the insects during o v i p o s i t i o n . The number of seeds obtained by extraction per cone (y) i s d i r e c t l y correlated with t o t a l number of seeds per cone exposed on the cut surface (XI). Also the percentage of seeds damaged by C.-, oregonensis per .cone on the cut surface (X2) i s correlated with the t o t a l number of damaged seeds per cone i n the following r e l a t i o n s h i p : Y = 0.249 x 1 - 1.080 x 2 + 30.560 This equation based on 110 cones, had a standard error of estimate of 7.45 seeds and a multiple c o r r e l a t i o n c o e f f i c i e n t (R) of 0.815 which i s highly s i g n i f i c a n t . The s i g n i f i c a n t reduction by Contarinia damage of the number of seeds obtained can be i l l u s t r a t e d as follows. Using the equation for a cone which has 20 seeds on the cut surface, i f there i s no damage, 35 seeds can be obtained. I f the damage i s 10 per cent (2 seeds on the cut surface), only 25 seeds per cone can be obtained. This reductions .of the seeds obtained i s caused by the fusion of the seeds to the scales by the Contarinia oregonensis g a l l s . The tree to tree v a r i a t i o n was highly s i g n i f i c a n t . - This v a r i a t i o n was affected by two major groups of factors of the tree c h a r a c t e r i s t i c s : tree size and the date when cones became pendent. The tree height, d.b.h. and crown width a l l were p o s i t i v e l y correlated with the damage of Contarinia oregonensis. However, i t i s very l i k e l y that the tree height was the only f a c t o r which affected the i n t e n s i t y of attack, and the other two were s i g n i f i c a n t because of the highly s i g n i f i c a n t r e l a t i o n s h i p between them and the tree height. The reason why the tree height i s thought to be the factor which influences the i n t e n s i t y of attack of Contarinia oregonensis i s that s i g n i f i c a n t l y more insects were caught by the traps i n the upper crown l e v e l than i n the mid crown, and s i g n i f i c a n t l y more i n the mid crown than i n the lower crown l e v e l . Also, i n a l l of the analyses s i g n i f i c a n t l y more seeds and s i g n i f i c a n t l y more cones were infested i n the upper crown l e v e l of the trees than i n the other two. From these r e s u l t s i t may be concluded that the egg-laying.Contarinia oregonensis f l y and search for o v i p o s i t i o n media more i n t e n s i v e l y i n the higher lev e l s from the ground than i n the lower ones, and because of t h i s the cones on a higher tree are much more exposed to the attack than on a lower tree. The other important factor which affected the tree-to-tree v a r i a t i o n was the date when cones became pendent. This factor was tested for the 1962 observation only; i n 1961 the date cones became erect was observed, which had no influence on the amount of damage by Contarinia. In 1962, the l a t e r the cones became pendent on a tree the heavier the damage by. Contarinia oregonensis. These r e s u l t s may suggest that i t does not matter when the female reproductive buds of Douglas f i r open because the length of time when the cones are open to receive p o l l e n varies from tree to tree. However, those trees on which the cones close e a r l i e r have less chance to be attacked than the l a t e r ones because probably the l i f e cycle of the major portion of Contarinia population i s synchronous with the l a t e r trees. The within«tree v a r i a t i o n of damage of Contarinia oregonensis was s i g n i f i c a n t e i t h e r from l e v e l to l e v e l or from outside to inside crown. However, the trend of damage was not consistent from outside to inside crown i n the two years nor from tree to tree. In the ho r i z o n t a l l e v e l s the damage increased with height i n the crown. S i g n i f i c a n t l y more cones were damaged i n 1961 (77.6 per cent) than in 1962 (73.2 per cent), but the percentage of seeds damaged was not d i f f e r e n t i n the two years (17.4 and 17.1 per cent, r e s p e c t i v e l y ) , which indicates no increase of the population from 1961 to 1962. A reduction of quantity of extracted seeds also was caused by D i o r y c t r i a a b i e t e l l a because they eat a large number of seeds as they eat t h e i r way through the cone. 1. Preventing of F e r t i l i z a t i o n In both years ei t h e r the number of cones or percentage of seeds damaged by Contarinia oregonensis per tree reduced to a highly s i g n i f i c a n t degree the percentage of f i l l e d seeds per tree. This r e s u l t may be explained i n two d i f f e r e n t ways: 1) the f e r t i l i z e d ovule does not develop into a seed because of the presence of Contarinia oregonensis larvae and g a l l s around i t . It i s very d i f f i c u l t to prove t h i s theory, because a large number of f i l l e d seeds can be found with Contarinia oregonensis damage around them i n 153 every year. For example, 4 per cent of the f i l l e d seeds were galled i n 1962. It i s much more l i k e l y that 2) the eggs l a i d by Contarinia  oregonensis at the .base of the scales prevent the f e r t i l i z a t i o n by cl o s i n g out the po l l e n grains from the ovule. I f t h i s hypothesis holds, those seeds which are f e r t i l i z e d before o v i p o s i t i o n , w i l l develop into a f i l l e d seed and those seeds which were not f e r t i l i z e d before the o v i p o s i t i o n , w i l l never be f e r t i l i z e d . Johnson (1963 b) found a s i m i l a r r e s u l t to t h i s from h i s study of eight trees on the Clemons Tree Farm near Elma, Washington. He observed a s i g n i f i c a n t vegetative c o r r e l a t i o n between the number of f i l l e d seeds per cut surface of the cones and the number of cone midges per cut surface, and concluded that Contarinia oregonensis has a detrimental e f f e c t on the f i l l e d seed content of the cone. The hypothesis which suggests the prevention of seed f e r t i l i z a t i o n should be considered whenever a r t i f i c i a l p o l l i n a t i o n i s to be applied to increase the percentage of f i l l e d seeds. If the theory holds, then the cones to be p o l l i n a t e d must be kept free from Contarinia eggs, otherwise the r e s u l t may not be s a t i s f a c t o r y . 2. Reducing of Quality of Extracted Seeds The q u a l i t y of extracted Douglas f i r seeds may be s i g n i f i c a n t l y reduced by the damage of.Megastigmus spermotrophus. . The tree to tree v a r i a t i o n of the damage by Megastigmus was highly s i g n i f i c a n t l y associated with percentage of seeds, percentage of f i l l e d seeds and number of cones damaged. A high proportion of t h i s v a r i a t i o n was accounted for by the amount of f i l l e d seeds produced by the tree. 154 Thus trees which produced more f i l l e d seeds were more heavily infested than those which produced l e s s . The highly s i g n i f i c a n t e f f e c t of t h i s f actor may lead to the conclusion that the Megastigmus female lays eggs into a f e r t i l i z e d seed only, and i s able to select those trees f o r o v i p o s i t i o n which have cones with a s u f f i c i e n t number of f e r t i l i z e d seeds. This hypothesis seems also to be supported by the fac t that the e f f e c t of the f i l l e d seed production of a tree was much stronger i n 1961 when the f i l l e d seed production was only 3.1 per cent per tree per cone than i n 1962 when much more f i l l e d seed was ava i l a b l e f o r o v i p o s i t i o n (6.5 per cent) per tree per cone. One may wonder by what means the female Megastigmus can detect the f i l l e d seeds. The simplest hypothesis i s that they recognize i t mechanically, as they drive i n t h e i r o v i p o s i t o r to a randomly chosen seed. I f they " f e e l " the seed tissue they lay the egg, and i f they do not, they t r y another seed. - Another p o s s i b i l i t y may be that since they spend a considerable amount of time to select the scale on a cone into which the ov i p o s i t o r w i l l be inserted ( M i l l e r , 1916), they possess an unknown sense f o r sending and rec e i v i n g return signals by which they can detect the presence of vi a b l e seed. The average cone length and width of a tree also affected the amount of damage by Megastigmus. The trees with longer and wider cones were more heavily attacked than those with smaller cones. Probably the length of ov i p o s i t o r i s the l i m i t i n g factor which stops the insect from laying eggs into a bigger cone having thicker scales overlying the ovules, or the f i l l e d seed detection i s more d i f f i c u l t on a bigger cone than on a small one because of depth to the ovules. 155 The amount of pollen buds per tree was negatively correlated with the per cent of seeds damaged by Megastigmus. This e f f e c t was rather secondary since the amount of f i l l e d seed per cone was negatively correlated with the amount of pollen per tree! This means that on a good pollen producer tree the f i l l e d seed production i s reduced, which reduces the Megastigmus damage. Some i n t r a - t r e e v a r i a t i o n was also observed for the damage of Megastigmus, but the trend of the i n t e n s i t y of attack was not consist-ent\rom year to year and from tree to tree. In a l l of the cases the percentage damaged followed the f i l l e d seed d i s t r i b u t i o n i n the crown. The percentage of seeds damaged by Megastigmus was not s i g n i f i c a n t -l y d i f f e r e n t i n 1961 (2.2 per cent) and 1962 (2.0 per cent). However, the percentage of f i l l e d seeds damaged i n 1961 (70.1 per cent) was s i g n i f i c a n t l y more than i n 1962 (33.5 per cent). This r e s u l t indicates that the population l e v e l of Megastigmus did not change from 1961 to 1962. The percentage of f i l l e d seeds damaged was s i g n i f i c a n t l y d i f f e r e n t because the f i l l e d seed production of the trees doubled from 1961 (3.1 per cent) to 1962 (6.5 per cent). 3. Improving Seed Production The f i r s t step i n improving seed production should be to improve the cone production per tree. Kozak-e_t. al. (1963) suggested that propagation of the best s i x per cent of the trees might increase cone production four times. They also indicated that the most promising method to increase cone production i s chemical f e r t i l i z a t i o n . Steinbrenner, D u f f i e l d and Campbell (I960), Stoate, Mahood and Crossin (1961), and 156 E b e l l (1962) reported excellent r e s u l t s from chemical f e r t i l i z a t i o n during the period of f l u s h i n g of vegetative buds. Steinbrenner et a l . showed an increase from 1,880 cones and 1.2 pounds of sound seeds per acre to 10,460 cones and 10.3 pounds of sound seeds per acre by chemi-ca l f e r t i l i z a t i o n . Stoate. _et _al. reported an increase of cone product-ion from 7 to 35 bushels per acre. Y i e l d of seed per cone might be increased 20 times with c o n t r o l l e d p o l l i n a t i o n (Allen and S z i k l a i , 1962) and might be doubled i f mass p o l l i n a t i o n were f e a s i b l e . The present study proves that the y i e l d of sound seeds per cone might at least be doubled by the control of Megastigmus only. The r e s u l t s also suggest that by c o n t r o l l i n g Contarinia oregonensis the percentage of f i l l e d seeds per cone could be increased from the e x i s t i n g 4.8 to a new l e v e l of 31 .per cent i n the average of the two years studied. The reduction by D i o r y c t r i a was 20 per cent of f i l l e d seeds i n 1961. It should be noted that a l l of these figures presented above are subject to a c e r t a i n range of error, because of the v a r i a t i o n of observations around the regression l i n e . However, these r e s u l t s suggest the need for b i o l o g i c a l or chemical control of insects to increase seed 'production. Kberber (1960) suggested a combination of i n s e c t i c i d e and c u l t u r a l or b i o l o g i c a l control methods against the cone and seed insects. The r e s u l t s of these studies open up new opportunities to obtain an adequate supply of seed by s p e c i a l d i r e c t i o n of seed c o l l e c t i o n e f f o r t rather than by attempting to combat the insects d i r e c t l y by i n s e c t i c i d e s . Various experiments have shown p o s s i b i l i t i e s i n the use of chemical 157 control for seed and cone ins e c t s . The method i s , however, fraught with d i f f i c u l t i e s and l i m i t a t i o n s . The d i f f i c u l t y with chemical control i s that the insects are protected during the l a r v a l stage i n the cones as well as during the pupal stage i n the s o i l or seed. The only p o s s i b i l i t y to reach them i s the adult stage, when they emerge from the s o i l or seed or search for o v i p o s i t i o n . Another problem here i s that the spraying has to be repeated a number of times because every speciesremerges and attacks at d i f f e r e n t times. Many of the spraying t r i a l s have f a i l e d i n the past. Rudinsky (1955) used DDT by ground sprayer, and found that the protection against Contarinia increased with the increased number of applications from May u n t i l July. Johnson and Winjum (1960) reported that a test of DDT and Guthion did not give s a t i s f a c t o r y r e s u l t s when applied from a h e l i c o p t e r . Out of ten i n s e c t i c i d e s tested on i n d i v i d u a l cone-bearing branches Guthion and Sevin gave best control of,Contarinia oregonensis (Johnson and Winjum, 1960). Johnson (1962 d) determined the concentration of Guthion necessary to give desired c o n t r o l . In 1962 Johnson (1963 e) used two Guthion formulations with a number of other chemicals and proved that both Guthion formulations were better than Sevin wettable powder. Koerber (1963 b) applied three i n s e c t i c i d e s to duff samples infested by Contarinia oregonensis. He found that Lindane was the most e f f e c t i v e , D i e l d r i n also caused a s i g n i f i c a n t reduction i n the number of emerged insects, but Sevin was no more e f f e c t i v e than d i e s e l o i l alone. Koerber's method, spraying the duff instead of cones, would avoid the timing problem 158 and the spray would not destroy the developing cones. However, the effectiveness of the spray applied to the duff may be reduced by the invasion of insects from unsprayed areas. Since the f l i g h t range of Contarinia i s unknown the operational use of duff spraying cannot be recommended. A dangerous point of spraying against cone and seed insects besides k i l l i n g parasites i s that some of the chemicals or solvents for the chemicals k i l l Douglas f i r cones. Johnson (1963 e) indicated that Guthion i n X-77 caused a c e r t a i n amount of abortion. Koerber (1963 a) stated that the d i e s e l o i l , which i s a commonly used solvent of i n s e c t i c i d e s , k i l l s cones when applied at rates i n the range of 5 to 10 gallons per acre. Since the chemical control of cone and seed insects does not seem to be solved yet, there are some other p o s s i b i l i t i e s by which the seed y i e l d can be increased. The present study has important implications for the planning of cone c o l l e c t i o n e f f o r t i n r e f o r e s t a t i o n work. If the cone c o l l e c t i o n i s planned to be - c a r r i e d out i n stands, the i n d i v i d u a l trees selected f o r the c o l l e c t i o n can be tested by a sequential sampling method to determine whether or not the seed c o l l e c t -ion from them w i l l be s a t i s f a c t o r y . In the chapter of Experimental Results, eight sequential sampling methods are presented, four to determine the l e v e l of i n f e s t a t i o n on i n d i v i d u a l trees and four to determine the i n f e s t a t i o n l e v e l for a stand. However, the plans for stand sampling are not p r a c t i c a l , because of the very large sample sizes required to f i n d out the l e v e l of i n f e s t a t i o n . The large.sample•sizes required are caused by the high tree-to-tree v a r i a t i o n of damage. 159 Those four plans which were worked out for i n d i v i d u a l trees can be used very e f f e c t i v e l y with a small number of samples. Two of these plans may become e s s e n t i a l f o r p r a c t i c a l f oresters to define the f i l l e d seed production and the t o t a l number of seeds damaged by various cone and seed insects. The other two plans may be useful i n research to determine the population l e v e l of Megastigmus spermotrophus and Contarinia oregonensis. For seed orchards, i t would be highly desirable to select trees which are r e s i s t a n t to cone and seed insects. None of the 154 trees studied was e n t i r e l y free from attack e i t h e r i n 1961 or 1962. A c e r t a i n number of Megastigmus and D i o r y c t r i a free trees were found, but a l l of the trees were infested by Contarinia oregonensis. The damaged and f i l l e d seed production of the best f i v e trees are summarized i n Table 57. The re s u l t s found i n t h i s table suggest that tree number 21 was the freest from attack, with only 4.8 per cent of i t s seeds damaged. Although the f i l l e d seed production of th i s tree was not the best (5.0 per cent), i t was less than the average of f i v e best trees (6.0 per cent), but more than the average of a l l trees (4.3 per cent) sampled. The best tree i n seed production was number 31 (10.4 per cent), but seeds of th i s tree were very badly damaged by Megastigmus (8.4 per cent). For seed orchards the reproduction of trees number 18 and 21 .is' suggested, because of the low damage and f a i r l y high f i l l e d seed production. The f i l l e d seed production could be increased by mass p o l l i n a t i o n ; however, control measures should be undertaken when using a r t i f i c i a l p o l l i n a t i o n because with the increased f i l l e d seed production an increased. Megastigmus damage i s associated. Although the percentage of f i l l e d seeds per tree was not s i g n i f i c a n t l y correlated with the number of cones produced, the f i v e best trees selected were better i n Table 57. Damages, f i l l e d observations). seed production on the best f i v e trees (averages of 1961 and 1962 Tree Total Damage by F i l l e d Average No. of years No. damage Contarinia Megastigmus D i o r y c t r i a seeds per cone number of with Percentages! cones per year3 cones^ 18 8.8 7.2 2.5 0.0 6.4 230 5 21 4.8 1.7 2.6 0.5 5.0 3 2 31 8.6 5.0 8.4 0.0 10.4 430 4 115 9.1 5.9 0.7 1.2 3.6 50 4 140 7.1 6.5 4.5 0.0 4.6 79 3 Averages 7.7 5.3 3.7 0.3 6.0 158 3.6 A l l over 2 averages 24.7 17.3 2.1 1.4 4.3 1094 2.9 4 1. Because some seeds were damaged by more than one species of insect the percentages of damage by species do not add to the t o t a l damage. 2. Averages f o r 48 trees sampled i n both years. 3. Averages for the period of 1957-1962. 4. Averages for a l l 154 trees. 161 cone production than the average (Table 57). This seems to be promising for tree breeding programs; however, the best tree, number 21 had only a very poor cone crop on i t every year. At least three major problems should be recognized besides the growth c h a r a c t e r i s t i c s and mechanical properties of the wood of the tree, 1) low insect damage, 2) good sound seed and 3) good cone production; . Obviously i t w i l l be very d i f f i c u l t to combine a high l e v e l of a l l desirable factors i n future programs of tree improvement. 4. Need f o r Further Studies The great range i n apparent resistance•of trees to attack by cone and seed insects should be recognized i n further studies and attempts made to determine b i o l o g i c a l l y sound reasons for the observed d i f f e r e n c e s . The causes of tree-to-tree v a r i a t i o n could not be f u l l y determined i n t h i s study, and much more work i s needed i n t h i s f i e l d . Dr. J.D. Beaton found that analysing the f o l i a g e of the 10 best and 10 worst cone producing trees, the trees were s i g n i f i c a n t l y d i f f e r e n t i n phosphorus and nitrogen contents. Also s i g n i f i c a n t crown l e v e l v a r i a t i o n was found i n phosphorus, calcium, potassium and sulphur content of the f o l i a g e . Since the cone scales are modified needles, i t i s very l i k e l y that some -of the tree-to-tree v a r i a t i o n of damage may be re l a t e d to some chemical content v a r i a t i o n s of the cones on the trees. Further study i s needed to define t h i s r e l a t i o n s h i p . The hardness of cone scales may a f f e c t laying of Megastigmus eggs. It would be desirable to f i n d the r e l a t i o n s h i p between the tree-to-tree v a r i a t i o n i n hardness of cone scales and the damage by Megastigmus. 162 Also, the means by which the Megastigmus female finds the f i l l e d seed should be investigated. Research i s needed to determine whether or not the Contarinia  oregonensis eggs prevent the f e r t i l i z a t i o n of the ovule i n the cone or whether the development of f i l l e d seed i s retarded by the larvae. An economical, probably sequential, sampling method would be useful i n determining the over-wintering cone and seed insect population i n the s o i l as a guide to the need f o r insect c o n t r o l . This might be combined with a sampling scheme to forecast cone crops i n late f a l l or e a r l y spring. 163 CONCLUSIONS Findings of the study are summarized as follows: 1. Highly s i g n i f i c a n t v a r i a t i o n occurred between trees f o r the damage of each insect species studied. 2. This v a r i a t i o n was affected mostly by tree size and date cones became pendent i n Contarinia oregonensis damage. 3. The tree v a r i a t i o n i n damage of Contarinia washingtonensis was s i g n i f i c a n t l y affected by the average cone size of the trees and by the colour of female flowers of the trees. 4. The tree-to-tree v a r i a t i o n i n damage of Megastigmus spermotrophus was p r i m a r i l y determined by the percentage of f i l l e d seed production of the trees and secondarily by the average cone s i z e . 5. The average cone size and duration of vegetative bud f l u s h i n g of the trees accounted f o r some of the v a r i a t i o n between trees i n the damage by D i o r y c t r i a a b i e t e l l a . 6. The i n t r a t r e e v a r i a t i o n of damage was s i g n i f i c a n t only for Contarinia  oregonensis. The damage of th i s species increased s i g n i f i c a n t l y from the bottom to the top of the l i v e crown. 7. The sequential sampling method was e f f e c t i v e i n se l e c t i n g trees f o r cone and seed c o l l e c t i o n . 8. Because of the high tree-to-tree v a r i a t i o n , the sequential sampling method was not useful to determine the i n f e s t a t i o n l e v e l i n stands. 9. Although no tree was wholly r e s i s t a n t to a l l three insects, the best with only 4.8 per cent of i t s seeds damaged was so much better than average that i t may be of considerable importance. 10. The great range i n apparent resistance of trees to attack by cone and seed insects should be recognized i n further studies and attempts be made to determine b i o l o g i c a l l y sound reasons for the observed di f f e r e n c e s . APPENDIX A Fig Mop showing location of Douglas fir trees Nos I — 158, University Research Forest Map showing l o c a t i o n of 154 Douglas f i r trees studied. APPENDIX B can IBP flfT* or poom ra Crown No. of f u l l M*d« Ho. of enBtT iaedi Ho. of scalai Cone No««a |' '" leval aed by N.-.I total Total isad by '•"**" 1 • .in.. otter Bar Con etbmt B.'IT 1 0". other 1 ' ! J 1 ! I 1 s 1 | j - j I 3 1 1 1 7 Y s -• I I } A J 4 t ; 1 r • i* -I —T 4 Form on which the information was c o l l e c t e d . APPENDIX C Five colour classes of female flowers on Douglas f i r . (The s l i d e of t h i s photograph i s a v a i l a b l e at The University of B r i t i s h Columbia, Faculty of Forestry) 168 APPENDIX D Counting of Megastigmus spermotrophus from one of the rearing boxes. Rearing box and s o i l temperature recorder. APPENDIX F Cones covered with cellophane bags. 171 APPENDIX G Trapping frames on tree No. 63 APPENDIX H AVE • t . t l »6666€ -CC V»H - 2 . ?7 6t--C C SU - 1 .*> 2 / H 6 I F_-0 C P - 1.1.434336-01 PI - 1.1bbSS7F-CS K - 9.923C61F-C1 O B S f v c p AND t X P P C T f n F R f O U f * C I F b I 0 8 S NOR b I N 001 N B : . : c 2 V . 7 2 1 " 3 . 5 0 9 <* 7 .651. 7 0 . 2 5 1 i.ee 36.CCC 3 9 . . 9 8 5 0 . 2 3 8 5 » . 6 » 3 3 7 . 3 7 3 2 . : : 17.CCC 3 3 . 8 1 3 39.TCb 3 1 . 3 2 0 19.9*9 3.CC 1 8 . 6 . 6 II ' 2 I 1 . 9 7 , 1C .060 ».CC 0 . 6 2 3 2 . 3 8 4 S . « 3 2 5 . 7 2 1 j .DCS 1 . 5 T b . 3 1 * .787 3.C55 6.CC i P c e o . 2 2 3 . 0 2 6 .153 1 . 6 3 7 T.BC 1 .CCS .C2 1 .CC 1 . 0 2 . . 8 7 6 e.oc .CCT c . cco . 0 0 3 . » 6 7 TOTALS 1 3 C . C 6 . I k 9 . 9 9 9 c» MM B I N POI N 8 1 .cc 2 . C C 5 . 9 5 . 7 3 C E . C I S . 6 b B » 2 6 E - C 2 8 . 3 6 3 0 9 1 6 - 0 0 1 .ft 1 2 8 2 1 F * 5 1 7 . C 3 3 2 C 8 E - C C b . 5 7 o 6 9 H E - C 3 1 . 3 9 7 H 1 0 F * 3 1 S . C 6 9 3 3 9 F - C C 6 . 5 S 3 6 2 9 F - 3 3 9 . C 2 1 3 9 R F - 3 9 1 .C99.07E-C2 * .*52S76E-S 1 3 . C C « . C C 5 . C C » . C C 9 2 5 C f - C C B . 5 2 8 3 9 C E - C I 1 . U S 9 2 T 9 E - C C 1 . 7 U 8 1 8 7 F - 0 1 1 . 8 3 2 2 7 9 E * C I 2 . 2 U 8 9 7 6 E » C 1 3 . 2 5 6 1 6 5 E - C 1 9 . C 2 9 C C 1 E - C C 6 . 2 I 9 9 H 9 E - C C . . 1 92. i i ' K - C 2 1 ,87863C?--3C 1 .32.C32E-C3 o . C C 7 . C C 6 . C C 1 . . 1 3 T b 9 E » C l k . 5 1 7 8 1 3 E . C 1 I . 2 9 6 b » 9 E - C 3 1 . U 5 0 8 9 5 F + C 2 7 . 8 C 5 8 5 3 E * C 2 2 . 6 7 2 1 9 8 E - C 5 2 . 2 7 3 6 T I E . C 1 3 . 8 5 9 U U B E . C 1 3 . 5 3 2 U 5 7 E-03 8.322212E-02 I.7357996-C2 ..A75755E-C1 T0T11 .S 1 . 2 8 0 3 2 T t * C 2 9 . 9 8 9 9 7 3 E » C 2 9 . 9 C C 9 2 V E - . C 1 2.033392E-33 D E M I S E S O F F O E E O P " 6 0 7 4 N A I V S 1 S C O . P L F T E , The output forms of the program f i t t i n g frequency d i s t r i b u t i o n s . APPENDIX I The percentage of seeds damaged by Contarinia oregonensis on four sides of the crown of five trees Legend^ | — 5 = trees N W Azimuth LITERATURE CITED A l l e n , G. S. 1941. A basis for forecasting seed crops of some coniferous trees. Jour. For. 39: 1014-6. A l l e n , G. S. and 0. S z i k l a i , 1962. P o l l i n a t i o n of Douglas f i r with water suspensions of p o l l e n . For. Sc. 8(1): 64-5. B l i s s , C. I. 1953. F i t t i n g the negative binomial d i s t r i b u t i o n to b i o l o g i c a l data. Biometrics. 9(2): 176-200. B l i s s , C. I. and A. R. G. Owen, 1958. Negative binomial d i s t r i b u t i o n with a common k. Biometrika. 45(1-2): 37-58. Chapman, J. A. 1962. F i e l d studies, on attack, height and log s e l e c t i o n by the ambrosia beetle, Trypodendron lineatum (Oliv.) (Coleoptera: s c o l y t i d a e ) . Can. Ent. 94(1): 74-92. de Sandt, C. 1930. Sur t r o i s microlepidopteres du genre D i o r y c t r i a . Rev. Zool. A g r i c . Appl. 29: 57-63. Duncan, D. B. 1951. A s i g n i f i c a n t test for differences between ranked treatments i n an analysis of variance. V i r g i n i a J . S c i . 2: 171-189. Duncan, D. B. 1955. M u l t i p l e range and multiple F t e s t s . Biometrics, 11: 1-42. Ebel, B. H. and Merkel, E. P. 1957. Insects destructive to the flowers, cones, and seeds of slash and longleaf pines. U.S. Forest Serv. Southeast. Forest Expt. Sta. Quarterly Report. 2 p. E b e l l , L. F. 1962. Growth and cone production responses of Douglas f i r to chemical f e r t i l i z a t i o n . Dept. of For. Forest Research Branch, B.C.: 62-68 mimeo., 43 p. F i n n i s , J . M., 1953. A note on the bud count method of forecasting cone crops of Douglas f i r . For. Chron. 29(2): 122-27. Foote, R. H.., 1956. G a l l midges associated with cones of western forest trees (Diptera: Itonididae). Jour. Wash. Acad. S c i . 46: 54 p. Garman, E. H. 1951. Seed production by conifers i n the Coastal Region of B r i t i s h Columbia., B.C. Forest Service, Res. D i v i s i o n , T. 35, 47 p. Graham, K., 1963. Concepts of Forest Entomology, Reinhold Publishing Corp., New York. 388 p. Graham, K., 1941. Investigations of the r o l e of insects i n r e l a t i o n to seed production i n Douglas f i r . Annual forest insect investigations and d e t a i l e d studies i n Coastal B r i t i s h Columbia. Can. Dept. Agr. Forest Biology Div. 21 p. 175 G r i f f i t h , B. G. 1963. Phenology, growth, and flower and cone production of open-grown Douglas f i r trees. For. Chron. Dec. (In press). Hedlin, A. F., 1958. Insects causing seed losses i n Douglas f i r on Vancouver Island i n 1957. Ent. Soc. B.C. Proc. 55: 37-39. Hedlin, A. F., 1959. Studies on cone and seed insects i n B r i t i s h Columbia. Can. Dept. Agr. S c i . Serv., Forest B i o l . Div. 20 p. Hedlin, A. F. 1961. The l i f e h i s t o r y and habits of a midge, Contarinia  oregonensis, Foote (Diptera: Cecidomyiidae) i n Douglas f i r cones. Can. Ent. 93: 952-67. Hedlin, A. F., 1962. Two systemic i n s e c t i c i d e s , phosphamidon and Systox, used against the Douglas f i r cone midge, Contarinia oregonensis, Foote. Can. Dept. of Forestry, Bi-monthly Progr. Rept. 18(1): 3-4. Hussey, N. W., 1954. Megastigmus f l i e s attacking conifer seed. Forestry Commission, London. Leaflet 8, 10 p. Hussey, N. ¥., 1955. The l i f e h i s t o r i e s of Megastigmus spermotrophus (Hymenoptera: Chalcidoidea) and i t s p r i n c i p a l parasite, with descriptions of the developmental stages. Trans. Roy. Ent. S o c , London. 106(2): 133-51. Johnson, N. E. 1962 a. D i s t r i b u t i o n of Douglas f i r cone midges i n forest l i t t e r beneath young, open-grown Douglas f i r . Can. Ent. 94(9): 915-21. Johnson, N. E. 1962 b. Tests of Guthion for the con t r o l of the Douglas f i r cone midge. Jour. Econ. Ent. 55(5): 613-16. Johnson, N. E. 1962 c. A possible sampling method for determining when to spray for control of the Douglas-fir cone midge. Weyerhaeuser Company Forestry Research Note 49. 10 p. Johnson, N. E. 1963 a. Contarinia washingtonensis (Diptera: Cecidomyiidae), new species i n f e s t i n g the cones of Douglas f i r . Ent. Soc. Amer. Ann. 56(1): 94-103. Johnson, N. E. 1963 b. Cone-scale necrosis and seed damage associated with attacks by Douglas f i r cone midge. For. S c i . 9(1): 44-51. Johnson, N. E. 1963 c. Compatability of p o l l e n with i n s e c t i c i d e s used i n Douglas f i r cones. Weyerhaeuser Company Forestry Research Note 52, 10 p. 176 Johnson, N.E., 1963 d. Time of attack of the Douglas-fir cone midge i n r e l a t i o n s h i p to cone development. Jour. For 61(5): 350-5. Johnson, N.E., 1963 e. Insecticides tested for control of the Douglas-f i r cone midge. Jour.Econ. Ent. 56(2): 236-7. Johnson, N.E. and H.J. Heikken, 1958. Damage to the seed of Douglas-f i r by the Douglas-fir cone midge. For. S c i . 4(4): 274-82. Johnson, N.E. and J.K. Winjum, 1960. Douglas-fir cone and seed insect b i o l o g i c a l and control studies: progress i n 1958, 1959. Weyerhaeuser Company Forestry Research Note 22. 23 p. Keen, F.P., 1958. Cone and seed insects of western forest trees. U.S. Dept. Agr. Tech. Bui. 1169. 168 p. Koerber, T.W. 1960 a. Douglas-fir cone and seed insect research, Progress Report, 1957-1958. U.S. Dept. Agr. 4500. 29 p. Koerber, T.W. 1960 b. Insects destructive to the Douglas-fir seed crop i n C a l i f o r n i a a problem a n a l y s i s . U.S. Dept. Agr. Technical Paper 45. 36 p. Koerber, T.W. 1963 a. The t o x i c i t y of d i e s e l o i l to Douglas-fir cones. U.S. Eorest Service Research Note PSW-6. 4 p. Koerber, T.W. 1963 b. Insecticide tests on the Douglas-fir cone midge, Contarinia oregonensis Foote. Can.. Ent. 95(6): 640-646. Kozak, A. and D.D. Munro, 1963. An I.B.M. 1620 computer program to f i t frequency d i s t r i b u t i o n s . For. Chron. 39(3): 377-8. Kozak, A., 0. S z i k l a i , B.G. G r i f f i t h and J.H.G. Smith, 1963. V a r i a t i o n i n cone and seed y i e l d from young, open-grown Douglas f i r s on the U.B.C. Research Forest. Univ. of B.C., Fac. of For., Research Paper 57, 8 p. Lyons, L.A. 1957. Insects a f f e c t i n g seed production i n red pine 11. , D i o r y c t r i a d i s c l u s a Heinrich, D. a b i e t e l l a (D and S), and D. Cambiieola (Dyar) (Lepidoptera: Phycitidae). Can. Ent. 89(2): 70-9. M i l l e r , J.M. 1914. Insect damage to the cones and seeds of P a c i f i c Coast c o n i f e r s . U.S. Dept. Agr. Bui. 95. 7 p. M i l l e r , J.M. 1916. O v i p o s i t i o n of Megastigmus spermotrophus i n the seed of Douglas-fir. Jour. Agr. Res. 6(2): 65-8. Munroe, E. 1959. Canadian species of D i o r y c t r i a Z e l l e r (Lepidoptera: Pyralidae) Can.. Ent. 91(2) 67-72. . Pettinger, L.F. and N.E. Johnson, 1962. The influence of overwintering s i t e on the m o r t a l i t y of the Douglas-fir cone midge (Contarinia oregonensis Foote). Weyerhaeuser Company Forestry Research Note 45. 5 p. 177 Robinson, B.A. 1963. V a r i a t i o n of seed c h a r a c t e r i s t i c s of Douglas f i r . Univ. of B.C., Fac. of For. Thesis for the B.S.F. degree.. 49 p. Ross, D.A. 1958. A l i s t of cone and seed insects of i n t e r i o r B r i t i s h Columbia. Ent. Soc. B.C. Proc. 55: 30-31. Rudinsky, J.A. 1955. Douglas f i r cone and seed insects (Progress Report). Weyerhaeuser Timber Company Forestry Research Note. 6 p. Steinbrenner, E.C., D u f f i e l d , J.W. and R.K. Campbell, 1960. Increased cone production of young Douglas f i r following nitrogen and phosphorus f e r t i l i z a t i o n . Jour. For. 58: 105-110. Stevens, R.E., 1957. Insect-caused damage to the 1956 Douglas f i r cone crop i n C a l i f o r n i a . U.S. Forest Serv. C a l i f . Forest and Range Expt. Sta., Forest Res. Note 120. 2 p. Stoate, T.N., I. Mahood, and E.C. Crossin, 1961. Cone production i n Douglas f i r (Pseudotsuga men z i e s i i ) . -Empire Forestry Review 40: 105-110. Tukey, J.W. 1949. Dyadicanova, an analysis of variance for vectors. Human Biology. 21: 65-110. U l l y e t , G.C. 1950. Competition for food and a l l i e d phenomena i n sheep-blowfly populations. Philosophical transactions of the Royal Society of London. 234: 77-174. Wald, A. 1947. Sequential analysis. John Wiley and Sons, Inc. New York. Waters, W.E. 1955. Sequential sampling i n forest insect surveys. For. S c i . 1: 68-79. . Winjum, J.K. and N.E. Johnson, 1960. A modified knife cone cutter for Douglas-fir seed studies. Jour. For. 58: 487-488. LITERATURE CONSULTED Anscombe, F.S. 1949. S t a t i s t i c a l analysis of insect counts based on the negative binomial d i s t r i b u t i o n . Biometrics, 5: 165-173. Anscombe, F.S. 1950. Sampling theory of the negative binomial and logarithmic series d i s t r i b u t i o n s . Biometrika 37: 358-382. Be a l l , G. 1940. The f i t and s i g n i f i c a n c e of contagious d i s t r i b u t i o n when applied to observations on l a r v a l insects. Ecology, 21: 460-474. B e a l l , G. and R.R. Reseia, 1953. A generalization of Neyman's contag-ious d i s t r i b u t i o n s . Biometrics, 9: 354-386. B l i s s , C.I., 1958. Analysis of insect counts as negative binomial d i s t r i b u t i o n s . 10th Int. Congress of Ent., Vol.2, 1015-1031. - Douglas, J.B. 1955. F i t t i n g the Neyman type A (two parameters) contagious d i s t r i b u t i o n . Biometrics. 11: 149-173. Foster, R.E. and A.L.S. Johnson, 1963. Assessments of pattern, frequency d i s t r i b u t i o n , and sampling of forest disease i n Douglas f i r plant-ations. Canada Department of Forestry. Studies i n Forest Pathology 25. Ives, W.G.H. 1954. Sequential sampling of insect populations. Forestry Chronicle. 30(3): 287-291. Knight, F.B. 1960. Sequential sampling of Engelmann spruce beetle i n f e s t a t i o n i n standing trees. Research Notes, No. 47. Forest Service, U.S. Dept. of A g r i c u l t u r e . Knight, F.B. 1960. Sequential sampling of black h i l l s beetle population. Research Notes, No. 48. Forest Service, U.S. Dept. of A g r i c u l t u r e . Koerber, T.W. 1963. Insecticide test on the Douglas-fir cone midge, Contarinia oregonensis Foote. Can. Ent. 95: 640-646. McGuire, J.U., T.A. Brindley and T.A. Bancroft. 1957. The d i s t r i b u t i o n of European corn borer larvae Pyrausta n u b i l a l i s (Hbn.) i n f i e l d corn. Biometrics. 13: 65-78. Morris, R.F. 1954. Sequential sampling technique for spruce budworm egg surveys. Canada J. Zool. 32: 302-313. Morris, R.F. 1955. The development of sampling techniques f o r forest insect d e f o l i a t o r s , with p a r t i c u l a r reference to the spruce budworm. Canada J. Zool. 33: 225-294. Oakland, G.B. 1951. Sequential analysis. Canada Dept. of A g r i c u l t u r e . Redmond, D.R. 1959. M o r t a l i t y of r o o t l e t s i n balsam f i r d e f o l i a t e d by the spruce budworm. Forest Science, 5(1): 65-69. Shumway, R. and J. Gurland, 1960. F i t t i n g the Poisson binomial d i s t r i b ution. Biometrics, 16: 522-533. Smith, J.H.G. and J.W. Ker. 1957. Some d i s t r i b u t i o n s encountered i n sampling forest stands. Forest Science. 3(2): 137-144. Smith, J.H.G. and J.W. Ker. 1958. Sequential sampling i n reproduction surveys. Journal of Forestry 56(2): 107-109. Smith, J.H.G., J.W. Ker, and J. Csizmazia, 1961. -Economic of reforest-ation of"Douglas f i r , western hemlock and western red cedar i n the Vancouver Forest D i s t r i c t . , Univ. of B.C., Fac. of For., For. B u l l 3, 144 p. Sprott, D.A. 1958. The method of maximum l i k e l i h o o d applied to the Poisson binomial d i s t r i b u t i o n . Biometrics. 14:97-106. Stark, R.W. 1952. Sequential sampling of the lodgepole needle miner. Forestry Chronicle. 28(1): 57-60. Walter, E.C. 1960. Sequential sampling i n spruce budworm control projects. Forest Science. 6(1): 51-58. Winjum, J.K. and N.E. Johnson, 1962. Estimating cone crops on young Douglas f i r . Weyerhaeuser Company, Forestry Research Note 46, 12 p. Figure I- Distribution of cone and seed insect damage in Douglas-fir cones-in •O a> a> co T3 a> o E o Q xi E Z o> o a> > < I Or 0-8-0-6 -0-4 0-2 Legend: Dioryctria damage Contarinia damage in all cones Contarinia damage in Dioryctria-free cones Contarinia damage in Megastigmus-free cones Megastigmus damage in all cones — * — Megastigmus damage in Dioryctria-free cones — Megastigmus damage in Contarinia-free cones 0 0 0 Base 30 40 50 60 Cone Length in Percentage 70 80 90 100 Top Figure 2 Distribution of cone and seed insect damage in Douglas-fir cones (middle crown)-Figure 3 - Distribution of cone and seed insect damage in Douglas-fir cones (upper crown)-Figure 4- Distribution of cone and seed insect damage in Douglas-fir cones (outside crown)-Figure 5- Distribution of cone and seed insect damage in Douglas-fir cones (inside crown)-Figure 6- Time of emergence of Contarinia oregonensis and Megastigmus spermotrophus in 1962 I5r-Contarinia oregonensis Legend Average number of Contarinia Total hours of sunshine 28 29 30 I 2 3 4 5 6 7 8 9 10 II 12 April May Time 15 a> jQ E c CP CT o >< w_ O > c -o -Legend Average number of Megastigmus Total hours of sunshine Megastigmus spermotrophus / \ / \ / \ i\ i \ i \ i \ l  1 Figure 7- Time of attack of Contarinia oregonensis and Megastigmus spermotrophus-i i i i i i i i i i i i i i i i i I I i i i i i i i I 25 27 29 I 3 5 7 9 11 13 15 17 19 21 April May I I l I I I I l I l I l I l i l I i I i I I I i I I I 31 I 3 5 7 9 II 13 15 17 19 21 23 25 May June Date of cones bagged or opened gure 8- Sequential graph for sampling cones damaged by Contarinia oregonensis-Figure 9- 0- C- and A- S N- curves for sampling cones damaged by Contarinia oregonensis-Legend1 P= probability of correct coll N= average sample number 0-C- Curves •00 0-50-O-OOf 0 Light vs-Medium \ Medium vs-Heavy I 4 6 Average number of seeds damaged a s A- S- N- Curves N 100-50 0 J Medium vs-. Heavy ^ ^ \ ^ Light vs- Medium -r 2 6 8 Average number of seeds damaged Figure 10- Sequential graph for sampling cones damaged by Megastigmus spermotrophus-Number of cones examined — y Figure II OC- and ASN- curves for sampling cones damaged by Megastigmus spermotrophus-Legend- P= probability of correct call IM= average sample number O- C Curves •00-0-50-Light vs-Medium Medium vs Heavy 000- "T* 6 2 4 Average number of seeds damaged 8 N ASN- Curves 100-50-i \ t \ Medium vs-y Heavy \ Light vs Medium 0- i — - — - ™ r-2 4 6 Average number of seeds damaged 8 Figure 12- Sequential graph for sampling total number of damaged cones Number of cones examined — n Figure 13- OC- and A-SN curves for sampling total number of damaged cones Legend1 P= probability of correct call N= average sample number 100 0-50-0 0 0 0 C Curves Light vs-Medium Medium vs-Heavy , —j i 1 0 2 4 6 8 Average number of seeds damaged N 150-100-50 A S- N Curves Light vs- Medium / \ / \ O-Hr 2 3 4 5 Average number of seeds damaged Medium vs Heavy -T 6 Figure 14- Sequential graph for sampling undamaged filled seeds-Number of cones examined — n Figure 15 OCand ASN- curves for sampling undamaged filled seeds Legend: P= probability of correct call N= average sample number 100-0-50-000 O C Curves Poor vs Medium Medium vs-Good i 1 0 2 4 6 Average number of undamaged filled seeds 8 N 150 100-50 A- S- N- Curves Poor vs- „ Medium/ J v - - " Medium vs-Good 1 1 1 1 2 4 6 8 Average number of undamaged fi l led seeds Figure 16- Sequential graph for sampling trees damaged by Megastigmus spermotrophus-Number of trees examined — n Figure 17- 0 0 and A-SN- curves for sampling trees damaged by Megastigmus spermotrophus Legend- P= probability of correct call N= average sample number OC- Curves •00 0-50-Light vs Medium 000-H T ~ i =T = 5 7 9 II Average number of cones damaged Medium vs Heavy 13 N 600-400 200-A- S N Curves Light vs /\ Medium / v, 0-Hr 5 9 -r II Average number of cones damaged Medium vs Heavy n 13 Figure 18- Sequential graph for sampling trees damaged by Dioryctria abietella-Number of trees examined — n Figure 19- OC- and A-SN- curves for sampling trees damaged by Dioryctria abietella-Legend- P= probability of correct call N= average sample number OC Curves IOO-0-50 OOO-fV 2 N 600-400-200-o4V Light vs-Medium Medium vs-Heavy 5 8 11 Average number of cones damaged A S N Curves n 14 Medium vs-Heavy \ ..." t"\ — ._-jr^""^-~. Light vs- Medium i i i 5 8 11 Average number of cones damaged 14 Figure 20- Sequential graph for sampling trees damaged by Contarinia oregonensis-Number of trees examined — n Figure 21- OC- and ASN- curves for sampling trees damaged by Contarinia oregonensis-Legend : P = probability of correct call N= average sample number i-oo-0-50-0-OO-Hr OC- Curves Light vs-Medium ^ Medium vs-Heavy \ N 60-4 0 -o-H, 5 8 11 Average number of cones damaged ASN- Curves 14 / i Light vs-Medium Medium vs-Heavy / / 20-5 8 11 Average number of cones damaged 14 Figure 22- Sequential graph for sampling trees for undamaged cones-Number of trees examined — n Figure 23- OC- and A-SN- curves for sampling undamaged cones-Legend : P= probability of correct call N= average sample number OC-Curves IOO----0-50-o-oo4V Poor vs-Medium Medium vs-Good 6 9 T-12 15 Average number of undamaged cones N 1000-500-A-SN- Curves . ^ - • ^ ^ Poor vs- Medium f\ i i Medium vs-/• 1 Good / i / » \ \ \ \ \ \ \ O-fir 3 - r 6 12 15 Average number of undamaged cones APPENDIX I The percentage of seeds damaged by Contarinia  oregonensis on four sides of the crown of five trees Legend' I — 5 = trees I -N E S W Azimuth 

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.831.1-0105603/manifest

Comment

Related Items